Archive-Date: xxx, 1 Aug 1993 14:04:29 -0400 Date: 1 Aug 93 14:04:29 -0400 Organization: Carnegie Mellon Computer Club Lines: 19 NNTP-Posting-Host: drycas.club.cc.cmu.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <744143136.703668.KLENSIN@INFOODS.UNU.EDU>, KLENSIN@INFOODS.UNU.EDU (John C Klensin) writes: > > It might be worth noting that the AT&T part of the Internic consortium > is running Netfind as a critical, advertised, production part of their > "white pages" strategy. > I find it disconcerting that a Network Information Center would invest the time /trouble/money in a flawed tool like netfind. Meanwhile they no longer allow "average users" to register in the WHOIS database. WHOIS is more reliable than netfind by a long shot. -- John 'FuzzFace\Fast-Eddie' McMahon fuzzface@drycas.club.cc.cmu.edu {use finger to get my PGP public key} "Any networking problem can be solved by the application of 75 miles of Unshielded Twisted Pair (UTP)" - InterOp Spring 1993 X-ST-Status: N Return-path: ================================================================================ Archive-Date: Sun, 1 Aug 1993 20:03:34 PDT Date: Sun, 1 Aug 93 20:03:34 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: netfind and finger To: fuzzface @ drycas.club.cc.cmu.edu cc: info-multinet @ TGV.COM > In article <744143136.703668.KLENSIN@INFOODS.UNU.EDU>, KLENSIN@INFOODS.UNU.EDU (John C Klensin) writes: >> >> It might be worth noting that the AT&T part of the Internic consortium >> is running Netfind as a critical, advertised, production part of their >> "white pages" strategy. >> > I find it disconcerting that a Network Information Center would invest the time > /trouble/money in a flawed tool like netfind. Meanwhile they no longer > allow "average users" to register in the WHOIS database. > WHOIS is more reliable than netfind by a long shot. Not true. Anyone can register with the new NIC because all you need to do is to register a domain name with you as a contact. And anyone can register a domain name. Don't tell them I sent you... Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 1 Aug 93 20:06:11 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet,comp.os.vms Subject: Printing - DCPS and TCPIP - yeah - again! Message-ID: <1993Aug2.120812.66596@qut.edu.au> From: palmer@qut.edu.au ================================================================================ Archive-Date: xxx, 2 Aug 1993 12:08:12 +1000 Date: 2 Aug 93 12:08:12 +1000 Organization: Queensland University of Technology Lines: 25 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET We are yet to install Dec Print Supervisor and up until now I have used LAT to all of our remote printers. Then I use LATSYM on some and DQS$PRTSMB on others because the LTA device behaves *like* a normal serial line. Now because of some things the communications guys have done, LAT won't get to a terminal server where a Printer was previously served by LAT. Yep - go to the multinet stream symbiont :-) OOPS - DCPS wants the location to look like a serial device, for device interrogation etc - formats the headers as postscripts etc - now how do we look for the stream symbiont :-( We have used CREATE_NTY to get some mail out of PMDF to other machines using a Phone_Net Channel. Now if I could create a permanent NTY then I could use any of my 'old favorite' symbionts or the one with DCPS. My understanding is that Create_NTY creates a transitory nty that when deallocated disappears. Did I miss something? Can Create_NTY create a 'permanent' nty like I can create a permanent LTA device with latcp? It would be nice if I could force Create_NTY to give me a device number that I specify not that it makes up but that isn't a major problem. Thanks in advance to all, Greg Palmer. Queensland University of Technology. X-ST-Status: N Return-path: Received: from Levels.UniSA.Edu.Au ([130.220.16.8]) by TGV.COM via INTERNET ; Sun, 1 Aug 93 22:06:48 PDT Received: from Levels.UniSA.Edu.Au by Levels.UniSA.Edu.Au (PMDF V4.2-13 #2428) id <01H19UOQ6KK09I4PYC@Levels.UniSA.Edu.Au>; Mon, 2 Aug 1993 14:36:33 +0930 ================================================================================ Archive-Date: Mon, 02 Aug 1993 14:36:33 +0930 Date: Mon, 02 Aug 1993 14:36:33 +0930 From: Rollo.Ross@Levels.UniSA.Edu.Au Subject: Where to find NPRV? Sender: Rollo Ross To: info-multinet@tgv.com Organization: University of South Australia X-Envelope-to: info-multinet@tgv.com X-VMS-To: IN%"info-multinet@tgv.com" X-VMS-Cc: CCAJR MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII The Internet Tool Catalog regularly posted to the comp.networks.noctools.tools newsgroup includes a description of a program called NPRV, written for Multinet 2.0 (!), which is supposedly available by anonymous FTP from CCC.NMFECC.GOV. When I try to collect it from there, I get the message "ftp disallowed." Mail to postmaster at that site, and to the author, doesn't get a reply. Searches with archie reveal only the text of the NOCtools message. Does anyone know where NPRV can be obtained? The abstract for NPRV describes it as follows: > NPRV is a full-screen, keypad-oriented utility that runs under > VAX/VMS. It allows the user to quickly scan through a user-defined list > of IP addresses (or domain names) and verify a node's reachability. The > node's reachability is determined by performing an ICMP echo, UDP echo > and a TCP echo at alternating three second intervals. The total number > of packets sent and received are displayed, as well as the minimum, > average and maximum round-trip times (in milliseconds) for each type of > echo. Additionally, a "trace route" function is performed to determine > the path from the local system to the remote host. Once all of the trace > route information has filled the screen, a "snapshot" of the screen can > be written to a text file. Upon exiting the utility, these text files > can be used to generate a logical network map showing host and gateway > interconnectivity. Rollo Ross Computer Centre, University of South Australia, The Levels, SA 5095, Australia Ph +61 8 302 3158 Fax 302 3385 DTE 505282622004 Rollo.Ross@UniSA.Edu.Au X-ST-Status: N Return-path: Received: from HUGBOX.SZTAKI.HU ([192.84.225.6]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 06:11:11 PDT Received: from aszi.sztaki.hu (novell.aszi.sztaki.hu) by HUGBOX.SZTAKI.HU (MX V3.3 VAX) with SMTP; Mon, 02 Aug 1993 15:12:52 gmt+1 Received: from ASZI/M_MAILQUEUE by aszi.sztaki.hu (Mercury 1.0); Mon, 2 Aug 93 15:11:42 GMT+1 To: TEX@rmcs.cranfield.ac.uk (Brian {Hamilton Kelly}), info-multinet@TGV.COM From: "Pasztor Miklos" Organization: MTA-SzTAKI / ASzI ================================================================================ Archive-Date: xxx, 2 Aug 1993 15:11:21 GMT+1 Date: 2 Aug 93 15:11:21 GMT+1 Subject: RE: host based router problem (revisited) Priority: normal X-mailer: Pegasus Mail v2.3 (R5). > > If there's anyone out there who knows of a Sun (or other machine) being > used in the manner that Miklos and ourselves are attempting to use MN on > a VAX, perhaps they could comment as to what address gets advertised in > the packets sent over the gateway path through the X25 tunnel. If the > address is that of the host, rather then the fictitious addresses of the > ends of the link, how does the system administrator specify this? I've seen some UNIX based routers, and they work the very same way. I thought that MultiNet is better than these but in this one and only case that does not seem to be true :-). I think that the problem is in the philosophy of IP. Unlike in DECNET, you do not have a node address, you have just interface addresses. The functionality we are looking for is something like "please use this interface address as a node address". Anyhow I concluded that if you have a host based router, you should use a subnet of (one of) your network(s) for the connection. Miklos ==================================================================== Pa'sztor Miklo's | E-mail: pasztor@hugbox.bitnet MTA SZTAKI/ASZI Budapest Victor H. u. 18-22 | Phone: (36)-(1)-149-75-32 Institute for Computation and Automation, Hungarian Academy of Sciences ==================================================================== X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 07:41:03 PDT X-Newsgroups: comp.networks.noctools.wanted,vmsnet.networks.management.misc, vmsnet.networks.misc,vmsnet.networks.tcp-ip.misc,vmsnet.networks.tcp-ip.multinet Subject: multi-platform mail, scheduling utilities Message-ID: <23j77dINNm8t@jhunix.hcf.jhu.edu> From: bmacwill@eureka.wbme.jhu.edu (Bruce MacWilliams) ================================================================================ Archive-Date: xxx, 2 Aug 1993 14:09:49 GMT Date: 2 Aug 1993 14:09:49 GMT Distribution: world Organization: Johns Hopkins University, Dept. of Biomedical Engineering NNTP-Posting-Host: eureka-gold.wbme.jhu.edu Lines: 9 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET We are trying to establish a network which will include UNIX, PC, and Mac machines. Most of the work just involves file transfer or terminal emulation which is easy, but the department want to start using it for some administrative work. They want to be able to email to any machine and do scheduling. Does anyone know of any software out there that would be suitable? Please email to: bmac@owl.med.jhu.edu Thanks. X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 09:38:47 PDT Via: uk.ac.tex; Mon, 2 Aug 1993 17:38:19 +0100 ================================================================================ Archive-Date: Mon, 2 Aug 1993 16:37 GMT Date: Mon, 2 Aug 93 16:37 GMT From: "UK TeX Archive Manager " To: INFO-MULTINET Subject: RE: host based router problem (revisited) In message dated 2 Aug 93 15:11:21 GMT+1, PASZTOR@HU.SZTAKI.ASZI (Pasztor Miklos) said: > I've seen some UNIX based routers, and they work the very same way. Yes: that's been confirmed for me by reading back over some early traffic related to hast-based routers on the janet-ip mailing list, when IP-over-Janet was first being set up. > I thought that MultiNet is better than these but in this one and only > case that does not seem to be true :-). I think that the problem is > in the philosophy of IP. Unlike in DECNET, you do not have a node address, > you have just interface addresses. The functionality we are looking for > is something like "please use this interface address as a node address". Exactly!! > Anyhow I concluded that if you have a host based router, you should > use a subnet of (one of) your network(s) for the connection. I reached the same conclusion: the only problem is, in the UK's instance, this isn't possible --- the administrators of Jips/Janet allocate the IP addresses for each end of the IP-over-X.25 tunnels, and their static routing is predicated upon this concept. Here's part of a reply I received from the JIPS management, when I copied back to them Ken Adelman's analysis of my problem: # All interesting stuff. The problem with X.25 and IP is that # each tends to have dedicated followers, and n'er the twain shall # meet. Consequently, if you try to do something involving the two, # you get strong views expressed from both sides, and neither is # terribly willing to help. # # We could have developed the JIPS using a separate subnet for # each leaf site connection off each NOC router. The trouble with # this is, it would have meant the NOC routers having many secondary # IP addresses on their X.25 interface, and a lot more routing # information to handle. This doesn't scale, obviously. I *do* appreciate his point there :-) # So we have the situation we have at the moment, which only # affects those people whose connection to the outside world # is via a host, and then only with packets sourced from that # host itself. It is possible to solve the problem by having # explicit routes for the 146.97 subnets, pointing at the local # NOC router. I think if both yourself and the Cranfield people # implement static routes for 146.97.160.0 pointing at the RAL # NOC router, this may well work. (Obviously if any other # sites attached to the RAL router don't have these routes in # place, you may also have trouble reaching them). And that seemed to be the conclusion reached during the correspondence last year on the Janet-ip list (which I obviously didn't understand and/or appreciate at the time:-). So each of the sites that enters the JIPS backbone through 146.97.160.1 (about 8--10) needs to put in explicit routing for 146.97.160.0 pointing to that router! Absolutely crazy, but looks like we're stuck with it. Life would be *so* much nicer if MultiNet would just pretend that packets going out on that perticular interface actually originated on the other (Ethernet) one. Ho hum!! Brian {Hamilton Kelly} System Manager for the UK TeX Archive (and also the guy who first moaned from RMCS ;-) X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 2 Aug 1993 10:50:00 PDT Date: Mon, 2 Aug 93 10:50:00 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: RE: host based router problem (revisited) To: SYSTEM @ TEX.AC.UK cc: info-multinet @ TGV.COM > I reached the same conclusion: the only problem is, in the UK's instance, this > isn't possible --- the administrators of Jips/Janet allocate the IP addresses > for each end of the IP-over-X.25 tunnels, and their static routing is > predicated upon this concept. > Here's part of a reply I received from the JIPS management, when I copied back > to them Ken Adelman's analysis of my problem: > # All interesting stuff. The problem with X.25 and IP is that > # each tends to have dedicated followers, and n'er the twain shall > # meet. Consequently, if you try to do something involving the two, > # you get strong views expressed from both sides, and neither is > # terribly willing to help. > # > # We could have developed the JIPS using a separate subnet for > # each leaf site connection off each NOC router. The trouble with > # this is, it would have meant the NOC routers having many secondary > # IP addresses on their X.25 interface, and a lot more routing > # information to handle. This doesn't scale, obviously. > I *do* appreciate his point there :-) It would mean that each of his routers would need to have one additional route PER connection they have. Or more specificially, one additional route PER physical network (an IP over X.25 link is a physical network for this purpose). This scheme scales quite well, the entire Internet is built on it! Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 2 Aug 1993 15:10:28 PDT Date: Mon, 2 Aug 93 15:10:28 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Printing - DCPS and TCPIP - yeah - again! To: palmer @ qut.edu.au cc: info-multinet @ TGV.COM, info-vax @ sri.com > We are yet to install Dec Print Supervisor and up until now I have used LAT > to all of our remote printers. Then I use LATSYM on some and DQS$PRTSMB on > others because the LTA device behaves *like* a normal serial line. Now because > of some things the communications guys have done, LAT won't get to a terminal > server where a Printer was previously served by LAT. > Yep - go to the multinet stream symbiont :-) > OOPS - DCPS wants the location to look like a serial device, for device > interrogation etc - formats the headers as postscripts etc - now how do we > look for the stream symbiont :-( > We have used CREATE_NTY to get some mail out of PMDF to other machines using > a Phone_Net Channel. Now if I could create a permanent NTY then I could use > any of my 'old favorite' symbionts or the one with DCPS. My understanding is > that Create_NTY creates a transitory nty that when deallocated disappears. > Did I miss something? Can Create_NTY create a 'permanent' nty like I can > create a permanent LTA device with latcp? It would be nice if I could force > Create_NTY to give me a device number that I specify not that it makes up > but that isn't a major problem. Create_NTY can't create a permanent one; there is no logic in the NTYDRIVER to retry (or even establish) the connection. The Create_NTY program passes an established connection to the driver. You could run Create_NTY and then set the NTY device /NOHANGUP/PERM, but if the connection is broken by the terminal server you'll need to re-run Create_NTY. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 15:16:27 PDT X-Newsgroups: vmsnet.sysmgt,vmsnet.misc,vmsnet.networks.tcp-ip.multinet Subject: Bootp problem with TCPIP (UCX) Message-ID: <23k0cp$641@Tut.MsState.Edu> From: ritter@CS.MsState.Edu (Tom Ritter) ================================================================================ Archive-Date: xxx, 2 Aug 1993 21:19:21 GMT Date: 2 Aug 1993 21:19:21 GMT Reply-To: ritter@CS.MsState.Edu Followup-To: vmsnet.networks.tcp-ip.ucx Distribution: world Organization: Mississippi State University, Computer Science NNTP-Posting-Host: mickey.cs.msstate.edu Lines: 31 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I have a problem on our Vaxstation 6000. (UXC 2.0D) I need some pointers on how to cure it. The problem manifests itself in BOOTP requests issued by PC/TCP failing about half the time. Using our Sniffer analyzer, we finally tracked the problem to two devices on our campus network, an ANNEX III terminal server, and our Vaxstation responding to these bootp broadcasts with "ICMP Error port 67 unreachable". If the issuing PC sees this response before seeing a legitimate bootp response, which happens about half the time, then the bootp fails. A SHOW SERVICES does not show the BOOTP service. (so I can't disable it...) This error message does not seem to be standard behavior. (+500 other machines do not answer this broadcast, even if they don't know about bootp...) How can I make UCX ignore bootp reqests without sending out an error message to the net???? One reply implied that the the bug was fixed in 2.0D. We just installed that kit with no change..... I'm forwarding followup's to the ucx newsgroup. Any suggestions? Tom Ritter ritter@cs.msstate.edu PS How much is an edu. price of Multi-net??? X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 2 Aug 1993 15:32:28 PDT Date: Mon, 2 Aug 93 15:32:28 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Bootp problem with TCPIP (UCX) To: ritter @ CS.MsState.Edu cc: info-multinet @ TGV.COM > I have a problem on our Vaxstation 6000. (UXC 2.0D) I need some pointers > on how to cure it. > The problem manifests itself in BOOTP requests issued by PC/TCP failing about > half the time. Using our Sniffer analyzer, we finally tracked the problem to two devices on our campus network, an ANNEX III terminal server, and our Vaxstation > responding to these bootp broadcasts with "ICMP Error port 67 unreachable". > If the issuing PC sees this response before seeing a legitimate bootp response, > which happens about half the time, then the bootp fails. > A SHOW SERVICES does not show the BOOTP service. (so I can't disable it...) > This error message does not seem to be standard behavior. (+500 other machines > do not answer this broadcast, even if they don't know about bootp...) > How can I make UCX ignore bootp reqests without sending out an error message to > the net???? > One reply implied that the the bug was fixed in 2.0D. We just installed that > kit with no change..... I'm forwarding followup's to the ucx newsgroup. > Any suggestions? Install MultiNet. IP implementations are not supposed to respond to broadcast requests with ANY sort of error (last sentence in Section 3.2, RFC1122, "Requirements for Internet Hosts -- Communication Layers") > How much is an edu. price of Multi-net??? Contact your salesperson; SALES@TGV.COM. We have a straight educational discount off list-price, and we also have a special program by which you can get a 1-year right-to-use MultiNet at your campus on all your machines for a single price. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 16:08:16 PDT From: oberman@ptavv.llnl.gov X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: netfind and finger Date: Mon, 2 Aug 93 21:50:12 GMT Organization: Lawrence Livermore National Laboratory Lines: 17 Message-ID: <23k260$cet@lll-winken.llnl.gov> NNTP-Posting-Host: ptavv.llnl.gov To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In Article <1993Aug1.140429.1@drycas.club.cc.cmu.edu> fuzzface@drycas.club.cc.cmu.edu (John McMahon) writes: >I find it disconcerting that a Network Information Center would invest the time >/trouble/money in a flawed tool like netfind. Meanwhile they no longer >allow "average users" to register in the WHOIS database. > >WHOIS is more reliable than netfind by a long shot. Actually, they should use Ph. More reliable, faster, and more user friendly! ANd the number of sites supporting Ph is growing fast. R. Kevin Oberman Lawrence Livermore National Laboratory Internet: koberman@llnl.gov (510) 422-6955 Disclaimer: Being a know-it-all isn't easy. It's especially tough when you don't know that much. But I'll keep trying. (Both) X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 16:40:05 PDT From: brodie@fps.mcw.edu (Kent C Brodie) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: TN3270 extended attributes.. how?!? Message-ID: <1993Aug2.102803.777@fps.mcw.edu> ================================================================================ Archive-Date: xxx, 2 Aug 1993 10:28:03 CST Date: 2 Aug 93 10:28:03 CST Organization: Medical College of Wisconsin, Milwaukee, WI Lines: 21 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi. We're just getting started using TN3270 to our campus IBM, and I am disappointed to find out that it's NOT going to work.... The reason is due tothe fact that the IBM uses "extended attributes", like CURSOR SELECT (etc). (term type= 3278-2-e) Multinets' TN3270 doesn't support this (heavy sigh!)... does anyone out there know of a package that *DOES* support ibm's extended attributes for tn3270? (I'm looking for BROAD answers here... any op system, including unix, pc, mac, ... maybe another tcp/ip/vms vendor..???) help! (According to TGV tech support, they do not plan to support this. argh.) -- Kent C. Brodie - Sr. Systems Manager InterNet: brodie@fps.mcw.edu Faculty Physicians & Surgeons uucpNet: fps!brodie Medical College of Wisconsin MaBellNet: +1 414 266 5080 "WARNING: Do not look into laser with remaining eye" (seen on a BIG laser) X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 17:05:24 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: bulletin? Message-ID: From: azog@world.std.com (azog) ================================================================================ Archive-Date: Mon, 2 Aug 1993 19:01:13 GMT Date: Mon, 2 Aug 1993 19:01:13 GMT Distribution: na Organization: The World Public Access UNIX, Brookline, MA Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I dont know exactly what one calls this, but someone once told me a while ago, about a peice of software that takes incomig mail messages, and then 'posts' them to a public area. This is for more like a mailing list type function: instead of allowing 500 people to subscribe to the sammailing list 500 times, just allow one 'global' user to do so. Am I making sense? And where can I find such a beast? Thanx -- Billy D'Augustine azog@world.std.com And I saw, and look, a pale horse, and the one seated upon it had the name Death. X-ST-Status: N Return-path: Received: from Black.Cerritos.EDU ([130.150.200.21]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 17:43:46 PDT Received: from Cerritos.EDU by Cerritos.EDU (PMDF V4.2-11 #4768) id <01H1A18E1F0WAIBQWQ@Cerritos.EDU>; Mon, 2 Aug 1993 17:43:30 PST ================================================================================ Archive-Date: Mon, 02 Aug 1993 17:43:30 -0800 Date: Mon, 02 Aug 1993 17:43:30 -0800 (PST) From: Bruce Tanner Subject: Re: netfind and finger To: info-multinet@tgv.com X-VMS-To: in%"info-multinet@tgv.com" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII In article <23k260$cet@lll-winken.llnl.gov>, oberman@ptavv.llnl.gov writes: > Actually, they should use Ph. More reliable, faster, and more user friendly! > ANd the number of sites supporting Ph is growing fast. I'd like to bring up the server (qi) for our Gopher-based CWIS. Has anybody tried porting qi to VMS and MultiNet? -Bruce -- Bruce Tanner (310) 860-2451 x 596 Tanner@Cerritos.EDU Cerritos College Norwalk, CA X-ST-Status: N Return-path: Received: from grouch.jpl.nasa.gov ([128.149.16.6]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 19:02:31 PDT ================================================================================ Archive-Date: Mon, 2 Aug 1993 18:49:33 -0700 Date: Mon, 2 Aug 1993 18:49:33 -0700 (PDT) From: "Better is the enemy of good... Voltaire" To: info-multinet@TGV.COM, SEF@grouch.jpl.nasa.gov Subject: Backup Bug? We have Multinet V3.2 and Client installed on a VAX 785 running VMS 5.3-1. The server system is an HP750. The problem is this: when I use the VMS BACKUP utility to restore files to a directory on a served disk the records in files end up all run-together on the UNIX side. There seem to be no line-feeds or carriage-control delimiters of any kind. They look fine on the VMS side. If I use the VMS COPY command instead of BACKUP then they look fine. If I use BACKUP and then COPY the restored files to themselves they look fine also. Any comments or suggestions? Thanks, Scott Fullner, JPL/Navigation X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Mon, 2 Aug 93 22:13:02 PDT ================================================================================ Archive-Date: Mon, 2 Aug 1993 22:12:55 -0700 Date: Mon, 2 Aug 1993 22:12:55 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: TN3270 extended attributes.. how?!? To: brodie@fps.mcw.edu (Kent C Brodie) Cc: INFO-MULTINET@TGV.COM In-Reply-To: <1993Aug2.102803.777@fps.mcw.edu> Mail-System-Version: Organization: TGV, Incorporated X-Phone: 408/427-4366 (work); 408/427-4365 (fax) X-Address: 603 Mission Street; Santa Cruz, CA 95060 (work) >Hi. We're just getting started using TN3270 to our campus IBM, and >I am disappointed to find out that it's NOT going to work.... > >The reason is due tothe fact that the IBM uses "extended attributes", >like CURSOR SELECT (etc). (term type= 3278-2-e) > >Multinets' TN3270 doesn't support this (heavy sigh!)... does anyone >out there know of a package that *DOES* support ibm's extended attributes >for tn3270? (I'm looking for BROAD answers here... any op system, >including unix, pc, mac, ... maybe another tcp/ip/vms vendor..???) > >help! > >(According to TGV tech support, they do not plan to support this. argh.) The main problem with supporting this in our tn3270 implementation, is that our current code base consists of the BSD tn3270 with MANY fixes. According to Bill, we'd have to pretty much rewrite it from scratch to be able to support extended attributes. This is not to say that we never will, but it's not at the top of the list of projects presently. Later, -----Stuart X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 2 Aug 1993 22:39:38 PDT Date: Mon, 2 Aug 93 22:39:38 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Backup Bug? To: SEF @ grouch.jpl.nasa.gov cc: info-multinet @ TGV.COM > We have Multinet V3.2 and Client installed on a VAX 785 running VMS 5.3-1. The > server system is an HP750. The problem is this: when I use the VMS BACKUP > utility to restore files to a directory on a served disk the records in files > end up all run-together on the UNIX side. There seem to be no line-feeds or > carriage-control delimiters of any kind. They look fine on the VMS side. If I > use the VMS COPY command instead of BACKUP then they look fine. If I use BACKUP > and then COPY the restored files to themselves they look fine also. Any > comments or suggestions? Sounds like the ".$fdl$" files which we use to hold special file attributes are getting lost. My guess is that the NFS server is choking on the odd filenames, but a TCPDUMP/SNAP=1500/RPC of this traffic should show you what is happening. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 06:10:08 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet,comp.os.vms From: portante@dngbat.dsg.dec.com (DTN 235-8372 / (508)635-8372 / MS DSG1-1/K8) Subject: Re: Printing - DCPS and TCPIP - yeah - again! Message-ID: <1993Aug3.123739.294@nntpd.lkg.dec.com> Lines: 17 Sender: usenet@nntpd.lkg.dec.com (USENET News System) Reply-To: portante@dngbat.dsg.dec.com (DTN 235-8372 / (508)635-8372 / MS DSG1-1/K8) Organization: Digital Equipment Corporation X-Newsreader: mxrn 6.18-4 ================================================================================ Archive-Date: Tue, 3 Aug 1993 12:37:39 GMT Date: Tue, 3 Aug 1993 12:37:39 GMT To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Could you explain more of your printing environment? How does Multinet get you connectivity to your printers? Is the printer attached to some IP box? As long as a device behaves like a serial line DCPS will drive it. So if NTYs look like serial lines, it should drive them, though I do not know much about Multinet. Willing to help if I can. -- Name: Peter A. Portante E-Mail: portante@hannah.enet.dec.com Phone: 508.635.8372 Address: MS DSG1-1/K8 Digital Equipment Corporation 4 Technology Park Drive Westford, MA 01886 X-ST-Status: N Return-path: Received: from HUGBOX.SZTAKI.HU ([192.84.225.6]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 06:57:06 PDT Received: from aszi.sztaki.hu (novell.aszi.sztaki.hu) by HUGBOX.SZTAKI.HU (MX V3.3 VAX) with SMTP; Tue, 03 Aug 1993 15:58:45 gmt+1 Received: from ASZI/M_MAILQUEUE by aszi.sztaki.hu (Mercury 1.0); Tue, 3 Aug 93 15:57:32 GMT+1 To: info-multinet@tgv.com From: "Pasztor Miklos" Organization: MTA-SzTAKI / ASzI ================================================================================ Archive-Date: xxx, 3 Aug 1993 15:56:56 GMT+1 Date: 3 Aug 93 15:56:56 GMT+1 Subject: RE: host based router problem (revisited) Priority: normal X-mailer: Pegasus Mail v2.3 (R5). Brian {Hamilton Kelly} quotes the JIPS people: # All interesting stuff. The problem with X.25 and IP is that # each tends to have dedicated followers, and n'er the twain shall # meet. Consequently, if you try to do something involving the two, # you get strong views expressed from both sides, and neither is # terribly willing to help. My understanding is that the problem is *not* related to X.25. You have the very same problem if you use DECNET, SLIP or even another IP over ethernet connection to connect your network to the Internet. # # We could have developed the JIPS using a separate subnet for # each leaf site connection off each NOC router. Choosing a subnet from *your* network instead of the common core seems to be more appropriate. Security is often based on network numbers, and it can happen, that even if you could reach a destination from your host, you fail because it does not turn out that your host is on that network. Miklos ==================================================================== Pa'sztor Miklo's | E-mail: pasztor@hugbox.bitnet MTA SZTAKI/ASZI Budapest Victor H. u. 18-22 | Phone: (36)-(1)-149-75-32 Institute for Computation and Automation, Hungarian Academy of Sciences ==================================================================== X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 07:14:31 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: hughes@logos.ucs.indiana.edu (larry j. hughes jr.) Subject: Re: netfind and finger Message-ID: Sender: news@usenet.ucs.indiana.edu (USENET News System) Nntp-Posting-Host: logos.ucs.indiana.edu Organization: University Computing Services News ================================================================================ Archive-Date: Tue, 3 Aug 1993 13:53:50 GMT Date: Tue, 3 Aug 1993 13:53:50 GMT Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <01H1A18E2R8YAIBQWQ@Cerritos.EDU>, TANNER@Cerritos.EDU (Bruce Tanner) writes: |> I'd like to bring up the server (qi) for our Gopher-based CWIS. Has anybody |> tried porting qi to VMS and MultiNet? qi, and the tools used to build a qi database, are extremely UNIX-specific. Portability was not on the author's mind when he wrote it. (This isn't a complaint, just an observation :-). //==================================================================\\ || Larry J. Hughes, Jr. | hughes@indiana.edu || || Indiana University | "The person who knows everything || || University Computing Services | has a lot to learn." || \\===================================================================// X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 07:42:36 PDT From: oberman@ptavv.llnl.gov X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: netfind and finger Date: Tue, 3 Aug 93 14:29:09 GMT Organization: Lawrence Livermore National Laboratory Lines: 50 Message-ID: <23lsn3$fpv@lll-winken.llnl.gov> NNTP-Posting-Host: ptavv.llnl.gov To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In Article <23k260$cet@lll-winken.llnl.gov> oberman@ptavv.llnl.gov writes: >Actually, they should use Ph. More reliable, faster, and more user friendly! >And the number of sites supporting Ph is growing fast. After recieveing many mail queries as to "what is Ph", here is a brief write-up. Ph is the University of Illinois electronic phone book. It is a client/server database system originally developed by the University for its own use and now distributed to many sites. The only server I am aware of is for Unix and I have built it on both Ultrix and SunOS systems. Clients are available for Unix, Macintosh, and OpenVMS systems with Multinet. The software is available from cso.uiuc.edu (ph-6.x.tar is the client and qi-2.x.tar is the server) in the /pub directory. The VMS-MultiNet Ph client is available from icaen.llnl.gov in the [.ph.vms] directory. The lookups at LLNL typically take about 2 seconds if no wildcards are used. Seldom more than 5 seconds with wildcards. Our database is about 15,000 entries. Basic syntax: "ph name" where name may be placed in any order (e.i. first name first, last name first, MI first) and the parts of the name are space delimited. $ PH oberman k* ---------------------------------------- name: R. Kevin Oberman phone: 510-422-6955 mailcode: L-156 office_location: Bld.131 Rm.2623A email: koberman@llnl.gov : oberman1@llnl.gov fax: 510-422-2495 affiliation: Electronics Eng - Operations Engineering & Services Division ---------------------------------------- Searching may also be qualified by other fields and those fields which are indexed may be primary search fields. For example: Ph ext=26955 would also find me. R. Kevin Oberman Lawrence Livermore National Laboratory Internet: koberman@llnl.gov (510) 422-6955 Disclaimer: Being a know-it-all isn't easy. It's especially tough when you don't know that much. But I'll keep trying. (Both) X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 3 Aug 1993 08:28:24 PDT Date: Tue, 3 Aug 93 08:28:24 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Printing - DCPS and TCPIP - yeah - again! To: portante @ hannah.enet.dec.com cc: info-multinet @ TGV.COM, info-vax @ sri.com > Could you explain more of your printing environment? How does Multinet get > you connectivity to your printers? Is the printer attached to some IP box? > As long as a device behaves like a serial line DCPS will drive it. So if NTYs > look like serial lines, it should drive them, though I do not know much about > Multinet. MultiNet normally gets connectivity to printers using it's own symbiont which makes the TCP connection. It is possible to make a serial-line-like-device that is hooked to a remote TCP port, but only an a temporary device basis; the the connection is broken user-intervention is needed to reestablish the connection. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 08:38:41 PDT From: eric@sejnet.sunet.se (Eric Thomas) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: TN3270 extended attributes.. how?!? ================================================================================ Archive-Date: xxx, 3 Aug 1993 17:11:18 WET Date: 3 Aug 93 17:11:18 WET Organization: SUNET, Stockholm, Sweden Lines: 22 Reply-To: ERIC@SEARN.SUNET.SE NNTP-Posting-Host: sejnet.sunet.se To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <744354775.726153.VANCE@TGV.COM>, VANCE@TGV.COM (L. Stuart Vance) writes: > The main problem with supporting this in our tn3270 implementation, is that > our current code base consists of the BSD tn3270 with MANY fixes. According > to Bill, we'd have to pretty much rewrite it from scratch to be able to support > extended attributes. This is not to say that we never will, but it's not at > the top of the list of projects presently. For what it's worth, I tried to add SFE support to tn3270 on unix a few years ago, and reached pretty much the same conclusion :-( Not that the code wouldn't benefit from a rewrite - the thing is so slow that, on a dedicated MV3400, I only get about 9.6kbps of effective speed when dialing in from home. The modem reports it is sending characters at about 28kbps thanks to compression, but the VAX isn't fast enough to paint them at that rate. I get better performance using telnet to a faster unix box and then tn3270 from that, but then I get to put up with all the bugs, not to mention 7-bit mode. I've been using Multinet's telnet every day for about a year and the only bug I've found is a gratuitous screen refresh that happens from time to time for unknown reasons. If you'd had to use the original you would know how much effort must have gone into fixing it, so I don't blame TGV for not wanting to touch the thing now that it works :-) Eric X-ST-Status: N Return-path: Received: from UTB.EDU ([192.150.93.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 08:44:11 PDT ================================================================================ Archive-Date: Tue, 3 Aug 1993 10:46:10 -0500 Date: Tue, 3 Aug 1993 10:46:10 -0500 (CDT) From: PHIL@UTB.EDU (NAME ) Subject: pathworks for vms To: info-multinet@tgv.com X-Vmsmail-To: SMTP%"info-multinet@tgv.com" Has anyone installed pathworks for vms without using DEC's VMS/Ultrix connection and subtituted Multinet for TCP/IP transport? Are there any instructions to do this? Any help is appreciated..... Phil Perez. Manager, Technical Operations X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 08:46:28 PDT ================================================================================ Archive-Date: Tue, 3 Aug 1993 8:46:20 -0700 Date: Tue, 3 Aug 1993 8:46:20 -0700 (PDT) From: Matt Madison To: azog@world.std.com CC: INFO-MULTINET@tgv.com Subject: Re: bulletin? >I dont know exactly what one calls this, but someone once told me >a while ago, about a peice of software that takes incomig mail >messages, and then 'posts' them to a public area. This is for more >like a mailing list type function: instead of allowing 500 people >to subscribe to the sammailing list 500 times, just allow one >'global' user to do so. > >Am I making sense? And where can I find such a beast? Thanx You can do this with MultiNet using BBOARD; there should be some documentation for it in the MultiNet System Administrators' Guide. A BBOARD file is just an MM-style mail file into which messages are placed; users can then access the bboard either through the BBOARD command or MM. Alternatively, you could get the BULLETIN package written by Mark London of MIT. It is available by e-mail from BULLETIN@NERUS.PFC.MIT.EDU (use SEND ALL in the body of the message). -Matt -- Matthew Madison | madison@tgv.com | +1 408 427 4366 TGV, Inc. | 603 Mission Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from SECS.UCSC.EDU ([128.114.141.60]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 09:23:32 PDT Received: by SECS.UCSC.EDU (MX V3.1C) id 13603; Tue, 03 Aug 1993 09:22:29 PDT ================================================================================ Archive-Date: Tue, 03 Aug 1993 09:21:10 PDT Date: Tue, 03 Aug 1993 09:21:10 PDT From: "W. Todd Wipke" To: info-multinet@tgv.com Subject: NFS mount still shows up, why? Once we accidentally had some mount points exported to world and a kind security auditor on campus noticed this and did a mount to see if it was really mountable before telling us about it. He claims it is no longer mounted, but our "MU SHOW/ALL" shows "ZOLTAR" still has something mounted served by MultiNet on node MOLENG. What must happen to cause the ZOLTAR mount to go away in the "MU SHO/ALL" listing of nodes mounting files from this server??? -Todd Wipke wipke@chemistry.ucsc.edu W. Todd Wipke wipke@secs.ucsc.edu Molecular Engineering Laboratory wipke@chemistry.ucsc.edu Department of Chemistry wipke@cats.bitnet University of California tel 408 459-2397 Santa Cruz, CA 95064 FAX 408 459-4716 ============= where innovation is a tradition =================== Banana Slug Chosen "Best College Mascot" in 1992 by National Directory of College Athletics X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 3 Aug 1993 10:04:04 -0700 Date: Tue, 3 Aug 1993 10:04:04 -0700 (PDT) From: "Bruce R'. Miller" Subject: Re: pathworks for vms To: PHIL@UTB.EDU Cc: info-multinet@TGV.COM In-Reply-To: <930803104610.d718@UTB.EDU> Mail-System-Version: >Has anyone installed pathworks for vms without using DEC's VMS/Ultrix >connection and subtituted Multinet for TCP/IP transport? Are there >any instructions to do this? Any help is appreciated..... Uh-huh. It should work without modification. Just follow the instructions for installing it with UCX. -bruce X-ST-Status: N Return-path: Received: from grouch.jpl.nasa.gov ([128.149.16.6]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 10:18:26 PDT ================================================================================ Archive-Date: Tue, 3 Aug 1993 10:05:27 -0700 Date: Tue, 3 Aug 1993 10:05:27 -0700 (PDT) From: "Better is the enemy of good... Voltaire" To: info-multinet@TGV.COM, SEF@grouch.jpl.nasa.gov Subject: Backup Bug Again >>We have Multinet V3.2 and Client installed on a VAX 785 running VMS 5.3-1. The >>server system is an HP750. The problem is this: when I use the VMS BACKUP >>utility to restore files to a directory on a served disk the records in files >>end up all run-together on the UNIX side. There seem to be no line-feeds or >>carriage-control delimiters of any kind. They look fine on the VMS side. If I >>use the VMS COPY command instead of BACKUP then they look fine. If I use >>BACKUP and then COPY the restored files to themselves they look fine also. Any >>comments or suggestions? > > Sounds like the ".$fdl$" files which we use to hold special >file attributes are getting lost. My guess is that the NFS server >is choking on the odd filenames, but a TCPDUMP/SNAP=1500/RPC of >this traffic should show you what is happening. > > Ken > But the problem is occurring on the UNIX side only. Everything looks A-OK on the VMS side including the ".$fdl$" files. My understanding is that only VMS/RMS uses the ".$fdl$" files? Am I wrong? Scott Fullner, JPL/Navigation X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 11:10:41 PDT Via: uk.co.ggr; Tue, 3 Aug 1993 19:09:19 +0100 From: mcj2412@ggr.co.uk Received: from relay.ggr.co.uk by uk0x08.ggr.co.uk; Tue, 3 Aug 93 19:10:20 BST Received: from UKSYSA.ggr.co.uk by mailhub.ggr.co.uk (5.59/imd-070593) id AA12436; Tue, 3 Aug 93 18:59:15 BST Message-Id: <9308031759.AA12436@mailhub.ggr.co.uk> ================================================================================ Archive-Date: xxx, 03 Aug 1993 18:42:00 BST Date: 03 Aug 93 18:42:00 BST Subject: RCP and RPC To: INFO-MULTINET@TGV.COM Does anyone know of an RCP or RPC implementation for an HP3000 running MPE/XL (2.2 and 3.1). Thanks, Mark. ps. I know almost nothing about these machines, I simply want to empower my users!! X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 3 Aug 1993 11:55:40 PDT Date: Tue, 3 Aug 93 11:55:40 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: NFS mount still shows up, why? To: wipke @ secs.ucsc.edu cc: info-multinet @ TGV.COM > Once we accidentally had some mount points exported to world and a kind > security auditor on campus noticed this and did a mount to see if it > was really mountable before telling us about it. He claims it is no longer > mounted, but our "MU SHOW/ALL" shows "ZOLTAR" still has something mounted > served by MultiNet on node MOLENG. What must happen to cause the ZOLTAR > mount to go away in the "MU SHO/ALL" listing of nodes mounting files from > this server??? Do a: $ multinet netcontrol rpcmount clear and then the list of things which are mounted will be zeroed. As various clients reboot, the list will populate with the correct information. Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 3 Aug 1993 11:57:28 PDT Date: Tue, 3 Aug 93 11:57:28 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Backup Bug Again To: SEF @ grouch.jpl.nasa.gov cc: info-multinet @ TGV.COM > >>We have Multinet V3.2 and Client installed on a VAX 785 running VMS 5.3-1. The > >>server system is an HP750. The problem is this: when I use the VMS BACKUP > >>utility to restore files to a directory on a served disk the records in files > >>end up all run-together on the UNIX side. There seem to be no line-feeds or > >>carriage-control delimiters of any kind. They look fine on the VMS side. If I > >>use the VMS COPY command instead of BACKUP then they look fine. If I use > >>BACKUP and then COPY the restored files to themselves they look fine also. Any > >>comments or suggestions? > > >> Sounds like the ".$fdl$" files which we use to hold special > >file attributes are getting lost. My guess is that the NFS server > >is choking on the odd filenames, but a TCPDUMP/SNAP=1500/RPC of > >this traffic should show you what is happening. > > >> Ken > > > But the problem is occurring on the UNIX side only. Everything looks A-OK on > the VMS side including the ".$fdl$" files. My understanding is that only > VMS/RMS uses the ".$fdl$" files? Am I wrong? Got it. The problem is that the logic in the NFS client which causes COPY and "normal" applications to write the file in record mode and thereby have RMS convert what would normally be VAR-CR data into Stream_LF data is bypassed by BACKUP's behavior of writing the file in Block mode and explicitly setting the attributes to VAR-CR. What you end up with is a VAR-CR on the UNIX side, which isn't very useful. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 12:14:59 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: hughes@logos.ucs.indiana.edu (larry j. hughes jr.) Subject: Re: TN3270 extended attributes.. how?!? Message-ID: Sender: news@usenet.ucs.indiana.edu (USENET News System) Nntp-Posting-Host: logos.ucs.indiana.edu Organization: University Computing Services News ================================================================================ Archive-Date: Tue, 3 Aug 1993 18:35:48 GMT Date: Tue, 3 Aug 1993 18:35:48 GMT Lines: 20 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug3.171118.1@sejnet.sunet.se>, eric@sejnet.sunet.se (Eric Thomas) writes: | For what it's worth, I tried to add SFE support to tn3270 on unix a few years | ago, and reached pretty much the same conclusion :-( The Berkeley tn3270 code is some of the worst I've ever seen. I hacked in a bug fix myself a few years ago, and it was pretty horrible time. If anybody is considering a rewrite or adding functionality to a UNIX- based tn3270 client (or is even just tired of Berkely tn3270 :-), I'd suggest taking a look at Mike Gerard's "3270" program. It can be found on dxcsftp.cern.ch in /pub/3270. Mike's address is jmg@dxcoms.cern.ch. Mike is extremely knowledegable, and makes bug fixes pronto. Last I checked, his code was still very readable, and smaller than Berkeley's. //==================================================================\\ || Larry J. Hughes, Jr. | hughes@indiana.edu || || Indiana University | "The person who knows everything || || University Computing Services | has a lot to learn." || \\===================================================================// X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 12:40:51 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: hyung@twg.com (Henry Yung (The Last In Line)) Subject: Re: TN3270 extended attributes.. how?!? Message-ID: <1993Aug3.183840.9739@twg.com> Lines: 29 Sender: hyung@speedy.wco.twg.com (Henry Yung (The Last In Line)) Reply-To: hyung@twg.com Organization: The Wollongong Group, West Coast Operations ================================================================================ Archive-Date: Tue, 3 Aug 1993 18:38:40 GMT Date: Tue, 3 Aug 1993 18:38:40 GMT To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug2.102803.777@fps.mcw.edu>, brodie@fps.mcw.edu (Kent C Brodie) writes: |> |>Hi. We're just getting started using TN3270 to our campus IBM, and |>I am disappointed to find out that it's NOT going to work.... |> |>The reason is due tothe fact that the IBM uses "extended attributes", |>like CURSOR SELECT (etc). (term type= 3278-2-e) |> |>Multinets' TN3270 doesn't support this (heavy sigh!)... does anyone |>out there know of a package that *DOES* support ibm's extended attributes |>for tn3270? (I'm looking for BROAD answers here... any op system, |>including unix, pc, mac, ... maybe another tcp/ip/vms vendor..???) |> Kent, Our TN3270 implementation supports "extended attributes" for Model 2, 3, 4 and 5. Hank PS. Oh, and we also support TN3179G. ------------------------------------------------------------------------ __ __/ | / _____/ Henry Yung Email: hyung@twg.com / | / / VMS Software Engineering / / / / -- / The Wollongong Group, Inc. _/ _/ _/ ______/ Voice: (415)962-7100 FAX: (415)969-5547 X-ST-Status: N Return-path: Return-Path: Received: from merck.com ([155.91.1.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 07:33:16 PDT Received: by igw.merck.com with rsmtp; Tue, 3 Aug 1993 10:36:26 EDT ================================================================================ Archive-Date: Tue, 3 Aug 1993 10:29:47 -0400 Date: Tue, 3 Aug 1993 10:29:47 -0400 (EDT) From: stgermain@merck.com To: info-multinet-relay@tgv.com Subject: Re: Alpha & Multinet ReSent-date: Tue, 3 Aug 1993 15:00:16 -0700 (PDT) ReSent-from: Trey Garlough ReSent-to: info-multinet@TGV.COM Do you forsee and problems running Multinet v3.2B on a single processor 400S Alpha class machine running VMS axp v1.5 ? Thx Robert St.Germain stgermain@merck.com Merck Frosst Canada MIS Telecommunications (514) 428-8671 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 3 Aug 93 15:48:17 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: hps@jach.hawaii.edu (Henry Stilmack (JAC System Manager)) Subject: Printing and forms... Message-ID: Sender: news@news.Hawaii.Edu Reply-To: hps@jach.hawaii.edu Organization: UK/Canada/Netherlands Joint Astronomy Centre, Hilo, HI ================================================================================ Archive-Date: Tue, 3 Aug 1993 22:09:38 GMT Date: Tue, 3 Aug 1993 22:09:38 GMT Lines: 20 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET We are using MultiNet to provide remote printing capabilities for remote VAXen to DEC LN03R printers connected via DECservers. (I know, it's wierd, but they used to be LAT connections, but now we have to use IP.) Anyway, we are seeing problems with some of the forms. Basically, the LN03R's are set up using DEC's Printing Services (the CPS symbiont) as execution queues, with generic queues for Postscript and ANSI text pointing (with the proper PRINT qualifiers) to the execution queue. The generic queues are set up as the targets of corresponding Multinet queues at the remote sites. The problem occurs when printing over the TCP/IP connection. Sometimes, a page will print a single line and then page feed. Also, when printing some (not all - it's very frustrating) Postscript graphics files, the job aborts in the middle with a stack overflow. The same files priont with no problems locally. Is there any way to get the Multinet symbiont to use the same forms as the DEC symbionts? I think this might solve the problems. Failing that, any other suggestions would be helpful. If more information is needed, please contact me directly. --- Henry Stilmack ) Henry Stilmack ) Computing Systems Manager ) Perform random kindnesses UK/Netherlands/Canada Joint Astronomy Centre ) and senseless acts of beauty 660 N. A'ohoku Place, Hilo, HI 96720 ) hps@jach.Hawaii.Edu 808-969-6530 ) ------------------------------------------------------------------------------ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 3 Aug 1993 16:46:37 PDT Date: Tue, 3 Aug 93 16:46:37 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Alpha & Multinet To: stgermain @ merck.com cc: info-multinet @ TGV.COM > Do you forsee and problems running Multinet v3.2B on a single > processor 400S Alpha class machine running VMS axp v1.5 ? No, but you will need to $ def multinet_override_version_check true prior to running the MultiNet installation. Note that V3.2 Rev C is required for multiprocessor AXP machines. Ken X-ST-Status: N Return-path: Received: from gimli.bio.purdue.edu ([128.210.18.10]) by TGV.COM via INTERNET ; Wed, 4 Aug 93 09:12:22 PDT Received: from localhost by gimli.bio.purdue.edu with SMTP id AA07330 (5.65b/IDA-1.4.3 for info-multinet@tgv.com); Wed, 4 Aug 93 11:12:21 -0500 Message-Id: <9308041612.AA07330@gimli.bio.purdue.edu> To: info-multinet@tgv.com Subject: harvard xload for Alpha AXP? ================================================================================ Archive-Date: Wed, 04 Aug 1993 11:12:20 -0500 Date: Wed, 04 Aug 93 11:12:20 -0500 From: mckay@gimli.bio.purdue.edu X-Mts: smtp Has anyone gotten the Harvard port of "xload" to compile and run on an Alpha AXP? It won't VEST... --ddm X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Wed, 4 Aug 93 11:02:30 PDT ================================================================================ Archive-Date: Wed, 4 Aug 1993 11:02:25 -0700 Date: Wed, 4 Aug 1993 11:02:25 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: Printing and forms... To: hps@jach.hawaii.edu Cc: INFO-MULTINET@TGV.COM In-Reply-To: Mail-System-Version: Organization: TGV, Incorporated X-Phone: 408/427-4366 (work); 408/427-4365 (fax) X-Address: 603 Mission Street; Santa Cruz, CA 95060 (work) >We are using MultiNet to provide remote printing capabilities for remote VAXen >to DEC LN03R printers connected via DECservers. (I know, it's wierd, but they >used to be LAT connections, but now we have to use IP.) Anyway, we are seeing >problems with some of the forms. Basically, the LN03R's are set up using DEC's >Printing Services (the CPS symbiont) as execution queues, with generic queues >for Postscript and ANSI text pointing (with the proper PRINT qualifiers) to the >execution queue. The generic queues are set up as the targets of corresponding >Multinet queues at the remote sites. > >The problem occurs when printing over the TCP/IP connection. Sometimes, a page >will print a single line and then page feed. Also, when printing some (not all >- it's very frustrating) Postscript graphics files, the job aborts in the >middle with a stack overflow. The same files priont with no problems locally. > >Is there any way to get the Multinet symbiont to use the same forms as the DEC >symbionts? I think this might solve the problems. Failing that, any other >suggestions would be helpful. If more information is needed, please contact me >directly. We do not have out own symbiont for remote printing. We merely replace PSM$OUTPUT (and a couple of other routines) with our our routines to handle network I/O. As such, VMS forms work just fine. What you will likely need to do is define a PostScript form similar to the following: $ show que /form /full post Form name Number Description --------- ------ ----------- POST (stock=DEFAULT) 3 Postscript input; white paper /LENGTH=66 /MARGIN=(BOTTOM=6) /SETUP=(POSTSCRIPT) /STOCK=DEFAULT /TRUNCATE /WIDTH=65535 The width is key; the default form will likely truncate or wrap the file at 80 or 132 characters (really wrecks 512 byte record PS files). Do a DIR /FULL on the files you are trying to print. What is the file/record format? What utility is generating the PS? Regards! -----Stuart X-ST-Status: N Return-path: Received: from ccs1.bbn.com ([128.89.4.28]) by TGV.COM via INTERNET ; Wed, 4 Aug 93 16:05:28 PDT Received: from ccs1.bbn.com by ccs1.bbn.com (PMDF V4.2-12 #3725) id <01H1CV5LL746001PTX@ccs1.bbn.com>; Wed, 4 Aug 1993 18:56:05 EDT ================================================================================ Archive-Date: Wed, 04 Aug 1993 18:56:05 -0400 Date: Wed, 04 Aug 1993 18:56:05 -0400 (EDT) From: Regis McEwen Subject: Re: MU SH/CONN=PROC behavior change To: CONNELLY@iscsvax.uni.edu Cc: info-multinet@tgv.com X-VMS-To: IN%"CONNELLY@iscsvax.uni.edu" X-VMS-Cc: IN%"info-multinet@tgv.com" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII >> > Any chance of adding one in the next release? >> >> Why? > > The old behavior allowed easier monitoring of active outbound sessions as > well as inbound sessions aimed at a particular process, e.g., an NNTP > connection. Adding the /CONTINUOUS qualifier provided an easy way to > observe send and receive queue numbers to ensure that data was flowing. > > The new behavior fills several screens with inbound telnet sessions, a > number of which at this site are in a CLOSING state due, for the most part, > to end-user ignorance. The fact that several screens are presented makes > it difficult, at best, to monitor a particular connection of interest. > > This isn't something I try to use on a daily basis, but I had grown > accustomed to the old behavior and found it useful in a number of > instances. Additionally, I'd like to see a qualifier to specify the duration in which each screen is displayed. /DISPLAY_TIME=20 or /INTERVAL=20 (seconds?) Not sure if others have noticed this or not, but, occassionly, when using the /ALL/CONTINUOUS qualifiers, I see (almost) a blank screen. No header, no page number, no information, except for the date and time at the bottom. (This occurs after display with the ARP tables and before the display with the Network Interface Statistics.) I've noticed this when just typing (MultiNet V3.2 Rev C): $ MULTINET SHOW/ALL/CONTINUOUS That (almost) blank screen appears to be displayed for the same length of time (10 seconds?) the others do. Admittedly, this is no big deal and is just cosmetic, but I get the "feeling" that I am missing a screen or some information while waiting, even though I think I am not. Perhaps this is an undocumented and optional screen that only gives you information if you have the "soon to be released" component or know the proper incantation/patch/person at TGV... -Regis X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Wed, 4 Aug 93 22:33:18 PDT Via: uk.ac.tex; Thu, 5 Aug 1993 06:22:23 +0100 ================================================================================ Archive-Date: Wed, 4 Aug 1993 17:05 GMT Date: Wed, 4 Aug 93 17:05 GMT From: "UK TeX Archive Manager " To: INFO-MULTINET Subject: Campus-wide prices for MultiNet In message <930802153228.6040016f@TGV.COM> dated Mon, 2 Aug 93 15:32:28 PDT, Adelman@COM.TGV (Kenneth Adelman) wrote (in an aside): > > How much is an edu. price of Multi-net??? > > Contact your salesperson; SALES@TGV.COM. We have a straight > educational discount off list-price, and we also have a special > program by which you can get a 1-year right-to-use MultiNet at > your campus on all your machines for a single price. Does the latter right-to-use licence apply outside the USA? When I recently spoke with your UK sales representatives, Harrier Softnet Ltd., I asked what was involved in running MultiNet on a number of VAXen and was told that I had to pay for each machine's licence separately. Errr... BTW, what happens *after* the end of that one year? Does one repeat the payment (in which case, I presume, there's no separate maintenance contract fee). Or, at that stage, having been locked-in to using MultiNet (and hating the idea of switching to anything else :-), does one have to pay for all the separate licences? Brian {Hamilton Kelly} System Manager for the UK TeX Archive at Aston University P.S. I was making the enquiry on behalf of another campus, not for Aston University, nor for the TeX Archive X-ST-Status: N Return-path: ================================================================================ Archive-Date: Wed, 4 Aug 1993 22:58:42 PDT Date: Wed, 4 Aug 93 22:58:42 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Campus-wide prices for MultiNet To: SYSTEM @ TEX.AC.UK cc: info-multinet @ TGV.COM > In message <930802153228.6040016f@TGV.COM> dated Mon, 2 Aug 93 15:32:28 PDT, > Adelman@COM.TGV (Kenneth Adelman) wrote (in an aside): >> > How much is an edu. price of Multi-net??? >> >> Contact your salesperson; SALES@TGV.COM. We have a straight >> educational discount off list-price, and we also have a special >> program by which you can get a 1-year right-to-use MultiNet at >> your campus on all your machines for a single price. Ok, but this is NOT an official answer. I haven't read the policies in a while and am not that familiar with them. If you would like an official answer, and are an educationa institution, contact SALES@TGV.COM or our sales people at 800-TGV-3440 or +1 408 427 4366. > Does the latter right-to-use licence apply outside the USA? When I recently > spoke with your UK sales representatives, Harrier Softnet Ltd., I asked what > was involved in running MultiNet on a number of VAXen and was told that I had > to pay for each machine's licence separately. Yes, it does apply outside of the USA. > Errr... BTW, what happens *after* the end of that one year? Does one repeat > the payment (in which case, I presume, there's no separate maintenance contract > fee). Or, at that stage, having been locked-in to using MultiNet (and hating > the idea of switching to anything else :-), does one have to pay for all the > separate licences? After the period you renew the subscription with another payment. Basically, what you're buying is a one-year right-to-use plus one-year maintenance. If you have traditional paid-up licenses prior to entering the educational subscription and wish to revert to the traditional licenses, we'll consider the maintenance on those licenses as paid-up (in other words, there is no penalty for not carrying the maintenance on those licenses during the years you were on the educational subscription). Ken X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Wed, 4 Aug 93 23:26:17 PDT ================================================================================ Archive-Date: Wed, 4 Aug 1993 23:26:13 -0700 Date: Wed, 4 Aug 1993 23:26:13 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: Campus-wide prices for MultiNet To: "UK TeX Archive Manager " Cc: INFO-MULTINET , sales@tgv.com Mail-System-Version: Organization: TGV, Incorporated X-Phone: 408/427-4366 (work); 408/427-4365 (fax) X-Address: 603 Mission Street; Santa Cruz, CA 95060 (work) >> > How much is an edu. price of Multi-net??? >> >> Contact your salesperson; SALES@TGV.COM. We have a straight >> educational discount off list-price, and we also have a special >> program by which you can get a 1-year right-to-use MultiNet at >> your campus on all your machines for a single price. > >Does the latter right-to-use licence apply outside the USA? When I recently >spoke with your UK sales representatives, Harrier Softnet Ltd., I asked what >was involved in running MultiNet on a number of VAXen and was told that I had >to pay for each machine's licence separately. > >Errr... BTW, what happens *after* the end of that one year? Does one repeat >the payment (in which case, I presume, there's no separate maintenance contract >fee). Or, at that stage, having been locked-in to using MultiNet (and hating >the idea of switching to anything else :-), does one have to pay for all the >separate licences? I've asked one of our international sales folks to post an authoritative answer to this question. Alternatively, you can always drop a note to sales@tgv.com. Regards! -----Stuart X-ST-Status: N Return-path: Received: from jarrah.itd.adelaide.edu.au ([129.127.40.12]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 00:30:13 PDT Received: by jarrah.itd.adelaide.edu.au with SMTP (5.61+IDA+MU/UA-5.28) id AA19944; Thu, 5 Aug 1993 16:59:57 +0930 Message-Id: <9308050729.AA19944@jarrah.itd.adelaide.edu.au> To: Adelman@TGV.COM (Kenneth Adelman) Cc: info-multinet@TGV.COM Subject: Re: netfind and finger In-Reply-To: Your message of "Sat, 31 Jul 1993 17:33:55 PDT." <930731173355.20800496@TGV.COM> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" ================================================================================ Archive-Date: Thu, 05 Aug 1993 16:59:56 +0930 Date: Thu, 05 Aug 1993 16:59:56 +0930 From: Andrew Rutherford Damm I'm miles behind! In message <930731173355.20800496@TGV.COM> you write: + > It might be worth noting that the AT&T part of the Internic consortium + > is running Netfind as a critical, advertised, production part of their + > "white pages" strategy. To the extent that it doesn't work, it would + > seem to me that they are obligated to fix it (or stop advertising it), + > and someone might try telling them that. There has never been a + > requirement that all finger servers return output in the same format. RFC 1288 states: - 2.5. Expected RUIP response - - For the most part, the output of an RUIP doesn't follow a strict - specification, since it is designed to be read by people instead of - programs. It should mainly strive to be informative. - - Output of ANY query is subject to the discussion in the security - section. Note its says "designed to be read by people instead of programs". Thus, if you want a program to read it, you probably shouldn't be using finger :-) + Or put up a FINGER server that returns an unlimited amount of + random data in the correct format. - 2.5.2. {U}{C} query - - A query of {U}{C} is a request for in-depth status of a specified - user {U}. If you really want to refuse this service, you probably - don't want to be running Finger in the first place. Ie, it is valid to not answer this, although it's not likely. In fact, there's not even a requirement that it returns correct information, although that is assumed. The only thing it says you must do, in 2.5.2, is: - An answer MUST include at least the full name of the user. If the - user is logged in, at least the same amount of information returned - by {C} for that user MUST also be returned by {U}{C}. {U} ::= username {C} ::= So "after" that you can say anything you want. Everything after that under the RFC falls under "recommended" or "optional" categories. Nowhere does it state output format, thus the "after" above - you can stick the real name at the end, or anywhere you feel like, just as long as it's there somewhere - even in the middle of the dictionary you're passing back to them. :-) Of course, Multinet currently (as of the release I'm looking at) doesn't conform to the spec in at least two places, so don't sling mud too hard. :-) I have a un*x based fingerd that actually allows the users to specify what their entry will look like - if they don't include the IRL name anywhere, it automatically presents this at the bottom of the output. Of course, this one breaks the RFC too as if it gets a connection from somewhere it doesn't like, it will not "actively refuse" to provide information, but just pretends it doesn't know about that user. This was because I got annoyed by people saying (effectively) "It doesn't look like SunOS, therefore it is broken". Death to such worms who make such dastardly claims! :-) The RFC also states that a user may be allowed to run a program to generate output for themselves being fingered (the inspiration for my templates, along with section 3.2.4 :-) So yes, you can stick in all the random garbage you feel like, as long as your name is in there somewhere. The RFC says so, and the RFC is always right. The RFC is your friend. Stay alert. Trust nobody. Keep your laser handy. (Apologies to "Paranoia" advertising. :-) Cya all, Andrew. /* * Andrew Rutherford andrewr@itd.adelaide.edu.au * +61 8 303 5669 Real Programmers always confuse Christmas and * Room 1060, Adelaide Uni Halloween because OCT 31 == DEC 25 ! */ X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 02:50:36 PDT From: terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr.) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Campus-wide prices for MultiNet Message-ID: <1993Aug5.052843.6707@spcvxb.spc.edu> ================================================================================ Archive-Date: xxx, 5 Aug 1993 05:28:43 EDT Date: 5 Aug 93 05:28:43 EDT Organization: St. Peter's College, US Lines: 19 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <14590985@MVB.SAIC.COM>, SYSTEM@TEX.AC.UK (UK TeX Archive Manager ) writes: > Errr... BTW, what happens *after* the end of that one year? Does one repeat > the payment (in which case, I presume, there's no separate maintenance contract > fee). Or, at that stage, having been locked-in to using MultiNet (and hating > the idea of switching to anything else :-), does one have to pay for all the > separate licences? We switched to MultiNet some time ago, using TGV's educational program. We have been very happy - you simply send TGV money annually, which covers the right-to-use as well as software support. I believe the cost depends on the number of systems involved. In our case, TGV's educational deal costs us less than software support for UCX, so we're actually saving money! I'd suggest you contact sales@tgv.com to get the exact details for your con- figuration (number and size of systems). Terry Kennedy Operations Manager, Academic Computing terry@spcvxa.bitnet St. Peter's College, Jersey City, NJ USA terry@spcvxa.spc.edu +1 201 915 9381 X-ST-Status: N Return-path: Received: from TEX.AC.UK ([134.151.40.18]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 03:42:09 PDT Received: by TeX.ac.uk (MX V3.3 VAX) id 4042; Thu, 05 Aug 1993 10:41:53 GMT ================================================================================ Archive-Date: Thu, 05 Aug 1993 10:41:47 GMT Date: Thu, 05 Aug 1993 10:41:47 GMT From: UK TeX Archive Manager To: info-multinet@tgv.com CC: system@TeX.ac.uk Subject: Configurability of the IP TTL value? Back in March, there was a query (initially in French) as to how to change the maximum number of hops (currently 30). Opinions varied concerning to what this query referred: it was stated that the IP packet time-to-live was 60. Eventually, it would seem, it was decided that the original inquirer had wanted to know about the /MAXIMUM_TTL qualifier for the traceroute command. I referred back to these messages today, because someone has fallen foul of UCX limiting their TTL value (for IP packets now, not route tracing) to 30! And apparently, this value is hard-wired into the code, and DEC are being very reluctant to do anything about it (although, in fairness to them, I understand that someone is looking into patching the image). The reason I looked back at the March correspondence was so that I could insert (yet another) plug for MultiNet into the discussion --- there's a big problem here in the UK in that universities can get UCX ``free'' under the DECcampus scheme --- but like EDLIN, which comes free with DOS, it's expensive at the price! It was never established, in March, whether one *could* alter the IP TTL value: I hope that it *is* possible, through NCU or a logical or whatever, because someone here has been quoting RFCs: > >Right. I've now had a reply back from the CSC telling me > >that this is not possible. To put it mildly I'm rather > >annoyed. > >This means that effectively they do not support TCP/IP, since RFC 791 >and the Hosts requirement RFC 1122 (Requirements for Internet Hosts -- >Communication Layers) state unambiguosly: > > "3.2.1.7 Time-to-Live: RFC-791 Section 3.2 > > A host MUST NOT send a datagram with a Time-to-Live (TTL) > value of zero. > > A host MUST NOT discard a datagram just because it was > received with TTL less than 2. > > The IP layer MUST provide a means for the transport layer to > set the TTL field of every datagram that is sent. When a > fixed TTL value is used, it MUST be configurable. The ^^^^^^^^^^^^^^^^^^^^^^^ > current suggested value will be published in the "Assigned > Numbers" RFC. > > DISCUSSION: > The TTL field has two functions: limit the lifetime of > TCP segments (see RFC-793 [TCP:1], p. 28), and > terminate Internet routing loops. Although TTL is a > time in seconds, it also has some attributes of a hop- > count, since each gateway is required to reduce the TTL > field by at least one. " > >DEC are wrong. it is trivial for them to fix, and should be made >economically non-trivial if they don't fix it. > >there are other products for VMS TCP/IP that dont this limitation, by the way, > >maybe DEC should buy them for you I *hope* that MultiNet is one of the products referred to! Brian {Hamilton Kelly} UK TeX Archive Manager at Aston University X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 06:10:02 PDT Via: uk.co.ggr; Thu, 5 Aug 1993 13:28:04 +0100 From: mcj2412@ggr.co.uk Received: from relay.ggr.co.uk by uk0x08.ggr.co.uk; Thu, 5 Aug 93 10:39:12 BST Received: from UKSYSA.ggr.co.uk by mailhub.ggr.co.uk (5.59/imd-070593) id AA19363; Thu, 5 Aug 93 10:29:26 BST Message-Id: <9308050929.AA19363@mailhub.ggr.co.uk> ================================================================================ Archive-Date: xxx, 05 Aug 1993 10:09:00 BST Date: 05 Aug 93 10:09:00 BST Subject: Frontier Super-TCP To: info-multinet@TGV.COM Does anyone have any information on Frontier Super-TCP. I appreciate this may not be the best place to ask.... Mark. Note, Frontier Super-TCP is a PC package and is probably its US name. It may be called something different in the UK (where I am). X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 5 Aug 1993 08:41:06 PDT Date: Thu, 5 Aug 93 08:41:06 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Configurability of the IP TTL value? To: system @ TeX.ac.uk cc: info-multinet @ TGV.COM > Back in March, there was a query (initially in French) as to how to change the > maximum number of hops (currently 30). Opinions varied concerning to what this > query referred: it was stated that the IP packet time-to-live was 60. > Eventually, it would seem, it was decided that the original inquirer had wanted > to know about the /MAXIMUM_TTL qualifier for the traceroute command. > I referred back to these messages today, because someone has fallen foul of UCX > limiting their TTL value (for IP packets now, not route tracing) to 30! And > apparently, this value is hard-wired into the code, and DEC are being very > reluctant to do anything about it (although, in fairness to them, I understand > that someone is looking into patching the image). > The reason I looked back at the March correspondence was so that I could insert > (yet another) plug for MultiNet into the discussion --- there's a big problem > here in the UK in that universities can get UCX ``free'' under the DECcampus > scheme --- but like EDLIN, which comes free with DOS, it's expensive at the > price! > It was never established, in March, whether one *could* alter the IP TTL value: > I hope that it *is* possible, through NCU or a logical or whatever, because > someone here has been quoting RFCs: > I *hope* that MultiNet is one of the products referred to! $ multinet set/kernel udp_ttl 80 $ multinet set/kernel tcp_ttl 80 You'd need to put these in your startup after starting MultiNet. This isn't an EASY way to change them, but at least you can change them. Btw, our defaults are 60 hops. Ken X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 09:00:05 PDT ================================================================================ Archive-Date: Thu, 5 Aug 1993 09:00:00 -0700 Date: Thu, 5 Aug 1993 09:00:00 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: Configurability of the IP TTL value? To: UK TeX Archive Manager Cc: info-multinet@tgv.com In-Reply-To: <00970905.96206160.4042@TeX.ac.uk> Mail-System-Version: Organization: TGV, Incorporated X-Phone: 408/427-4366 (work); 408/427-4365 (fax) X-Address: 603 Mission Street; Santa Cruz, CA 95060 (work) >It was never established, in March, whether one *could* alter the IP TTL value: >I hope that it *is* possible, through NCU or a logical or whatever, because >someone here has been quoting RFCs: >. >. >. >>there are other products for VMS TCP/IP that dont this limitation, by the way, >> >>maybe DEC should buy them for you > >I *hope* that MultiNet is one of the products referred to! It is, although it is not particularly well documented. It requires making a trivial kernel patch (one variable for TCP, one variable for UDP) or using a particular MULTINET SET command. It does, however, fall into the range of things that we only like to give out with a "prescription". I.e., if not used carefully, you can REALLY hose yourself... The current MultiNet default is 60 hops. If anyone actually has a network currently with more hops than that, not only will I tell you how to fix the problem, I'll also treat you to dinner the next time I see you so that you can tell me how you built and managed such a network. Although, I guess that I should be careful with my offer. I have actually seen 45 hops on the Internet.... Regards! -----Stuart X-ST-Status: N Return-path: Received: from ZCIAS1.ZIFF.COM ([140.244.1.69]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 09:59:22 PDT Received: from DN-ARNOR by zcias1.ziff.com (PMDF #3828 ) id <01H1DY5JOUSG00235V@zcias1.ziff.com>; Thu, 5 Aug 1993 12:58:18 EST Received: from arnor.zis.ziff.com by arnor.zis.ziff.com (PMDF V4.2-10 #3828) id <01H1DXRDIAW0KT5P58@arnor.zis.ziff.com>; Thu, 5 Aug 1993 12:57:48 EST ================================================================================ Archive-Date: xxx, 05 Aug 1993 12:57:47 -0500 Date: 05 Aug 1993 12:57:47 -0500 (EST) From: D_CF@zis.ziff.com Subject: Re: Frontier Super-TCP To: info-multinet@tgv.com Cc: mcj2412@ggr.co.uk X-Envelope-to: info-multinet@tgv.com X-VMS-To: IN%"info-multinet@tgv.com" X-VMS-Cc: IN%"mcj2412@ggr.co.uk",D_CF MIME-version: 1.0 >From: IN%"mcj2412@ggr.co.uk" >Subj: Frontier Super-TCP > >Does anyone have any information on Frontier Super-TCP. I appreciate this >may not be the best place to ask.... > >Mark. > >Note, Frontier Super-TCP is a PC package and is probably its US name. It >may be called something different in the UK (where I am). > Try: Frontier Technologies Corp. 10201 N. Port Washington Road Mequon, WI 53092 FAX FOR INFO: 414-241-7084 EMAIL FOR INFO: tcp@frontiertech.com VOICE: 414-241-4555 BBS: 414-241-7083 -Carlos Fuentes, Jr. Lead Systems Engineer Citicorp*I/B/E/S X-ST-Status: N Return-path: Received: from CADRE1.CALTECH.EDU ([131.215.154.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 10:35:02 PDT ================================================================================ Archive-Date: Thu, 5 Aug 1993 10:35:17 -0700 Date: Thu, 5 Aug 1993 10:35:17 -0700 (PDT) From: JDM@CADRE1.CALTECH.EDU To: INFO-MULTINET@TGV.COM CC: JDM@CADRE1.CALTECH.EDU Subject: signature trailers Is there a way to get VMSMAIL to append a signature to every message as does various internet mail packages? jdm@caltech.edu JON MELVIN X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 12:46:14 PDT From: fuzzface@drycas.club.cc.cmu.edu (John McMahon) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: signature trailers ================================================================================ Archive-Date: xxx, 5 Aug 1993 15:10 EDT Date: 5 Aug 1993 15:10 EDT Organization: Carnegie Mellon University Computer Club Lines: 18 Distribution: world NNTP-Posting-Host: drycas.club.cc.cmu.edu News-Software: VAX/VMS VNEWS 1.41 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930805103517.248002c2@CADRE1.CALTECH.EDU>, JDM@CADRE1.CALTECH.EDU writes... >Is there a way to get VMSMAIL to append a signature to every message as >does various internet mail packages? It depends on the editor you use. I use EVE/TPU. When I am in Mail I assign the logical TPU$COMMAND to issue a bunch of commands which include resetting the margins, adding a signature file and doing indenting. The trick is to only use that when in mail. I do this by running mail in a subprocess. John 'FuzzFace\Fast-Eddie' McMahon fuzzface@drycas.club.cc.cmu.edu {use finger to get my PGP public key} "Any networking problem can be solved by the application of 75 miles of Unshielded Twisted Pair (UTP)" - InterOp Spring 1993 X-ST-Status: N Return-path: Received: from SECS.UCSC.EDU ([128.114.141.60]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 14:30:30 PDT Received: by SECS.UCSC.EDU (MX V3.1C) id 14406; Thu, 05 Aug 1993 14:29:45 PDT ================================================================================ Archive-Date: Thu, 05 Aug 1993 14:29:47 PDT Date: Thu, 05 Aug 1993 14:29:47 PDT From: "W. Todd Wipke" To: info-multinet@tgv.com Subject: VMS backup of a DEC Ultrix system via MultiNet -------------------- ------------------- |DEC 5000/240 | | Vax 4000/500 | | no tape |-------------------------------| TF85 tape | -------------------- TCPIP | MultiNet | ------------------- What is the optimal way to back up disks on the DEC 5000/240 to the tape unit on the VAX 4000/500 such that they can actually be successfully restored in full or selective files? What are the desired SEMANTICS to make it work right? 1. The DEC Ultrix system could tar and compress to the vax disks and then the vax could backup that file to tape. 2. The vax could mount the Ultrix file structure and back it up as though it were a disk on the vax. 3. Somehow use the remote tape unit directly? I would especially like to hear from someone who does this regularly and has had to use it for recovery purposes. There are too many variables to work out by experimenting. -Todd Wipke wipke@chemistry.ucsc.edu X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 14:47:56 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Configurability of the IP TTL value? Message-ID: <23rrq5$3gl@lll-winken.llnl.gov> From: oberman@ptavv.llnl.gov Date: Thu, 5 Aug 93 20:50:33 GMT Organization: Lawrence Livermore National Laboratory NNTP-Posting-Host: ptavv.llnl.gov Lines: 20 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In Article <744566400.265336.VANCE@tgv.com> VANCE@TGV.COM (L. Stuart Vance) writes: >The current MultiNet default is 60 hops. If anyone actually has a network >currently with more hops than that, not only will I tell you how to fix the >problem, I'll also treat you to dinner the next time I see you so that you can >tell me how you built and managed such a network. Although, I guess that I >should be careful with my offer. I have actually seen 45 hops on the >Internet.... Someone should point out to DEC that the official recommended TTL default is 64. It was increased from 30 years ago. 64 was recommneded in RFC1340 which is dated July, 1992, so UCX is not meeting the standard. (Wow! I'm shocked!) :-) But, in this case, neither is MultiNet. (OK. I'm picking nits.) R. Kevin Oberman Lawrence Livermore National Laboratory Internet: koberman@llnl.gov (510) 422-6955 Disclaimer: Being a know-it-all isn't easy. It's especially tough when you don't know that much. But I'll keep trying. (Both) X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 5 Aug 1993 15:22:24 PDT Date: Thu, 5 Aug 93 15:22:24 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Configurability of the IP TTL value? To: oberman @ ptavv.llnl.gov cc: info-multinet @ TGV.COM > In Article <744566400.265336.VANCE@tgv.com> > VANCE@TGV.COM (L. Stuart Vance) writes: > >The current MultiNet default is 60 hops. If anyone actually has a network > >currently with more hops than that, not only will I tell you how to fix the > >problem, I'll also treat you to dinner the next time I see you so that you can > >tell me how you built and managed such a network. Although, I guess that I > >should be careful with my offer. I have actually seen 45 hops on the > >Internet.... > Someone should point out to DEC that the official recommended TTL default is > 64. It was increased from 30 years ago. 64 was recommneded in RFC1340 which is > dated July, 1992, so UCX is not meeting the standard. (Wow! I'm shocked!) :-) > But, in this case, neither is MultiNet. (OK. I'm picking nits.) Oops. Missed this one. This will be fixed (raised from 60 to 64) in MultiNet V3.3). Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 5 Aug 1993 15:38:09 PDT Date: Thu, 5 Aug 93 15:38:09 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: VMS backup of a DEC Ultrix system via MultiNet To: wipke @ chemistry.ucsc.edu cc: service @ TGV.COM, info-multinet @ TGV.COM > -------------------- ------------------- > |DEC 5000/240 | | Vax 4000/500 | > | no tape |-------------------------------| TF85 tape | > -------------------- TCPIP | MultiNet | > ------------------- > What is the optimal way to back up disks on the DEC 5000/240 to the > tape unit on the VAX 4000/500 such that they can actually be successfully > restored in full or selective files? What are the desired SEMANTICS to make > it work right? > 1. The DEC Ultrix system could tar and compress to the vax disks and then > the vax could backup that file to tape. > 2. The vax could mount the Ultrix file structure and back it up as though > it were a disk on the vax. > 3. Somehow use the remote tape unit directly? > I would especially like to hear from someone who does this regularly and has > had to use it for recovery purposes. There are too many variables to work > out by experimenting. Use rdump on the UNIX machine to remotely access the VMS tape drive and write a dump-format tape directly to it. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 17:14:40 PDT From: fuzzface@drycas.club.cc.cmu.edu (John McMahon) X-Newsgroups: comp.os.vms,vmsnet.networks.tcp-ip.multinet Subject: Off into the sunrise (Was: How to buy a used VAX ?) ================================================================================ Archive-Date: xxx, 5 Aug 1993 19:09 EDT Date: 5 Aug 1993 19:09 EDT Organization: Carnegie Mellon University Computer Club Lines: 46 Distribution: world NNTP-Posting-Host: drycas.club.cc.cmu.edu News-Software: VAX/VMS VNEWS 1.41 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Non-Technical Drek Follows... "Are you the Zaphod Beeblebrox ?" "No, I'm a Zaphod Beeblebrox. Didn't you hear ? We come in six packs." - Douglas Adams, The Hitchikers Guide to the Galaxy trilogy In article <1993Aug1.133017.1@drycas.club.cc.cmu.edu>, fuzzface@drycas.club.cc.cmu.edu (John McMahon) writes... >Well, after 7 years in the VMS arena I find myself going to a new job that only >has one VMS system. And that machine has been relegated to the obsolete >backwater of the machine room, so I don't expect it to be a big part of my >career. :-( Well, I've received a number of messages from folks on "How to buy a used VAX?" (summary forthcoming to an INFO-VAX source near you). I've also received messages from people asking "Are you the same guy who works for TGV ?" and offering well wishes on my new "non-VMS" job. It's great hearing from folks, and thanks much (from me and my wife) for the nice thoughts. Having caused a great amount of confusion among a lot of people I figured I ought to clear this up. I currently work in Technical Support, where my job is to confuse... oops... not confuse our customers :-) FUZZFACE@DRYCAS.CLUB.CC.CMU.EDU is the same guy who works for TGV and is known to some of you as MCMAHON@TGV.COM. I've been with TGV for three years as of last month. I've been on Info-Vax/Comp.Os.Vms using different accounts at TGV and NASA since 1987. I've been lucky to meet a lot of you through the network, DECUS and Interop. It's been great fun, well except for that last Info-Vax flamewar... and there was that time at DECUS... and... :-) Well, all good things have to come to an end. I think there is a law of thermodynamics requiring ends. So, at the end of next week (Friday 13 August 1993) I'll be leaving TGV and California. I'm headed off to work on TCP/IP software for Macintoshes at InterCon Systems in Virginia. I'll still be floating around DECUS, Interop, the Internet and any place they haven't thrown me out of yet. :-) I won't be doing much VMS anymore, which will be weird... but then again life usually is :-) Cheers! John 'FuzzFace\Fast-Eddie' McMahon fuzzface@drycas.club.cc.cmu.edu {use finger to get my PGP public key} "Any networking problem can be solved by the application of 75 miles of Unshielded Twisted Pair (UTP)" - InterOp Spring 1993 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 17:29:02 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: dwing@uh01.colorado.edu (Dan Wing) Subject: Re: signature trailers Message-ID: <1993Aug5.231043.11032@colorado.edu> Sender: news@colorado.edu (The Daily Planet) Nntp-Posting-Host: buckie.hsc.colorado.edu Reply-To: dwing@uh01.colorado.edu Organization: University of Colorado Hospital Authority, Denver ================================================================================ Archive-Date: Thu, 5 Aug 1993 23:10:43 GMT Date: Thu, 5 Aug 1993 23:10:43 GMT Lines: 28 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930805103517.248002c2@CADRE1.CALTECH.EDU>, JDM@CADRE1.CALTECH.EDU writes: >Is there a way to get VMSMAIL to append a signature to every message as >does various internet mail packages? Nope. You can, however, tell VMSmail to use a .COM file when editing, and this .COM file can append a signature before giving the edited mail message back to VMSmail. You can use the file SYS$SYSTEM:MAILEDIT.COM as a template, and maybe do something like: $ OPEN/APPEND FILE 'P2' $ WRITE FILE "Signed, user@host" $ CLOSE FILE after the EDIT statements in that .COM file. Then define the *unsupported* logical MAIL$EDIT to point to your newly-modified MAILEDIT.COM, and you should be set. I haven't tried this, so there might be a problem with EDT's filetype and the filetype that OPEN/APPEND will work with.... (The INFO-VAX/comp.os.vms FAQ contains some code written by Kevin Cole which will append signatures to your mail messages and prefix lines with ">" using EDT. You might want to take a look at that stuff, too.) -Dan Wing, Systems Administrator, University Hospital, Denver dwing@uh01.colorado.edu or wing_d@ucolmcc.bitnet X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 19:28:36 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Help needed to convert VMS WAIS server code Message-ID: <1993Aug6.113212.66656@qut.edu.au> From: ling@qut.edu.au ================================================================================ Archive-Date: xxx, 6 Aug 1993 11:32:12 +1000 Date: 6 Aug 93 11:32:12 +1000 Organization: Queensland University of Technology Lines: 35 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hello Multinet TCPIP experts, This is about compiling and linking the wais server and indexer on VMS. I am having troubles compiling and linking the VMS wais server and indexer on a VAXstation using TGV/Multinet TCP/IP software. I got the VMS wais software from sunsite.unc.edu. The version (8b2) is based on TCP/IP and my site only got the Multinet TCP/IP. When I tried to link the server code with the library UCX$IPC.OLB, I got the error: %LINK-W-NUDFSYMS, 1 undefined symbol: %LINK-I-UDFSYM, TCPWARE_SERVER When I tried to link the indexer code with the library UCX$IPC.OLB, I got the error: %LINK-W-NUDFSYMS, 2 undefined symbols: %LINK-I-UDFSYM, LOG_WRITE %LINK-I-UDFSYM, OUT_WRITE I would be most grateful if persons who have converted the software to run on Multinet TCP/IP or persons who know how to convert can show me what to do. I am a raw beginner in C. Thanks in advance. How-Hie Ling. (h.ling@qut.edu.au) Queensland University of Technology, Brisbane, Queensland, Australia. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 5 Aug 93 22:55:09 PDT X-Newsgroups: vmsnet.networks.desktop.pathworks,vmsnet.networks.tcp-ip.multinet From: lynam@maize.qut.edu.au (Barry Lynam) Subject: Pathworks for Mac, HP Jetdirect and Multinet Streams Message-ID: <1993Aug6.052717.4090@news.qut.edu.au> Lines: 61 Sender: news@news.qut.edu.au (USENET News System) Reply-To: B.Lynam@qut.edu.au Organization: Queensland University of Technology ================================================================================ Archive-Date: Fri, 6 Aug 1993 05:27:17 GMT Date: Fri, 6 Aug 93 05:27:17 GMT To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi, I have a problem and I'm wondering if there is a solution. We have a HP LaserJet 4m with a JetDirect TCPIP card. We have a queue setup on our VAXen using the Stream Protocol. All that works fine. PC can print using Pathworks. VMS users can print normally. The printer autodetects postscript so we can even have text headers and postscript output. Now the problem. We also have Mac's and Pathworks for Mac. If I create an Appletalk queue to print to the VMS queue we get an error printed after the header, %PSM-E-MODNOTFND, library module MSAP$PROCSET5 not found in record 0 -LBR-E-KEYNOTFND, key not found From what I understand, the problem is that the Device control Lib and the symbiont aren't talking. (My VMS is not great, but my unix, mac, pc and networking is OK.) Has anybody got this situation working? Does anyone have any ideas? (aside from making the printer talk LAT) Much thanks in advance Barry PS Below is how the printer is setup and what the failing jobs look like. command>> show queue test_gp /full/all Printer queue TEST_GP, stopped, on REDGUM::NLP6:, mounted form DEFAULT /BASE_PRIORITY=4 /DEFAULT=(FEED,FLAG,FORM=DEFAULT) /OWNER=[SYSTEM] /PROCESSOR=MULTINET_STREAM_SYMBIONT /PROTECTION=(S:E,O:D,G:R,W:W) /SEPARATE=(FLAG) Entry Jobname Username Blocks Status ----- ------- -------- ------ ------ 2731 "" MSAP$ACCOUNT 76 Pending (queue stopped) Submitted 5-AUG-1993 17:34 /FORM=DEFAULT /NOTE="PrintMonitor document "Desktop" received from Macintosh user """ /PARAM=("DATA=POSTSCRIPT") /PRIORITY=30 File: _$1$DUA0:[SYS1.MSA.MSAP$SPOOL]MSAP$RCVR0_TEMP.PS;1 /DELETE /SETUP=(MSAP$PROCSET5) +--------------------------------------------------------------------- ---+ | Barry Lynam EMail: B.Lynam@qut.edu.au | | Communications Software Support Fax: +61 7 864 1343 | | Computing Services Postal: GPO Box 2434 | | Queensland University of Technology Brisbane 4001 | | Brisbane AUSTRALIA AUSTRALIA | +--------------------------------------------------------------------- ---+ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 5 Aug 1993 23:00:29 PDT Date: Thu, 5 Aug 93 23:00:29 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Pathworks for Mac, HP Jetdirect and Multinet Streams To: B.Lynam @ qut.edu.au cc: info-multinet @ TGV.COM > I have a problem and I'm wondering if there is a solution. We have > a HP LaserJet 4m with a JetDirect TCPIP card. We have a queue setup > on our VAXen using the Stream Protocol. All that works fine. PC can > print using Pathworks. VMS users can print normally. The printer > autodetects postscript so we can even have text headers and postscript > output. Now the problem. We also have Mac's and Pathworks for Mac. > If I create an Appletalk queue to print to the VMS queue we get an > error printed after the header, > %PSM-E-MODNOTFND, library module MSAP$PROCSET5 not found in record 0 > -LBR-E-KEYNOTFND, key not found > From what I understand, the problem is that the Device control Lib and > the symbiont aren't talking. (My VMS is not great, but my unix, mac, > pc and networking is OK.) Has anybody got this situation working? > Does anyone have any ideas? (aside from making the printer talk LAT) > Much thanks in advance > Barry > PS Below is how the printer is setup and what the failing jobs look > like. > command>> show queue test_gp /full/all > Printer queue TEST_GP, stopped, on REDGUM::NLP6:, mounted form DEFAULT > /BASE_PRIORITY=4 /DEFAULT=(FEED,FLAG,FORM=DEFAULT) /OWNER=[SYSTEM] > /PROCESSOR=MULTINET_STREAM_SYMBIONT /PROTECTION=(S:E,O:D,G:R,W:W) > /SEPARATE=(FLAG) > Entry Jobname Username Blocks Status > ----- ------- -------- ------ ------ > 2731 "" MSAP$ACCOUNT 76 Pending (queue stopped) > Submitted 5-AUG-1993 17:34 /FORM=DEFAULT > /NOTE="PrintMonitor document "Desktop" received from > Macintosh user """ > /PARAM=("DATA=POSTSCRIPT") /PRIORITY=30 > File: _$1$DUA0:[SYS1.MSA.MSAP$SPOOL]MSAP$RCVR0_TEMP.PS;1 > /DELETE > /SETUP=(MSAP$PROCSET5) Whatever is submitting this job is doing so with the qualifier /SETUP=MSAP$PROCSET5. Presumbly it expects the queue to be initialized with a setup library; you'll need to create a MULTINET:INITIALIZE_TEST_GP.COM file to replace the commands in MULTINET:REMOTE-PRINTER-QUEUES.COM with a new initialize that includes a /LIBRARY= to specify the appopriate library. Ken X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 02:19:55 PDT Via: uk.co.ggr; Fri, 6 Aug 1993 10:18:15 +0100 From: mcj2412@ggr.co.uk Received: from relay.ggr.co.uk by uk0x08.ggr.co.uk; Fri, 6 Aug 93 10:19:18 BST Received: from UKSYSA.ggr.co.uk by mailhub.ggr.co.uk (5.59/imd-070593) id AA23922; Fri, 6 Aug 93 10:11:48 BST Message-Id: <9308060911.AA23922@mailhub.ggr.co.uk> ================================================================================ Archive-Date: xxx, 06 Aug 1993 09:59:00 BST Date: 06 Aug 93 09:59:00 BST Subject: RCP To: info-multinet@TGV.COM Hi. I've looked through the Multinet manuals (maybe not in the right places mind you) and don't find much on RCP. I did find the VMS HELP. Can you confirm: 1. RCP uses .RHOSTS just like RLOGIN and RSHELL. 2. RCP fails if it gets any unexpected output from the remote end, ie because the SYS$SYLOGIN procedure on the target machine in my case outputs some 'nice' user information. Thanks, Mark. ps. As my manager's just given a highly political priority zero task maybe someone can help. Does anyone know of a TCP/IP product that allows you to watch someone elses terminal session? I think there was a VMS product called Control or something. Any information will be gratefully received. Thanks again. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 02:55:11 PDT From: nmmnh@apsicc.aps.edu (Not a bug, a feature...) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Slip problems ================================================================================ Archive-Date: xxx, 5 Aug 1993 05:35 MST Date: 5 Aug 1993 05:35 MST Organization: Albuquerque Public Schools - Career Enrichment Center Lines: 21 Distribution: world NNTP-Posting-Host: apsicc.aps.edu News-Software: VAX/VMS VNEWS 1.41 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET We are having problems getting a SLIP connection to worrk properly and need some help. We are using a subnet mask of 255.255.255.192. The dynamic slip line configuration looks like this: VAX1 (IP 192.133.100.65) ==> SLIP port (192.133.100.130) to modem to modem SLIP port (192.133.100.129) ==> VAX2 (IP 192.133.100.5) The modems stay connected, but data does not flow through. On a ping to the remote site the modem sends and indicates it is receiving data. All the routing tables are set up properly. When the phone line drops I get an error message that no route is available to the remote host. With the phone connection Pings and other attempts to share data fail, with no explanation. Any ideas? john chadwick nmmnh@apsicc.aps.edu X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 05:13:47 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: iversen@dsfys1.fi.uib.no (Per Steinar Iversen) Subject: Re: Configurability of the IP TTL value? Message-ID: <1993Aug6.110914.13099@alf.uib.no> Lines: 57 Sender: iversen@vsfys1.fi.uib.no (Per Steinar Iversen) Reply-To: iversen@vsfys1.fi.uib.no Organization: Department of Physics, University of Bergen, Norway X-Newsreader: mxrn 6.18-3 ================================================================================ Archive-Date: Fri, 6 Aug 1993 11:09:14 GMT Date: Fri, 6 Aug 93 11:09:14 GMT Lines: 57 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <23rrq5$3gl@lll-winken.llnl.gov>, oberman@ptavv.llnl.gov writes: >In Article <744566400.265336.VANCE@tgv.com> >VANCE@TGV.COM (L. Stuart Vance) writes: > >>The current MultiNet default is 60 hops. If anyone actually has a network >>currently with more hops than that, not only will I tell you how to fix the >>problem, I'll also treat you to dinner the next time I see you so that you can >>tell me how you built and managed such a network. Although, I guess that I >>should be careful with my offer. I have actually seen 45 hops on the >>Internet.... > >Someone should point out to DEC that the official recommended TTL default is >64. It was increased from 30 years ago. 64 was recommneded in RFC1340 which is >dated July, 1992, so UCX is not meeting the standard. (Wow! I'm shocked!) :-) >But, in this case, neither is MultiNet. (OK. I'm picking nits.) > >R. Kevin Oberman Lawrence Livermore National Laboratory >Internet: koberman@llnl.gov (510) 422-6955 In the group comp.protocols.tcp-ip there was recently a pointer to a tool "netconfig" that can print out and also modify a number of tcp/ip parameters for a number of popular Unix systems. For my DECstations running Ultrix 4.2 I got: dsfys1.fi.uib.no> netconfig IP: Routing (ipforwarding): on ICMP: Redirects (ipsendredirects): on Network Unreachables (ipgateway): off TCP: Receive Checksum (tcpcksum): on Output Buffer Size (tcp_sendspace): 16384 Input Buffer Size (tcp_recvspace): 16384 UDP: Checksum (udpcksum): on Max Sndbuf and Msg Size (udp_sendspace): 9000 Max Rcvbuf and Msg Size (udp_recvspace): 9000 TTL (udp_ttl): 30 It seems that it is not only UCX that has a TTL of 30 ... Fortunately netconfig lets me modify this immediately to 64, and I increased my in/output buffersizes too. -psi +------------------------------------------------------------------------------+ ! Per Steinar Iversen ! Internet: iversen@vsfys1.fi.uib.no ! ! Fysisk Institutt ! BITnet: iversen@cernvm.bitnet ! ! Universitetet i Bergen ! HEPnet: VSFYS::IVERSEN (VSFYS=21.341=21845) ! ! Allegaten 55 ! Phone: +47 5212770 ! ! N-5007 Bergen ! Fax: +47 5318334 ! ! NORWAY ! ! +------------------------------------------------------------------------------+ X-ST-Status: N Return-path: Received: from JADE.BUCKNELL.EDU ([134.82.1.21]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 07:09:18 PDT ================================================================================ Archive-Date: Fri, 6 Aug 1993 10:09:22 -0400 Date: Fri, 6 Aug 1993 10:09:22 -0400 (EDT) From: LYNCH@jade.bucknell.edu To: info-multinet@tgv.com Subject: FTP site for patch to MAILSHR Could someone please send me the address of an FTP site that has the patch to mailshr.exe that allows users to enter addresses without specifying SMTP%""? I am running Multinet version 3.2B and VMS version 5.5-2 on a VAX 4400. Thanks, Mike ------------------------------------------------------------------------------ Michael Lynch Systems Librarian, Bertrand Library (717) 524-3565 Bucknell University LYNCH@JADE.BUCKNELL.EDU Lewisburg, PA. 17837 ------------------------------------------------------------------------------ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 6 Aug 1993 08:19:18 PDT Date: Fri, 6 Aug 93 08:19:18 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: RCP To: mcj2412 @ ggr.co.uk cc: info-multinet @ TGV.COM > Hi. I've looked through the Multinet manuals (maybe not in the right > places mind you) and don't find much on RCP. I did find the VMS HELP. > Can you confirm: > 1. RCP uses .RHOSTS just like RLOGIN and RSHELL. Actually, RCP uses RSHELL (or REXEC) to fire up the RCP server. It is RSHELL which in turn uses .RHOSTS> > 2. RCP fails if it gets any unexpected output from the remote end, ie > because the SYS$SYLOGIN procedure on the target machine in my case > outputs some 'nice' user information. Correct. But you could put a check in your LOGIN.COM or SYLOGIN to see if you are an RSHELL job and exit without printing anything in this case. > ps. As my manager's just given a highly political priority zero task > maybe someone can help. Does anyone know of a TCP/IP product that > allows you to watch someone elses terminal session? I think there > was a VMS product called Control or something. Any information > will be gratefully received. The WATCH utility available from DECUS. Ken X-ST-Status: N Return-path: Received: from vms1.cc.uop.edu ([138.9.1.2]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 08:22:43 PDT Received: by vms1.cc.uop.edu (MX V3.1B) id 30769; Fri, 06 Aug 1993 08:22:20 PDT Sender: ebates@uop.edu ================================================================================ Archive-Date: Fri, 06 Aug 1993 08:22:18 PDT Date: Fri, 06 Aug 1993 08:22:18 PDT From: Ed Bates Reply-To: ebates@uop.edu To: dwing@uh01.colorado.edu CC: INFO-MULTINET@TGV.COM Subject: Re: signature trailers On Thu, 5 Aug 1993 at 23:10:43 GMT, dwing@uh01.colorado.edu (Dan Wing) wrote: >In article <930805103517.248002c2@CADRE1.CALTECH.EDU>, JDM@CADRE1.CALTECH.EDU >writes: > >>Is there a way to get VMSMAIL to append a signature to every message as >>does various internet mail packages? > >Nope. You can, however, tell VMSmail to use a .COM file when editing, and >this .COM file can append a signature before giving the edited mail message >back to VMSmail. For completeness, someone recently mentioned that you could use TPU to do something similar. Here's another option. I have a different sig for local mail than for non-local mail. To accomplish this, I did the following: - Create a file for each sig. - Define a logical for each file, and add these definitions to LOGIN.COM. I used SIGL and SIGO for my logical names. - At the end of each mail message, manually insert the appropriate sig. (In EDT, " INC logical", in EVE, "READ logical".) Why do I do this? I don't want to send an Internet sig to local users, and visa-versa. Also, it's easy to edit the sigs. (In or out of MAIL, "EDIT logical".) -- Ed - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Edwin J. (Ed) Bates VAX Admin/Internet Mgr/ Senior Academic Consultant Macintosh Consultant . . _ _ Office of Computing Services Internet: ebates@uop.edu |#| / \ | \ University of the Pacific AppleLink: U1441 |#| | | |_/ 877 W. Stadium Dr. Telephone: (209) 946-2251 |#| | | | Stockton, CA 95211 Fax: (209) 946-2898 \_/ \_/ | X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 6 Aug 1993 08:28:46 PDT Date: Fri, 6 Aug 93 08:28:46 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Configurability of the IP TTL value? To: iversen @ vsfys1.fi.uib.no cc: info-multinet @ TGV.COM > In article <23rrq5$3gl@lll-winken.llnl.gov>, oberman@ptavv.llnl.gov writes: > >In Article <744566400.265336.VANCE@tgv.com> > >VANCE@TGV.COM (L. Stuart Vance) writes: > > > >>The current MultiNet default is 60 hops. If anyone actually has a network > >>currently with more hops than that, not only will I tell you how to fix the > >>problem, I'll also treat you to dinner the next time I see you so that you can > >>tell me how you built and managed such a network. Although, I guess that I > >>should be careful with my offer. I have actually seen 45 hops on the > >>Internet.... > > > >Someone should point out to DEC that the official recommended TTL default is > >64. It was increased from 30 years ago. 64 was recommneded in RFC1340 which is > >dated July, 1992, so UCX is not meeting the standard. (Wow! I'm shocked!) :-) > >But, in this case, neither is MultiNet. (OK. I'm picking nits.) > > > >R. Kevin Oberman Lawrence Livermore National Laboratory > >Internet: koberman@llnl.gov (510) 422-6955 > In the group comp.protocols.tcp-ip there was recently a pointer to a tool > "netconfig" that can print out and also modify a number of tcp/ip parameters > for a number of popular Unix systems. For my DECstations running Ultrix 4.2 > I got: > dsfys1.fi.uib.no> netconfig > IP: > Routing (ipforwarding): on > ICMP: > Redirects (ipsendredirects): on > Network Unreachables (ipgateway): off > TCP: > Receive Checksum (tcpcksum): on Yuck! This utility should be modified to NEVER let someone turn this off. > Output Buffer Size (tcp_sendspace): 16384 > Input Buffer Size (tcp_recvspace): 16384 > UDP: > Checksum (udpcksum): on > Max Sndbuf and Msg Size (udp_sendspace): 9000 > Max Rcvbuf and Msg Size (udp_recvspace): 9000 > TTL (udp_ttl): 30 > It seems that it is not only UCX that has a TTL of 30 ... > Fortunately netconfig lets me modify this immediately to 64, and I increased my > in/output buffersizes too. You may want to check your sources. The 4.3BSD tahoe releases and earlier had the TCP TTL constant hardcoded with a #define of 30. We modified this in MultiNet to put it into a "cell" which was easily patched (called tcp_ttl). Changing udp_ttl isn't going to affect the TCP TTL, so, depending on your implementation, netconfig may not let you change the TCP TTL. On raising the buffer sizes -- For UDP this is a no-brainer with little cost. The real meaning of the sendspace is the maximum size of a UDP datagram you can transmit; typically applications expecting to send more than 4K will do with own setsockopt() to raise this. For TCP raising these affects the flow control window. If you have a very big flow control window interactive logins like TELNET and RLOGIN can get real obnoxious because so much data get buffered. Then you need to consider a BSD bug -- in tcp_output.c there is some code like: /* * Compare available window to amount of window * known to peer (as advertised window less * next expected input). If the difference is at least two * max size segments or at least 35% of the maximum possible * window, then want to send a window update to peer. */ if (win > 0) { int adv = win - (tp->rcv_adv - tp->rcv_nxt); if (so->so_rcv.sb_cc == 0 && adv >= 2 * tp->t_maxseg) goto send; if (!so->so_rcv.sb_hiwat) goto send; /* Fix divide by zero if (100 * adv / so->so_rcv.sb_hiwat >= 35) goto send; If you use setsockopt() to LOWER the TCP window size after the connection is established then the calculation of "adv" could go negative, causing (100 * adv / so->so_rcv.sb_hiwat) to be a very small value, never >= 35. (This is fixed in MultiNet). Lowering the window size could be done accidentally by something that thinks it is "raising" it from 4K to 8K rather than lowering it from 16K to 8K. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 10:29:15 PDT From: fuzzface@drycas.club.cc.cmu.edu (John McMahon) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: RCP ================================================================================ Archive-Date: xxx, 6 Aug 1993 12:21 EDT Date: 6 Aug 1993 12:21 EDT Organization: Carnegie Mellon University Computer Club Lines: 35 Distribution: world NNTP-Posting-Host: drycas.club.cc.cmu.edu News-Software: VAX/VMS VNEWS 1.41 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <9308060911.AA23922@mailhub.ggr.co.uk>, mcj2412@ggr.co.uk writes... >Hi. I've looked through the Multinet manuals (maybe not in the right >places mind you) and don't find much on RCP. I did find the VMS HELP. MultiNet Users' Guide V3.2B - Chapter 4 >1. RCP uses .RHOSTS just like RLOGIN and RSHELL. Yes. >2. RCP fails if it gets any unexpected output from the remote end, ie > because the SYS$SYLOGIN procedure on the target machine in my case > outputs some 'nice' user information. Yes. >ps. As my manager's just given a highly political priority zero task > maybe someone can help. Does anyone know of a TCP/IP product that > allows you to watch someone elses terminal session? I think there > was a VMS product called Control or something. Any information > will be gratefully received. There are both public domain products and commercial products that run on VMS for monitoring other terminal sessions on the same VMS system. Control is one of these products and it is available from RAXCO. Check the DECUS library for the public domain tools. I'd suggest that before you buy/install one of these things that you have a firm written policy on terminal monitoring. Avoid lawsuits later :-) John 'FuzzFace\Fast-Eddie' McMahon fuzzface@drycas.club.cc.cmu.edu {use finger to get my PGP public key} "Any networking problem can be solved by the application of 75 miles of Unshielded Twisted Pair (UTP)" - InterOp Spring 1993 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 10:54:30 PDT From: gkn@nic.cerf.net (Gerard K. Newman) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Slip problems ================================================================================ Archive-Date: xxx, 6 Aug 1993 17:33:07 GMT Date: 6 Aug 1993 17:33:07 GMT Organization: TGV, Inc. Lines: 34 Distribution: world NNTP-Posting-Host: nic.cerf.net Summary: Some sort of modem trouble Keywords: SLIP, flow control, modems To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <5AUG199305355507@apsicc.aps.edu> nmmnh@apsicc.aps.edu (Not a bug, a feature...) writes: >We are having problems getting a SLIP connection to worrk >properly and need some help. We are using a subnet mask of >255.255.255.192. The dynamic slip line configuration looks >like this: > >VAX1 (IP 192.133.100.65) ==> SLIP port (192.133.100.130) >to modem >to modem >SLIP port (192.133.100.129) ==> VAX2 (IP 192.133.100.5) > >The modems stay connected, but data does not flow through. On >a ping to the remote site the modem sends and indicates it >is receiving data. All the routing tables are set up properly. >When the phone line drops I get an error message that no route >is available to the remote host. With the phone connection Pings >and other attempts to share data fail, with no explanation. As we discussed yesterday, I believe that the modems are not passing all of the data (i.e., are perhaps configured for flow control or data compression which would result in their not being able to pass all data values thru). To get any further will probably require putting some sort of communications analyzer on the line; the packets are flowing out of the machines appropriately, but are not being passed up to the other side. Gerard K. Newman | TGV, Inc. 101 Cooper St. | 800.848.3440 (voice) gkn@tgv.com | Santa Cruz, CA 95060-4526 | 408.427.4365 (fax) -- Gerard K. Newman, TGV, Inc. 603 Mission St. Santa Cruz, CA 95060 gkn@tgv.com | 619.591.4717 (voice+fax) | 800.TGV.3440 | 408.427.4366 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 14:43:44 PDT From: kcb@post.its.mcw.edu (Kent C. Brodie) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: VMS backup of a DEC Ultrix system via MultiNet ================================================================================ Archive-Date: xxx, 6 Aug 1993 16:02:37 -0500 Date: 6 Aug 1993 16:02:37 -0500 Organization: Medical College of Wisconsin; Milwaukee Wisconsin Lines: 76 NNTP-Posting-Host: post.its.mcw.edu X-Newsreader: TIN [version 1.2 PL1] To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET W. Todd Wipke (wipke@SECS.UCSC.EDU) wrote: : -------------------- ------------------- : |DEC 5000/240 | | Vax 4000/500 | : | no tape |-------------------------------| TF85 tape | : -------------------- TCPIP | MultiNet | : ------------------- : I would especially like to hear from someone who does this regularly and has : had to use it for recovery purposes. There are too many variables to work : out by experimenting. Well, we use the "rdump" facility as Ken suggested. It works like a charm. We haven't used it for RECOVERY yet, but the "rrestore" function should work as easily. Note: thank god SOMEONE realized "r" utilities are useful. Too bad dec didn't put it in THEIR product. sigh. -another satisfied TGV customer...(!) The following dcl has these limitations: 1) it ALWAYS does a level 0 dump ("full backup"). 2) it depends heavily on ".rhosts" entries being in place. Specifically, the ".rhosts" files are kinda like proxies. Read the docs on "r" command use and .rhosts... 3) This procedure was designed by me, for my *VMS* operators, to backup my ultrix systems (which have no tape drive). $! check for specific user, otherwise the 'rhosts' entries fail $ if f$edit(f$getjpi("","USERNAME"),"TRIM") .nes. "OPERATORB" $ then $ say "This procedure MUST be run from the OPERATORB account" $ exit $ endif $ type sys$input This backup procedure will backup 4 different file systems from the clinical unix systems. *4* tape cartridges are needed in drive MIA0:. Put them in the following order: UNIX1 "/" UNIX1 "/USR" UNIX2 "/" UNIX2 "/USR" $ inquire ans "Press RETURN when the tapes are loaded..." $! $ write sys$output "Dumping unix1 root filesystem...",f$time() $ rshell /user=root unix1.fps.mcw.edu - rdump 0nouf vaxwel.fps.mcw.edu:mia0: / $! $ write sys$output "Dumping unix1 usr filesystem...",f$time() $ rshell /user=root unix1.fps.mcw.edu - rdump 0nouf vaxwel.fps.mcw.edu:mia0: /usr $! $ write sys$output "Dumping unix2 root filesystem...",f$time() $ rshell /user=root unix2.fps.mcw.edu - rdump 0nouf vaxwel.fps.mcw.edu:mia0: / $! $ write sys$output "Dumping unix2 usr filesystem...",f$time() $ rshell /user=root unix2.fps.mcw.edu - rdump 0nouf vaxwel.fps.mcw.edu:mia0: /usr $ exit -- -------------- Kent C. Brodie - Sr. Systems/Network Manager brodie@post.its.mcw.edu Information Technology Systems Medical College of Wisconsin (414) 257-8769 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 17:04:28 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: pse525@siemens.co.at (Siemens Oesterreich PSE525) Subject: Public domain TCP/IP Message-ID: <1993Aug6.152421.4035@siemens.co.at> ================================================================================ Archive-Date: Fri, 6 Aug 1993 15:24:21 GMT Date: Fri, 6 Aug 1993 15:24:21 GMT Organization: SIEMENS Austria Corp., A-1100 Wien, Gudrunstr. 11 X-Newsreader: TIN [version 1.1 PL8] Lines: 8 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Can anybody answer my following questions: -- Is CMUIP the only public domain TCP/IP implementation for VAX/VMS? -- The other way round: What is UCX and Multinet? -- What is the average price of a commercial TCP/IP (for a MVAX 3100/30) Thanks in advance Erik Margraf / Siemens Austria erik@asgard.gud.siemens.co.at X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 6 Aug 93 17:18:22 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: signature trailers Message-ID: <6AUG199316415120@eql.caltech.edu> From: rankin@eql.caltech.edu (Pat Rankin) ================================================================================ Archive-Date: xxx, 6 Aug 1993 16:41 PDT Date: 6 Aug 1993 16:41 PDT Distribution: world Organization: California Institute of Technology NNTP-Posting-Host: eql.caltech.edu News-Software: VAX/VMS VNEWS 1.41 Lines: 8 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET >[...] Then define the *unsupported* > logical MAIL$EDIT to point to your newly-modified MAILEDIT.COM, [...] MAIL$EDIT is fully documented in section 6.11.2 and appendix C.7 of the V6.0 _OpenVMS_User's_Manual_. Lack of such documentation in the past was probably due to an oversight rather than non-support. Pat Rankin, rankin@eql.caltech.edu X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 6 Aug 1993 21:58:33 PDT Date: Fri, 6 Aug 93 21:58:33 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: VMS backup of a DEC Ultrix system via MultiNet To: kcb @ post.its.mcw.edu cc: info-multinet @ TGV.COM > W. Todd Wipke (wipke@SECS.UCSC.EDU) wrote: > : -------------------- ------------------- > : |DEC 5000/240 | | Vax 4000/500 | > : | no tape |-------------------------------| TF85 tape | > : -------------------- TCPIP | MultiNet | > : ------------------- > : I would especially like to hear from someone who does this regularly and has > : had to use it for recovery purposes. There are too many variables to work > : out by experimenting. > Well, we use the "rdump" facility as Ken suggested. It works like a charm. > We haven't used it for RECOVERY yet, but the "rrestore" function should > work as easily. Please please please always try a rrestore (at least an rrestore tv to list the files) on a tape (and if you put multiple savesets on a tape, try listing each one in succession) before counting on rdump/rrestore. We've found some implementations of other people's clients which don't speak the RMT protocols correct and I'm worried that you may be able to dump but not restore (although in these cases you should have a valid dump format tape, just no way of reading it). You should do this with ANY backup scheme. I can't recall whose implementations offhand, so I won't attempt to name names, but it was some of the more obscure UNIX boxes. Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 6 Aug 1993 22:20:43 PDT Date: Fri, 6 Aug 93 22:20:43 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Public domain TCP/IP To: pse525 @ siemens.co.at cc: info-multinet @ TGV.COM > Can anybody answer my following questions: > -- Is CMUIP the only public domain TCP/IP implementation for VAX/VMS? Certainly the only public domain TCP/IP implementation which is worth considering. There may be some others (I think Excelan may have put their version in the public domain years ago, but it is not generally worth considering). > -- The other way round: What is UCX and Multinet? UCX (aka DEC TCP/IP Services for OpenVMS) is the TCP/IP package that DEC sells. MultiNet is the TCP/IP package that we (TGV) sell. I'll let others expand on the differences. > -- What is the average price of a commercial TCP/IP (for a MVAX 3100/30) You can get pricing for our product by calling us at +1 408 427 4366 (800-TGV-3440), or in Austria by contacting our distributor: Tekekec Airtronic Gmgh Kapuzinerstrasse 9 8000 Munich 2 GERMANY 49 (89) 5164 0 49 (89) 5164 110 FAX Contact: Josef Baumler Ken X-ST-Status: N Return-path: Received: from post.its.mcw.edu ([141.106.32.10]) by TGV.COM via INTERNET ; Sat, 7 Aug 93 20:00:37 PDT Received: by post.its.mcw.edu (5.65/MCW-ITS-1.6) id AA09425; Sat, 7 Aug 1993 22:00:52 -0500 ================================================================================ Archive-Date: Sat, 7 Aug 1993 21:57:52 -0500 Date: Sat, 7 Aug 1993 21:57:52 -0500 (CDT) From: "Kent C. Brodie" Subject: Re: VMS backup of a DEC Ultrix system via MultiNet To: Kenneth Adelman Cc: info-multinet@TGV.COM Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Fri, 6 Aug 1993, Kenneth Adelman wrote: > > Well, we use the "rdump" facility as Ken suggested. It works like a charm. > > We haven't used it for RECOVERY yet, but the "rrestore" function should > > work as easily. > > Please please please always try a rrestore (at least an rrestore > tv to list the files) on a tape (and if you put multiple savesets on a > tape,.... I should have qualified the above posting. We first started using the above "rdump" *on the same day I did that post*. i.e., it's ALL brand new. heh. I was planning on testing the restore next week. I will of course let you know how it works...! (fyi: decstatoin 5000-240, Ultrix 4.2) > You should do this with ANY backup scheme. I usually test my backups when they're needed. :-) I like to live dangerously.... (just last week we restored a full and *8* mumps-based incrementals. talk about SCARY... :-) In all seriousness, the testing WAS on my list. stay tuned! -------------- Kent C. Brodie - Sr. Systems/Network Manager brodie@post.its.mcw.edu Information Technology Systems Medical College of Wisconsin (414) 257-8769 X-ST-Status: N Return-path: Received: from ACF7.NYU.EDU ([128.122.128.17]) by TGV.COM via INTERNET ; Sat, 7 Aug 93 23:18:19 PDT Received: from ACFcluster.NYU.EDU by ACFcluster.NYU.EDU (PMDF V4.2-13 #2884) id <01H1HFE9EJVKVRPS62@ACFcluster.NYU.EDU>; Sun, 8 Aug 1993 01:56:05 EST ================================================================================ Archive-Date: Sun, 08 Aug 1993 01:55:36 -0500 Date: Sun, 08 Aug 1993 01:55:36 -0500 (EST) From: Stephen Tihor Subject: Re: RCP To: info-multinet@tgv.com MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII I must conconcur onthe written policy note. And also point out the supervisor series avaialble from Hunter goatley's fileserve and its shadows like Terry's ftp server. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 8 Aug 93 10:38:46 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: dwing@uh01.colorado.edu (Dan Wing) Subject: Re: FTP site for patch to MAILSHR Message-ID: <1993Aug8.170509.8295@colorado.edu> Sender: news@colorado.edu (The Daily Planet) Nntp-Posting-Host: buckie.hsc.colorado.edu Reply-To: dwing@uh01.colorado.edu Organization: University of Colorado Hospital Authority, Denver ================================================================================ Archive-Date: Sun, 8 Aug 1993 17:05:09 GMT Date: Sun, 8 Aug 1993 17:05:09 GMT Lines: 13 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930806100922.6fa@jade.bucknell.edu>, LYNCH@jade.bucknell.edu writes: >Could someone please send me the address of an FTP site that >has the patch to mailshr.exe that allows users to enter addresses >without specifying SMTP%""? I am running Multinet version 3.2B >and VMS version 5.5-2 on a VAX 4400. Check ftp.spc.edu or fileserv@wkuvx1.bitnet. -dan -Dan Wing, Systems Administrator, University Hospital, Denver dwing@uh01.colorado.edu or wing_d@ucolmcc.bitnet X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 8 Aug 93 17:16:23 PDT From: munroe@dmc.com (Dick Munroe) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Still trying for SLIP connectivity. Message-ID: <1993Aug8.194233.2059@dmc.com> ================================================================================ Archive-Date: xxx, 8 Aug 1993 19:42:33 EDT Date: 8 Aug 93 19:42:33 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 34 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Still trying to get a SLIP link up, I'm getting closer, but it looks as if output from my system goes into the bit bucket. I got the SL device defined: $ mu sho/interface sl1 Device sl1: flags=4071 IP Address = 140.186.89.2 IP Sub-Net Mask = 255.255.0.0 IP Point to Point Destination = 192.9.200.34 This was taken after the caller had logged in and established the connection. I'm suspicious of the sub-net mask. I took the default when I defined the line, shouldn't it be 255.255.255.0? In any event, to see if I could see any traffic, I fired up TCPDUMP and had the caller ping me. This is what I got: $ mu tcpdump/interface=sl1 19:22:25.91 192.9.200.34 > 140.86.88.1: 192.9.200.34 > 140.86.88.1: icmp: echo request 19:23:58.47 192.9.200.34 > 140.86.88.1: 192.9.200.34 > 140.86.88.1: icmp: echo request Since the output format for TCPDUMP doesn't appear to be documented anywhere in the Multinet document set I'm not too sure what its trying to tell me, but I THINK its saying that it got a packet from 192.9.200.34 for 140.86.88.1. All well and good, but with tracing on at the other end, it doesn't look as if the echo response is getting out the SL1 interface. What should I do to debug this further??? -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: Received: from BIO.EMBNET.SE ([130.238.39.6]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 01:28:06 PDT ================================================================================ Archive-Date: Mon, 9 Aug 1993 10:28:03 +0200 Date: Mon, 9 Aug 1993 10:28:03 +0200 (MET-DST) From: PETER@bmc.uu.se (Peter Gad, Computer Dept@BMC, Uppsala) Subject: RE: signature trailers To: info-multinet@tgv.com X-Vmsmail-To: SMTP%"info-multinet@tgv.com" Hi, for those of you thast prefer to use EDT as the maileditor of choice here's a neat way of getting signatures included in your mail. Two files are included below, extract those as MAILEDIT.COM (place it somewhere convenient, like SYS$MANAGER) and EDTINI.MAIL (also placed somewhere convenient; NOTE: this is a modified version of the wellknown(??) EDTINI.EDT) dont forget 'SET PROTECTION=(W:RE)' on both these files, then do (either in SYSTARTUP, SYLOGIN.COM or in LOGIN.COM) $ define MAIL$EDIT SYS$MANAGER:MAILEDIT.COM ! make it 'define/sys' if you use SYSTARTUP) Then when you do either SEND/EDIT or SEND/EXTRACT you will get your signature included at the end of your message. Kindly note that the MAILEDIT.COM assumes the EDTINI.MAIL to be placed in sys$specific:[sysmgr], if you want it somewhere else you'll have to change the directoryname accordingly. Peter Gad (Sys($)Manager) ! peter@bio.embnet.SE Computing Dept., ! peter@bmc.uu.se Biomedical Center ! Voice: +46 18 174016 University of Uppsala, Sweden ! Fax : +46 18 551759 "enbart omedelbart medelbara meddelanden meddelas omedelbart" =========== MAILEDIT.COM =============== $ ! $ ! Command procedure to invoke an editor for MAIL. $ ! $ ! Inputs: $ ! $ ! P1 = Input file name. $ ! P2 = Output file name. $ ! $ ! If MAIL$EDIT is undefined, MAIL will invoke callable EDT. $ ! If MAIL$EDITis defined to be a command procedure, $ ! MAIL will create a subprocess to edit the mail. $ ! $ ! Note that this procedure is run in the context of a subprocess. $ ! LOGIN.COM is not executed. However, all process logical names $ ! and DCL global symbols are copied. $ ! $ ! The default directory is the same as the parent process $ ! $ DEFINE /USER SYS$INPUT 'F$TRNLNM("SYS$OUTPUT")' $ IF P1 .EQS. "" THEN GOTO NOINPUT $ ED = "" ! if you have an 'ED == "EDIT/EDT..." defined already ! $ EDIT/EDT/COMMAND=sys$specific:[sysmgr]EDTINI.MAIL/OUTPUT='P2' 'P1' $ EXIT $NOINPUT: $ EDIT/EDT/COMMAND=sys$specific:[sysmgr]EDTINI.MAIL 'P2 $ EXIT ================== EDTINI.MAIL ====================== ! Set-up: ! ! Suppress the bell or buzzer that signals errors SET QUIET ! Define delimiters for the word entity SET ENTITY WORD ' .,?!;:[]()<>*-+=/\' ! Define delimiters for the sentence entity SET ENTITY SENTENCE '. ?!' ! Suppress line numbers in line mode SET NONUMBERS ! Wrap full words to the next line at 78 characters SET WRAP 78 ! ! Key Definitions: ! ! Change to the specified buffer DEFINE KEY GOLD B AS "EXT CHANGE =?'CHANGE TO BUFFER: '." ! Change to the last current line in the previous buffer DEFINE KEY GOLD C AS "EXT CHANGE LAST." ! Change the case of the current word DEFINE KEY GOLD D AS "+CHGCW." ! Locate the last current line in the MAIN buffer DEFINE KEY GOLD G AS "EXT FIND=MAIN.." ! Prompt for a buffer name and then locate the last current line ! in the specified buffer DEFINE KEY GOLD H AS "EXT FIND=?*'BUFFER: '.." ! Exit file DEFINE KEY GOLD E AS "EXT EXIT." ! Fill the text of the current paragraph to the right margin DEFINE KEY GOLD P AS "FILLPAR." ! Quit file DEFINE KEY GOLD Q AS "EXT QUIT." ! Shift screen left 1 tab (8 columns) DEFINE KEY GOLD L AS "SHL." ! Shift screen right 1 tab (8 columns) DEFINE KEY GOLD R AS "SHR." ! Replace all occurrences of the specified string in the ! current buffer with the specified string DEFINE KEY GOLD S AS "EXT S/?*'REPLACE: '/?*' WITH: '/WHOLE." ! Transpose the two characters to the left of the cursor DEFINE KEY GOLD T AS "BACK C DC ADV C UNDC." ! Write the current buffer to the specified file DEFINE KEY GOLD W AS "EXT WRITE ?*'WRITE TO FILE: '." ! Include the specified file at the current line DEFINE KEY GOLD X AS "EXT INCLUDE ?*'INCLUDE FILE: '." ! Include the specified file at the beginning of ! the specified buffer DEFINE KEY GOLD Y AS "EXT INCLUDE ?*'INCLUDE FILE: ' =?*' BUFFER: ';FIND BEGIN." ! INCLUDE sys$login:MAIL.SIGNATURE =MAIN END FIND BEG ! ! Select editing mode: SET MODE CHANGE X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 04:55:38 PDT X-Newsgroups: comp.os.vms,vmsnet.networks.tcp-ip.multinet From: chris@csvax1.ucc.ie (Chris Higgins - System Administrator) Subject: RWHO / RUPTIME Daemon for VMS/MULTINET Message-ID: <1993Aug09.090346.18293@donau.et.tudelft.nl> Sender: news@donau.et.tudelft.nl (UseNet News System) Nntp-Posting-Host: csvax1.ucc.ie Reply-To: chris@csvax1.ucc.ie Organization: Computer Science Dept. University College Cork, Ireland. ================================================================================ Archive-Date: Mon, 09 Aug 1993 09:03:46 GMT Date: Mon, 09 Aug 1993 09:03:46 GMT Lines: 21 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I hope this goes through, because our news feed is VERY unstable at the mo. I'm looking for a RWHO Daemon for VMS / Multinet. I want to be able to do a/ rwho b/ ruptime from a local Ultrix LAN, and have the VMS machines appear. (Multinet 3.2b / VMS 5.5-2 ) Any suggestions would be greatly appreciated. Thanks. Chris. + J.C. Higgins, + + If you love something, set it + + VMS Sys. Admin, + Chris@csvax1.ucc.ie + free. If it doesn't come back + + Comp.Sc.Dept. + Chris@odyssey.ucc.ie + to you, hunt it down and + + UCC, Ireland + C.Higgins@bureau.ucc.ie + KILL it. -- Me. + X-ST-Status: N Return-path: Received: from gimli.bio.purdue.edu ([128.210.18.10]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 06:28:24 PDT Received: from localhost by gimli.bio.purdue.edu with SMTP id AA13843 (5.65b/IDA-1.4.3 for info-multinet@TGV.COM); Mon, 9 Aug 93 08:28:21 -0500 Message-Id: <9308091328.AA13843@gimli.bio.purdue.edu> To: Adelman@TGV.COM (Kenneth Adelman) Cc: info-multinet@TGV.COM Subject: Re: VMS backup of a DEC Ultrix system via MultiNet In-Reply-To: Your message of Fri, 06 Aug 93 21:58:33 -0700. <930806215833.20800aeb@TGV.COM> ================================================================================ Archive-Date: Mon, 09 Aug 1993 08:28:20 -0500 Date: Mon, 09 Aug 93 08:28:20 -0500 From: mckay@gimli.bio.purdue.edu X-Mts: smtp ---- Excerpts from your message of August 6, 1993 ---- Please please please always try a rrestore (at least an rrestore tv to list the files) on a tape (and if you put multiple savesets on a tape, try listing each one in succession) before counting on rdump/rrestore. We've found some implementations of other people's clients which don't speak the RMT protocols correct and I'm worried that you may be able to dump but not restore (although in these cases you should have a valid dump format tape, just no way of reading it). ---- End of message ---- I can save you a little time. I've tried the following and they work for me with Multinet's RMT: rdd (part of the rtape package from comp.sources.unix) rtar (part of the same package) rrestore (SunOS 4.1.2) rrestore (Ultrix 4.2a) rrestore (ES/os 2.3 (Evans & Sutherland ESV workstation)) You should use this information as one test case only. Ken is quite right when he says you ought to test it yourself. --Dwight D. McKay, Purdue University, Department of Biological Sciences --Office: LILY B-145, Phone: (317) 494-4481 --mckay@gimli.bio.purdue.edu X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 10:49:33 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: brack@ccs.yorku.ca (Tony Brack) Subject: MultiNet & DEC POSIX Coexistence ... ? Message-ID: <1993Aug9.131657.1@ccs.yorku.ca> Lines: 32 Sender: news@newshub.ccs.yorku.ca (USENET News System) Organization: York University, Toronto ================================================================================ Archive-Date: Mon, 9 Aug 1993 18:16:57 GMT Date: Mon, 9 Aug 1993 18:16:57 GMT To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Greetings ... I am attempting to fire-up DEC POSIX on a MicroVAX 3100 that is running the current release of MultiNet, and am reluctant to "just give it a stab". POSIX itself tries to bring up UCX when running on a node that is using that product (I presume for NFS). I do not currently start MultiNet's NFS on this node, but could (it IS licensed)... in other words, I am running base MultiNet 3.2C and MultiWare. When I try to fire up POSIX, I get the message: Full UCX must be started before POSIX, POSIX startup aborted. Has anybody done this, and if so, what do I need to do? At some point I will be using MultiNet NFS client & server. How does this change things? Any help would be appreciated... Thanks in advance!!! -- Tony Brack ][ Usenet: ......!utzoo!yunexus!brack VMS & UNIX Technical Support ][ brack@ccs.yorku.ca Computing & Communications Services ][ Bitnet: brack@yulibra York University. ][ Voice: (416) 736-2100 x22687 4700 Keele St. || North York, Ontario M3J 1P3 || What's this coming up... ... sure hope it's friendly! X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 11:23:58 PDT From: terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr.) X-Newsgroups: comp.os.vms,vmsnet.networks.tcp-ip.multinet Subject: Re: RWHO / RUPTIME Daemon for VMS/MULTINET Message-ID: <1993Aug9.134728.6745@spcvxb.spc.edu> ================================================================================ Archive-Date: xxx, 9 Aug 1993 13:47:28 EDT Date: 9 Aug 93 13:47:28 EDT Organization: St. Peter's College, US Lines: 21 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug09.090346.18293@donau.et.tudelft.nl>, chris@csvax1.ucc.ie (Chris Higgins - System Administrator) writes: > I'm looking for a RWHO Daemon for VMS / Multinet. > I want to be able to do > a/ rwho > b/ ruptime > > from a local Ultrix LAN, and have the VMS machines appear. You can get my rwho/ruptime client and server from ftp.spc.edu in the [.ucx] subdirectory. It works under both UCX and MultiNet. In a cluster, each of the nodes broadcasts information, but only one (master) node keeps track of incoming broadcasts. It is easy to configure for a single interface. Unlike the Unix version, if you have multiple interfaces it doesn't determine this on the fly, so you have to make some coding changes. Feel free to write if you have any problems installing it. Terry Kennedy Operations Manager, Academic Computing terry@spcvxa.bitnet St. Peter's College, Jersey City, NJ USA terry@spcvxa.spc.edu +1 201 915 9381 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 12:56:16 PDT X-Newsgroups: comp.os.vms,vmsnet.networks.tcp-ip.multinet Subject: Re: RWHO / RUPTIME Daemon for VMS/MULTINET Message-ID: <2466t2INN10c1@rs1.rrz.Uni-Koeln.DE> From: moravec@ph1.Uni-Koeln.DE (Andreas Moravec) ================================================================================ Archive-Date: xxx, 9 Aug 1993 19:00:50 GMT Date: 9 Aug 1993 19:00:50 GMT Reply-To: moravec@ph1.Uni-Koeln.DE Organization: University of Cologne, I. Physik. Inst., Germany NNTP-Posting-Host: newton.ph1.uni-koeln.de Lines: 20 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug09.090346.18293@donau.et.tudelft.nl>, chris@csvax1.ucc.ie (Chris Higgins - System Administrator) writes: >I'm looking for a RWHO Daemon for VMS / Multinet. >I want to be able to do > a/ rwho > b/ ruptime rwho/ruptime client and server for UCX/MultiNet in ftp.spc.edu:[.ucx]rwho.bck or european mirror site: nic.switch.ch:mirror/vms/spc/ucx/rwho.bck -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ KOSMA = Koelner Observatorium fuer SubMillimeter-Astronomie Andreas Moravec TEL +49 (221) 470 4529 I. Physikalisches Institut FAX +49 (221) 470 5162 Universitaet zu Koeln TELEX 888 2291 UNIK D Zuelpicher Str. 77 INTERNET moravec@ph1.Uni-Koeln.DE 50937 Koeln, Germany X.400 S=Moravec;OU=Ph1;P=Uni-Koeln;A=D400;C=DE X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 13:52:48 PDT From: ronk@cascade.ens.tek.com (Ron Kirkpatrick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Better response to VRFY and EXPN commands. ================================================================================ Archive-Date: xxx, 9 Aug 1993 20:19:07 GMT Date: 9 Aug 1993 20:19:07 GMT Organization: Tektronix, Inc., Beaverton, OR Lines: 16 Distribution: world Reply-To: Ron.C.Kirkpatrick@tek.com NNTP-Posting-Host: cascade.ens.tek.com To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Has anyone modified USER_SMTP_DISPATCH.C to: 1. Return an error code of 550 instead of 250 in response to VRFY and EXPN commands if the user is flagged as DISMAIL and/or DISUSER? 2. Return the forwarding address in response to either the VRFY or EXPN commands? I would rather not reinvent the wheel. -- Thanks, Ron Kirkpatrick News Administrator/Postmaster Tektronix, Inc 503-627-6707 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 15:04:48 PDT From: lang@wharfin.mit.edu X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Gnus Help Needed ================================================================================ Archive-Date: xxx, 9 Aug 1993 17:40:06 EST Date: 9 Aug 93 17:40:06 EST Organization: Massachvsetts Institvte of Technology Lines: 18 NNTP-Posting-Host: wharfin.mit.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi, Has anyone that is a current user of Multinet been able to get GNUs to work on a VAX with Multinet? It looks as though it should be pretty easy to do, but so far I haven't been able to figure out exactly how to get it to work. I poked around a bit, but I couldn't find anything. One of the other systems types at my site said that he tried to get it to work, but that there was a feature of Multinet that did not seem to work...I don't know exactly what. Any help would be appreciated. rml ------------------------------------------------------------------------------- Bob Lang Room 11-221 | This space for rent. Consultant MIT Computer Services | Voice: (617)253-7438 FAX: (617)258-9535 | ------------------------------------------------------------------------------- X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 16:48:30 PDT From: system@dmc.com X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Still trying for SLIP connectivity. Message-ID: <1993Aug9.192524.2061@dmc.com> ================================================================================ Archive-Date: xxx, 9 Aug 1993 19:25:24 EDT Date: 9 Aug 93 19:25:24 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 74 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug8.194233.2059@dmc.com>, munroe@dmc.com (Dick Munroe) writes: > $ mu sho/interface sl1 > Device sl1: flags=4071 > IP Address = 140.186.89.2 > IP Sub-Net Mask = 255.255.0.0 > IP Point to Point Destination = 192.9.200.34 > Well, thanks to Ken I hope I will NEVER forget that IP numbers denote interfaces, not nodes. Given that, the caller can ping me, telnet, etc. Unfortunately, I don't seem to be able to get anything to happen the other way. The route information set up is: $ MultiNet Set/Route/Add=(Dest:140.186.89.2,Gateway:127.0.0.1) Which makes sense (this is a locally connected interface), but when I try to ping the node on the other end of the interface (192.9.200.34) it looks like the packets go out my default route: $ mu ping 192.9.200.34/n=5/verb PING 192.9.200.34 (192.9.200.34): 56 data bytes 36 bytes from enss.near.net (192.54.222.6): Destination Host Unreachable Vr HL TOS Len ID Flg off TTL Pro cks Src Dst Data 4 5 00 0054 199b 0 0000 fb 01 a7b7 140.186.88.1 140.186.88.1 36 bytes from enss.near.net (192.54.222.6): Destination Host Unreachable Vr HL TOS Len ID Flg off TTL Pro cks Src Dst Data 4 5 00 0054 1f9b 0 0000 fb 01 a1b7 140.186.88.1 140.186.88.1 36 bytes from enss.near.net (192.54.222.6): Destination Host Unreachable Vr HL TOS Len ID Flg off TTL Pro cks Src Dst Data 4 5 00 0054 249b 0 0000 fb 01 9cb7 140.186.88.1 140.186.88.1 36 bytes from enss.near.net (192.54.222.6): Destination Host Unreachable Vr HL TOS Len ID Flg off TTL Pro cks Src Dst Data 4 5 00 0054 299b 0 0000 fb 01 97b7 140.186.88.1 140.186.88.1 36 bytes from enss.near.net (192.54.222.6): Destination Host Unreachable Vr HL TOS Len ID Flg off TTL Pro cks Src Dst Data 4 5 00 0054 2d9b 0 0000 fb 01 93b7 140.186.88.1 140.186.88.1 A traceroute seems to confirm this: $ mu traceroute 192.9.200.34 traceroute to 192.9.200.34 (192.9.200.34), 30 hops max, 38 byte packets 1 CENTNET-GW.DMC.COM (140.186.88.6) 50 ms 40 ms 30 ms 2 CENTNET-GW.CENT.NET (140.186.1.1) 40 ms 30 ms 30 ms 3 bbn1-gw.near.net (131.192.52.1) 410 ms 200 ms 450 ms 4 mit2-gw.near.net (131.192.2.2) 510 ms 240 ms 60 ms 5 enss.near.net (192.54.222.6) 280 ms !H 380 ms !H 290 ms !H Yet the routing tables appear to be set up properly: $ mu sho/route MultiNet IP Routing tables: Destination Gateway Flags Refcnt Use Interface ------------ ------------ ----- ------ ----- --------- 140.186.89.1 LOCALHOST Up,Gateway,H 0 0 lo0 LOCALHOST LOCALHOST Up,Host 2 8012 lo0 140.186.89.2 LOCALHOST Up,Gateway,H 0 0 lo0 198.175.187.3 140.186.89.1 Up,Host 0 0 sl0 192.9.200.34 140.186.89.2 Up,Host 1 25 sl1 DEFAULT-GATEWAY 140.186.88.6 Up,Gateway 4 155068 se0 140.186.88.0 DMC.COM Up 0 7692 se0 Why isn't traffic being attempted through interface 140.186.89.2 (which, admittedly is down at the moment)? And how should I set up my DNS so that everybody will know that DMC.COM is reachable via not just my ethernet interfaces but all the slip interfaces as well. Is it as simple as a bunch of A records in the DNS? I really can't figure out what I need to do to get routing information broadcast to the rest of the world. I can't for the life of me figure out whether I need to run GATED or not and if so, how to get it configured properly. A few examples in the manual would have been a big help. Dick Munroe X-ST-Status: N Return-path: Received: from VESTA.SUNQUEST.COM ([149.138.1.42]) by TGV.COM via INTERNET ; Mon, 9 Aug 93 17:41:13 PDT ================================================================================ Archive-Date: Mon, 9 Aug 1993 17:37:06 -0700 Date: Mon, 9 Aug 1993 17:37:06 -0700 (MST) From: Ehud Gavron To: INFO-MULTINET@TGV.COM CC: GAVRON@VESTA.SUNQUEST.COM Subject: See Dick GET CONNECTED!!! Date sent: 9-AUG-1993 17:33:44 That whining consultant, Dick "Get Connected!!!" Munroe proves yet again that there's a reason some of us get paid to be network consultants. >Subj: Re: Still trying for SLIP connectivity. ... >> Device sl1: flags=4071 >> IP Address = 140.186.89.2 >> IP Sub-Net Mask = 255.255.0.0 >> IP Point to Point Destination = 192.9.200.34 >> >$ mu ping 192.9.200.34/n=5/verb ... >Destination Gateway Flags Refcnt Use Interface >------------ ------------ ----- ------ ----- --------- >140.186.89.1 LOCALHOST Up,Gateway,H 0 0 lo0 >LOCALHOST LOCALHOST Up,Host 2 8012 lo0 >140.186.89.2 LOCALHOST Up,Gateway,H 0 0 lo0 >198.175.187.3 140.186.89.1 Up,Host 0 0 sl0 >192.9.200.34 140.186.89.2 Up,Host 1 25 sl1 >DEFAULT-GATEWAY 140.186.88.6 Up,Gateway 4 155068 se0 >140.186.88.0 DMC.COM Up 0 7692 se0 > >Why isn't traffic being attempted through interface 140.186.89.2 (which, >admittedly is down at the moment)? Duh. You answered your own question. Next? >And how should I set up my DNS so that everybody will know that DMC.COM is >reachable via not just my ethernet interfaces but all the slip interfaces as >well. Is it as simple as a bunch of A records in the DNS? Duh. You answered that one correctly too. Next? >I really can't figure out what I need to do to get routing information >broadcast to the rest of the world. I can't for the life of me figure out >whether I need to run GATED or not and if so, how to get it configured >properly. A few examples in the manual would have been a big help. Duh. Ask your upstream provider what THEY use for routing info. Static routes, RIP, IGRP, BGP, or whatever. >Dick Munroe GET CONNECTED!!! Hire real network consultants who can solve your connectivity problems without filling up mailing lists!!! Only $19.94. Call 1-900-GET-SLIP. Children ask your parents before calling. Dick, you too. Ehud -- Ehud Gavron (EG76) gavron@vesta.sunquest.com X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 10 Aug 93 05:06:49 PDT From: munroe@dmc.com (Dick Munroe) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Still trying for SLIP connectivity. Message-ID: <1993Aug10.070713.2063@dmc.com> ================================================================================ Archive-Date: xxx, 10 Aug 1993 07:07:13 EDT Date: 10 Aug 93 07:07:13 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 17 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug9.192524.2061@dmc.com>, system@dmc.com writes: > In article <1993Aug8.194233.2059@dmc.com>, munroe@dmc.com (Dick Munroe) writes: >> $ mu sho/interface sl1 >> Device sl1: flags=4071 >> IP Address = 140.186.89.2 >> IP Sub-Net Mask = 255.255.0.0 >> IP Point to Point Destination = 192.9.200.34 It WAS the subnet mask that was the problem. 255.255.255.0 and the routing works just fine. -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 10 Aug 93 15:27:02 PDT From: munroe@dmc.com (Dick Munroe) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Problems with stopping SL interface devices. Message-ID: <1993Aug10.163811.2066@dmc.com> ================================================================================ Archive-Date: xxx, 10 Aug 1993 16:38:11 EDT Date: 10 Aug 93 16:38:11 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 36 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Further issues in the attempts to get SLIP up and running. I accidentally attempted to stop a slip interface while someone was still using it. The interface DIDN'T stop and I got the following: $ multinet set/interface SL1/link=slip/vms_device=_LTA5158:/down %MULTINET-F-SETPORT, unable to connect VMS device _LTA5158: to MultiNet kernel for sl1 -MULTINET-F-EBUSY, Mount device busy The reference count for the device is hosed: Terminal LTA5158:, device type unknown, is offline, record-oriented device, carriage control, network device. Error count 0 Operations completed 2 Owner process "" Owner UIC [SYSTEM] Owner process ID 20A00DA7 Dev Prot Reference count 1 Default buffer size 80 And the owner process id is bogus: $ stop/id=20A00DA7 %SYSTEM-W-NONEXPR, nonexistent process Somehow this doesn't seem right (I'm running Multinet 3.2B, VMS 5.5-2). In any event, now I can't get the interface stopped at all. Any suggestions (other than rebooting?) -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 10 Aug 93 15:27:03 PDT From: munroe@dmc.com (Dick Munroe) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Still trying for SLIP connectivity. Message-ID: <1993Aug10.171545.2067@dmc.com> ================================================================================ Archive-Date: xxx, 10 Aug 1993 17:15:44 EDT Date: 10 Aug 93 17:15:44 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 24 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug10.070713.2063@dmc.com>, munroe@dmc.com (Dick Munroe) writes: >>> $ mu sho/interface sl1 >>> Device sl1: flags=4071 >>> IP Address = 140.186.89.2 >>> IP Sub-Net Mask = 255.255.0.0 >>> IP Point to Point Destination = 192.9.200.34 > > It WAS the subnet mask that was the problem. 255.255.255.0 and the routing > works just fine. At least between 192.9.200.34 and 140.186.89.2. When 192.9.200.34 attempts to send traffic (ping for example) "out" of my system through 140.186.89.2 and then, supposedly, out through my default route to the rest of the world (140.186.88.1) nothing happens. As I've said, I'm fuzzy on exactly how routing works, so I've likely gotten something screwed up, but I haven't a clue what. Any suggestions? -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 10 Aug 1993 16:06:02 PDT Date: Tue, 10 Aug 93 16:06:02 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Problems with stopping SL interface devices. To: munroe @ dmc.com cc: info-multinet @ TGV.COM > Further issues in the attempts to get SLIP up and running. I accidentally > attempted to stop a slip interface while someone was still using it. The > interface DIDN'T stop and I got the following: > $ multinet set/interface SL1/link=slip/vms_device=_LTA5158:/down > %MULTINET-F-SETPORT, unable to connect VMS device _LTA5158: to MultiNet kernel > for sl1 > -MULTINET-F-EBUSY, Mount device busy > The reference count for the device is hosed: > Terminal LTA5158:, device type unknown, is offline, record-oriented device, > carriage control, network device. > Error count 0 Operations completed 2 > Owner process "" Owner UIC [SYSTEM] > Owner process ID 20A00DA7 Dev Prot > Reference count 1 Default buffer size 80 > And the owner process id is bogus: > $ stop/id=20A00DA7 > %SYSTEM-W-NONEXPR, nonexistent process > Somehow this doesn't seem right (I'm running Multinet 3.2B, VMS 5.5-2). > In any event, now I can't get the interface stopped at all. > Any suggestions (other than rebooting?) Check for OPCOM messages. If you don't get any, then the problem is that MultiNet has a transmit pending on the device and is waiting for the completion interrupt. If this condition persists, it sounds like the device is wedged and a reboot will probably be necessary to clear it. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 10 Aug 93 17:26:20 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: cornwaca@ldoc03.ldc.dupont.com ( ) Subject: Copying files and subdirectories with MultiNet Message-ID: <1993Aug10.194912.25952@es.dupont.com> Sender: news@es.dupont.com (USENET News System) Nntp-Posting-Host: ldoc01.lvs.dupont.com Reply-To: cornwaca@ldoc03.ldc.dupont.com Organization: DuPont (Opinions are those of the writer only) ================================================================================ Archive-Date: Tue, 10 Aug 1993 19:49:12 GMT Date: Tue, 10 Aug 1993 19:49:12 GMT Lines: 5 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Is there a way to copy all the files in a directory plus it's subdirectories using MGET or MPUT. I know a basic FTP will copy multiple files, I'm not sure about the best approach for a directory tree. Am I stuck with creating a VMS Backup saveset and moving it to the remote node (via IP or DECNET)? Thanks in advance. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 10 Aug 93 20:49:09 PDT From: francois@italia.boeing.com (Francois VanSteertegem) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Handy doo-dads for vms-mail (using EVE). Message-ID: <103041@bcsaic.boeing.com> ================================================================================ Archive-Date: xxx, 10 Aug 1993 14:34:49 GMT Date: 10 Aug 93 14:34:49 GMT Sender: francois@bcsaic.boeing.com Distribution: na Organization: Boeing Computer Services, Seattle Lines: 125 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Well, MultiNet it ain't. But everybody here is VMS, and I've seen these questions posted more than once on this thread. So here goes... I've often heard VMS-mail users ask "How do you get those ">" thingies at the beginning of every line when you reply to a message?" Another common question is "How can I get automatic-signature-block-appending?" I've put together a few lines of code to implement these functions and included them (and instructions for using them) in this posting. Here we go... 1) Place this line in your login.com: $ MAIL=="MAIL/EDIT=(SEND,REPLY=EXTRACT,FORWARD)" This will cause VMS-mail to invoke the default editor whenever you send, forward or reply to a message. It will also cause the replied-to message to be included in your reply. 2) Set your default mail-editor to EVE. Get into mail, and at the MAIL> prompt enter the command: SET EDIT TPU Now--when you send a message--you'll use the EVE editor to compose it. If you're an EDT fan (like me) don't worry; EVE can be made to act very much like EDT. 3) Create the file SYS$LOGIN:EVE$INIT.EVE and fill it with the code I've appended to the end of this posting. 4) Create the file SYS$LOGIN:TPU$COMMAND.TPU and fill it with the code I've appended to the end of this posting. And that's it. Now, when you go to send a mail message F11 will put a ">" at the beginning of each line in the message, and F12 will cause the contents of SYS$LOGIN:SIGNATURE.TXT to be appended to the end of the message. I haven't done extensive testing on this, but I'm pretty sure it'll run on any vanilla VMS system (v5.4-2 or later) Feel free to modify the code or send me critisms. Good Luck! Francois VanSteertegem 206-965-1384 (voice) 206-237-0052 (fax) francois@boeing.com or francois@bmt.ca.boeing.com !/\/\/\/\/\/\ This is the code for SYS$LOGIN:EVE$INIT.EVE \/\/\//\/\/\/\/\/\/\ ! This is the file sys$login:eve$init.eve The TPU commands in this file ! get executed each time you start EVE. (Unless you have the logical EVE$INIT ! define to point at some othe initialization file.) ! These lines make EVE look a lot like EDT. They are optional set key edt set curs bound set nowrap ! ! define keys to execute user-defined-TPU-procedures ! !The "stick a > in front of every line" procedure define key=f11 tpu user_mark_text ! define key=f12 tpu user_append_signature !/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ !/\/\/\/\/\/ This is the code for SYS$LOGIN:TPU$COMMAND.TPU /\/\/\/\/\/\/\/\/\ ! This is the file sys$login:tpu$command.tpu The procedures in this file ! are made available to your EVE editing session. (Unless you've defined ! the logical tpu$command to point to something else.) To execute these ! procedures enter the command TPU XXXX (where XXXX is the procedure-name) ! at the EVE "Command:" prompt. ! ! PROCEDURE user_mark_text ! ! This procedure will stick a ">" in front of every line in the current buffer. ! It leaves the cursor in its original position. !Local variables... LOCAL starting_location, numlines, i; !Make note of the current cursor position. starting_location:=MARK (NONE); !How many lines are in the buffer? numlines:=GET_INFO (CURRENT_BUFFER,"record_count"); !For every line in the buffer, go to the beginning of that line and !insert a ">". i:=0; LOOP EXITIF i>=numlines; i:=i+1; POSITION (i); POSITION (LINE_BEGIN); COPY_TEXT (">"); ENDLOOP; !Put the cursor back where we found it. POSITION (starting_location); ENDPROCEDURE; PROCEDURE user_append_signature ! This procedure appends the contents of sys$login:signature.txt to ! the end of the current buffer. !Go to the end of the buffer. POSITION (BUFFER_END); !And include the file. READ_FILE ("sys$login:signature.txt"); ENDPROCEDURE; !/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 10 Aug 1993 21:21:53 PDT Date: Tue, 10 Aug 93 21:21:53 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Copying files and subdirectories with MultiNet To: cornwaca @ ldoc03.ldc.dupont.com cc: info-multinet @ TGV.COM > Is there a way to copy all the files in a directory plus it's subdirectories using MGET or MPUT. I know a basic FTP will copy multiple files, I'm not sure > about the best approach for a directory tree. Am I stuck with creating a > VMS Backup saveset and moving it to the remote node (via IP or DECNET)? Nope. There is no system-independent way to transfer directory trees with MPUT/MGET, and we haven't added any VMS extensions to do this. I typically use BACKUP savesets. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 11 Aug 93 12:28:08 PDT From: bill@main.morris.org (Billy D'Augustine) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: IRC server? ================================================================================ Archive-Date: xxx, 11 Aug 1993 19:03:11 GMT Date: 11 Aug 1993 19:03:11 GMT Organization: Morris Automated Information Network Lines: 14 Reply-To: bill@main.morris.org NNTP-Posting-Host: main.morris.org To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I found a nice little IRC client, and the best part about it is it was already compiled and linked for Multinet. (big concern for me, since we have no language compilers, execpt Macro32). Anyways, I am wondering there exists a Multinet IRC server? Thanx Billy D'Augustine bill@main.morris.org Until the blood on your sword is the blood of a king. Hail and kill! X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 11 Aug 93 12:39:48 PDT From: bwllibmod@sgcl1.unisg.ch (David-Michael Lincke) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: signature trailers Message-ID: <1993Aug11.141650.924@sgcl1.unisg.ch> ================================================================================ Archive-Date: xxx, 11 Aug 1993 14:16:50 GMT Date: 11 Aug 93 14:16:50 GMT Organization: University of St.Gallen, Switzerland Lines: 26 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <009709BB.443AAE4E.30769@vms1.cc.uop.edu>, ebates@vms1.cc.uop.edu (Ed Bates) writes: > On Thu, 5 Aug 1993 at 23:10:43 GMT, dwing@uh01.colorado.edu (Dan Wing) wrote: >>In article <930805103517.248002c2@CADRE1.CALTECH.EDU>, JDM@CADRE1.CALTECH.EDU >>writes: >> >>>Is there a way to get VMSMAIL to append a signature to every message as >>>does various internet mail packages? >> >>Nope. You can, however, tell VMSmail to use a .COM file when editing, and >>this .COM file can append a signature before giving the edited mail message >>back to VMSmail. > For those of you having installed PMDF at your site, you can have your signature appended on the fly to every non local outgoing (i.e. SMTP Mail) mail by defining one of the following logicals: $ DEFINE PMDF_SIGNATURE "text" or $ DEFINE PMDF_SIGNATURE "@" David -- David-Michael Lincke ******** University of St. Gallen HSG ******** ------------------------------------------------------------------- bwllibmod@sgcl1.unisg.ch, ae456@freenet.carleton.ca 92636828s@sgcl1.unisg.ch, lincke@ezinfo.vmsmail.ethz.ch X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 11 Aug 93 16:31:43 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Copying files and subdirectories with MultiNet Message-ID: <24bl8m$pr9@zippy.Telcom.Arizona.EDU> From: leonard@telcom.arizona.edu (Aaron Leonard) ================================================================================ Archive-Date: xxx, 11 Aug 1993 20:36:37 GMT Date: 11 Aug 1993 20:36:37 GMT Reply-To: Leonard@Arizona.EDU Distribution: world Organization: University of Arizona Telecommunications NNTP-Posting-Host: rena.telcom.arizona.edu Lines: 11 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug10.194912.25952@es.dupont.com>, cornwaca@ldoc03.ldc.dupont.com ( ) writes: | Is there a way to copy all the files in a directory plus it's subdirectories using MGET or MPUT. | I know a basic FTP will copy multiple files, I'm not sure | about the best approach for a directory tree. Am I stuck with creating a | VMS Backup saveset and moving it to the remote node (via IP or DECNET)? RCP is supposed to be able to copy a directory tree with the /RECURSIVE flag (but I can never figure out how the hell to make this work.) Aaron X-ST-Status: N Return-path: Received: from ERICH.TRIUMF.CA ([128.189.128.2]) by TGV.COM via INTERNET ; Wed, 11 Aug 93 18:19:20 PDT Received: by erich.triumf.ca (MX V3.3 VAX) id 2336; Wed, 11 Aug 1993 18:21:58 PST ================================================================================ Archive-Date: Wed, 11 Aug 1993 18:21:55 PST Date: Wed, 11 Aug 1993 18:21:55 PST From: "David Scott Cunningham, TRIUMF, 604 222-1047" To: info-multinet@tgv.com Subject: problem with No Passall on queue device Hello, The NLP2: device for one of our networked print queues displays as thus: Printer NLP2:, device type unknown, is online, allocated, record-oriented device, carriage control, device is spooled through an intermediate device. Error count 0 Operations completed 0 Owner process "SYMBIONT_192" Owner UIC [0,0] Owner process ID 2620018B Dev Prot S:RWLP,O:RWLP,G:RWLP,W:RWLP Reference count 1 Default buffer size 0 Page width 0 Page Length 66 No Carriage_return Formfeed Lowercase No Passall No Wrap No Printall No Fallback No Tab Truncate Intermediate device: MTI0$DIA0: Associated queue: LAS8 But it is not passing all data for binary files which are bitmaps for HP plots. How can I change the device characteristic to Passall and/or Printall from the current No Passall/No Printall? We are using Multinet 3.2 on VMS 5.4-2, and the queue is sonfigured as a raw IP STREAM type. Dave X-ST-Status: N Return-path: ================================================================================ Archive-Date: Wed, 11 Aug 1993 20:36:20 PDT Date: Wed, 11 Aug 93 20:36:20 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: problem with No Passall on queue device To: omicron @ erich.triumf.ca cc: info-multinet @ TGV.COM > The NLP2: device for one of our networked print queues displays as thus: > Printer NLP2:, device type unknown, is online, allocated, record-oriented > device, carriage control, device is spooled through an intermediate device. > Error count 0 Operations completed 0 > Owner process "SYMBIONT_192" Owner UIC [0,0] > Owner process ID 2620018B Dev Prot S:RWLP,O:RWLP,G:RWLP,W:RWLP > Reference count 1 Default buffer size 0 > Page width 0 Page Length 66 > No Carriage_return Formfeed Lowercase > No Passall No Wrap No Printall > No Fallback No Tab Truncate > Intermediate device: MTI0$DIA0: > Associated queue: LAS8 > But it is not passing all data for binary files which are bitmaps for HP plots. > How can I change the device characteristic to Passall and/or Printall from > the current No Passall/No Printall? > We are using Multinet 3.2 on VMS 5.4-2, and the queue is sonfigured as a > raw IP STREAM type. The VMS symbiont ignores the setting of Passall on the device; you need to PRINT/PASSALL. Ken X-ST-Status: N Return-path: Received: from ALPHA.NOAAPMEL.GOV ([192.68.161.2]) by TGV.COM via INTERNET ; Thu, 12 Aug 93 10:12:24 PDT Received: from ALPHA.PMEL.NOAA.GOV by ALPHA.PMEL.NOAA.GOV (PMDF V4.2-11 #2604) id <01H1NK6KKK2O91XTXB@ALPHA.PMEL.NOAA.GOV>; Thu, 12 Aug 1993 10:11:37 PDT ================================================================================ Archive-Date: Thu, 12 Aug 1993 10:11:37 -0700 Date: Thu, 12 Aug 1993 10:11:37 -0700 (PDT) From: Mike Knezevich Subject: NTP To: info-multinet@tgv.com X-VMS-To: IN%"info-multinet@tgv.com" MIME-version: 1.0 When using NTP to synchronize the VAX time, does anybody have any advice on either just turning ntp on and letting the VAX time change (it's 2+ minutes off) all at once, or using the waytoobig parameter to force a more gradual change in the vax time (say 5-10 seconds at a time)? thanks, Mike X-ST-Status: N Return-path: Received: from REG.TRIUMF.CA ([128.189.128.4]) by TGV.COM via INTERNET ; Thu, 12 Aug 93 16:25:34 PDT Received: by erich.triumf.ca (MX V3.3 VAX) id 3869; Thu, 12 Aug 1993 16:28:05 PST ================================================================================ Archive-Date: Thu, 12 Aug 1993 16:28:07 PST Date: Thu, 12 Aug 1993 16:28:07 PST From: "David Scott Cunningham, TRIUMF, 604 222-1047" To: info-multinet@tgv.com Subject: further to printing plot files to STREAM que (was Passall) Hello again, Printing a plot file to a remote queue defined as STREAM with /passall on the print command did not work; it still stripped characters from the plot file. However, if I set up a second queue to point to the same device, and configure it as an LPR queue, then do print/passall with a plot file, the file does print correctly. Yest, if I send a text file to the LPR queue, it strips off carriage returns without /passall, and if I do include a /passall, then nothing comes out at all. In other words: /PASSALL /NOPASSALL -------------+--------------------------+--------------------------- | Binary file trashed | Binary file trashed STREAM queue | | | Text file trashed | Text file WORKS -------------+--------------------------+--------------------------- | Binary file WORKS | Binary file trashed LPR queue | | | Text file trashed | Text file trashed So, I can have two queues pointing to the same device, and let users know the difference, but I'd also like to know if this is a bug in multinet, or perhaps with the network printer. We are using Multinet 3.2, on VMS 5.4-2, and are talking to an HP Laserjet 4m connected to the network with Microplex M200 TCP/IP printer adapter? Dave X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 12 Aug 93 16:40:58 PDT From: haro@sable.jpl.nasa.gov (Raul Haro) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: UDP/checksum ================================================================================ Archive-Date: xxx, 12 Aug 1993 22:58:18 GMT Date: 12 Aug 1993 22:58:18 GMT Organization: Jet Propulsion Laboratory Lines: 15 Distribution: world Reply-To: haro@sable.jpl.nasa.gov (Raul Haro) NNTP-Posting-Host: 137.79.114.217 X-Newsreader: mxrn 6.18-4 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi there, Could the UDP checksum be turned off on a VAX? If so, how would I go about doing it? Thanks c-ya Raul -- Raul Haro haro@sable.jpl.nasa.gov X-ST-Status: N Return-path: Received: from hamlet.caltech.edu ([131.215.139.3]) by TGV.COM via INTERNET ; Thu, 12 Aug 93 19:07:56 PDT Received: from venom (venom.jpl.nasa.gov) by HAMLET.CALTECH.EDU (PMDF V4.2-13 #3650) id <01H1O33DAWNK8X3PLA@HAMLET.CALTECH.EDU>; Thu, 12 Aug 1993 19:07:48 PDT Received: by venom (1.37.109.4/16.2) id AA16890; Thu, 12 Aug 93 19:07:34 -0700 ================================================================================ Archive-Date: Thu, 12 Aug 1993 19:07:34 -0700 Date: Thu, 12 Aug 1993 19:07:34 -0700 From: "Jack J. Tuszynski" Subject: print otptions on multinet remote print queue To: INFO-MULTINET@TGV.COM X-Envelope-to: INFO-MULTINET@TGV.COM Dear TGV Multinet technical support, Vax Name: Grouch Serial Number: SVA2220 On our Grouch Vax 11/785 we have TGV Multinet 3.2 running. We used the MULTINET CONFIGURE/PRINTER utility to set up a remote printer queue to a Unix workstation. The Unix workstation is an HP 9000/735 running HP-UX 9.01 with an HP LaserJet IV Si MX attached to it. This is a postscript printer by the way. The software on the HP used to drive the printer is the new JetDirect software that works with the ethernet card installed in the printer. A script on the HP workstation determines if a file sent to it is ASCII or Postscript. On the VAX/VMS side I use PRINT/QUEUE=NICOLE_LJ1 to send text files to the Unix workstation/printer. It works fine. Anyway, my question to you is how does one use the PRINT command to specify PRINT/LANDSCAPE or PRINT/WIDTH=132 like we used to be able to do back in the old days. Before we surplused the Imagen attached to our Vax, which was driven by Northlake's Imprint software, we were able to do this. Is this something that should be configured on the VAX side, with the PRINT command, or on the Unix side in the script? And if on the Unix side, how do I send it commands to do things through PRINT? Thank you for assisting me in this matter if you can. Jack Tuszynski jjt@venom.jpl.nasa.gov X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 12 Aug 1993 22:24:57 PDT Date: Thu, 12 Aug 93 22:24:57 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: further to printing plot files to STREAM que (was Passall) To: omicron @ erich.triumf.ca cc: info-multinet @ TGV.COM > Printing a plot file to a remote queue defined as STREAM with /passall > on the print command did not work; it still stripped characters from > the plot file. > However, if I set up a second queue to point to the same device, and > configure it as an LPR queue, then do print/passall with a plot file, > the file does print correctly. > Yest, if I send a text file to the LPR queue, it strips off carriage > returns without /passall, and if I do include a /passall, then nothing > comes out at all. > In other words: > /PASSALL /NOPASSALL > -------------+--------------------------+--------------------------- > | Binary file trashed | Binary file trashed > STREAM queue | | > | Text file trashed | Text file WORKS > -------------+--------------------------+--------------------------- > | Binary file WORKS | Binary file trashed > LPR queue | | > | Text file trashed | Text file trashed > So, I can have two queues pointing to the same device, and let users > know the difference, but I'd also like to know if this is a bug in > multinet, or perhaps with the network printer. > We are using Multinet 3.2, on VMS 5.4-2, and are talking to an > HP Laserjet 4m connected to the network with Microplex M200 TCP/IP > printer adapter? No ideas. I'd like to see minimal-case examples of the trashing comparing the original file with the output as viewed by a network analyzer. I have no idea what transformations the printer adapter might apply. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 12 Aug 93 22:54:56 PDT From: fuzzface@drycas.club.cc.cmu.edu (John McMahon) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: UDP/checksum ================================================================================ Archive-Date: xxx, 13 Aug 1993 01:42 EDT Date: 13 Aug 1993 01:42 EDT Organization: Carnegie Mellon University Computer Club Lines: 15 Distribution: world NNTP-Posting-Host: drycas.club.cc.cmu.edu News-Software: VAX/VMS VNEWS 1.41 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <24ehua$8gf@elroy.jpl.nasa.gov>, haro@sable.jpl.nasa.gov (Raul Haro) writes... > >Could the UDP checksum be turned off on a VAX? > Turning off UDP checksums means that your data is no longer checked for possible errors. Under MultiNet, disabling UDP Checksums does not give you a performance boost since the checksum algorithms are already pretty fast. Why do you want to do this ? John 'FuzzFace\Fast-Eddie' McMahon fuzzface@drycas.club.cc.cmu.edu {use finger to get my PGP public key} "Any networking problem can be solved by the application of 75 miles of Unshielded Twisted Pair (UTP)" - InterOp Spring 1993 X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 12 Aug 1993 23:00:15 PDT Date: Thu, 12 Aug 93 23:00:15 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: TGV moving; INFO-MULTINET down this weekend To: info-multinet @ TGV.COM TGV is going to be moving to our new offices this coming weekend (Saturday & Sunday August 14th and 15th, PDT). From 5pm Friday until sometime over the weekend, our Internet connection and hence INFO-MULTINET will be down. Starting Monday, August 16th our new address will be: TGV, Inc. 101 Cooper Street Santa Cruz, CA 95060 and new phone numbers will be: (800) TGV-3440 +1 408 457 5200 (main operator) +1 408 457 5201 (*new* technical support hotline) +1 408 457 5205 (FAX) Ken X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Fri, 13 Aug 93 07:59:09 PDT Via: uk.co.ggr; Fri, 13 Aug 1993 15:58:43 +0100 Received: from uk0x12.ggr.co.uk by uk0x08.ggr.co.uk; Fri, 13 Aug 93 15:59:45 BST Received: from uk0x12.ggr.co.uk by mailhub.ggr.co.uk (5.59/imd-070593) id AA21705; Fri, 13 Aug 93 15:53:13 BST Received: by uk0x12.ggr.co.uk (4.1/imd110593) id AA07902; Fri, 13 Aug 93 15:58:37 BST ================================================================================ Archive-Date: Fri, 13 Aug 1993 15:49:14 +0100 Date: Fri, 13 Aug 1993 15:49:14 +0100 (BST) From: Lord Wodehouse Subject: Searching for domains To: TGV list Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi, On most unix systems, you can edit resolv.conf to have a line to search multiple domains. domain ggr.co.uk search ggr.co.uk another_name.com nameserver xx.xx.xx.xx nameserver yy.yy.yy.yy This means you can telnet, ftp or whatever to a node in either ggr.co.uk or another_name.com and just supply the short host name (the names are all different in both domains). The question is "Can you do this with Multinet? It is easy with CMUTEK (yes I ghave done it too). However I am not clear what to change in Multinet. Have tried changing multinet_localdomain to be a search list like "MULTINET_LOCALDOMAIN" = "ggr.co.uk" = "another_name.com" but this does not seem to work (well if it does, I did not find the key). All help gratefully accepted. We run DNS on various SUNs which can contact the DNS server in the other domain as required. Lord John - The Programming Peer mail - w0400@ggr.co.uk tlx - 8951942 GLXPRI G Have network - will mail! fax - +44 81 423 4070 ============================================================================= | | | The contents of this message *may* reflect my personal opinion. They are | | *not* intended to reflect those of my employer, or anyone else, unless | | otherwise stated. If your machine crashes after using any methods, code | | or whatever suggested by this mail, it is *your* problem and not mine. | | | ============================================================================= X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 13 Aug 93 12:03:14 PDT From: brodie@fps.mcw.edu (Kent C Brodie) X-Newsgroups: vmsnet.networks.tcp-ip.multinet,comp.sys.dec Subject: need assistance with rdump stuff Message-ID: <1993Aug13.130448.779@fps.mcw.edu> ================================================================================ Archive-Date: xxx, 13 Aug 1993 13:04:48 CST Date: 13 Aug 93 13:04:48 CST Organization: Medical College of Wisconsin, Milwaukee, WI Lines: 25 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I am using the "rdump" command to store dumps of my ultrix system on a TGV (vax) server tape. I have it working nicely, but I'd like to explore the ability to store MULTIPLE dumps to the same tape. Can someone explain how the internals of this works? Specifically: 1) I assume I use the (tgv) rmt tape parameters to /NOMOUNT and /NORWEIND and /NOUNLOAD, so that the tape stays where it is. 2) I assume that multiple "rdump" commands can be done in a row. 3) How then, is the "rrestore" done? Specifically, say I dump "/" as the first dump, and then "/usr" as the second dump. When restoring, how do I get the SECOND dump-file? -- Kent C. Brodie - Sr. Systems Manager InterNet: brodie@fps.mcw.edu Faculty Physicians & Surgeons uucpNet: fps!brodie Medical College of Wisconsin MaBellNet: +1 414 266 5080 "WARNING: Do not look into laser with remaining eye" (seen on a BIG laser) X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 13 Aug 93 12:03:15 PDT From: haro@sable.jpl.nasa.gov (Raul Haro) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: UDP/checksum ================================================================================ Archive-Date: xxx, 13 Aug 1993 18:50:07 GMT Date: 13 Aug 1993 18:50:07 GMT Organization: Jet Propulsion Laboratory Lines: 48 Distribution: world Reply-To: haro@sable.jpl.nasa.gov (Raul Haro) NNTP-Posting-Host: 137.79.114.217 X-Newsreader: mxrn 6.18-4 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <13AUG199301421461@drycas.club.cc.cmu.edu>, fuzzface@drycas.club.cc.cmu.edu (John McMahon) writes: > From: fuzzface@drycas.club.cc.cmu.edu (John McMahon) > Newsgroups: vmsnet.networks.tcp-ip.multinet > Subject: Re: UDP/checksum > Date: 13 Aug 1993 01:42 EDT > Organization: Carnegie Mellon University Computer Club > Lines: 15 > News-Software: VAX/VMS VNEWS 1.41 > > In article <24ehua$8gf@elroy.jpl.nasa.gov>, haro@sable.jpl.nasa.gov (Raul Haro) writes... > > > >Could the UDP checksum be turned off on a VAX? > > > > Turning off UDP checksums means that your data is no longer checked for > possible errors. Under MultiNet, disabling UDP Checksums does not give you a > performance boost since the checksum algorithms are already pretty fast. Why > do you want to do this ? > > John 'FuzzFace\Fast-Eddie' McMahon > fuzzface@drycas.club.cc.cmu.edu {use finger to get my PGP public key} > > "Any networking problem can be solved by the application of 75 miles of > Unshielded Twisted Pair (UTP)" - InterOp Spring 1993 > The reason for inquiring about turning off the checksums is because I suspected that the reasons that we were missing blocks was due to the checksums being calculated incorrectly from the sending machine. This does not seem likely now. The problem we are having is that we are missing blocks at the receiving machine (VAX). What I have been able to determine so far is that the blocks are making it into the machine, but when we read them from the UDP port, ~25% of the blocks are missing. We have tried running on a faster VAX machine, increasing the UDP port buffer to its max value and increasing the priority on the process which is reading the blocks, but the number of blocks missing as remained the same. Any recommendations would be aprreciated. Thanks c-ya Raul -- Raul Haro haro@sable.jpl.nasa.gov X-ST-Status: N Return-path: Received: from jpl-mil ([137.78.160.9]) by TGV.COM via INTERNET ; Fri, 13 Aug 93 12:31:35 PDT Received: from venom (venom.jpl.nasa.gov) by jpl-mil (4.1/SMI-4.1+xDXRm2.2) id AA02230; Fri, 13 Aug 93 12:28:23 PDT Received: by venom (1.37.109.4/1.34) id AA20082; Fri, 13 Aug 93 12:31:24 -0700 ================================================================================ Archive-Date: Fri, 13 Aug 1993 12:31:24 -0700 Date: Fri, 13 Aug 93 12:31:24 -0700 From: jjt@venom.Jpl.Nasa.Gov (Jack J. Tuszynski) To: INFO-MULTINET@TGV.COM Subject: request for tech support Dear TGV Multinet technical support, Vax Name: Grouch Serial Number: SVA2220 On our Grouch Vax 11/785 we have TGV Multinet 3.2 running. We used the MULTINET CONFIGURE/PRINTER utility to set up a remote printer queue to a Unix workstation. The Unix workstation is an HP 9000/735 running HP-UX 9.01 with an HP LaserJet IV Si MX attached to it. This is a postscript printer by the way. The software on the HP used to drive the printer is the new JetDirect software that works with the ethernet card installed in the printer. A script on the HP workstation determines if a file sent to it is ASCII or Postscript. On the VAX/VMS side I use PRINT/QUEUE=NICOLE_LJ1 to send text files to the Unix workstation/printer. It works fine. Anyway, my question to you is how does one use the PRINT command to specify PRINT/LANDSCAPE or PRINT/WIDTH=132 like we used to be able to do back in the old days. Before we surplused the Imagen attached to our Vax, which was driven by Northlake's Imprint software, we were able to do this. Is this something that should be configured on the VAX side, with the PRINT command, or on the Unix side in the script? And if on the Unix side, how do I send it commands to do things through PRINT? Thank you for assisting me in this matter if you can. Jack Tuszynski jjt@venom.jpl.nasa.gov X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Fri, 13 Aug 93 13:36:27 PDT ================================================================================ Archive-Date: Fri, 13 Aug 1993 13:36:23 -0700 Date: Fri, 13 Aug 1993 13:36:23 -0700 (PDT) From: Matt Madison To: jjt@venom.Jpl.Nasa.Gov CC: info-multinet@TGV.COM Subject: RE: request for tech support > On our Grouch Vax 11/785 we have TGV Multinet 3.2 running. We >used the MULTINET CONFIGURE/PRINTER utility to set up a remote printer queue >to a Unix workstation. The Unix workstation is an HP 9000/735 running >HP-UX 9.01 with an HP LaserJet IV Si MX attached to it. This is a postscript >printer by the way. The software on the HP used to drive the printer is the >new JetDirect software that works with the ethernet card installed in the >printer. A script on the HP workstation determines if a file sent to it >is ASCII or Postscript. On the VAX/VMS side I use PRINT/QUEUE=NICOLE_LJ1 >to send text files to the Unix workstation/printer. It works fine. > Anyway, my question to you is how does one use the PRINT command >to specify PRINT/LANDSCAPE or PRINT/WIDTH=132 like we used to be able to >do back in the old days. Before we surplused the Imagen attached to our Vax, >which was driven by Northlake's Imprint software, we were able to do this. >Is this something that should be configured on the VAX side, with the PRINT >command, or on the Unix side in the script? And if on the Unix side, how do >I send it commands to do things through PRINT? > Thank you for assisting me in this matter if you can. Neither /LANDSCAPE nor /WIDTH are qualifiers for the PRINT command on VMS. However, there are a couple of ways you could try and get the printer to do special things for you. One is to dump LPD and drive the LaserJet directly, creating modules in a device control library to send the necessary control sequences to the printer to get it to do what you want. We do this with our Laser Jet 3si; the device control library we use is available by anonymous FTP from PUBLIC.TGV.COM in directory [.HPLJ3SI]. If you still want to use LPD, you may still be able to use a device control library to insert the necessary control sequences -- provided the software on the UNIX system doesn't barf when it sees them. Alternatively, you could use the USER_LPD_CLIENT user exit to convert PRINT command parameters into commands that go into the LPD control file for the print job, then have the UNIX system convert those into the necessary control sequences to change the printer settings. Source code for a template USER_LPD_CLIENT.C is in the [.EXAMPLES] subdirectory under the MULTINET: directory. I'm not sure what would be involved in getting this sort of thing setup on the UNIX system. -Matt -- Matthew Madison | madison@tgv.com | +1 408 427-4366 (457-5200 after 13 Aug) TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 13 Aug 1993 21:49:18 PDT Date: Fri, 13 Aug 93 21:49:18 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: need assistance with rdump stuff To: brodie @ fps.mcw.edu cc: info-multinet @ TGV.COM { I am using the "rdump" command to store dumps of my ultrix system on > a TGV (vax) server tape. > I have it working nicely, but I'd like to explore the ability > to store MULTIPLE dumps to the same tape. > Can someone explain how the internals of this works? > Specifically: > 1) I assume I use the (tgv) rmt tape parameters to /NOMOUNT and /NORWEIND > and /NOUNLOAD, so that the tape stays where it is. Correct > 2) I assume that multiple "rdump" commands can be done in a row. Correct~? > 3) How then, is the "rrestore" done? Specifically, say I dump "/" > as the first dump, and then "/usr" as the second dump. When > restoring, how do I get the SECOND dump-file? You ask it to skip the first file. For example: % rsh vax mt -f mua0:/nomount/nounload/norewind fsf 1 "fsf" means forward skip file, and 1 means 1 file to skip. THEN you do the rrestore (PLEASE try this!). Basically, our RSHELL server "emulates" the UNIX mt command for magtape manipulation! Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 13 Aug 1993 21:50:46 PDT Date: Fri, 13 Aug 93 21:50:46 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: UDP/checksum To: haro @ sable.jpl.nasa.gov cc: info-multinet @ TGV.COM > The reason for inquiring about turning off the checksums is because I suspected > that the reasons that we were missing blocks was due to the checksums being > calculated incorrectly from the sending machine. This does not seem likely now. > The problem we are having is that we are missing blocks at the receiving machine > (VAX). What I have been able to determine so far is that the blocks are making > it into the machine, but when we read them from the UDP port, ~25% of > the blocks are missing. We have tried running on a faster VAX machine, > increasing the UDP port buffer to its max value and increasing the priority on > the process which is reading the blocks, but the number of blocks missing > as remained the same. If you have a disparity in the speed of the transmitting machine and the receiving machine and no flow control in your protocol, a packet loss over UDP like this is going to be "normal" (that is, expected). Ken X-ST-Status: N Return-path: Received: from Grubnik.TGV.COM ([134.24.12.126]) by TGV.COM via INTERNET ; Sat, 14 Aug 93 02:06:55 PDT Received: from MVB.SAIC.COM by Grubnik.TGV.COM with INTERNET ; Fri, 13 Aug 93 20:19:07 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet,comp.sys.dec Subject: Re: need assistance with rdump stuff Message-ID: <21815.2c6bf93c@ecs.umass.edu> From: rrooks@ecs.umass.edu ================================================================================ Archive-Date: xxx, 13 Aug 1993 20:27:40 GMT Date: 13 Aug 93 20:27:40 GMT Lines: 31 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET > > 1) I assume I use the (tgv) rmt tape parameters to /NOMOUNT and /NORWEIND > and /NOUNLOAD, so that the tape stays where it is. > Yes, this works for me. > 2) I assume that multiple "rdump" commands can be done in a row. Yes. > > 3) How then, is the "rrestore" done? Specifically, say I dump "/" > as the first dump, and then "/usr" as the second dump. When > restoring, how do I get the SECOND dump-file? > To rrestore a file, you have to specify which save set to use with the s parameter. For instance, let's say your tape drive is mkb500: and your VAX machine is chevax: rrestore -iosf 2 chevax:mkb500: This uses the interactive restoration program with saveset #2. i - interactive restoration o - foreign compatibility s - "saveset" number f - foreign device name Ray X-ST-Status: N Return-path: Received: from Grubnik.TGV.COM ([134.24.12.126]) by TGV.COM via INTERNET ; Sat, 14 Aug 93 02:08:27 PDT Received: from MVB.SAIC.COM by Grubnik.TGV.COM with INTERNET ; Fri, 13 Aug 93 16:44:54 PDT From: bill@main.morris.org (Billy D'Augustine) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: TELNET usage summaries? ================================================================================ Archive-Date: xxx, 13 Aug 1993 23:28:41 GMT Date: 13 Aug 1993 23:28:41 GMT Organization: Morris Automated Information Network Lines: 45 Reply-To: bill@main.morris.org NNTP-Posting-Host: main.morris.org To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I have a rather uncomfortable situation on my hands. We are a public library consortium, and our net connection is being funded by an outside party as part of an experiment in electronic communications (I know, sounds lofty, but I dont care, it gets the net to my desk :-) As part of the deal, the outside party wants a usage summary of what people are doing. So I got a call today by one of the people involved, and he started asking me questions, which boiled down to: can we see WHERE people are telnetting to? I told him by using VMS accounting, I can basiclly only see a person telnetting, not where. Just the fact that the telnet image was activated. I also told him that I could get info on people telnetting into the system, but just the system name they are coming from. Needless to say, he was quite unhappy that I couldnt help him. I feel that such info they want (who is telnetting out, what machine are they logging into) seems to be an invasion on privacy. Before I blow him off, I wanted to know if there is any way I can get such info? Ive done accounting, and as I said above, can see users activiating the telnet image. By setting the following for the telnet server: Service "TELNET": TCP socket (AF_INET,SOCK_STREAM), Port 23 Socket Options = SO_KEEPALIVE INIT() = Merge_Image LISTEN() = TCP_Listen CONNECTED() = TCP_Connected SERVICE() = Internal_Telnet Program = "MULTINET:LOADABLE_KERBEROS_TELNET" Log File for Accepts & Rejects = MULTINET:SERVER.LOG This seems to only log incoming telnet sessions, and just tells me where the origin was. Thanx Billy D'Augustine VAX/VMS system admin bill@main.morris.org Morris County Info Systems Until the blood on your sword is the blood of a king. Hail and kill! X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sat, 14 Aug 93 11:04:17 PDT From: francois@italia.boeing.com (Francois VanSteertegem) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: VMS 5.4-3 through 5.5-2 may break MultiNet. Message-ID: <103327@bcsaic.boeing.com> ================================================================================ Archive-Date: xxx, 14 Aug 1993 06:07:17 GMT Date: 14 Aug 93 06:07:17 GMT Sender: francois@bcsaic.boeing.com Distribution: na Organization: Boeing Computer Services, Seattle Lines: 51 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET The VMS 5.5 upgrade broke my Multinet 3.2-C. (TGV said that not everybody has this problem. I got the impression that it was mostly DEBNA-cards that got hosed.) My configuration: Hardware: A single-system-disk VAX cluster consisting of a 6320 and an 8350--both with DEBNA ethernet controllers. Software: Multinet v3.2-C. Upgrading from VMS 5.4-2 to VMS 5.5 The symptoms: I couldn't telnet from one node of the cluster to the other. I couldn't receive SMTP mail from other sites. No one was able to telnet in to me. I WAS able to send mail and telnet to other sites. I was also able to telnet out to another site and then telnet back to myself--I'm not sure why that was. The problem (as told to me by TGV): Faulty ethernet drivers in VMS 5.4-3, VMS 5.5, VMS 5.5-1 and VMS 5.5-2. The fix: Apply CSCPAT-0552 (that's zero-five-five-two) whenever you upgrade to one of these versions. I had to get it from DEC, but maybe TGV would be willing to keep the latest version of it on their anonymous-FTP? Kudos to TGV for their quick response to my cries for help--I LOVE their technical support. (But I wish they would have warned me about this, somehow.) I hope this proves useful. Regards -- Francois VanSteertegem 206-965-1384 (voice) 206-237-0052 (fax) Internet address: francois@boeing.com or francois@bmt.ca.boeing.com X-ST-Status: N Return-path: ================================================================================ Archive-Date: Sat, 14 Aug 1993 15:22:54 PDT Date: Sat, 14 Aug 93 15:22:54 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: TELNET usage summaries? To: bill @ main.morris.org cc: info-multinet @ TGV.COM > I have a rather uncomfortable situation on my hands. > We are a public library consortium, and our net connection is > being funded by an outside party as part of an experiment in > electronic communications (I know, sounds lofty, but I dont > care, it gets the net to my desk :-) > As part of the deal, the outside party wants a usage > summary of what people are doing. So I got a call today by > one of the people involved, and he started asking me questions, > which boiled down to: can we see WHERE people are telnetting > to? > I told him by using VMS accounting, I can basiclly only > see a person telnetting, not where. Just the fact that the > telnet image was activated. I also told him that I could get > info on people telnetting into the system, but just the system > name they are coming from. > Needless to say, he was quite unhappy that I couldnt > help him. I feel that such info they want (who is telnetting > out, what machine are they logging into) seems to be an > invasion on privacy. > Before I blow him off, I wanted to know if there > is any way I can get such info? Incoming TELNETs there is no way to get the username, just the origin machine. The username is not conveyed by the protocol. Outgoing connections, take a look at the USER_ACCESS.C module in the [MULTINET.EXAMPLES] directory. You can customize this to do what you want. Ken X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Sat, 14 Aug 93 15:49:11 PDT ================================================================================ Archive-Date: Sat, 14 Aug 1993 15:49:07 -0700 Date: Sat, 14 Aug 1993 15:49:07 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: TELNET usage summaries? To: bill@main.morris.org Cc: INFO-MULTINET@TGV.COM In-Reply-To: <24h839$9oh@c3po.jvnc.net> Mail-System-Version: Organization: TGV, Incorporated X-Phone: 408/427-4366 (work); 408/427-4365 (fax) X-Address: 603 Mission Street; Santa Cruz, CA 95060 (work) > I have a rather uncomfortable situation on my hands. >We are a public library consortium, and our net connection is >being funded by an outside party as part of an experiment in >electronic communications (I know, sounds lofty, but I dont >care, it gets the net to my desk :-) > As part of the deal, the outside party wants a usage >summary of what people are doing. So I got a call today by >one of the people involved, and he started asking me questions, >which boiled down to: can we see WHERE people are telnetting >to? > I told him by using VMS accounting, I can basiclly only >see a person telnetting, not where. Just the fact that the >telnet image was activated. I also told him that I could get >info on people telnetting into the system, but just the system >name they are coming from. > Needless to say, he was quite unhappy that I couldnt >help him. I feel that such info they want (who is telnetting >out, what machine are they logging into) seems to be an >invasion on privacy. > Before I blow him off, I wanted to know if there >is any way I can get such info? Ive done accounting, and >as I said above, can see users activiating the telnet >image. By setting the following for the telnet server: > >Service "TELNET": > TCP socket (AF_INET,SOCK_STREAM), Port 23 > Socket Options = SO_KEEPALIVE > INIT() = Merge_Image > LISTEN() = TCP_Listen > CONNECTED() = TCP_Connected > SERVICE() = Internal_Telnet > Program = "MULTINET:LOADABLE_KERBEROS_TELNET" > Log File for Accepts & Rejects = MULTINET:SERVER.LOG > > This seems to only log incoming telnet sessions, and >just tells me where the origin was. Check out the file MULTINET_ROOT:[MULTINET.EXAMPLES]USER_ACCESS.C. Each MultiNet client utility is linked against it, and will call the routines in it. This should give you what you want. Note, however, that it does not prevent someone from writing their own TELNET client and using that to connect out. Regards! -----Stuart X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 16 Aug 93 03:12:22 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Stream Printing limitations Message-ID: <1993Aug16.123156.27365@state.systems.sa.gov.au> From: enisaxg@state.systems.sa.gov.au ================================================================================ Archive-Date: xxx, 16 Aug 1993 12:31:56 ACST Date: 16 Aug 93 12:31:56 ACST Organization: State Systems, South Australia Lines: 41 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi! Someone (from Innosoft, I think) posted a note here a while ago about a home-grown VMS print symbiont that works with Multinet Stream Printing and recognises Postscript vs Text on-the-fly. This seems genuinely useful if it exists; we have been trying to get an Alpha machine here by justifying it as a Print Server using Multinet; unfortunately, we have a lot of HP 3Si and 4M printers, which requires us to use the (somewhat limited) Multinet stream symbiont. These are probably FAQ's about the stream symbiont, but: 1. We'd dearly like to see its usefulness enhanced by allowing easily-defined banner and flag pages (maybe per queue, so that text and PS printing don't clash). 2. It would be nice if there was some way of pre-processing files queued for the stream so that PS/text could be automatically recognised, and the appropriate action could be taken (eg issue mode-switching commands to printers that support this, like our HP's). These things are easy to do within Unix (using eg HP's JetDirect software on HP-UX or Solaris/SUNOS and just a bit of shell programming), and I believe that it's just too hard under VMS+Multinet (modifying a print symbiont is no light task!), harder than it probably needs to be. Although we run the HP JetDirect/Unix software in-house, we believe VMS (with Multinet) gives a much more robust and commercial-strength capability for printing (= our operators find Unix printer admin to be too basic, with not enough info returned to the user about job status etc; however, at least it's easy to program!). We'd ideally like ALL printing to be done under VMS/Multinet control. Does anyone out there know of any fixes/already-written-symbionts??? Any help would be appreciated. Thanks in advance. Alexander Gunjko [ gunjkoa@dep.sa.gov.au ] Systems Administrator South Australian Office of Planning and Urban Development. X-ST-Status: N Return-path: Received: from Butler.EDU ([159.242.1.5]) by TGV.COM via INTERNET ; Mon, 16 Aug 93 09:31:59 PDT Received: from Butler.EDU by Butler.EDU (PMDF V4.2-12 #3291) id <01H1T3E8HH8M9EDG34@Butler.EDU>; Mon, 16 Aug 1993 11:31:45 EST ================================================================================ Archive-Date: Mon, 16 Aug 1993 11:31:45 -0500 Date: Mon, 16 Aug 1993 11:31:45 -0500 (EST) From: "David J. Knapke" Subject: Re: VMS 5.4-3 through 5.5-2 may break MultiNet. To: info-multinet@tgv.com X-VMS-To: IN%IMULTINET MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII >Kudos to TGV for their quick response to my cries for help--I LOVE their >technical support. (But I wish they would have warned me about this, >somehow.) This has been a 'well known' problem with that Ethernet card for quite some time. We got SEVERAL bits of info related to this from Digital. If you don't have it, a great tool to keep you informed from Digital is DSIN. The only requirement is that you carry software maintenance from DEC. David J. Knapke Knapke@Butler.EDU Assistant Director, Knapke@ButlerU.BITNET Information Resources Butler University Voice: (317) 283-9420 Indianapolis, IN 46208 Fax: (317) 283-9740 X-ST-Status: N Return-path: Received: from PFC.MIT.EDU ([18.77.0.113]) by TGV.COM via INTERNET ; Mon, 16 Aug 93 10:34:46 PDT ================================================================================ Archive-Date: Mon, 16 Aug 1993 13:34:44 -0400 Date: Mon, 16 Aug 1993 13:34:44 -0400 (EDT) From: Mark London To: info-multinet@tgv.com Subject: rmtalloc with f11cd Does RMTALLOC/CD work with CD's mounted using F11CD? I get the following error: %RMT-F-CONNERR, Error connecting to remote host -RMT-I-REMINFO, Remote error code 0, "E98962" X-ST-Status: N Return-path: Received: from PFC.MIT.EDU ([18.77.0.113]) by TGV.COM via INTERNET ; Mon, 16 Aug 93 11:35:08 PDT ================================================================================ Archive-Date: Mon, 16 Aug 1993 14:35:01 -0400 Date: Mon, 16 Aug 1993 14:35:01 -0400 (EDT) From: Mark London To: info-multinet@tgv.com Subject: cdrom and f11cd Ignore my previous question. I decide you to use DFS to access the cdrom. X-ST-Status: N Return-path: Received: from cdev.cfsmo.honeywell.com ([129.30.30.3]) by TGV.COM via INTERNET ; Mon, 16 Aug 93 14:22:22 PDT ================================================================================ Archive-Date: Mon, 16 Aug 1993 16:22:19 -0500 Date: Mon, 16 Aug 1993 16:22:19 -0500 (CDT) From: MLEO@cdev.cfsmo.honeywell.com (Mike Leo, 612-785-4219) Subject: Ultrix questions To: info-multinet@tgv.com X-Vmsmail-To: SMTP%"info-multinet@tgv.com" Hi, I know this isn't exactly a Multinet question... I am having some problems with an Ultrix mailer, and I need some Ultrix and sendmail.cf advice. Is there a mailing reflector for Ultrix and/or sendmail.cf questions? Maybe something similiar to info-vax@sri.com? Thanks, mleo@cdev.cfsmo.honeywell.com X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 16 Aug 93 19:03:38 PDT From: francois@italia.boeing.com (Francois VanSteertegem) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Egg on my face. Message-ID: <103447@bcsaic.boeing.com> ================================================================================ Archive-Date: xxx, 16 Aug 1993 16:10:37 GMT Date: 16 Aug 93 16:10:37 GMT Sender: francois@bcsaic.boeing.com Distribution: na Organization: Boeing Computer Services, Seattle Lines: 22 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET A couple of days ago I wrote... > > The VMS 5.5 upgrade broke my Multinet 3.2-C. (TGV said that not everybody > has this problem. I got the impression that it was mostly DEBNA-cards that > got hosed.) > >Kudos to TGV for their quick response to my cries for help--I LOVE their >technical support. (But I wish they would have warned me about this, >somehow.) > Well, they DID warn me. The problem is described in the Multinet V3.2 installation guide (appendix A)--which I should have read before upgrading VMS. My apologies to TGV. -- Francois VanSteertegem 206-965-1384 (voice) 206-237-0052 (fax) Internet address: francois@boeing.com or francois@bmt.ca.boeing.com X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 06:29:12 PDT Via: uk.co.ggr; Tue, 17 Aug 1993 14:28:23 +0100 From: mcj2412@ggr.co.uk Received: from relay.ggr.co.uk by uk0x08.ggr.co.uk; Tue, 17 Aug 93 14:29:24 BST Received: from UKSYSA.ggr.co.uk by mailhub.ggr.co.uk (5.59/imd-070593) id AA04131; Tue, 17 Aug 93 14:20:30 BST Message-Id: <9308171320.AA04131@mailhub.ggr.co.uk> ================================================================================ Archive-Date: xxx, 17 Aug 1993 14:07:00 BST Date: 17 Aug 93 14:07:00 BST Subject: Problem with Stream printing To: info-multinet@TGV.COM I have an LN03R connected to a DECserver 700. I have created a STREAM queue on the VAX using MU CONF/PRINTER. The queue looks thus: Printer queue IP_UK2T51, idle, on UKGV01::NLP1:, mounted form DEFAULT /BASE_PRIORITY=4 /DEFAULT=(FEED,FORM=DEFAULT) Lowercase /OWNER=[SYSTEM] /PROCESSOR=MULTINET_STREAM_SYMBIONT /PROTECTION=(S:E,O:D,G:R,W:W) The device NLP1 is, Printer NLP1:, device type unknown, is online, allocated, record-oriented device, carriage control, device is spooled through an intermediate device. Error count 0 Operations completed 0 Owner process "SYMBIONT_104" Owner UIC [0,0] Owner process ID 2021A66A Dev Prot S:RWLP,O:RWLP,G:RWLP,W:RWLP Reference count 1 Default buffer size 0 Page width 0 Page Length 66 No Carriage_return Formfeed Lowercase No Passall No Wrap No Printall No Fallback No Tab Truncate Intermediate device: UKGH04$DUA169: Associated queue: IP_UK2T51 A show in mu conf/print shows, Queue Name IP Destination Remote Queue Name ---------- --------------- ----------------- IP_UK2T51 147.184.208.10 TCP port 2013 However the problem is that when I, $ print/queue=ip_uk2t51 file.ext the printer leds show the symbol 6 and the flashing lightening symbol but nothing get printed. Now thats interesting.... is the "TCP port 2013" significant? Like, if on the DECserver I show all ports I see, Port Access Status Services Offered . . 13 Remote Idle Listener 2013 . . Now should the service offered column and the remote queue name column be the same? If thats not significant then has anyine got any suggestions? Thanks, Mark. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 07:44:13 PDT From: teague@acavax.lynchburg.edu X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Is there public domain Telnet for VMS? Message-ID: <1993Aug17.100105.263@acavax.lynchburg.edu> ================================================================================ Archive-Date: xxx, 17 Aug 1993 10:01:05 -0500 Date: 17 Aug 93 10:01:05 -0500 Organization: Lynchburg College Lines: 7 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I want to provide Telnet access to a VAX which has no networking software installed. Is there something short of a full-blown Multinet installation I can do to allow this? Is there any public domain software for VMS which provides simple Telnet access? Please E-mail responses to teague@acavax.lynchburg.edu. Thanks. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 08:21:37 PDT From: roseberry@eisner.decus.org X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: MSDOS / Windows equiv of Eudora Message-ID: <1993Aug17.074828.341@eisner.decus.org> ================================================================================ Archive-Date: xxx, 17 Aug 1993 07:48:28 -0400 Date: 17 Aug 93 07:48:28 -0400 Organization: DECUServe Lines: 9 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET The guys with Macs really like how Eudora works the the Multinet POP Server. Is there some software for the IBM PCs that is just as good as Eudora and will work with Multinet's POP Server ? - Bert Roseberry roseberry@eisner.decus.org - or - US Coast Guard bert@mailstorm.dot.gov X-ST-Status: N Return-path: Received: from heechee.meediv.lanl.gov ([128.165.196.2]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 09:17:52 PDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 10:17:50 -0600 Date: Tue, 17 Aug 1993 10:17:50 -0600 (MDT) From: Jim Whitfill - Los Alamos Subject: Re: MSDOS / Windows equiv of Eudora To: info-multinet@tgv.com Cc: roseberry@eisner.decus.org Mail-System-Version: Organization: Los Alamos National Laboratory > From: roseberry@eisner.decus.org > X-Newsgroups: vmsnet.networks.tcp-ip.multinet > Subject: MSDOS / Windows equiv of Eudora > Message-ID: <1993Aug17.074828.341@eisner.decus.org> > Date: 17 Aug 93 07:48:28 -0400 > Organization: DECUServe > Lines: 9 > To: INFO-MULTINET@TGV.COM > X-Gateway-Source-Info: USENET > > > The guys with Macs really like how Eudora works the the Multinet POP Server. > > Is there some software for the IBM PCs that is just as good as Eudora and > will work with Multinet's POP Server ? > > - Bert Roseberry roseberry@eisner.decus.org - or - > US Coast Guard bert@mailstorm.dot.gov > > There is a version of Eudora for the PC's. Contact sdorner@qualcomm.com for more information. Steve Dorner is the author of Eudora, both the Mac and PC versions. || --------------------------------------------------------------------------- Jim A. Whitfill Mechanical & Electronic Eng. Division (505) 667-9282 whitfill@meediv.lanl.gov Los Alamos National Laboratory Opinions are MINE ONLY!! X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 09:25:26 PDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 11:23:56 -0500 Date: Tue, 17 Aug 1993 11:23:56 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: roseberry@eisner.decus.org CC: info-multinet@tgv.com, METZE@VMETZE.MRL.UIUC.EDU Subject: RE: MSDOS / Windows equiv of Eudora Yes; we have very good results with NUpop, which is public domain, but seems to be an excellent product. I don't normally go for public domain stuff, either. If you like I will have someone contact you about where to get a copy. Ginny Metze ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + Virginia Metze, not speaking for anyone but herself. + + metze@vmetze.mrl.uiuc.edu + ======================================================== | I remember you, Prairie Bayou! I knew you would | | die before you 'threw a clunker.' | ======================================================== From: SMTP%"roseberry@eisner.decus.org" 17-AUG-1993 11:21:36.10 To: METZE CC: Subj: MSDOS / Windows equiv of Eudora From: roseberry@eisner.decus.org X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: MSDOS / Windows equiv of Eudora Message-ID: <1993Aug17.074828.341@eisner.decus.org> ================================================================================ Archive-Date: xxx, 17 Aug 1993 07:48:28 -0400 Date: 17 Aug 93 07:48:28 -0400 Organization: DECUServe Lines: 9 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET The guys with Macs really like how Eudora works the the Multinet POP Server. Is there some software for the IBM PCs that is just as good as Eudora and will work with Multinet's POP Server ? - Bert Roseberry roseberry@eisner.decus.org - or - US Coast Guard bert@mailstorm.dot.gov X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 09:38:43 PDT Received: from psulias.psu.edu by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H1UOOH6J6OQPEX8L@psulias.psu.edu>; Tue, 17 Aug 1993 12:38:21 EDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 12:38:21 -0400 Date: Tue, 17 Aug 1993 12:38:21 -0400 (EDT) From: "J.Lance Wilkinson, 814-865-1818" Subject: Determining vendor of TCP/IP transport To: INFO-MULTINET@psulias.psu.edu, INFO-VAX@psulias.psu.edu Organization: Penn State University / University Libraries X-VMS-To: IMULTINET X-VMS-Cc: INFO_VAX,JLW MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII I have an application which might use TGV Multinet, or UCX, or Wollongong, or CMU/Tek or some other vendor's TCP/IP for its communications. Is there any way I can tell, *at run time*, which vendor's TCP/IP is being used? That is, is there a "getvendor()" socket call, or a global symbol with a specific value, or some such other heuristic that will let me determine which vendor's TCP/IP facilities are associated with the image? While the application is to run on both Unix and VMS, this particular aspect is only defined for VMS so non-Unix methods are permissable. +-"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+ | J.Lance Wilkinson ("Lance") BitNet: JLW@PSULIAS.BITNET | | Systems Design Specialist - Lead InterNet: jlw@psulias.psu.edu | | Library Computing Services AT&T:(814) 865-1818 FAX:(814)863-3560 | | E8 Pattee Library "I'd rather be dancing..." | | Penn State University A host is a host from coast to coast, | | University Park, PA 16802 And no one will talk to a host that's close | | Unless the host that isn't close | | Is busy, hung or dead. | +------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+ X-ST-Status: N Return-path: Received: from VGER.NIDDK.NIH.GOV ([128.231.60.11]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 10:15:38 PDT Received: from VGER.NIDDK.NIH.GOV by VGER.NIDDK.NIH.GOV (PMDF V4.2-13 #3741) id <01H1UQ6WO6FS8WW0VM@VGER.NIDDK.NIH.GOV>; Tue, 17 Aug 1993 13:15:32 EDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 13:15:32 -0400 Date: Tue, 17 Aug 1993 13:15:32 -0400 (EDT) From: Gerson H Cohen Subject: Re: MSDOS / Windows equiv of Eudora To: roseberry@eisner.decus.org Cc: info-multinet@tgv.com X-VMS-To: IN%"roseberry@eisner.decus.org" X-VMS-Cc: IN%INFO-MULTINET,GHC MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII > The guys with Macs really like how Eudora works the the Multinet POP Server. > > Is there some software for the IBM PCs that is just as good as Eudora and > will work with Multinet's POP Server ? > > - Bert Roseberry roseberry@eisner.decus.org - or - > US Coast Guard bert@mailstorm.dot.gov I have been using NUPOP from ftp.acns.nwu.edu and find it very good on the pc. Version 2.0 has been in beta for a while now and is due to be finished within weeks. Try it. Gerson Cohen - Dr Gerson H Cohen National Institutes of Health, Bldg 5, Room 335 9000 Rockville Pike, Bethesda, MD 20892 USA 'Phone: (301)402-4495 [Voice] (301)496-0201 [FAX] Internet: ghc@vger.niddk.nih.gov BITNet: ghc@nihklmb NIH DECNet: VGER::GHC X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 11:43:33 PDT Received: from THOR.INNOSOFT.COM by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H1UTB8WKLS8WX8N6@psulias.psu.edu>; Tue, 17 Aug 1993 14:43:26 EDT Received: from INNOSOFT.COM by INNOSOFT.COM (PMDF V4.2-14 #1336) id <01H1O5WK740W984I58@INNOSOFT.COM>; Tue, 17 Aug 1993 11:42:55 PST ================================================================================ Archive-Date: Tue, 17 Aug 1993 11:35:16 -0800 Date: Tue, 17 Aug 1993 11:35:16 -0800 (PST) From: Ned Freed Subject: Re: Determining vendor of TCP/IP transport To: JLW@psulias.psu.edu Cc: INFO-MULTINET@psulias.psu.edu, INFO-VAX@psulias.psu.edu MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII > I have an application which might use TGV Multinet, or UCX, or Wollongong, or > CMU/Tek or some other vendor's TCP/IP for its communications. Is there any > way I can tell, *at run time*, which vendor's TCP/IP is being used? That is, > is there a "getvendor()" socket call, or a global symbol with a specific value, > or some such other heuristic that will let me determine which vendor's TCP/IP > facilities are associated with the image? While the application is to run on > both Unix and VMS, this particular aspect is only defined for VMS so non-Unix > methods are permissable. I strongly recommend that you use the UCX socket interface. With the exception of CMU, all of these packages support this interface. This eliminates the need for any run time check. There really isn't any other reasonable way to do with a single image. Use of the private vendor interfaces usually involves linking against vendor-supplied shareable images, which in turns means your image won't work when the right transport isn't there. You could attempt to work around this with LIB$FIND_IMAGE_SYMBOL but it would be really nasty. You could try to do everything at the $QIO level, of course. This would mean duplicating a significant number of runtime routines supplied by most vendors in your application. This sort of thing is very very painful. But if you elect to do this the vendor name would be implicit in the device you assign a channel to. In conclusion, there isn't any getvendor routine because such a routine really isn't useful. Either you are dealing with a particular vendor's private interface, in which case the vendor is known implicitly, or you are dealing with a generic interface to multiple packages, in which case you should not need to know which vendor package it is. Ned X-ST-Status: N Return-path: Received: from po2.andrew.cmu.edu ([128.2.10.102]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 13:06:13 PDT Received: from localhost (postman@localhost) by po2.andrew.cmu.edu (8.5/8.5) id QAA10843; Tue, 17 Aug 1993 16:06:09 -0400 Received: via switchmail; Tue, 17 Aug 1993 16:06:07 -0400 (EDT) Received: from stern.phys.cmu.edu via qmail ID ; Tue, 17 Aug 1993 16:05:50 -0400 (EDT) Received: from stern.phys.cmu.edu via qmail ID ; Tue, 17 Aug 1993 16:05:41 -0400 (EDT) Received: from mms.0.1.23.EzMail.2.0.CUILIB.3.45.SNAP.NOT.LINKED.stern.phys.cmu.edu.sun4c.411 via MS.5.6.stern.phys.cmu.edu.sun4c_411; Tue, 17 Aug 1993 16:05:40 -0400 (EDT) Message-ID: ================================================================================ Archive-Date: Tue, 17 Aug 1993 16:05:41 -0400 Date: Tue, 17 Aug 1993 16:05:41 -0400 (EDT) From: Shing-Te Li To: info-multinet@tgv.com Subject: mm read and finger Cc: Hi, I just installed multinet v3.2 in our vax cluster (vax3200's running vms5.2), and I have some problems with the behavior of mm & finger. 1. Even I active mm to receive all the mails, for those mails sent locally from "mail", they will goto "mail" instead of "mm". 2. In mm, when I read a long mail, it won't stop at every page as "mail" did, is there a way to read page by page in mm ? 3. For people still use "mail" to receive their mails, the finger program will not tell them that they have new mails. I think these are very basic questions, since I am new to multinet, and I can't find the solutions from the online help, any command or suggestion will be appreciated. Shing-Te Li Dept. of Physics Carnegie Mellon University X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 14:01:30 PDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 14:01:29 -0700 Date: Tue, 17 Aug 1993 14:01:29 -0700 (PDT) From: Matt Madison To: sl2x+@andrew.cmu.edu CC: info-multinet@TGV.COM Subject: RE: mm read and finger > I just installed multinet v3.2 in our vax cluster (vax3200's running >vms5.2), and I have some problems with the behavior of mm & finger. >1. Even I active mm to receive all the mails, for those mails sent >locally from "mail", they will goto "mail" instead of "mm". Correct. You can fix this by having each MM user enter the following command in VMS MAIL: MAIL> SET FORWARD SMTP%username This will force VMS MAIL to forward local mail into MultiNet's mailer, which will in turn deliver mail to MM. Just make sure that if users decide to stop using MM, they use SET NOFORWARD to disable this forwarding. >2. In mm, when I read a long mail, it won't stop at every page as "mail" >did, is there a way to read page by page in mm ? MM> SET MORE-PROCESSING 1 >3. For people still use "mail" to receive their mails, the finger >program will not tell them that they have new mails. Correct. Finger's new mail display only works if you use MM. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 14:13:24 PDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 16:12:04 -0500 Date: Tue, 17 Aug 1993 16:12:04 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: system-wide FINGER Does anyone have a 'system-wide' FINGER implementation they would be willing to share. I would like to type FINGER and have it, by default, print out who is logged on to all of our 50 VMS and UNIX machines. Ginny Metze ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + Virginia Metze, not speaking for anyone but herself. + + metze@vmetze.mrl.uiuc.edu + ======================================================== | I remember you, Prairie Bayou! I knew you would | | die before you 'threw a clunker.' | ======================================================== End of returned message X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 17:24:15 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: gopher client problem Message-ID: <1993Aug17.155600.1@admvax.sonoma.edu> From: jpz@admvax.sonoma.edu ================================================================================ Archive-Date: xxx, 17 Aug 1993 15:56:00 -0800 Date: 17 Aug 93 15:56:00 -0800 Organization: Sonoma State University Nntp-Posting-Host: admvax.sonoma.edu Lines: 23 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I'm trying to get a gopher client running on my VMS 5.5-2/Multinet 3.2C system. I've tried both clients from boombox.micro.umn.edu, 0.6 and (gopher+)1.12. Both compile and link OK. When I run, both give the error: Cannot connect to host any.gopher.server I've tried several known good hosts with the same result. Also, $ telnet any.gopher.server /port=70 gives me the (unformatted) main gopher menu from the specified host. Am I missing something obvious that needs to be done (to Multinet?) for the gopher client to work correctly? E-mail or post. Thanks in advance for any help you can provide. Jack Ziegler | Computing Services | internet: jpz@sonoma.edu Sonoma State University | phone : (707)664-3098 Rohnert Park, CA 94928 | FAX : (707)664-2505 X-ST-Status: N Return-path: Received: from vax2.cstp.umkc.edu ([134.193.2.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 18:05:10 PDT Received: from vax2.cstp.umkc.edu by vax2.cstp.umkc.edu (PMDF #3182 ) id <01H1V4BN0YM89D4C9O@vax2.cstp.umkc.edu>; Tue, 17 Aug 1993 20:05:04 CDT ================================================================================ Archive-Date: xxx, 17 Aug 1993 20:05:04 -0500 Date: 17 Aug 1993 20:05:04 -0500 (CDT) From: Brian McKeever Subject: CKermit compilation problem. To: INFO-MULTINET@TGV.COM Organization: University of Missouri - Kansas City, CSTP X-Envelope-to: INFO-MULTINET@TGV.COM X-VMS-To: IN%"INFO-MULTINET@TGV.COM" MIME-version: 1.0 I'm trying to compile C-KERMIT 5A(188) (Dated Dec. 1992) on a DECstation 3000-400 and keep getting the following: typedef long time_t; .............^ %CC-E-DECREDEC, In this declaration, time_t has no linkage and has a prior declaration in this scope at line number 33 in file SYS$COMMON:[SYSLIB]DECC$RTLDEF.TLB;1. at line number 33 in file MULTINET_COMMON_ROOT:[MULTINET.INCLUDE.SYS]TYPES.H;1 Anyone know what's going on here? All the other C source compiled without any major problems (just a couple of warnings about a macro redefinition). Thanks! -=Brian X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 17 Aug 1993 18:14:56 PDT Date: Tue, 17 Aug 93 18:14:56 PDT From: mahan @ TGV.COM (Patrick L. Mahan) Subject: RE: CKermit compilation problem. To: info-multinet @ TGV.COM, mckeever @ vax2.cstp.umkc.edu X-ST-Vmsmail-To: ST%"info-multinet" # # I'm trying to compile C-KERMIT 5A(188) (Dated Dec. 1992) on a DECstation # 3000-400 and keep getting the following: # # typedef long time_t; # .............^ # %CC-E-DECREDEC, In this declaration, time_t has no linkage and has a # prior declaration in this scope at line number 33 in file # SYS$COMMON:[SYSLIB]DECC$RTLDEF.TLB;1. # at line number 33 in file # MULTINET_COMMON_ROOT:[MULTINET.INCLUDE.SYS]TYPES.H;1 # # Anyone know what's going on here? All the other C source compiled without # any major problems (just a couple of warnings about a macro redefinition). # # Thanks! # Because the MULTINET_ROOT:[MULTINET.INCLUDE.SYS]TYPES.H needs is incorrect. A corrected version is at the end of this message. It is fixed in the next release of MultiNet. Patrick L. Mahan --- TGV Window Washer ------------------------------- Mahan@TGV.COM --------- Waking a person unnecessarily should not be considered - Lazarus Long a capital crime. For a first offense, that is From the Notebooks of Lazarus Long ================ MULTINET_ROOT:[MULTINET.INCLUDE.SYS]TYPES.H ================= /* * Copyright (c) 1982, 1986 Regents of the University of California. * All rights reserved. The Berkeley software License Agreement * specifies the terms and conditions for redistribution. * * @(#)types.h 7.1 (Berkeley) 6/4/86 */ #ifndef _TYPES_ #define _TYPES_ /* * Basic system types and major/minor device constructing/busting macros. */ typedef unsigned char u_char; typedef unsigned short u_short; typedef unsigned int u_int; typedef unsigned long u_long; typedef unsigned short ushort; /* SYS III compatibility */ #if !defined(__CADDR_T) && !defined(CADDR_T) #ifndef __CADDR_T #define __CADDR_T #endif /* !__CADDR_T */ #ifndef CADDR_T #define CADDR_T #endif /* !CADDR_T */ typedef char * caddr_t; #endif /* !__CADDR_T && !CADDR_T */ #ifdef __ALPHA #ifndef __TIME_T #define __TIME_T 1 typedef unsigned long int time_t; #endif /* __TIME_T */ #else #ifndef __TYPES #ifndef __TYPES_LOADED #ifdef __ALPHA typedef unsigned long time_t; #else typedef long time_t; #endif #define __TYPES #define __TYPES_LOADED #endif /* __TYPES_LOADED */ #endif /* __TYPES */ #endif /* __ALPHA */ #define NBBY 8 /* number of bits in a byte */ /* * Select uses bit masks of file descriptors in longs. * These macros manipulate such bit fields (the filesystem macros use chars). * FD_SETSIZE may be defined by the user, but the default here * should be >= CHANNELCNT (SYSGEN parameter) */ #ifndef FD_SETSIZE #define FD_SETSIZE 512 #endif #define CHANNELSIZE 16 /* Size of a channel */ typedef long fd_mask; #define NFDBITS (sizeof(fd_mask) * NBBY) /* bits per mask */ #ifndef howmany #define howmany(x, y) (((x)+((y)-1))/(y)) #endif typedef struct fd_set { fd_mask fds_bits[howmany(FD_SETSIZE, NFDBITS)]; } fd_set; #define FD_SET(n, p) ((p)->fds_bits[(n)/CHANNELSIZE/NFDBITS] |= (1 << (((n)/CHANNELSIZE) % NFDBITS))) #define FD_CLR(n, p) ((p)->fds_bits[(n)/CHANNELSIZE/NFDBITS] &= ~(1 << (((n)/CHANNELSIZE) % NFDBITS))) #define FD_ISSET(n, p) ((!((n) % CHANNELSIZE)) && ((p)->fds_bits[(n)/CHANNELSIZE/NFDBITS] & (1 << (((n)/CHANNELSIZE) % NFDBITS)))) #define FD_ZERO(p) bzero((char *)(p), sizeof(*(p))) #endif /* _TYPES_ */ X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 19:56:58 PDT From: terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr.) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: system-wide FINGER Message-ID: <1993Aug17.222638.6793@spcvxb.spc.edu> ================================================================================ Archive-Date: xxx, 17 Aug 1993 22:26:38 EDT Date: 17 Aug 93 22:26:38 EDT Organization: St. Peter's College, US Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930817161204.69@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: > Does anyone have a 'system-wide' FINGER implementation they would > be willing to share. I would like to type FINGER and have it, > by default, print out who is logged on to all of our 50 VMS and > UNIX machines. You probably want something like "rwho" rather than FINGER. Rwho works by broadcasting status updates, while FINGER would have to query each host when somebody ran the program. You can grab a copy of my UCX/MultiNet rwho port from ftp.spc.edu in the [.ucx] directory. Terry Kennedy Operations Manager, Academic Computing terry@spcvxa.bitnet St. Peter's College, Jersey City, NJ USA terry@spcvxa.spc.edu +1 201 915 9381 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 20:17:27 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: kelly_d@summer.chem.su.oz.au Subject: problem receiving mail from mailserver Message-ID: <1993Aug18.114826.1@summer.chem.su.oz.au> Sender: news@ucc.su.OZ.AU Nntp-Posting-Host: summer.chem.su.oz.au Organization: School of Chemistry, University of Sydney ================================================================================ Archive-Date: Wed, 18 Aug 1993 16:48:26 GMT Date: Wed, 18 Aug 1993 16:48:26 GMT Lines: 31 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi Guys, We have a problem receiving some mail items from our mail server. We're running Multinet V3.2 Rev B on VMS 5.5-2 . Our mail server is an Apollo running Domain/OS 10.4 with a slightly modified sendmail version 5.61 . Tha Apollo (adder.chem.su.oz.au) manages to connect to our Vax (summer.chem.su.oz.au), receives its banner back and sends a HELO command as follows : 220 summer.chem.su.oz.au TGV MultiNet V3.2 Rev B SMTP service ready at Tue, 17 Aug 1993 9:41:32 +1000 (EST) >>> HELO adder.maths.su.oz.au After that, the session freezes, either because the HELO command never gets to summer, or because summer's response never comes back. This mailserver worked fine for us with Multinet V3.0 and VMS 5.4-3 . Our Multinet configuration under V3.2 is identical to the V3.0 configuration. Any suggestions much appreciated. Thanks, Dave. ------------------------------------------------------------------- Dr David Kelly e-mail: kelly_d@summer.chem.su.oz.au School of Chemistry, phone : +61 2 692 4852 University of Sydney. fax : +61 2 692 3329 X-ST-Status: N Return-path: Received: from relay2.geis.com ([192.77.188.3]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 21:34:51 PDT Received: by relay2.geis.com (1.37.109.4/15.6) id AA11733; Wed, 18 Aug 93 05:34:41 +0100 From: c.ward7@genie.geis.com Message-Id: <9308180434.AA11733@relay2.geis.com> ================================================================================ Archive-Date: Wed, 18 Aug 1993 05:01:00 BST Date: Wed, 18 Aug 93 05:01:00 BST To: info-multinet@tgv.com Subject: MM SEND Question X-Genie-Id: 4859964 X-Genie-From: C.WARD7 Is it possible to get MM SEND to accept all arguments from the command line ala VMS MAIL? I'd like to include something like: $ MM SEND /SUBJECT="Some Text" MESSAGE.FILE ADDRESS_LIST in a DCL script. Thanks, Craig -- c.ward7@genie.geis.com 08:25 p.m. Tuesday 08/17/93. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 03:29:00 PDT From: tor@orwell.tmih.no (Tor-Arne N|stdal 2r) X-Newsgroups: vmsnet.networks.tcp-ip.multinet,comp.sys.dec Subject: Re: need assistance with rdump stuff ================================================================================ Archive-Date: xxx, 18 Aug 1993 10:21:35 GMT Date: 18 Aug 1993 10:21:35 GMT Organization: UniNett Lines: 40 Distribution: world Reply-To: tor@tmih.no NNTP-Posting-Host: orwell.tmih.no To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET -- rdump and multiple partitions: SITE: Ultrix 4.x / DECstations / csh / EXAbyte scsi tape drive I am using rdump with multiple partitions from multiple sites into one 8 mm tape. Using : commands: mt, dd, rdump, rrestore device: /dev/nrmt0h connected to a 8mm EXAbyte drive approx. 5GB Labeling tape: echo $LABEL | dd obs=1024 of=/dev/nrmt0h This $LABEL contains a list of dumped partitions and order. Dumping: rdump 0uf / /dev/nrmt0h (remember to use a norewind device) and so on... Then i use mt command with fsf parameter to skip to the appropriate partition to restore. set LABEL=`dd ibs=1024 if=/dev/nrmt0h` echo out a menu for a specific partition and skip forward to it with mt fsf # run rrestore -if /dev/nrmt0h Surely scripts around everything to keep properly order/track of things is vital (my opinion thus i have to do it manually from time to time 8-) +----------------------------------+ | Tor-Arne Nostdal |To be or not to be - W.Shakespeare | Telemark Institute of Technology | 2B&^2B - ASCII | Mail: tor@tmih.no | do be do ba - F.Sinatra +----------------------------------+ X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 04:34:50 PDT Received: from psulias.psu.edu by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H1VSIJHERK8ZDVQL@psulias.psu.edu>; Wed, 18 Aug 1993 07:34:39 EDT ================================================================================ Archive-Date: Wed, 18 Aug 1993 07:34:39 -0400 Date: Wed, 18 Aug 1993 07:34:39 -0400 (EDT) From: "J.Lance Wilkinson, 814-865-1818" Subject: Re: gopher client problem To: jpz@sonoma.edu, INFO-MULTINET@psulias.psu.edu Organization: Penn State University / University Libraries X-VMS-To: IN%"jpz@sonoma.edu" X-VMS-Cc: IMULTINET,JLW MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Jack Zeigler asked about the gopher clients on boombox. Please don't use the 0.6 client on boombox. aFTP to Niord.SHSU.edu and pick up the .ZIP files there for a gopher client that works. A gopher+ client that works will be on boombox shortly (I'm told it's been uploaded, it just must not be accessible yet). We on the VMSGopher-L@trln.lib.unc.edu "VMS Gopher Porting Interest Group" have been busy trying to insure that both a client *and* a server work well on VMS with Multinet, UCX or Wollongong. Archives exist at Niord.SHSU.edu for both Gopher and aFTP, while Gopher only access for a server archive is at psulias.psu.edu, and Gopher only access for both is at trln.lib.unc.edu. +-"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+ | J.Lance Wilkinson ("Lance") BitNet: JLW@PSULIAS.BITNET | | Systems Design Specialist - Lead InterNet: jlw@psulias.psu.edu | | Library Computing Services AT&T:(814) 865-1818 FAX:(814)863-3560 | | E8 Pattee Library "I'd rather be dancing..." | | Penn State University A host is a host from coast to coast, | | University Park, PA 16802 And no one will talk to a host that's close | | Unless the host that isn't close | | Is busy, hung or dead. | +------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+ X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 05:25:46 PDT ================================================================================ Archive-Date: Wed, 18 Aug 1993 7:24:26 -0500 Date: Wed, 18 Aug 1993 7:24:26 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: rwho Maybe that would work better. Generally the rwho info returned from the UNIX machines has been appallingly incorrect... but I will try again. Ginny From: SMTP%"terry@spcvxb.spc.edu" 17-AUG-1993 22:27:02.73 To: METZE CC: Subj: Re: system-wide FINGER From: terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr.) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: system-wide FINGER Message-ID: <1993Aug17.222638.6793@spcvxb.spc.edu> ================================================================================ Archive-Date: xxx, 17 Aug 1993 22:26:38 EDT Date: 17 Aug 93 22:26:38 EDT Organization: St. Peter's College, US Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930817161204.69@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: > Does anyone have a 'system-wide' FINGER implementation they would > be willing to share. I would like to type FINGER and have it, > by default, print out who is logged on to all of our 50 VMS and > UNIX machines. You probably want something like "rwho" rather than FINGER. Rwho works by broadcasting status updates, while FINGER would have to query each host when somebody ran the program. You can grab a copy of my UCX/MultiNet rwho port from ftp.spc.edu in the [.ucx] directory. Terry Kennedy Operations Manager, Academic Computing terry@spcvxa.bitnet St. Peter's College, Jersey City, NJ USA terry@spcvxa.spc.edu +1 201 915 9381 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 07:41:31 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet,vmsnet.sources.d Subject: Where could I get Archie sources? Message-ID: <1993Aug18.072047.7253@mcclb0.med.nyu.edu> From: smith@mcclb0.med.nyu.edu ================================================================================ Archive-Date: xxx, 18 Aug 1993 12:20:47 GMT Date: 18 Aug 93 12:20:47 GMT Organization: NYU Medical Center, New York, NY 10016, USA Lines: 8 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Two questions: Does anyone know where I could get the sources for Archie, both client and server (ideally for Multinet, but not absolutely necessary). Also, does anyone know how people register the fact that they have s/w that they want made available though the Archie d/b? Thanks. +----------------------------------------------------------------------------+ |Ross Smith, Research Computing Resource, Department of Cell Biology, NYU-MC| |E-Mail: SMITH@NYUMED.BITNET (BITNET), SMITH@MCCLB0.MED.NYU.EDU (Internet)| +----------------------------------------------------------------------------+ X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 07:43:25 PDT ================================================================================ Archive-Date: Wed, 18 Aug 1993 7:43:07 -0700 Date: Wed, 18 Aug 1993 7:43:07 -0700 (PDT) From: Matt Madison To: kelly_d@summer.chem.su.oz.au CC: INFO-MULTINET@TGV.COM Subject: RE: problem receiving mail from mailserver >We have a problem receiving some mail items from our mail server. >We're running Multinet V3.2 Rev B on VMS 5.5-2 . >Our mail server is an Apollo running Domain/OS 10.4 with a slightly modified >sendmail version 5.61 . > >Tha Apollo (adder.chem.su.oz.au) manages to connect to our Vax >(summer.chem.su.oz.au), receives its banner back and sends a HELO command as >follows : > >220 summer.chem.su.oz.au TGV MultiNet V3.2 Rev B SMTP service ready at Tue, 17 >Aug 1993 9:41:32 +1000 (EST) >>>> HELO adder.maths.su.oz.au > >After that, the session freezes, either because the HELO command never gets to >summer, or because summer's response never comes back. > >This mailserver worked fine for us with Multinet V3.0 and VMS 5.4-3 . Our >Multinet configuration under V3.2 is identical to the V3.0 configuration. The problem is probably that the Apollo is ending its HELO command with just a bare linefeed rather than a carriage return/linefeed pair, like it is supposed to. You can get MultiNet to accept this by defining a logical name: $ DEFINE/SYSTEM/EXEC MULTINET_SMTP_ACCEPT_UNIX_LF_BRAIN_DAMAGE TRUE -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 07:49:27 PDT ================================================================================ Archive-Date: Wed, 18 Aug 1993 7:49:16 -0700 Date: Wed, 18 Aug 1993 7:49:16 -0700 (PDT) From: Matt Madison To: c.ward7@genie.geis.com CC: INFO-MULTINET@TGV.COM Subject: RE: MM SEND Question >Is it possible to get MM SEND to accept all arguments from the >command line ala VMS MAIL? I'd like to include something like: > >$ MM SEND /SUBJECT="Some Text" MESSAGE.FILE ADDRESS_LIST > >in a DCL script. Unfortunately, no. I'll look into allowing something like this for a future release, if it all possible. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 09:25:17 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Multinet LPR support Message-ID: <11AUG199312435370@ariel.lerc.nasa.gov> From: uugblum@ariel.lerc.nasa.gov (Greg Blumers) ================================================================================ Archive-Date: xxx, 11 Aug 1993 12:43 EST Date: 11 Aug 1993 12:43 EST Sender: uugblum@ariel.lerc.nasa.gov (Greg Blumers [3-6777]) Distribution: world Organization: NASA Lewis Research Center NNTP-Posting-Host: ariel.lerc.nasa.gov News-Software: VAX/VMS VNEWS 1.41 Lines: 11 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Is there a way to cancel a remote print job (lprm) which was submitted through a Multinet LPD print queue? -------------------------------------------------------------------------------- Greg Blumers Sverdrup Technology, Inc. Systems Programmer c/o NASA Lewis Research Center (216)433-6777 Mail Stop 142-2 21000 Brookpark Road uugblum@scivax.lerc.nasa.gov Cleveland, OH 44135 -------------------------------------------------------------------------------- X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 09:25:18 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: VAX/VMS DNS NameServer ? Message-ID: From: martineau@MacMartineau.ccr.hydro.qc.ca (Alain Martineau) ================================================================================ Archive-Date: Wed, 18 Aug 1993 14:58:08 GMT Date: Wed, 18 Aug 1993 14:58:08 GMT Sender: news@ireq.hydro.qc.ca (Netnews Admin) Followup-To: vmsnet.networks.tcp-ip.multinet Organization: Hydro Quebec Lines: 10 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET The group that provided nameservice for the company decided it is too much work and will dump us shortly. I must find a way to run a nameserver here. We don't have any unix systems available for that, only MVS and VMS servers. Is there a nameserver implementation on VMS ? Thanks. Alain Martineau Hydro Quebec amartineau@nccr.ccr.hydro.qc.ca X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 10:05:51 PDT ================================================================================ Archive-Date: Wed, 18 Aug 1993 10:05:48 -0700 Date: Wed, 18 Aug 1993 10:05:48 -0700 (PDT) From: Matt Madison To: uugblum@ariel.lerc.nasa.gov CC: info-multinet@TGV.COM Subject: RE: Multinet LPR support >Is there a way to cancel a remote print job (lprm) which was submitted >through a Multinet LPD print queue? Not currently. We will be adding an LPRM command in V3.3. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 10:43:35 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: VAX/VMS DNS NameServer ? Message-ID: <1993Aug18.161943.15245@news.arc.nasa.gov> From: madison@tgv.com (Matt Madison) ================================================================================ Archive-Date: Wed, 18 Aug 1993 16:19:43 GMT Date: Wed, 18 Aug 1993 16:19:43 GMT Reply-To: madison@tgv.com Sender: usenet@news.arc.nasa.gov Organization: The Foundation for Dinner Peas Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article , martineau@MacMartineau.ccr.hydro.qc.ca (Alain Martineau) writes: >The group that provided nameservice for the company decided it is too much >work and will dump us shortly. I must find a way to run a nameserver here. >We don't have any unix systems available for that, only MVS and VMS >servers. Is there a nameserver implementation on VMS ? MultiNet comes with a full name server implementation. Check out the System Administrators' Guide (the chapter on DNS and host tables) for more information. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from TEX.AC.UK ([134.151.40.18]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 10:43:58 PDT Received: by TeX.ac.uk (MX V3.3 VAX) id 5112; Wed, 18 Aug 1993 17:43:51 GMT ================================================================================ Archive-Date: Wed, 18 Aug 1993 17:43:49 GMT Date: Wed, 18 Aug 1993 17:43:49 GMT From: UK TeX Archive Manager To: info-multinet@tgv.com Subject: RE: VAX/VMS DNS NameServer In message dated Wed, 18 Aug 1993 14:58:08 GMT, martineau@MacMartineau.ccr.hydro.qc.ca (Alain Martineau) wrote: > The group that provided nameservice for the company decided it is too much > work and will dump us shortly. I must find a way to run a nameserver here. > We don't have any unix systems available for that, only MVS and VMS > servers. Is there a nameserver implementation on VMS ? But of course! Your message was circulated on the info-multinet mailing list, presumably gatewayed from vmsnet.networks.tcp-ip.multinet, so you're aware of the existence of MultiNet; perhaps you even use it? Anyway, it's got a fully- functional named implementation, which *this* machine TeX.ac.uk is using quite happily to serve the TeX.ac.uk and 78.63.193.in-addr.arpa domains. > Thanks. > > Alain Martineau > Hydro Quebec > amartineau@nccr.ccr.hydro.qc.ca Two points to make here: 1) by ftp'ing to nccr.ccr.hydro.qc.ca, I discovered that that machine, at least, is running that UCX rubbish, V2. Chuck it out, install MultiNet in its place, and then you'll have the nameserver you want ;-) 2) If you want your mail to go there, why don't you arrange to send it with a From or Reply-To header saying so; then I wouldn't have received it as being from MacMartineau.ccr.hydro.qc.ca. I wouldn't be surprised to hear that that machine isn't designed to receive mail directly. (In fact, I've just realized that I started off sending this with REPLY/ED/EX, so I'll now have to do a SEND/LAST instead to info-multinet. Come on TGV, why don't you add something to your mail exploder so that *you* add a Reply-To pointing to the list instead of the contributor? In my experience, most other lists do this, and it makes replying to postings much easier.) Brian {Hamilton Kelly} System Manager for the (VMS) UK TeX Archive at Aston University X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 13:25:28 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Determining vendor of TCP/IP transport Message-ID: <1993Aug18.143332.2101@dmc.com> From: munroe@dmc.com (Dick Munroe) ================================================================================ Archive-Date: xxx, 18 Aug 1993 14:33:32 EDT Date: 18 Aug 93 14:33:32 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 15 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <01H1UN0IZGNY984I58@INNOSOFT.COM>, NED@INNOSOFT.COM (Ned Freed) writes: > I strongly recommend that you use the UCX socket interface. With the exception > of CMU, all of these packages support this interface. This eliminates the need > for any run time check. Instead of the UCX socket interface, try NETLIB, available from ftp.spc.edu or fileserv@wkuvx1.bitnet. It's an interface to ALL TCP/IP packages and it doesn't rely on the "goodness" of someone's UCX emulation. -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 16:02:47 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: XDM on MicroVAX Message-ID: <1993Aug18.204440.1232@mksol.dseg.ti.com> From: pyron@skndiv.dseg.ti.com (Dillon Pyron) ================================================================================ Archive-Date: Wed, 18 Aug 1993 20:44:40 GMT Date: Wed, 18 Aug 1993 20:44:40 GMT Reply-To: pyron@skndiv.dseg.ti.com Sender: usenet@mksol.dseg.ti.com (Usenet News) Organization: TI/DSEG VAX Support Nntp-Posting-Host: skndiv.dseg.ti.com Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I'm trying to get NCD's XDM fired up on a MicroVAX, but all it does is flash for a few seconds, then put the chooser back up. Any suggestions? The MV is running MultiNet, and this group is the most knowledgable in this area. Thanks. -- Dillon Pyron | The opinions expressed are those of the TI/DSEG Lewisville VAX Support | sender unless otherwise stated. (214)462-3556 (when I'm here) | (214)492-4656 (when I'm home) |Please send mail to pyron@dseg.ti.com pyron@dseg.ti.com |since skndiv is going away. Thanks PADI AI-54909 | X-ST-Status: N Return-path: ================================================================================ Archive-Date: Wed, 18 Aug 1993 17:32:42 PDT Date: Wed, 18 Aug 93 17:32:42 PDT From: mahan @ TGV.COM (Patrick L. Mahan) Subject: RE: XDM on MicroVAX To: info-multinet @ TGV.COM, pyron @ skndiv.dseg.ti.com X-ST-Vmsmail-To: ST%"info-multinet" # # I'm trying to get NCD's XDM fired up on a MicroVAX, but all it does is flash # for a few seconds, then put the chooser back up. Any suggestions? # # The MV is running MultiNet, and this group is the most knowledgable in this # area. Thanks. # Check to make sure the XDM server is running on the MicroVAX. Also, you may need to modify the file NCD_ROOT:[BIN]NCD_XDM_TCPIP_INSTALL.COM so that it creates system logicals as opposed to group logicals. You can contact support@ncd.com for more help. Patrick L. Mahan --- TGV Window Washer ------------------------------- Mahan@TGV.COM --------- Waking a person unnecessarily should not be considered - Lazarus Long a capital crime. For a first offense, that is From the Notebooks of Lazarus Long X-ST-Status: N Return-path: Received: from DCV4KD.PHS.COM ([149.111.1.4]) by TGV.COM via INTERNET ; Wed, 18 Aug 93 19:36:22 PDT Received: from DCV4KD.PHS.COM by DCV4KD.PHS.COM (PMDF V4.2-10 #4056) id <01H1WHRO5C0000095T@DCV4KD.PHS.COM>; Wed, 18 Aug 1993 19:36:18 PST ================================================================================ Archive-Date: Wed, 18 Aug 1993 19:36:18 -0800 Date: Wed, 18 Aug 1993 19:36:18 -0800 (PST) From: Urban Surfer Subject: Re: XDM on MicroVAX To: info-multinet@tgv.com X-VMS-To: IN%"info-multinet@tgv.com" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII >I'm trying to get NCD's XDM fired up on a MicroVAX, but all it does is flash >for a few seconds, then put the chooser back up. Any suggestions? What version of NCDware and VMS are you running? I've had some problems with the XDM server in the past, but I have it running now with NCDware 3.0 and VMS 6.0. Matt Holdrege holdrege@phs.com MH235 X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 04:25:51 PDT Via: uk.ac.tex; Thu, 19 Aug 1993 12:25:28 +0100 ================================================================================ Archive-Date: Thu, 19 AUG 1993 11:25:15 GMT Date: Thu, 19 AUG 93 11:25:15 GMT From: SYSTEM@TEX.AC.UK To: info-multinet Subject: RE: MULTINET_SERVER looping Sender: JANET "SYSTEM@UK.AC.TEX" Originally-to: INTERNET%"info-multinet@com.tgv" Originally-from: SYSTEM "UK TeX Archive Manager " Mailer: Janet_Mailshr V3.5 ( 13-OCT-1989 14:07:27 ) In a private mailing to me of Tue, 1 Jun 1993 10:37:28 +0200 Ben Segal wrote: > Hallo Brian, > After reading your plea on the mailing list, we had quite a > bad time with this problem at CERN and tracked it down here to the BIND > server looping. If you'd like more info, please let me know - it could > well be your problem if you run local cacheing-only DNS servers with > forwarders: do you? Well, at the time (when MULTINET_SERVER went into a loop about 1--2 times a week) we were indeed running a caching-only server, with forwarders. For the past month, I've been primary nameserver for my own domain, and the server hasn't once gone into a race state. So it looks as though there may be something significant in what Ben wrote. Brian {Hamilton Kelly} System Manager for the (VMS) UK TeX Archive at Aston University PS I expect the damned thing will prove me wrong just as soon as this mail leaves me :-J X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 10:01:07 PDT Received: from psulias.psu.edu by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H1XIBNBWWWHXIRGL@psulias.psu.edu>; Thu, 19 Aug 1993 13:00:53 EDT ================================================================================ Archive-Date: Thu, 19 Aug 1993 13:00:53 -0400 Date: Thu, 19 Aug 1993 13:00:53 -0400 (EDT) From: "J.Lance Wilkinson, (814) 865-1818" Subject: Re: Determining vendor of TCP/IP transport To: INFO-MULTINET@psulias.psu.edu Cc: monroe@dmc.com, ned@innosoft.com Organization: Penn State University / University Libraries X-VMS-To: IMULTINET X-VMS-Cc: in%"monroe@dmc.com", in%"ned@innosoft.com" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII On INFO-MULTINET, in response to my query on "Determining vendor of TCP/IP transport": munroe@dmc.com (Dick Munroe) replied to >In article <01H1UN0IZGNY984I58@INNOSOFT.COM>, NED@INNOSOFT.COM (Ned Freed) > writes: >> I strongly recommend that you use the UCX socket interface. With the exception >> of CMU, all of these packages support this interface. This eliminates the need >> for any run time check. > >Instead of the UCX socket interface, try NETLIB, available from ftp.spc.edu or >fileserv@wkuvx1.bitnet. It's an interface to ALL TCP/IP packages and it >doesn't rely on the "goodness" of someone's UCX emulation. Ned & Dick, I appreciate the concern, but at present a lot of people did a lot of work to get the software to work specifically with these different services. My sole concern is for one part of the code which is reporting what agent it is running with that is part of a response to a query from a client that it's servicing. Even if and when we incorporate NETLIB into the package, I'll still want it to be able to determine if the base TCP/IP product its running over is Multinet, or UCX, or whatever and *tell the client that asked.* Mike Kelsey (I think on INFO-VAX) indicated that Miguel Marquina at CERN had developed a .COM file that tests for which agent is running and reports it. This involves looking for various agent- specific logicals or some such thing. I'm waiting for a response from the message Mike forwarded to Miguel at CERN to see if this will solve the requirements. Since each TCP/IP agent currently requires inclusion of a agent-specific options file in the link, I've considered some kind of symbol in the options file, that the program could test for using nlist(). But when I saw that the symbols are numeric values only, it put a damper on that. For the moment I'm living with compile-time #ifdef's since there's a TCP/IP agent specification required to compile most modules of the source code. I may leave it at that, but I'd really like to be able to determine not just Multinet or UCX, for example, but Multinet 3.2 and UCX 3.0. That requires more than just compile-time #ifdef's. But maybe there's agent-specific header file that could be included to get the information, too? +-"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+ | J.Lance Wilkinson ("Lance") BitNet: JLW@PSULIAS.BITNET | | Systems Design Specialist - Lead InterNet: jlw@psulias.psu.edu | | Library Computing Services AT&T:(814) 865-1818 FAX:(814)863-3560 | | E8 Pattee Library "I'd rather be dancing..." | | Penn State University A host is a host from coast to coast, | | University Park, PA 16802 And no one will talk to a host that's close | | Unless the host that isn't close | | Is busy, hung or dead. | +------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+ X-ST-Status: N Return-path: Received: from jpl-mil ([137.78.160.9]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 11:06:16 PDT Received: from venom (venom.jpl.nasa.gov) by jpl-mil (4.1/SMI-4.1+xDXRm2.2) id AA06685; Thu, 19 Aug 93 11:03:02 PDT Received: by venom (1.37.109.4/1.34) id AA01615; Thu, 19 Aug 93 11:05:50 -0700 ================================================================================ Archive-Date: Thu, 19 Aug 1993 11:05:50 -0700 Date: Thu, 19 Aug 93 11:05:50 -0700 From: jjt@venom.Jpl.Nasa.Gov (Jack J. Tuszynski) To: INFO-MULTINET@TGV.COM Subject: mail forwarding from VMS to Unix workstation Dear TGV Multinet technical support, Vax Name: Grouch Serial Number: SVA2220 On our Vax VMS 5.3, Multinet 3.1, several users forward their mail to their Unix workstations. They do so using the following command in mail. MAIL> SET FORWARD SMTP%"""JJT@VENOM.JPL.NASA.GOV""" Notice the tripple quotation marks. Without three quotes, the forward won't work. Anyway, the forwarding works fine for individual messages. But when I send a message to all users on the Vax, and there are over 100, then the "to: " message on the Unix workstation includes all of the Vax users that I send the mail to. My question to you is how to forward mail properly from a Vax to a Unix workstation? Also, how does one avoid including everybody on the Vax in the "to: " message, when I send a mail message to everyone on the Vax? Below is the command (proper VMS) that I use on the Vax to mail to everyone. $ MAIL/SUBJECT="USER DISK 98% Full" MESSAGE98.TXT "@SYS$MANAGER:ALLUSERS.DIS" Jack Tuszynski jjt@venom.jpl.nasa.gov X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 12:17:10 PDT From: ironsd@aphst1.saic.com (Don Irons) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: MTU setting on SLIP port Message-ID: <1993Aug19.132758.66@aphst1.saic.com> ================================================================================ Archive-Date: xxx, 19 Aug 1993 13:27:58 -05:00 Date: 19 Aug 93 13:27:58 -05:00 Organization: Science Applications International Corp. Lines: 13 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Is the MTU settable on Multinet SLIP circuits configured on VAX TXA terminal ports? The System Admin guide (ch.2) makes no mention of MTU when configuring SLIP circuits. When I configured my circuit between two networks (a VAX at one end, a DECserver running SLIP at the other) the Multinet circuit port came up with an MTU of 296. I configured the DECserver end (which is settable) to that same value even though the DECserver end has a maximum of 1006. Is there a reason why it comes up set to 296? I have read several discussions recently about performance of SLIP circuits all of which recommended much higher settings for MTU. If the MTU on Multinet SLIP interfaces is settable, does the change require a reboot of the VAX? Don Irons "Opinions are mine alone." ironsd@aphst1.saic.com X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 12:28:50 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: tn3270, numeric keypad Message-ID: <19AUG199312525288@vz.cis.umn.edu> From: bbostrom@vz.cis.umn.edu (Brad Bostrom) ================================================================================ Archive-Date: Thu, 19 Aug 1993 17:52:00 GMT Date: Thu, 19 Aug 1993 17:52:00 GMT Sender: bbostrom Organization: Minn Higher Educ Coordinating Board News-Software: VAX/VMS VNEWS 1.41 Nntp-Posting-Host: vz.cis.umn.edu Lines: 22 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Greetings- I am attempting to redefine what keystroke(s) on my vt320 keyboard translate to pf keys for tn3270, and also allow the numeric keypad to be used for entering numbers. I edited multinet:map3270.dat and succeeded in defining new keystrokes for pf keys, but 'undefining' the numeric keypad keys didn't work (they don't send numbers like the top row keys). For example, in map3270.dat I changed pfk1 = '\EOq' # kp1 to pfk1 = '\EOP\EOq' # pf1 + kp1 Now, pressing pf1 + kp1 does send a pfk1, but pressing kp1 doesn't send a 1, it sends some kind of control code. Any ideas? Thanks, Brad Bostrom (bostrom@hecb.state.mn.us) X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 13:50:17 PDT ================================================================================ Archive-Date: Thu, 19 Aug 1993 13:50:13 -0700 Date: Thu, 19 Aug 1993 13:50:13 -0700 (PDT) From: Matt Madison To: jjt@venom.Jpl.Nasa.Gov CC: info-multinet@TGV.COM Subject: RE: mail forwarding from VMS to Unix workstation > On our Vax VMS 5.3, Multinet 3.1, several users forward their mail >to their Unix workstations. They do so using the following command in mail. > >MAIL> SET FORWARD SMTP%"""JJT@VENOM.JPL.NASA.GOV""" > > Notice the tripple quotation marks. Without three quotes, the forward >won't work. Yup, that's VMS MAIL for you. > Anyway, the forwarding works fine for individual messages. But >when I send a message to all users on the Vax, and there are over 100, >then the "to: " message on the Unix workstation includes all of the Vax users >that I send the mail to. > My question to you is how to forward mail properly from a Vax to a Unix >workstation? Also, how does one avoid including everybody on the Vax in the >"to: " message, when I send a mail message to everyone on the Vax? Below is >the command (proper VMS) that I use on the Vax to mail to everyone. > >$ MAIL/SUBJECT="USER DISK 98% Full" MESSAGE98.TXT "@SYS$MANAGER:ALLUSERS.DIS" Ugh. How about setting up your mailing list using the MULTINET:SMTP_ALIASES. file instead? something like allusers : user1, user2, user3; then $ MAIL/SUBJECT=subject file SMTP%ALLUSERS Eliminating the VMS MAIL distribution list will eliminate the massive To: header. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 19 Aug 1993 15:08:03 PDT Date: Thu, 19 Aug 93 15:08:03 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: MTU setting on SLIP port To: ironsd @ aphst1.saic.com cc: info-multinet @ TGV.COM > Is the MTU settable on Multinet SLIP circuits configured on VAX TXA terminal > ports? The System Admin guide (ch.2) makes no mention of MTU when configuring > SLIP circuits. When I configured my circuit between two networks (a VAX at one > end, a DECserver running SLIP at the other) the Multinet circuit port came up > with an MTU of 296. I configured the DECserver end (which is settable) to that > same value even though the DECserver end has a maximum of 1006. Is there a > reason why it comes up set to 296? I have read several discussions recently > about performance of SLIP circuits all of which recommended much higher > settings for MTU. If the MTU on Multinet SLIP interfaces is settable, does the > change require a reboot of the VAX? We use a low MTU because the affect on throughput is negligible (<1%) but there is a drastic improvement in perceived interactive response while a file transfer is also running over the same line (we do prioritized queueing of packets based on TCP port number in the SLIP driver). The MTU is not configurable, but we do, however, run with an MRU just short of 2K, so you should feel free to run the DECserver end at 1006 if you want. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 15:27:59 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: RE: MULTINET_SERVER looping Message-ID: <1993Aug19.153611@usho91.hou281.chevron.com> From: hwpwh@usho91.hou281.chevron.com (Pat Whitnel) ================================================================================ Archive-Date: xxx, 19 Aug 1993 20:36:11 GMT Date: 19 Aug 93 20:36:11 GMT Sender: news@nntpserver.chevron.com (USENET News System) Organization: Chevron Lines: 35 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <00000F58_00075F48.0097140BF8DA5C32$16_1@UK.AC.TEX>, SYSTEM@TEX.AC.UK writes: |> In a private mailing to me of Tue, 1 Jun 1993 10:37:28 +0200 |> Ben Segal wrote: |> |> > Hallo Brian, |> > After reading your plea on the mailing list, we had quite a |> > bad time with this problem at CERN and tracked it down here to the BIND |> > server looping. If you'd like more info, please let me know - it could |> > well be your problem if you run local cacheing-only DNS servers with |> > forwarders: do you? |> |> Well, at the time (when MULTINET_SERVER went into a loop about 1--2 times a |> week) we were indeed running a caching-only server, with forwarders. For the |> past month, I've been primary nameserver for my own domain, and the server |> hasn't once gone into a race state. |> I have one node that has the same looping scenario, and it is also configured the same way... a caching-only server, with forwarders...it is the only one that has ever had the problem...pat |> |> So it looks as though there may be something significant in what Ben wrote. |> |> |> |> Brian {Hamilton Kelly} |> System Manager for the (VMS) UK TeX Archive at Aston University |> |> PS I expect the damned thing will prove me wrong just as soon as this mail |> leaves me :-J -- Pat Whitnel hwpwh@chevron.com Multidisciplinary System Administration Chevron Information Technology Company BTD/Unix Systems*These are MY opinions;* 2811 Hayes Road Houston, TX 77082-2642 V: 713-596-3190 *however,WE suffer from* P. O. Box 42832 Houston, TX 77242-2832 F: 713-596-3041 * split personalities. * X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 16:52:41 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: tn3270, numeric keypad Message-ID: <1993Aug19.212010.5993@mksol.dseg.ti.com> From: pyron@skndiv.dseg.ti.com (Dillon Pyron) ================================================================================ Archive-Date: Thu, 19 Aug 1993 21:20:10 GMT Date: Thu, 19 Aug 1993 21:20:10 GMT Reply-To: pyron@skndiv.dseg.ti.com Sender: usenet@mksol.dseg.ti.com (Usenet News) Organization: TI/DSEG VAX Support Nntp-Posting-Host: skndiv.dseg.ti.com Lines: 30 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <19AUG199312525288@vz.cis.umn.edu>, bbostrom@vz.cis.umn.edu (Brad Bostrom) writes: >I am attempting to redefine what keystroke(s) on my vt320 >keyboard translate to pf keys for tn3270, and also allow the >numeric keypad to be used for entering numbers. > >I edited multinet:map3270.dat and succeeded in defining new >keystrokes for pf keys, but 'undefining' the numeric keypad keys >didn't work (they don't send numbers like the top row keys). > >For example, in map3270.dat I changed > pfk1 = '\EOq' # kp1 >to > pfk1 = '\EOP\EOq' # pf1 + kp1 > >Now, pressing pf1 + kp1 does send a pfk1, but pressing kp1 >doesn't send a 1, it sends some kind of control code. > >Any ideas? You may need to do a SET KEYPAD/NUMERIC. Of course, as a side effect, the keypad will also not work in your prefered manner. -- Dillon Pyron | The opinions expressed are those of the TI/DSEG Lewisville VAX Support | sender unless otherwise stated. (214)462-3556 (when I'm here) | (214)492-4656 (when I'm home) |Please send mail to pyron@dseg.ti.com pyron@dseg.ti.com |since skndiv is going away. Thanks PADI AI-54909 | X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 17:44:06 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Help needed to compile/link wais server with multinet tcp/ip. Message-ID: <1993Aug20.093217.66793@qut.edu.au> From: ling@qut.edu.au ================================================================================ Archive-Date: xxx, 20 Aug 1993 09:32:16 +1000 Date: 20 Aug 93 09:32:16 +1000 Organization: Queensland University of Technology Lines: 34 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hello Multinet TCPIP experts, This is about compiling and linking the wais server and indexer on VMS. I am having troubles compiling and linking the VMS wais server and indexer using TGV/Multinet TCP/IP software. I got the VMS wais software from sunsite.unc.edu. The version (8b2) is based on TCP/IP and my site only got the Multinet TCP/IP. When I tried to link the server code with the library UCX$IPC.OLB, I got the error: %LINK-W-NUDFSYMS, 1 undefined symbol: %LINK-I-UDFSYM, TCPWARE_SERVER When I tried to link the indexer code with the library UCX$IPC.OLB, I got the error: %LINK-W-NUDFSYMS, 2 undefined symbols: %LINK-I-UDFSYM, LOG_WRITE %LINK-I-UDFSYM, OUT_WRITE I would be most grateful if persons who have converted the software to run on Multinet TCP/IP or persons who know how to convert can show me what to do. Thanks in advance. How-Hie Ling. (h.ling@qut.edu.au) Queensland University of Technology, Brisbane, Queensland, Australia. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 19 Aug 93 19:29:03 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: MultiWare questions Message-ID: <1993Aug19.125308.5275@verifone.com> From: brian_u@hnlv4.verifone.com ================================================================================ Archive-Date: Thu, 19 Aug 1993 02:52:58 GMT Date: Thu, 19 Aug 93 02:52:58 GMT Organization: VeriFone Inc. Keywords: MultiWare,NetWare,MultiNet,TGV Nntp-Posting-Host: hnlv4 Lines: 15 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I read the MultiWare spec sheet but have a few questions. 1. Can MultiWare send VMS print jobs to printers with Novell-only Ethernet print server cards installed? I heard this feature was on a wish list. 2. Does MultiWare support "raw" 802.3 Ethernet framing which is the default for NetWare 3.x and 2.x? Or does it require Ethernet II framing protocol-type 8137? Thanks, Brian K. Uechi | brian_u@verifone.com (Internet) VeriFone Inc. | uunet!verifone!brian_u (UUCP) 100 Kahelu Ave. | 808-623-2911 (Voice) Mililani, HI 96789 | 808-625-3201 (FAX) X-ST-Status: N Return-path: Received: from jarrah.itd.adelaide.edu.au ([129.127.40.12]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 02:37:12 PDT Received: by jarrah.itd.adelaide.edu.au with SMTP (5.61+IDA+MU/UA-5.28) id AA20380; Fri, 20 Aug 1993 19:06:18 +0930 Message-Id: <9308200936.AA20380@jarrah.itd.adelaide.edu.au> To: kelly_d@summer.chem.su.oz.au Cc: info-multinet@tgv.com Subject: Re: problem receiving mail from mailserver In-Reply-To: Your message of "Wed, 18 Aug 1993 16:48:26 GMT." <1993Aug18.114826.1@summer.chem.su.oz.au> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" ================================================================================ Archive-Date: Fri, 20 Aug 1993 19:06:17 +0930 Date: Fri, 20 Aug 1993 19:06:17 +0930 From: Andrew Rutherford In message <1993Aug18.114826.1@summer.chem.su.oz.au> you write: + We have a problem receiving some mail items from our mail server. + We're running Multinet V3.2 Rev B on VMS 5.5-2 . + Our mail server is an Apollo running Domain/OS 10.4 with a slightly modified + sendmail version 5.61 . Further to Matthew's posting, you're better of fixing the problem on the Apollo. Check the ether/tcp/smtp mailer spec and check it has "E=\r\n" on it. This is a common problem (one of our departments got bitten by it) where a braindamaged Unix sendmail sends out line feeds on the end of lines and not as required. The guys in TGV fixed a problem with other boxes (some machines send and all sorts of other combinations) by baiscally looking for the . So telling sendmail that the end-of-line should be fixes the problem. Eg, one around here looks like this: Mtcp, P=[IPC], F=mDFMuCXLR, S=15, R=15, A=IPC $h, E=\r\n Basically, Multinet is waiting for the and the Apollo is waiting for a responce, so you see that multinet logical name is actually justifiable. :-) Of course, we define the logical on most of our boxes, otherwise we get the standard run of Unix weenies who can't configure their own boxes who think "It's a VMS machine, it works to everything else, therefore it can't be our fault and they must be at fault". Don't get me wrong by the "Unix weenie" statement - I'm a Unix hacker and think it's great, but Unix seems to breed more morons than VMS does, probably because it's ubiquitous. :-( Hope this helps, Andrew. /* * Andrew Rutherford andrewr@itd.adelaide.edu.au * +61 8 303 5669 Real Programmers always confuse Christmas and * Room 1060, Adelaide Uni Halloween because OCT 31 == DEC 25 ! */ Of course this is my opinion! Can you imagine even a University rubber stamping this? I'm not an academic! X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 04:35:36 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Passing VMS qualifiers to VAX LPD? Message-ID: <1993Aug20.155922.27379@state.systems.sa.gov.au> From: enisaxg@state.systems.sa.gov.au ================================================================================ Archive-Date: xxx, 20 Aug 1993 15:59:21 ACST Date: 20 Aug 93 15:59:21 ACST Organization: State Systems, South Australia Lines: 37 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi! This is just a question/suggestion : with the Multinet print serving, it would be nice if it were possible to send a range of typical "PRINT/xxx" parameters to the [LPD or Stream] queue on the VAX, much as Novell allow. eg On Unix: lp -dlj0 -J/hold/form=newform file [file...] which on the VAX would act like PRINT/JOB=jobname/HOLD/FORM=newform file[,file...] (Other useful qualifiers might be /ff, /banner etc). Now I can achieve some of this by modifying the code in USER_LPD_SERVER.C, but I notice that the example says that the parameter to -J is "VMSified" so that it contains only legal VMS characters; it would be nice if the user could strip off what they wanted (ie the /param values) and THEN allow VMSification to occur. This would be a nifty way of allowing users to get more control over the symbiont, and would get a LOT of users off my back! So I guess I'd like to know how to grab the parameter, or perhaps, how to put the VMS parameters into a line of the LPR control file that gets sent over to the VAX LPD server (this seems a bit less ideal, as it means modifying the lpr clients on our multitude of HP/SUN/PC Unix machines). BTW, I think the print facilities offered by Multinet are great, so much so that my advocacy means that the users keep expecting more and more! Thanks in advance ... Alexander Gunjko [ gunjkoa@dep.sa.gov.au ] Systems Administrator South Australian Office of Planning and Urban Development. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 04:59:28 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: tn3270, numeric keypad Message-ID: <1993Aug20.104122.12137@celsiustech.se> From: erl@celsiustech.se (Erik Lindgren) ================================================================================ Archive-Date: Fri, 20 Aug 1993 10:41:22 GMT Date: Fri, 20 Aug 1993 10:41:22 GMT Organization: CelsiusTech AB Lines: 40 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <19AUG199312525288@vz.cis.umn.edu> bbostrom@vz.cis.umn.edu (Brad Bostrom) writes: >Greetings- > >I am attempting to redefine what keystroke(s) on my vt320 >keyboard translate to pf keys for tn3270, and also allow the >numeric keypad to be used for entering numbers. > >I edited multinet:map3270.dat and succeeded in defining new >keystrokes for pf keys, but 'undefining' the numeric keypad keys >didn't work (they don't send numbers like the top row keys). > >For example, in map3270.dat I changed > pfk1 = '\EOq' # kp1 >to > pfk1 = '\EOP\EOq' # pf1 + kp1 > >Now, pressing pf1 + kp1 does send a pfk1, but pressing kp1 >doesn't send a 1, it sends some kind of control code. > >Any ideas? > >Thanks, >Brad Bostrom (bostrom@hecb.state.mn.us) There is an undocumented (=unsupported) logical name that can give you a numeric keypad in MultiNet TN3270. Define MULTINET_TN3270_APPLICATION_KEYPAD to OFF before you start TN3270 to get a numeric keypad. This will do the trick together with the modified MAP3270.DAT in the example above. (Tested and verified on MultiNet TELNET-32 Version 3.2) Use it on your own risk since it is an undocumented feature. Regards Erik Lindgren, UPEC systemutvecklings AB, Stockholm, Sweden working for CelsiusTech, speaking for my self. X-ST-Status: N Return-path: Received: from CACD1.CACD.CR.ROCKWELL.COM ([131.198.129.51]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 06:27:12 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 8:26:57 -0500 Date: Fri, 20 Aug 1993 8:26:57 -0500 (CDT) From: RAS@CACD1.CACD.CR.ROCKWELL.COM To: info-multinet@TGV.COM Subject: RE: mail forwarding from VMS to Unix workstation |jjt@venom.Jpl.Nasa.Gov wrote and MADISON@TGV.COM responded: | |> On our Vax VMS 5.3, Multinet 3.1, several users forward their mail |>to their Unix workstations. They do so using the following command in mail. |> |>MAIL> SET FORWARD SMTP%"""JJT@VENOM.JPL.NASA.GOV""" |> |> Notice the tripple quotation marks. Without three quotes, the forward |>won't work. | |Yup, that's VMS MAIL for you. | |> Anyway, the forwarding works fine for individual messages. But |>when I send a message to all users on the Vax, and there are over 100, |>then the "to: " message on the Unix workstation includes all of the Vax users |>that I send the mail to. |> My question to you is how to forward mail properly from a Vax to a Unix |>workstation? Also, how does one avoid including everybody on the Vax in the |>"to: " message, when I send a mail message to everyone on the Vax? Below is |>the command (proper VMS) that I use on the Vax to mail to everyone. |> |>$ MAIL/SUBJECT="USER DISK 98% Full" MESSAGE98.TXT "@SYS$MANAGER:ALLUSERS.DIS" | |Ugh. How about setting up your mailing list using the MULTINET:SMTP_ALIASES. |file instead? something like | | allusers : user1, user2, user3; | |then | | $ MAIL/SUBJECT=subject file SMTP%ALLUSERS | |Eliminating the VMS MAIL distribution list will eliminate the massive To: |header. Madison's response may be appropriate for a small site where the user registrations aren't very volatile, but for a large site (where the number of users is in the 100's or 1000's), the valid username list can change multiple times in a day. It isn't practical to keep a fixed list of all the system users; at our site, the ALLUSERS.DIS file is generated just prior to use. Sending mail to all users is accomplished by creating the message, then invoking a procedure to create the distribution file and send the message. Although stripping out the current "allusers" entry from MULTINET:SMTP_ALIASES, then adding a new allusers entry created from the ALLUSERS.DIS file is doable, it is not very clean. It would be much better to have some sort of flag (e.g. unique entry in the .DIS file [SMTP%"NOEXPAND_DISTRIBUTION"], a mail qualifier, a process logical, or a system logical) that caused the SMTP interface to behave the same as standard VMS and not expand the distribution list. Bob Schneider Enterprise Core Network Team ras@cacd1.cacd.cr.rockwell.com Design Support Engineering ras@131.198.129.51 Rockwell International ras%27746.decnet@consort.rockwell.com 400 Collins Road NE M/S 106-180 raschnei@crems.cr.rockwell.com Cedar Rapids, IA 52498 Voice: 319/395-3863 FAX: 319/395-3841 X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 07:07:22 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 9:05:59 -0500 Date: Fri, 20 Aug 1993 9:05:59 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: multinet nameserving I would like to end our reliance on campus nameservers for a number of reasons. But this campus appears to use/promote all sorts of addressing which appears to me to be weird. Does anyone there have a feeling for how well the multinet name servers would respond to unusual addresses? Ginny X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 07:36:00 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 9:34:37 -0500 Date: Fri, 20 Aug 1993 9:34:37 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: name server Perhaps instead of 'weird' I should have used 'unusual.' How good is multinet nameserver at dealing with more unusual addressing formats? My comment was occasioned because apparently this campus had some difficulty with the multinet nameserver in the past because it did not accept some address forms in use here. Ginny X-ST-Status: N Return-path: Received: from Cone-Of-Silence.TGV.COM ([161.44.128.112]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 07:48:48 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 7:48:46 -0700 Date: Fri, 20 Aug 1993 7:48:46 -0700 (PDT) From: Matt Madison To: brian_u@hnlv4.verifone.com CC: INFO-MULTINET@TGV.COM Subject: RE: MultiWare questions >1. Can MultiWare send VMS print jobs to printers with Novell-only Ethernet >print server cards installed? I heard this feature was on a wish list. The upcoming second release will support VMS->NetWare printing, but only to NetWare queues, not directly to printers via RPRINTER. >2. Does MultiWare support "raw" 802.3 Ethernet framing which is the >default for NetWare 3.x and 2.x? Or does it require Ethernet II framing >protocol-type 8137? The second release will support raw 802.3 framing. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 08:10:39 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: chris@csvax1.ucc.ie (Chris Higgins - System Administrator) Subject: Re: mail forwarding from VMS to Unix workstation Message-ID: <1993Aug20.110349.19709@donau.et.tudelft.nl> Sender: news@donau.et.tudelft.nl (UseNet News System) Nntp-Posting-Host: csvax1.ucc.ie Reply-To: chris@csvax1.ucc.ie Organization: Computer Science Dept. University College Cork, Ireland. ================================================================================ Archive-Date: Fri, 20 Aug 1993 11:03:49 GMT Date: Fri, 20 Aug 1993 11:03:49 GMT Lines: 22 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <9308191805.AA01615@venom>, jjt@venom.Jpl.Nasa.Gov (Jack J. Tuszynski) writes: > My question to you is how to forward mail properly from a Vax to a Unix >workstation? Also, how does one avoid including everybody on the Vax in the >"to: " message, when I send a mail message to everyone on the Vax? Below is >the command (proper VMS) that I use on the Vax to mail to everyone. > >$ MAIL/SUBJECT="USER DISK 98% Full" MESSAGE98.TXT "@SYS$MANAGER:ALLUSERS.DIS" Separate issue: Why waste disk space mailing users ? Why not a login message ? Why not use something like BULLETIN ? > > Jack Tuszynski > jjt@venom.jpl.nasa.gov Chris. + J.C. Higgins, + + If you love something, set it + + VMS Sys. Admin, + Chris@csvax1.ucc.ie + free. If it doesn't come back + + Comp.Sc.Dept. + Chris@odyssey.ucc.ie + to you, hunt it down and + + UCC, Ireland + C.Higgins@bureau.ucc.ie + KILL it. -- Me. + X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 20 Aug 1993 08:20:49 PDT Date: Fri, 20 Aug 93 08:20:49 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: name server To: METZE @ VMETZE.MRL.UIUC.EDU cc: info-multinet @ TGV.COM > Perhaps instead of 'weird' I should have used 'unusual.' How good is > multinet nameserver at dealing with more unusual addressing formats? > My comment was occasioned because apparently this campus had some difficulty > with the multinet nameserver in the past because it did not accept some > address forms in use here. Exactly the same as BIND 4.8.3, soon to be exactly the same as BIND 4.9. Could you give us an example? Ken X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 08:31:19 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 10:29:56 -0500 Date: Fri, 20 Aug 1993 10:29:56 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: Adelman@TGV.COM CC: METZE@VMETZE.MRL.UIUC.EDU, info-multinet@tgv.com Subject: Re: name server I don't remember if I heard any details at the time; I can go back and look at my old mail. And I suppose I can just try it. Multinet support is so good that if there is a problem I am sure it would be resolved. I just didn't want lots of mail to go back. From: SMTP%"Adelman@TGV.COM" 20-AUG-1993 10:19:36.47 To: METZE CC: Subj: Re: name server ================================================================================ Archive-Date: Fri, 20 Aug 1993 08:20:49 PDT Date: Fri, 20 Aug 93 08:20:49 PDT From: adelman@TGV.COM (Kenneth Adelman) Reply-To: Adelman@TGV.COM (Kenneth Adelman) Subject: Re: name server 20-Aug-1993 To: METZE@VMETZE.MRL.UIUC.EDU cc: info-multinet@TGV.COM 20-Aug-1993 To: METZE@VMETZE.MRL.UIUC.EDU cc: info-multinet@TGV.COM > Perhaps instead of 'weird' I should have used 'unusual.' How good is > multinet nameserver at dealing with more unusual addressing formats? > My comment was occasioned because apparently this campus had some difficulty > with the multinet nameserver in the past because it did not accept some > address forms in use here. Exactly the same as BIND 4.8.3, soon to be exactly the same as BIND 4.9. Could you give us an example? Ken X-ST-Status: N Return-path: Received: from kosmos.agu.org ([192.102.233.2]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 08:45:12 PDT Received: from KOSMOS.AGU.ORG by KOSMOS.AGU.ORG (PMDF V4.2-11 #3854) id <01H1YTWCJT8W8WWA6Y@KOSMOS.AGU.ORG>; Fri, 20 Aug 1993 11:45:42 EST ================================================================================ Archive-Date: Fri, 20 Aug 1993 11:45:42 -0500 Date: Fri, 20 Aug 1993 11:45:42 -0500 (EST) From: "Shailesh Shah, ext. 208" Subject: Gopher info. To: info-multinet@tgv.com X-VMS-To: IN%"info-multinet@tgv.com" X-VMS-Cc: SINPARAJ,SSHAH MIME-version: 1.0 We are looking for commercial packages for gopher server on Unix. Please let us know as soon as you can. Thank you for your help. PS: Unix meaning Sparc station. Shailesh Shah. X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 09:24:14 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 11:21:51 -0500 Date: Fri, 20 Aug 1993 11:21:51 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: chris@csvax1.ucc.ie CC: info-multinet@tgv.com, METZE@VMETZE.MRL.UIUC.EDU Subject: Re: mail forwarding from VMS to Unix workstation Perhaps their users are like ours; most of them will never read the login messages... Ginny From: SMTP%"chris@csvax1.ucc.ie" 20-AUG-1993 11:19:42.81 To: METZE CC: Subj: Re: mail forwarding from VMS to Unix workstation X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: chris@csvax1.ucc.ie (Chris Higgins - System Administrator) Subject: Re: mail forwarding from VMS to Unix workstation Message-ID: <1993Aug20.110349.19709@donau.et.tudelft.nl> Sender: news@donau.et.tudelft.nl (UseNet News System) Nntp-Posting-Host: csvax1.ucc.ie Reply-To: chris@csvax1.ucc.ie Organization: Computer Science Dept. University College Cork, Ireland. ================================================================================ Archive-Date: Fri, 20 Aug 1993 11:03:49 GMT Date: Fri, 20 Aug 1993 11:03:49 GMT Lines: 22 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <9308191805.AA01615@venom>, jjt@venom.Jpl.Nasa.Gov (Jack J. Tuszynski) writes: > My question to you is how to forward mail properly from a Vax to a Unix >workstation? Also, how does one avoid including everybody on the Vax in the >"to: " message, when I send a mail message to everyone on the Vax? Below is >the command (proper VMS) that I use on the Vax to mail to everyone. > >$ MAIL/SUBJECT="USER DISK 98% Full" MESSAGE98.TXT "@SYS$MANAGER:ALLUSERS.DIS" Separate issue: Why waste disk space mailing users ? Why not a login message ? Why not use something like BULLETIN ? > > Jack Tuszynski > jjt@venom.jpl.nasa.gov Chris. + J.C. Higgins, + + If you love something, set it + + VMS Sys. Admin, + Chris@csvax1.ucc.ie + free. If it doesn't come back + + Comp.Sc.Dept. + Chris@odyssey.ucc.ie + to you, hunt it down and + + UCC, Ireland + C.Higgins@bureau.ucc.ie + KILL it. -- Me. + X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 10:23:08 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: multinet nameserving Message-ID: <252sq3$3da@zippy.Telcom.Arizona.EDU> From: leonard@telcom.arizona.edu (Aaron Leonard) ================================================================================ Archive-Date: xxx, 20 Aug 1993 16:06:27 GMT Date: 20 Aug 1993 16:06:27 GMT Reply-To: Leonard@Arizona.EDU Distribution: world Organization: University of Arizona Telecommunications NNTP-Posting-Host: rena.telcom.arizona.edu Lines: 15 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930820090559.69@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: | I would like to end our reliance on campus nameservers for a number | of reasons. But this campus appears to use/promote all sorts of | addressing which appears to me to be weird. Does anyone there have | a feeling for how well the multinet name servers would respond | to unusual addresses? Could you be more explicit regarding what you consider to be "weird" or "unusual"? We've been operating our campus' nameservers on a MultiNet platform for the last 3-4 years, and they have been performing well (i.e. correctly) to queries both outlandish and banal. Aaron X-ST-Status: N Return-path: Received: from VGER.NIDDK.NIH.GOV ([128.231.60.11]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 12:42:51 PDT Received: from VGER.NIDDK.NIH.GOV by VGER.NIDDK.NIH.GOV (PMDF V4.2-13 #3741) id <01H1Z1WST8I88WW2IQ@VGER.NIDDK.NIH.GOV>; Fri, 20 Aug 1993 15:42:45 EDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 15:42:45 -0400 Date: Fri, 20 Aug 1993 15:42:45 -0400 (EDT) From: Gerson H Cohen Subject: Cluster mounting of NFSmounted volumes To: info-multinet@tgv.com X-VMS-To: IN%INFO-MULTINET X-VMS-Cc: GHC MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Two questions relating to nfs volume references: Is there any plan to serve cluster-wide volumes which are mounted via nfsmount? A sort of nfsmount/cluster is what I have in mind. Must a machine have an nfs server before having an nfs client? Application: I am looking for a cost-effective and efficient way to provide bidirectional nfs service to and from a workstation. If the drives on the workstation are cluster mounted and one node can be a nfs server, they are effectively served. It seems, however, that client nfs is required to see nfs volumes served from the outside world. Have I missed something? Is there another configuration I should consider? Thanks for your thoughts, Gerson Cohen - Dr Gerson H Cohen National Institutes of Health, Bldg 5, Room 335 9000 Rockville Pike, Bethesda, MD 20892 USA 'Phone: (301)402-4495 [Voice] (301)496-0201 [FAX] Internet: ghc@vger.niddk.nih.gov BITNet: ghc@nihklmb NIH DECNet: VGER::GHC X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 20 Aug 1993 13:10:12 PDT Date: Fri, 20 Aug 93 13:10:12 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Cluster mounting of NFSmounted volumes To: ghc @ vger.niddk.nih.gov cc: info-multinet @ TGV.COM > Is there any plan to serve cluster-wide volumes which are mounted > via nfsmount? A sort of nfsmount/cluster is what I have in mind. No. MSCP serves disks at a block-level, much lower than is available via NFS. There is no way MSCP could be used to cluster-serve NFS volumes, and if we wrote a special protocol to do it it would not have any advantages over just using NFSMOUNT on each machine individually. > Must a machine have an nfs server before having an nfs client? No. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 13:50:10 PDT From: kcb@post.its.mcw.edu (Kent C. Brodie) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Passing VMS qualifiers to VAX LPD? ================================================================================ Archive-Date: xxx, 20 Aug 1993 15:14:29 -0500 Date: 20 Aug 1993 15:14:29 -0500 Organization: Medical College of Wisconsin; Milwaukee Wisconsin Lines: 22 NNTP-Posting-Host: post.its.mcw.edu X-Newsreader: TIN [version 1.2 PL1] To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET enisaxg@state.systems.sa.gov.au wrote: : This is just a question/suggestion : with the Multinet print serving, it would : be nice if it were possible to send a range of typical "PRINT/xxx" parameters : to the [LPD or Stream] queue on the VAX, much as Novell allow. eg Here here. I have the same problem--- If, for example, I want to print using a different vms FORM, I have to handle that request by creating a DIFFERENT (generic) queue on the vax with the defaults using the new form, and then I gotta create a new unix print queue to point to it. I.e., the only way I've figured out to bypass this issue is by means of creating MULTIPLE queues. aaaarrrrggggh.... -- -------------- Kent C. Brodie - Sr. Systems/Network Manager brodie@post.its.mcw.edu Information Technology Systems Medical College of Wisconsin (414) 257-8769 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 14:13:16 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: bbostrom@vz.cis.umn.edu (Brad Bostrom) Subject: Re: tn3270, numeric keypad Message-ID: <20AUG199315535615@vz.cis.umn.edu> News-Software: VAX/VMS VNEWS 1.41 Sender: bbostrom Nntp-Posting-Host: vz.cis.umn.edu Organization: Minn Higher Educ Coordinating Board ================================================================================ Archive-Date: Fri, 20 Aug 1993 20:53:00 GMT Date: Fri, 20 Aug 1993 20:53:00 GMT Lines: 51 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug20.104122.12137@celsiustech.se>, erl@celsiustech.se (Erik Lindgren) writes... >In article <19AUG199312525288@vz.cis.umn.edu> bbostrom@vz.cis.umn.edu (Brad Bostrom) writes: >>Greetings- >> >>I am attempting to redefine what keystroke(s) on my vt320 >>keyboard translate to pf keys for tn3270, and also allow the >>numeric keypad to be used for entering numbers. >> >>I edited multinet:map3270.dat and succeeded in defining new >>keystrokes for pf keys, but 'undefining' the numeric keypad keys >>didn't work (they don't send numbers like the top row keys). >> >>For example, in map3270.dat I changed >> pfk1 = '\EOq' # kp1 >>to >> pfk1 = '\EOP\EOq' # pf1 + kp1 >> >>Now, pressing pf1 + kp1 does send a pfk1, but pressing kp1 >>doesn't send a 1, it sends some kind of control code. >> >>Any ideas? >> >>Thanks, >>Brad Bostrom (bostrom@hecb.state.mn.us) > >There is an undocumented (=unsupported) logical name that can give you >a numeric keypad in MultiNet TN3270. > >Define MULTINET_TN3270_APPLICATION_KEYPAD to OFF before you start >TN3270 to get a numeric keypad. > >This will do the trick together with the modified MAP3270.DAT in the >example above. (Tested and verified on MultiNet TELNET-32 Version 3.2) > >Use it on your own risk since it is an undocumented feature. > >Regards > >Erik Lindgren, UPEC systemutvecklings AB, Stockholm, Sweden >working for CelsiusTech, speaking for my self. > Yes, setting that logical name does make the numeric keys work. However, now it seems that the numeric keys can't be used in a multiple-keystroke definition for a pfk (in my example, pf1 + kp1 no longer works as a pfk1). So, it seems I can have one or the other, but not both. Was this your experience? Tack s mycket for your help. Brad Bostrm (bostrom@hecb.state.mn.us) X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 14:36:02 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: palter@tgv.com (Bill Palter) Subject: Re: tn3270, numeric keypad Message-ID: <1993Aug20.205739.3557@news.arc.nasa.gov> Sender: usenet@news.arc.nasa.gov Reply-To: palter@tgv.com Organization: TGV Incorporated ================================================================================ Archive-Date: Fri, 20 Aug 1993 20:57:39 GMT Date: Fri, 20 Aug 1993 20:57:39 GMT Lines: 73 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <20AUG199315535615@vz.cis.umn.edu>, bbostrom@vz.cis.umn.edu (Brad Bostrom) writes: >Newsgroups: vmsnet.networks.tcp-ip.multinet >Path: ames!elroy.jpl.nasa.gov!usc!howland.reston.ans.net!gatech!news-feed-1.peachnet.edu!umn.edu!vz.cis.umn.edu!bbostrom >From: bbostrom@vz.cis.umn.edu (Brad Bostrom) >Subject: Re: tn3270, numeric keypad >Message-ID: <20AUG199315535615@vz.cis.umn.edu> >News-Software: VAX/VMS VNEWS 1.41 >Sender: bbostrom >Nntp-Posting-Host: vz.cis.umn.edu >Organization: Minn Higher Educ Coordinating Board >References: <19AUG199312525288@vz.cis.umn.edu> <1993Aug20.104122.12137@celsiustech.se> >Date: Fri, 20 Aug 1993 20:53:00 GMT >Lines: 51 > >In article <1993Aug20.104122.12137@celsiustech.se>, erl@celsiustech.se (Erik Lindgren) writes... >>In article <19AUG199312525288@vz.cis.umn.edu> bbostrom@vz.cis.umn.edu (Brad Bostrom) writes: >>>Greetings- >>> >>>I am attempting to redefine what keystroke(s) on my vt320 >>>keyboard translate to pf keys for tn3270, and also allow the >>>numeric keypad to be used for entering numbers. >>> >>>I edited multinet:map3270.dat and succeeded in defining new >>>keystrokes for pf keys, but 'undefining' the numeric keypad keys >>>didn't work (they don't send numbers like the top row keys). >>> >>>For example, in map3270.dat I changed >>> pfk1 = '\EOq' # kp1 >>>to >>> pfk1 = '\EOP\EOq' # pf1 + kp1 >>> >>>Now, pressing pf1 + kp1 does send a pfk1, but pressing kp1 >>>doesn't send a 1, it sends some kind of control code. >>> >>>Any ideas? >>> >>>Thanks, >>>Brad Bostrom (bostrom@hecb.state.mn.us) >> >>There is an undocumented (=unsupported) logical name that can give you >>a numeric keypad in MultiNet TN3270. >> >>Define MULTINET_TN3270_APPLICATION_KEYPAD to OFF before you start >>TN3270 to get a numeric keypad. >> >>This will do the trick together with the modified MAP3270.DAT in the >>example above. (Tested and verified on MultiNet TELNET-32 Version 3.2) >> >>Use it on your own risk since it is an undocumented feature. >> >>Regards >> >>Erik Lindgren, UPEC systemutvecklings AB, Stockholm, Sweden >>working for CelsiusTech, speaking for my self. >> > >Yes, setting that logical name does make the numeric keys work. >However, now it seems that the numeric keys can't be used in a >multiple-keystroke definition for a pfk (in my example, pf1 + kp1 >no longer works as a pfk1). So, it seems I can have one or the >other, but not both. Was this your experience? > Your definitions should change to pfk1 = '\EOP1'; # pf1 + kp1 Bill Palter TGV X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 15:23:43 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Problem with POP3 and huge messages. Message-ID: <1993Aug20.161131.7255@mcclb0.med.nyu.edu> From: smith@mcclb0.med.nyu.edu ================================================================================ Archive-Date: xxx, 20 Aug 1993 21:11:31 GMT Date: 20 Aug 93 21:11:31 GMT Organization: NYU Medical Center, New York, NY 10016, USA Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET A moron user just dumped a 2Mb mail message onto one of our secretaries. Eudora tried to pick it up and got a message like " Can't malloc() 15000000 bytes" (I hope I got this right..). And then quit. This meant that the user did not get any mail since it all got aborted after the failure on the huge message. Is there any way to avoid this problem: e.g. by skipping messages that can not get sent due to memory, or other, limitations? +----------------------------------------------------------------------------+ |Ross Smith, Research Computing Resource, Department of Cell Biology, NYU-MC| |E-Mail: SMITH@NYUMED.BITNET (BITNET), SMITH@MCCLB0.MED.NYU.EDU (Internet)| +----------------------------------------------------------------------------+ X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 16:21:16 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: khhunt@ss3.magec.com (Kevin H. Hunt x7343) Subject: PC TCP/IP Packages with Multinet and SUNS Message-ID: Followup-To: vmsnet.networks.tcp-ip.multinet Organization: Magnavox Electronics Systems Co. Distribution: na ================================================================================ Archive-Date: Fri, 20 Aug 1993 21:36:48 GMT Date: Fri, 20 Aug 1993 21:36:48 GMT Lines: 17 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET We're in the evaluation process for a PC windows based tcp/ip and X server combo to work with Multinet and SUNs. We're thinking about PC-NFS 4.0 and XVision5 from VisionWare. Anyone have an opinion on these?? Anyone else have a combo that works great. In the past we have had problems with fonts on the PC's. Thanks in advance. Please respond by email. -- ---------------------------------------------------------------------- Kevin H. Hunt Magnavox Corp. 1313 Production RD MS 10-55 Fort Wayne IN, 46808 X-ST-Status: N Return-path: Received: from PFC.MIT.EDU ([18.77.0.113]) by TGV.COM via INTERNET ; Fri, 20 Aug 93 16:47:47 PDT ================================================================================ Archive-Date: Fri, 20 Aug 1993 19:47:45 -0400 Date: Fri, 20 Aug 1993 19:47:45 -0400 (EDT) From: Mark London To: INFO-MULTINET@TGV.COM Subject: Using MAIL$ routines I'm using MAIL$ routines to send mail and am setting the from address using the MAIL$ calls (from a privileged account). If I use the program to send to local addresses, it works fine. However, if I specify to send mail via SMTP, the from address is ignored. Is this a multinet or vms problem? Thanks. Mark X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 05:08:29 PDT From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: name server ================================================================================ Archive-Date: xxx, 21 Aug 1993 06:55:06 GMT Date: 21 Aug 1993 06:55:06 GMT Organization: HST Wide Field/Planetary Camera Lines: 30 Distribution: world Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930820102956.5bb@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: = =I don't remember if I heard any details at the time; I can go back and =look at my old mail. = =And I suppose I can just try it. Multinet support is so good that if =there is a problem I am sure it would be resolved. = =I just didn't want lots of mail to go back. So then why not take the trouble to get one of the system managers of your nameservers to describe what sorts of records they're using? Gee, I guess that would be a big imposition. It's just not fair to ask someone who's apparently doing something non-standard what the hell they're doing. Instead, ask a ridiculously vague question of the vendor. How the hell do you expect TGV to answer a question of the form, "The folks here are doing something. I won't tell you what it is, but rumor has it that MULTInet couldn't handle it in the past. Will it handle it now?" Get your head out of your ass, and if you want an answer, give enough information to permit an answer. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 05:59:21 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: henrym@sacto.mp.usbr.gov Subject: Re: gopher client problem Sender: news@sacsa3.mp.usbr.gov ( News Login Account) Message-ID: <0097156A.A82C4330@sacto.mp.usbr.gov> ================================================================================ Archive-Date: Sat, 21 Aug 1993 12:15:31 GMT Date: Sat, 21 Aug 1993 12:15:31 GMT Reply-To: henrym@sacto.mp.usbr.gov Organization: United States Bureau of Reclamation, Mid Pacific Region, Sacramento, CA Lines: 37 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug17.155600.1@admvax.sonoma.edu>, jpz@admvax.sonoma.edu writes: >I'm trying to get a gopher client running on my VMS 5.5-2/Multinet >3.2C system. I've tried both clients from boombox.micro.umn.edu, 0.6 >and (gopher+)1.12. Both compile and link OK. When I run, both give the >error: > > Cannot connect to host any.gopher.server > >I've tried several known good hosts with the same result. Also, > >$ telnet any.gopher.server /port=70 > >gives me the (unformatted) main gopher menu from the specified host. > >Am I missing something obvious that needs to be done (to Multinet?) >for the gopher client to work correctly? > >E-mail or post. Thanks in advance for any help you can provide. > >Jack Ziegler | >Computing Services | internet: jpz@sonoma.edu >Sonoma State University | phone : (707)664-3098 >Rohnert Park, CA 94928 | FAX : (707)664-2505 > Jack, I ran into the same problem. Make sure you are compiling with the MultiNet header files, not the VMS header files. -HWM ---------- Henry W. Miller Assistant Systems and Network Manager U.S. Bureau of Reclamation, Mid Pacific Region 2800 Cottage Way MP1130 Sacramento, CA 95825 (916) 978-5108 X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 06:31:19 PDT ================================================================================ Archive-Date: Sat, 21 Aug 1993 8:29:55 -0500 Date: Sat, 21 Aug 1993 8:29:55 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: general question re multinet name serving capabilities I see that someone has taken offense to my general question about multinet nameserving capabilities. I am sorry to have offended him, but I don't believe my question was out of line. I was wanting impressions from others as to how well the multinet name server functioned. I didn't want SPECIFIC information. One reason I didn't give specific information is that I don't HAVE specific information. A warning was sent out to a number of people about problems with the multinet address parser, which, it was said, 'choked on legal addresses' which one of the campus machines was 'fond of generating.' Those problems whatever they were were resolved, and I don't really care what they were. If general questions cannot be asked here, perhaps it should be stated. But personally I don't find ridiculous general questions any more irritating than ridiculous specific questions. I am sorry if anyone doesn't want their bandwidth cluttered, but I have as much right to clutter it as anyone else. There was recently done on this campus a study of the intimidation of active women posters on internet. Maybe I should add this example to their collection. Ginny Metze From: SMTP%"carl@SOL1.GPS.CALTECH.EDU" 21-AUG-1993 07:46:54.84 To: METZE CC: Subj: Re: name server From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: name server ================================================================================ Archive-Date: xxx, 21 Aug 1993 06:55:06 GMT Date: 21 Aug 1993 06:55:06 GMT Organization: HST Wide Field/Planetary Camera Lines: 30 Distribution: world Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930820102956.5bb@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: = =I don't remember if I heard any details at the time; I can go back and =look at my old mail. = =And I suppose I can just try it. Multinet support is so good that if =there is a problem I am sure it would be resolved. = =I just didn't want lots of mail to go back. So then why not take the trouble to get one of the system managers of your nameservers to describe what sorts of records they're using? Gee, I guess that would be a big imposition. It's just not fair to ask someone who's apparently doing something non-standard what the hell they're doing. Instead, ask a ridiculously vague question of the vendor. How the hell do you expect TGV to answer a question of the form, "The folks here are doing something. I won't tell you what it is, but rumor has it that MULTInet couldn't handle it in the past. Will it handle it now?" Get your head out of your ass, and if you want an answer, give enough information to permit an answer. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from WLV.IIPO.GTEGSC.COM ([192.26.147.1]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 07:20:00 PDT Received: by WLV.IIPO.GTEGSC.COM (5.65/1.35) id AA09949; Sat, 21 Aug 93 07:19:50 -0700 ================================================================================ Archive-Date: Sat, 21 Aug 1993 07:19:50 -0700 Date: Sat, 21 Aug 93 07:19:50 -0700 From: mcc@WLV.IIPO.GTEGSC.COM (Merton Campbell Crockett) To: carl@SOL1.GPS.CALTECH.EDU Subject: Re: name server Cc: info-multinet@tgv.com Carl: As frustrating as vague questions are, the last sentence of your message is in poor taste and unnecessary. Metze appeared to be getting to the nub of the apparent problem. The handling of mail. MultiNet will not append the domain component to a host name on a mail item that is forwarded or relayed through a MultiNet host. This presents a problem for networks that are configured not to use the domain component on mail among hosts within the domain and only use fully qualified names when one or more recipients are external to the current domain or subdomain. MultiNet's reaction to host names that are not fully-qualified is to return the mail to the sender--only if the sender used a fully-qualified name--or send it to the Postmaster on the MultiNet host. The latter being the most probable action. The solution is to allow the MultiNet host to remain the Domain Name Service provider but to move the Mail Exchange Service to a Unix platform where one has control over sendmail.cf. Merton Campbell Crockett GTE Government Systems X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 08:09:33 PDT ================================================================================ Archive-Date: Sat, 21 Aug 1993 8:09:30 -0700 Date: Sat, 21 Aug 1993 8:09:30 -0700 (PDT) From: Matt Madison To: MRL@PFC.MIT.EDU CC: INFO-MULTINET@TGV.COM Subject: RE: Using MAIL$ routines >I'm using MAIL$ routines to send mail and am setting the from address using >the MAIL$ calls (from a privileged account). If I use the program to send to >local addresses, it works fine. However, if I specify to send mail via SMTP, >the from address is ignored. Is this a multinet or vms problem? Thanks. You're doing this with the MAIL$_SEND_FROM_LINE item code, right? Either that information doesn't get passed to a foreign protocol handler, or if it does, it does so with an item code that our handler doesn't recognize yet. I'll look into it. -Matt -- Matthew Madison | madison@tgv.com | +1 408 457 5200 TGV, Inc. | 101 Cooper Street | Santa Cruz, CA 95060 USA X-ST-Status: N Return-path: ================================================================================ Archive-Date: Sat, 21 Aug 1993 08:51:04 PDT Date: Sat, 21 Aug 93 08:51:04 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: name server To: mcc @ WLV.IIPO.GTEGSC.COM cc: info-multinet @ TGV.COM > As frustrating as vague questions are, the last sentence of your message is > in poor taste and unnecessary. > Metze appeared to be getting to the nub of the apparent problem. The handling > of mail. MultiNet will not append the domain component to a host name on a > mail item that is forwarded or relayed through a MultiNet host. This presents > a problem for networks that are configured not to use the domain component on > mail among hosts within the domain and only use fully qualified names when one > or more recipients are external to the current domain or subdomain. Huh? Wouldn't the solution to be to conform to RFC1123: "5.2.2 Canonicalization: RFC-821 Section 3.1 The domain names that a Sender-SMTP sends in MAIL and RCPT commands MUST have been "canonicalized," i.e., they must be fully-qualified principal names or domain literals, not nicknames or domain abbreviations. ..." It would be IMPROPER for a mail transfer agent to qualify domain names that arrive unqualified. How could you tell the difference between ADELMAN@NET which was meant to be ADELMAN@NET.TGV.COM and the address ADELMAN@NET where "NET" it the top-level domain NET (Btw, this is a perfectly valid address!)? On the basis of RFC1123, MultiNet assumes that an address like ADELMAN@NET is already canonical and refering to the top-level domain. The fix to this problem is to configure your networks to conform to RFC1123 and canonicalize all names before introducing them into the SMTP system. > MultiNet's reaction to host names that are not fully-qualified is to return > the mail to the sender--only if the sender used a fully-qualified name--or > send it to the Postmaster on the MultiNet host. The latter being the most > probable action. Not quite -- we attempt to deliver the mail first (to the top-level domain), and if that fails, we attempt to return the mail (possibly to the top-level domain), and if the return attempt fails, we deliver it to the local postmaster. Ken X-ST-Status: N Return-path: Received: from THOR.INNOSOFT.COM ([192.160.253.66]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 12:31:56 PDT Received: from INNOSOFT.COM by INNOSOFT.COM (PMDF V4.2-14 #1336) id <01H1Y0TG5AI8984I58@INNOSOFT.COM>; Sat, 21 Aug 1993 12:31:41 PST ================================================================================ Archive-Date: Sat, 21 Aug 1993 12:11:40 -0800 Date: Sat, 21 Aug 1993 12:11:40 -0800 (PST) From: Ned Freed Subject: Handling of shortform names (was: name server) To: mcc@WLV.IIPO.GTEGSC.COM Cc: carl@SOL1.GPS.CALTECH.EDU, info-multinet@TGV.COM MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII > Metze appeared to be getting to the nub of the apparent problem. The handling > of mail. I note in passing that this has nothing whatsoever to do with nameservers. > MultiNet will not append the domain component to a host name on a > mail item that is forwarded or relayed through a MultiNet host. This presents > a problem for networks that are configured not to use the domain component on > mail among hosts within the domain and only use fully qualified names when one > or more recipients are external to the current domain or subdomain. First of all, Multinet is *not* supposed to do this. As Ken has already pointed out, RFC1123 forbids the use of shortform names in SMTP transactions. The proper solution is to fix the broken software that is emitting such things to begin with. Second, in practice the solution of blithely appending domain information to shortform names is unworkable in practice. The fact of the matter is that you cannot expect other Internet hosts to do this and hence changing Multinet does not solve the real problem in any way -- you may make local mail work in some limited cases, but the hosts emitting the short form names will still be broken and unable to interoperate with other hosts that comply with the standards. Moreover, making a mail system do this can do more harm than good even in the local case. We have lots of experience with this stuff in PMDF (which has all sorts of facilities to do this sort of address rewriting) and our experience is that all too often the addition of default domain information has serious side effects. Fixing these side effects introduces more side effects, and on and on and on. Ultimately this leads to extensive amounts of host-specific information in many files on many hosts. This result may work locally but is a nightmare to maintain. I have seen sites go so far as to set up a complex series of mail servers to automatic the distribution and updating of the resulting tables. This is precisely the sort of nonsense the DNS was supposed to get rid of. > MultiNet's reaction to host names that are not fully-qualified is to return > the mail to the sender--only if the sender used a fully-qualified name--or > send it to the Postmaster on the MultiNet host. The latter being the most > probable action. Multinet is simply conforming to the standards here, which more or less mandate such handling. > The solution is to allow the MultiNet host to remain the Domain Name Service > provider but to move the Mail Exchange Service to a Unix platform where one > has control over sendmail.cf. I beg to differ. The solution is to correct the configuration on the hosts that are illegally emitting shortform names to begin with. Solving the problem by using a host with extensive rewrite facilities (be it sendmail or PMDF or whatever) may work but isn't a good idea. Incidentally, sendmail itself is all too frequently the culprit in these scenarios. Not only do many vendor-supplied sendmail configurations leak shortform names to the network, all too often header addresses contain no host information at all: To: user1@host1 Cc: user2, user3 (Yes, PMDF has facilities to handle this sort of garbage as well but we STRONGLY recommend that sites refrain from using this capability.) I should emphasize that sendmail the program isn't the problem here (although the fact that sendmail is terribly difficult to configure properly is a major factor). The problem is broken sendmail configurations, which unfortunately are commonplace. As such, even if you decide to opt for a sendmail-based message exchanger, make sure your sendmail configuration is reasonable or you will only make the problem much much worse. Ned X-ST-Status: N Return-path: Received: from THOR.INNOSOFT.COM ([192.160.253.66]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 12:53:11 PDT Received: from INNOSOFT.COM by INNOSOFT.COM (PMDF V4.2-14 #1336) id <01H1Y0TG5AI8984I58@INNOSOFT.COM>; Sat, 21 Aug 1993 12:52:59 PST ================================================================================ Archive-Date: Sat, 21 Aug 1993 12:37:34 -0800 Date: Sat, 21 Aug 1993 12:37:34 -0800 (PST) From: Ned Freed Subject: Re: general question re multinet name serving capabilities To: METZE@VMETZE.MRL.UIUC.EDU Cc: info-multinet@TGV.COM, METZE@VMETZE.MRL.UIUC.EDU MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII > I see that someone has taken offense to my general question about > multinet nameserving capabilities. I am sorry to have offended > him, but I don't believe my question was out of line. I was > wanting impressions from others as to how well the multinet > name server functioned. I didn't want SPECIFIC information. For what it's worth, my experience has been that the Multinet nameserver functions very well indeed. I have run nameservers on a variety of systems and Multinet's is the best implementation I have seen. Frankly, this isn't saying as much as it might seem. Most vendors don't ship anything resembling a recent BIND version, don't fix known BIND bugs, often make stupid, senseless and undocumented changes to BIND that make it extremely difficult to configure, and are totally uninterested in dealing with resulting customer problems. Yes, I know that TGV's software and support is great, but this is less of an issue than the fact that most other vendor's software and support is so lousy. > One reason I didn't give specific information is that I don't > HAVE specific information. A warning was sent out to a number > of people about problems with the multinet address parser, > which, it was said, 'choked on legal addresses' which one of > the campus machines was 'fond of generating.' Those problems > whatever they were were resolved, and I don't really care what > they were. As I stated in a previous reply, this isn't a nameserver issue. While nameservers play a part in providing domain information, they are not at all involved in address rewriting. > There was recently done on this campus a study of the intimidation > of active women posters on internet. Maybe I should add this example > to their collection. Assuming that this collection is for intimidation that singles out women specifically, I would suggest that before you proceed you compare it with previous exchanges Carl has had with other people. I suspect that you'll find yourself to be the subject of equality rather than inequality ;-) I myself don't pay much attention to gender in email, especially since names are not necessarily a good determining factor and gender really is a nonissue in discussions of this sort. All I saw was a posting and Carl's response. I suspect most other readers did the same. Ned X-ST-Status: N Return-path: Received: from VESTA.SUNQUEST.COM ([149.138.1.42]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 13:16:34 PDT ================================================================================ Archive-Date: Sat, 21 Aug 1993 13:16:28 -0700 Date: Sat, 21 Aug 1993 13:16:28 -0700 (MST) From: Ehud Gavron To: INFO-MULTINET@TGV.COM CC: GAVRON@VESTA.SUNQUEST.COM Subject: Morons strike again Date sent: 21-AUG-1993 13:14:29 Ginny Metze posts a vague question with no details. When asked for details she says there aren't any but she wanted an answer anyway. When asked for what she was expecting she whines: >There was recently done on this campus a study of the intimidation >of active women posters on internet. Maybe I should add this example >to their collection. Perhaps if I posted a really stupid vague question and got flamed I should yell "'Scrimination! 'Scrimination." Would that ALL STUPID PEOPLE would be intimidated into nonposting. >Ginny Metze As Carl put it best: >Get your head out of your ass, and if you want an answer, give enough >information to permit an answer. Ehud -- Ehud Gavron (EG76) gavron@vesta.sunquest.com "The world bores you when you're cool." X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 16:18:50 PDT ================================================================================ Archive-Date: Sat, 21 Aug 1993 18:17:20 -0500 Date: Sat, 21 Aug 1993 18:17:20 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: mcc@WLV.IIPO.GTEGSC.COM CC: info-multinet@tgv.com, METZE@VMETZE.MRL.UIUC.EDU Subject: Re: name server Thank you for taking the time to address what the likely concerns would be. I really appreciate it. Ginny From: SMTP%"mcc@WLV.IIPO.GTEGSC.COM" 21-AUG-1993 09:52:14.32 To: METZE CC: Subj: Re: name server ================================================================================ Archive-Date: Sat, 21 Aug 1993 07:19:50 -0700 Date: Sat, 21 Aug 93 07:19:50 -0700 From: mcc@WLV.IIPO.GTEGSC.COM (Merton Campbell Crockett) To: carl@SOL1.GPS.CALTECH.EDU Subject: Re: name server cc: info-multinet@TGV.COM Carl: As frustrating as vague questions are, the last sentence of your message is in poor taste and unnecessary. Metze appeared to be getting to the nub of the apparent problem. The handling of mail. MultiNet will not append the domain component to a host name on a mail item that is forwarded or relayed through a MultiNet host. This presents a problem for networks that are configured not to use the domain component on mail among hosts within the domain and only use fully qualified names when one or more recipients are external to the current domain or subdomain. MultiNet's reaction to host names that are not fully-qualified is to return the mail to the sender--only if the sender used a fully-qualified name--or send it to the Postmaster on the MultiNet host. The latter being the most probable action. The solution is to allow the MultiNet host to remain the Domain Name Service provider but to move the Mail Exchange Service to a Unix platform where one has control over sendmail.cf. Merton Campbell Crockett GTE Government Systems X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 16:51:23 PDT ================================================================================ Archive-Date: Sat, 21 Aug 1993 18:49:49 -0500 Date: Sat, 21 Aug 1993 18:49:49 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: GAVRON@VESTA.SUNQUEST.COM CC: INFO-MULTINET@TGV.COM, METZE@VMETZE.MRL.UIUC.EDU Subject: MORONS I agree... would that all stupid morons would be intimidated into not posting. X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 17:13:31 PDT ================================================================================ Archive-Date: Sat, 21 Aug 1993 19:12:06 -0500 Date: Sat, 21 Aug 1993 19:12:06 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: thanks for info/opinions about name serving/mailers I think I must have somehow given the impression I wanted more specific information than I did... I merely wondered if people were happy with their multinet nameserving capabilities. Then when someone pressed for specifics, I could only come up with the mailer problem, which is also of interest to me, but just muddied the waters. However, all of the opinions and information I found interesting and helpful. Since there was a mailer problem on campus that was 'blamed on' the multinet software, that triggered my concern about whether it would be best to 'stick to' campus services. There are not many of us using multinet on this campus, and virtually everything is unix. I was very hesitant to try to run a VMS machine as a name server... Why didn't I just put out a questionnaire? On a scale of 1 to 5, where 1 is rotten and 5 is fantastic, how would you rate multinet nameserving capabilities? That is the kind of question I get from surveys all of time... I would still enjoy hearing the experiences and observations of those who have time. There were several posts about name server problems earlier this week that I found interesting and useful as well. In fact, I guess that is what triggered my question. If you do have time to share experiences and opinions, please just send me email -- no one else seems interested. Ginny X-ST-Status: N Return-path: Received: from gawain.umkc.edu ([134.193.1.2]) by TGV.COM via INTERNET ; Sat, 21 Aug 93 21:45:27 PDT Received: by gawain.umkc.edu (5.65/DEC-Ultrix/4.3) id AA18174; Sat, 21 Aug 1993 23:45:18 -0500 Message-Id: <9308220445.AA18174@gawain.umkc.edu> To: METZE@VMETZE.MRL.UIUC.EDU Cc: info-multinet@TGV.COM, archerb@gawain.umkc.edu Subject: Re: thanks for info/opinions about name serving/mailers In-Reply-To: Your message of "Sat, 21 Aug 93 19:12:06 CDT." <930821191206.822@VMETZE.MRL.UIUC.EDU> ================================================================================ Archive-Date: Sat, 21 Aug 1993 23:45:18 -0500 Date: Sat, 21 Aug 93 23:45:18 -0500 From: "Barry Archer, UMKC NOC" X-Mts: smtp > > Why didn't I just put out a questionnaire? On a scale of 1 to 5, > where 1 is rotten and 5 is fantastic, how would you rate multinet > nameserving capabilities? That is the kind of question I get from > surveys all of time... > > I would still enjoy hearing the experiences and observations of > those who have time. There were several posts about name server > problems earlier this week that I found interesting and useful as > well. In fact, I guess that is what triggered my question. > > If you do have time to share experiences and opinions, please > just send me email -- no one else seems interested. > > Ginny > > I'd rate Multinet at at least a 4, if not a 5. We used Multinet for primary nameservice for our campus for our first year or so on the Internet. As a then neophyte I still required a little help from TGV, but it was pretty painless, and we only moved the nameservice to a dedicated UNIX for other reasons. We still have Multinet secondary nameservers in some criticial placeces on campus. We've had no major problems ( Brian McKeever will jump in, if he's seen any ). And I'm *not* just saying this just because I'll be seeing the TGV gang next week...;-). ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Barry Archer, Network Operations Manager University of Missouri - Kansas City archerb@gawain.umkc.edu 816.235.1187 "Madness is the result not of uncertainty but certainty." - Friedrich Nietzsche - Internetworking ninja for hire ( have resume, will travel ) X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Sun, 22 Aug 93 04:27:30 PDT ================================================================================ Archive-Date: Sun, 22 Aug 1993 6:26:05 -0500 Date: Sun, 22 Aug 1993 6:26:05 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: women posters I made the following comment, to which several have responded that I was not treated differently because I am a woman. That wasn't the point I was trying to make. I don't know if the internet bullies care whether the target is a man or a woman. But the end result of what you all believe is normal free-for-all apparently is that women become intimidated. Look to see how many posters are women, and how many women continue after receiving treatment that men seem to find is ignorable. Someone else is probably worrying about the intimidation of men; the campus study was about the effect on women. This was my original comment: There was recently done on this campus a study of the intimidation of active women posters on internet. Maybe I should add this example to their collection. Ginny Metze X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 22 Aug 93 11:10:16 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: general question re multinet name serving capabilities Message-ID: <2567hfINNca4@gap.caltech.edu> From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) ================================================================================ Archive-Date: xxx, 21 Aug 1993 22:27:59 GMT Date: 21 Aug 1993 22:27:59 GMT Reply-To: carl@SOL1.GPS.CALTECH.EDU Distribution: world Organization: HST Wide Field/Planetary Camera NNTP-Posting-Host: sol1.gps.caltech.edu Lines: 39 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930821082955.7e2@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: =I see that someone has taken offense to my general question about =multinet nameserving capabilities. I am sorry to have offended =him, but I don't believe my question was out of line. I was =wanting impressions from others as to how well the multinet =name server functioned. I didn't want SPECIFIC information. = =One reason I didn't give specific information is that I don't =HAVE specific information. A warning was sent out to a number =of people about problems with the multinet address parser, =which, it was said, 'choked on legal addresses' which one of =the campus machines was 'fond of generating.' Those problems =whatever they were were resolved, and I don't really care what =they were. Ah. So the problem had nothing to do with a name server at all. It was a problem with a mailer. Now, if you'd just had the brains to mention this earlier, you'd probably have received better answers to your question. =If general questions cannot be asked here, perhaps it should be =stated. But personally I don't find ridiculous general questions =any more irritating than ridiculous specific questions. I am =sorry if anyone doesn't want their bandwidth cluttered, but I have =as much right to clutter it as anyone else. = =There was recently done on this campus a study of the intimidation =of active women posters on internet. Maybe I should add this example =to their collection. Before you do that, perhaps you should consider the fact that I didn't know you were a woman. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 22 Aug 93 11:10:19 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: PC TCP/IP Packages with Multinet and SUNS Message-ID: <1993Aug21.204749.1@slacvx.slac.stanford.edu> From: fairfield@slacvx.slac.stanford.edu ================================================================================ Archive-Date: Sun, 22 Aug 1993 04:47:49 GMT Date: Sun, 22 Aug 1993 04:47:49 GMT Sender: news@unixhub.SLAC.Stanford.EDU Followup-To: vmsnet.networks.tcp-ip.multinet Distribution: na Organization: Stanford Linear Accelerator Center Lines: 29 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article , khhunt@ss3.magec.com (Kevin H. Hunt x7343) writes: > We're in the evaluation process for a PC windows based tcp/ip and X server > combo to work with Multinet and SUNs. > > We're thinking about PC-NFS 4.0 and XVision5 from VisionWare. > > Anyone have an opinion on these?? Anyone else have a combo that works > great. We use FTP Software's PCTPC. It's a very good product and IMHO, much easier to configure than PC-NFS (well, we sort brought down a segment of the Ethernet when we _tried_ to configure PC-NFS :-( ) I've done extensive evaluations of XVision (V4.1) and HCL eXceed/w (v3.2). While both are good products, I much prefer eXceed [caveat: I haven't tried XVision V5]. For the two versions evaluated, eXceed was noticably faster and provided diagnostics missing in XVision (e.g. log files, although I understand the latter are available in XVisions5). The thing I really didn't like about XVision is that _both_ ALT keys are sent to _either_ X _or_ Windows, while eXceed allows one to configure, e.g. the left-ALT to go to Windows and the right-ALT go to X, or vice versa, etc. -Ken -- Dr. Kenneth H. Fairfield | Internet: Fairfield@Slac.Stanford.Edu SLAC, P.O.Box 4349, MS 98 | DECnet: 45537::FAIRFIELD (45537=SLACVX) Stanford, CA 94309 | FAX: (415) 926-4335 ---------------------------------------------------------------------------- These opinions are mine, not SLAC's, Stanford's, nor the DOE's... X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 22 Aug 93 15:27:34 PDT From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: name server ================================================================================ Archive-Date: xxx, 22 Aug 1993 22:05:28 GMT Date: 22 Aug 1993 22:05:28 GMT Organization: HST Wide Field/Planetary Camera Lines: 20 Distribution: world Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930821181720.822@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: =From: mcc@WLV.IIPO.GTEGSC.COM (Merton Campbell Crockett) =MultiNet's reaction to host names that are not fully-qualified is to return =the mail to the sender--only if the sender used a fully-qualified name--or =send it to the Postmaster on the MultiNet host. The latter being the most =probable action. Actually, MultiNet SHOULD do neither. During the SMTP transaction, when MultiNet's SMTP server sees an invalid From: or To: field, it should reject the message. Disposition of the message then falls to the machine that tried sending the message with the bogus address. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 04:54:02 PDT Via: uk.ac.tex; Mon, 23 Aug 1993 12:53:18 +0100 ================================================================================ Archive-Date: Mon, 23 AUG 1993 11:53:02 GMT Date: Mon, 23 AUG 93 11:53:02 GMT From: SYSTEM@TEX.AC.UK To: INFO-MULTINET Subject: Two minor queries about ftp serving Sender: JANET "SYSTEM@UK.AC.TEX" Originally-to: INTERNET%COM.TGV::INFO-MULTINET Originally-from: SYSTEM "UK TeX Archive Manager " Mailer: Janet_Mailshr V3.5 ( 13-OCT-1989 14:07:27 ) The common behaviour of a Unix-based ftp client, after passing the username `anonymous' to the server, is to say: 331 anonymous user ok. Send real ident as password. Much has been written about brain-damaged client software. How about those brain-damaged liveware clients? You'd be amazed how many dickheads give the string `REAL IDENT' as the password. I guess that some of the other stereo- typed ``user'' names that my logs reveal are due to the wording of the similar response message with other ftp client programs --- you know, things like E-MAIL, USER and GUEST! I attempt to detect some of these in my FTP_SERVER.COM, not actually blocking access, but suggesting that next time they might really reveal who they are. I suppose that there's no way that I could control the "Password: " prompt that the MultiNet ftp server displays, because I guess that it really comes from LOGINOUT.EXE, doesn't it? But if there *is* some way of controlling it, I'd sure like the opportunity. You'll observe that I've shown those examples of stupid user responses in capital letters: that's because that's the way FTP_SERVER.EXE passes them to me through the MULTINET_ANONYMOUS_PASSWORD logical. Does the upcasing get performed by LOGINOUT.EXE (in which case [pun intended], there's obviously nothing more to be said) or is it done by FTP_SERVER.EXE? If the latter, is there any hope that I could be passed *exactly* what the user typed, and not a processed version thereof: sometimes, one might want to e-mail a user of an archive such as ours; whilst, generally speaking, Unix sites nowadays tend to be less case-sensitive than was the norm a few years ago, there are still some sites that bounce mail addressed to FRED when it should have gone to Fred. Brian {Hamilton Kelly} System Manager for the (VMS) UK TeX Archive at Aston University X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 23 Aug 1993 08:20:35 PDT Date: Mon, 23 Aug 93 08:20:35 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: name server To: carl @ SOL1.GPS.CALTECH.EDU cc: info-multinet @ TGV.COM > In article <930821181720.822@VMETZE.MRL.UIUC.EDU>, METZE@VMETZE.MRL.UIUC.EDU writes: > =From: mcc@WLV.IIPO.GTEGSC.COM (Merton Campbell Crockett) > =MultiNet's reaction to host names that are not fully-qualified is to return > =the mail to the sender--only if the sender used a fully-qualified name--or > =send it to the Postmaster on the MultiNet host. The latter being the most > =probable action. > Actually, MultiNet SHOULD do neither. During the SMTP transaction, when > MultiNet's SMTP server sees an invalid From: or To: field, it should reject the > message. Disposition of the message then falls to the machine that tried > sending the message with the bogus address. Actually, RFC1123 says that a receiving SMTP *MAY* validate destination addresses as they are received and reject the destination, and says nothing about validating the source address. Ken X-ST-Status: N Return-path: Return-Path: ================================================================================ Archive-Date: Thu, 12 Aug 1993 10:35:32 PDT Date: Thu, 12 Aug 93 10:35:32 PDT From: gkn (Gerard K. Newman) @ TGV.COM Reply-To: service @ tgv.com Subject: Re: NTP To: info-multinet-relay @ TGV.COM cc: service @ TGV.COM Comment: Please send service mail to SERVICE@TGV.COM ReSent-date: Mon, 23 Aug 1993 08:21:06 -0700 (PDT) ReSent-from: Trey Garlough ReSent-to: info-multinet@TGV.COM > When using NTP to synchronize the VAX time, does anybody have any advice on > either just turning ntp on and letting the VAX time change (it's 2+ > minutes off) all at once, or using the waytoobig parameter to force a more > gradual change in the vax time (say 5-10 seconds at a time)? NTP should slew the clock gradually all by itself (either by running it a bit fast to catch up with what time it really is, or by running it slow to let what time it really is catch up with it). I wouldn't mess with WAYTOOBIG. Cheers, Gerard K. Newman | TGV, Inc. 101 Cooper St. | 800.848.3440 (voice) gkn@tgv.com | Santa Cruz, CA 95060-4526 | 408.427.4365 (fax) X-ST-Status: N Return-path: Return-Path: Received: from CADRE1.CALTECH.EDU ([131.215.154.1]) by TGV.COM via INTERNET ; Tue, 17 Aug 93 10:25:11 PDT ================================================================================ Archive-Date: Tue, 17 Aug 1993 10:24:53 -0700 Date: Tue, 17 Aug 1993 10:24:53 -0700 (PDT) From: JDM@CADRE1.CALTECH.EDU To: info-multinet-relay@tgv.com CC: JDM@CADRE1.CALTECH.EDU Subject: RE: Stream Printing limitations ReSent-date: Mon, 23 Aug 1993 08:24:21 -0700 (PDT) ReSent-from: Trey Garlough ReSent-to: info-multinet@TGV.COM We have a VAX-VMS print symbiont (have not tried it on Alpha) which recognizes %! and routes data straight through to a PostScript printer, but if the file does not start with %!, converts it to PostScript assuming that it contains straight text or a few Diablo control codes at worst. Also included is a true page counting scheme which we use for accounting; that involves information sent back from the PostScript printer. It is not something we have packages for distribution, but code is commented! JON MELVIN jdm@caltech.edu X-ST-Status: N Return-path: Received: from MAX.U.WASHINGTON.EDU ([128.95.137.4]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 08:37:21 PDT Received: from MAX.U.WASHINGTON.EDU.L by MAX.U.WASHINGTON.EDU.L (PMDF V4.2-12 #4607) id <01H22U2CO7FQ9GVAFD@MAX.U.WASHINGTON.EDU.L>; Mon, 23 Aug 1993 08:40:30 PDT ================================================================================ Archive-Date: Mon, 23 Aug 1993 08:40:30 -0700 Date: Mon, 23 Aug 1993 08:40:30 -0700 (PDT) From: DEREK@U.WASHINGTON.EDU Subject: Re: Handling of shortform names (was: name server) Sender: "Derek S. Haining, Univ. of Washington, (206) 543-5579" To: NED@INNOSOFT.COM Cc: info-multinet@TGV.COM X-Envelope-to: info-multinet@TGV.COM X-VMS-To: IN%"NED@INNOSOFT.COM" X-VMS-Cc: IN%"info-multinet@TGV.COM",DEREK MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII As a disinterested bystander in this topic (or should that be a stander-by?) I wish to second comments made by Ned Freed, who wrote: >Second, in practice the solution of blithely appending domain information to >shortform names is unworkable in practice. We ran into this about a year ago. It was decided quite a while back by our networking folk that we should automatically expand "partial" mail addresses. Thus, an address such as: derek@max became derek@max.u.washington.edu This worked OK for a long time, until Czechoslovakia got onto the internet. Suddenly, folks trying to e-mail their collegaues there had their mail rejected. Why? Because of a rule which expanded "short-form" names. I.e., "derekh@june.cs" became "derekh@june.cs.washington.edu" ("cs" is for the computer science department, and is also the country identifier for Czechoslovakia.) Not to be discouraged, some fancy footwork was done and life went back to "normal". Guess what happened when Estonia joined the internet... Yep, Electrical Engineering (ee) started having troubles. I think that the current "fix" is to allow names like "derekh@june.cs." to denore (in)complete addresses. (I forget which way it goes.) -Derek S. Haining University Computing Services University of Washington Seattle, Washington 98195 (206) 543-5579 DEREK@MAX.BITNET DEREK@MAX.U.WASHINGTON.EDU X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 23 Aug 1993 11:08:28 PDT Date: Mon, 23 Aug 93 11:08:28 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Two minor queries about ftp serving To: SYSTEM @ TEX.AC.UK cc: info-multinet @ TGV.COM > The common behaviour of a Unix-based ftp client, after passing the username > `anonymous' to the server, is to say: > 331 anonymous user ok. Send real ident as password. > Much has been written about brain-damaged client software. How about those > brain-damaged liveware clients? You'd be amazed how many dickheads give the > string `REAL IDENT' as the password. I guess that some of the other stereo- > typed ``user'' names that my logs reveal are due to the wording of the similar > response message with other ftp client programs --- you know, things like > E-MAIL, USER and GUEST! Actually, I send the password GUEST quite a lot because it works universally. My fingers are pre-programmed to type "LOGIN ANONYMOUS GUEST" without waiting for the prompt back. > I attempt to detect some of these in my FTP_SERVER.COM, not actually blocking > access, but suggesting that next time they might really reveal who they are. I > suppose that there's no way that I could control the "Password: " prompt that > the MultiNet ftp server displays, because I guess that it really comes from > LOGINOUT.EXE, doesn't it? But if there *is* some way of controlling it, I'd > sure like the opportunity. This prompt is wired into the FTP *client*. > You'll observe that I've shown those examples of stupid user responses in > capital letters: that's because that's the way FTP_SERVER.EXE passes them to me > through the MULTINET_ANONYMOUS_PASSWORD logical. Does the upcasing get > performed by LOGINOUT.EXE (in which case [pun intended], there's obviously > nothing more to be said) or is it done by FTP_SERVER.EXE? It is done by the FTP server process. I could fix it, but am afraid to since I might break people's command procedures which check for particular passwords. Ken X-ST-Status: N Return-path: Received: from SACMGR.MP.USBR.GOV ([140.214.12.4]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 13:15:18 PDT Received: by sacto.mp.usbr.gov (MX V3.3 VAX) id 23343; Mon, 23 Aug 1993 13:15:05 PDT ================================================================================ Archive-Date: Mon, 23 Aug 1993 13:15:00 PDT Date: Mon, 23 Aug 1993 13:15:00 PDT From: Jim Wheeler # 916-978-5089 To: METZE@VMETZE.MRL.UIUC.EDU CC: info-multinet@tgv.com, jimww@sacto.mp.usbr.gov Subject: RE: general question re multinet name serving capabilities >From: MX%"METZE@VMETZE.MRL.UIUC.EDU" 21-AUG-1993 06:54:19.54 >Subject: general question re multinet name serving capabilities > > >I see that someone has taken offense to my general question about >multinet nameserving capabilities. I am sorry to have offended >him, but I don't believe my question was out of line. I was >wanting impressions from others as to how well the multinet >name server functioned. I didn't want SPECIFIC information. > >One reason I didn't give specific information is that I don't >HAVE specific information. A warning was sent out to a number >of people about problems with the multinet address parser, >which, it was said, 'choked on legal addresses' which one of >the campus machines was 'fond of generating.' Those problems >whatever they were were resolved, and I don't really care what >they were. > >If general questions cannot be asked here, perhaps it should be >stated. But personally I don't find ridiculous general questions >any more irritating than ridiculous specific questions. I am >sorry if anyone doesn't want their bandwidth cluttered, but I have >as much right to clutter it as anyone else. > >There was recently done on this campus a study of the intimidation >of active women posters on internet. Maybe I should add this example >to their collection. > >Ginny Metze > ....garbage deleted....... Ginny, Don't hesitate to post it, but, Carl is just as abusive to anyone that asks a question he disaproves of. Why does he even expend the effort if he thinks someone is beneath him ? However, there is a price to pay for everyting, and he is one of the prices you pay for all the valuable information obtained from this list. That is my opinion anyway jimww # James W. Wheeler # # # Regional Systems & WAN Manager # voice: 916 978-5089 fax: 916-978-5284 # # Mid Pacific Region # # # U.S. Bureau of Reclamation # inet: jimww@sacto.mp.usbr.gov # # 2800 Cottage Way MP1130 # # # Sacramento, CA 95825 # # X-ST-Status: N Return-path: Received: from SECS.UCSC.EDU ([128.114.141.60]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 13:59:09 PDT Received: by SECS.UCSC.EDU (MX V3.1C) id 18574; Mon, 23 Aug 1993 13:59:01 PDT ================================================================================ Archive-Date: Mon, 23 Aug 1993 13:58:56 PDT Date: Mon, 23 Aug 1993 13:58:56 PDT From: "W. Todd Wipke" To: info-multinet@tgv.com Subject: NFSMount serving A node in a vms cluster running the NFS server can serve any disk it can see in the cluster to the world, but for a node in the cluster to access disks outside the cluster that are NFS served by remote hosts, that node must run the NFS Client. Thus typically a cluster would run one server for exporting vms disks, but many clients--one for each node that wanted access to remote NFS non-cluster disks. Correct? -Todd Wipke wipke@chemistry.ucsc.edu X-ST-Status: N Return-path: Received: from mprgate.mpr.ca ([134.87.131.13]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 14:11:17 PDT Received: from magellan.mpr.ca by mprgate.mpr.ca with SMTP id AA00321 (5.65c/IDA-1.4.4 for ); Mon, 23 Aug 1993 14:08:38 -0700 Received: by magellan.mpr.ca (4.1/SMI-4.1) id AA18426; Mon, 23 Aug 93 14:08:38 PDT From: hamm@mprgate.mpr.ca (Rob Hamm) Message-Id: <9308232108.AA18426@magellan.mpr.ca> Subject: RE: general question re multinet name serving capabilities To: jimww@sacto.mp.usbr.gov (Jim Wheeler # 916-978-5089) ================================================================================ Archive-Date: Mon, 23 Aug 1993 14:08:37 PDT Date: Mon, 23 Aug 93 14:08:37 PDT Cc: METZE@VMETZE.MRL.UIUC.EDU, info-multinet@TGV.COM, jimww@sacto.mp.usbr.gov X-Mailer: ELM [version 2.3 PL11] > >Don't hesitate to post it, but, Carl is just as abusive to anyone that asks a >question he disaproves of. Why does he even expend the effort if he thinks >someone is beneath him ? However, there is a price to pay for everyting, and >he is one of the prices you pay for all the valuable information obtained from >this list. That is my opinion anyway > >jimww > > I have to disagree with this. Carl *used* to be very abusive, he has calmed down *VERY* much lately and has been providing excellent technical information lately. Ehud on the otherhand... :-) ...rob X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 14:14:39 PDT From: itopsjc@meteor.syscon.hii.com X-Newsgroups: vmsnet.networks.tcp-ip.multinet,vmsnet.mail.mx Subject: Making the transition from MultiNet SMTP to MX Message-ID: <1993Aug23.164044.23@meteor> ================================================================================ Archive-Date: xxx, 23 Aug 1993 16:40:44 -0700 Date: 23 Aug 93 16:40:44 -0700 Organization: SYSCON Corporation Lines: 19 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET We've just recently switched Internet mail handlers from SMTP, provided with MultiNet, to the freeware package, MX. Our users are used to sending mail out over the Internet using SMTP. Does anyone have any recommendations on how we can make the transition easier for them? As a temporary solution, I've copied MX_EXE:MX_MAILSHR.EXE to SYS$LIBRARY:SMTP_MAILSHR.EXE and INSTALLed it /OPEN/SHARED/HEADER_RES. This appears to work okay for our purposes, but is there a better way? Also, where are MX_MAILSHR and SMTP_MAILSHR normally kept? Neither are in SYS$LIBRARY, where MAIL appears to be looking. Of course, there's no documentation on foreign mail transports, so I can't just look it up... Thanks in advance to all who respond. -- Steve Coile SYSCON Corporation SCOILE@SYSCON.HII.COM 1000 Thomas Jefferson St., NW (202) 342-4096 Washington, DC 20007 X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 23 Aug 1993 14:30:54 PDT Date: Mon, 23 Aug 93 14:30:54 PDT From: mahan @ TGV.COM (Patrick L. Mahan) Subject: RE: NFSMount serving To: info-multinet @ TGV.COM, wipke @ chemistry.ucsc.edu X-ST-Vmsmail-To: ST%"info-multinet" # # A node in a vms cluster running the NFS server can serve any disk it can # see in the cluster to the world, but for a node in the cluster to access # disks outside the cluster that are NFS served by remote hosts, that node # must run the NFS Client. Thus typically a cluster would run one server # for exporting vms disks, but many clients--one for each node that wanted # access to remote NFS non-cluster disks. Correct? # -Todd Wipke # wipke@chemistry.ucsc.edu # Correct. X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 23 Aug 1993 15:03:08 PDT Date: Mon, 23 Aug 93 15:03:08 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: NFSMount serving To: wipke @ chemistry.ucsc.edu cc: info-multinet @ TGV.COM > A node in a vms cluster running the NFS server can serve any disk it can > see in the cluster to the world, but for a node in the cluster to access > disks outside the cluster that are NFS served by remote hosts, that node > must run the NFS Client. Thus typically a cluster would run one server > for exporting vms disks, but many clients--one for each node that wanted > access to remote NFS non-cluster disks. Correct? Correct. We recommend you run the NFS Server on the machine which physically has the disks attached to reduce network/cluster traffic (especially in the case of an LAVC), but this could be affected by other factors, such as if you have a much faster machine with more memory that must access the disks remotely. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 16:09:14 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet,vmsnet.mail.mx Subject: Re: Making the transition from MultiNet SMTP to MX Message-ID: <1993Aug23.223719.16798@colorado.edu> From: dwing@uh01.colorado.edu (Dan Wing) ================================================================================ Archive-Date: Mon, 23 Aug 1993 22:37:19 GMT Date: Mon, 23 Aug 1993 22:37:19 GMT Reply-To: dwing@uh01.colorado.edu Sender: news@colorado.edu (The Daily Planet) Followup-To: vmsnet.mail.mx Organization: University of Colorado Hospital Authority, Denver Nntp-Posting-Host: buckie.hsc.colorado.edu Lines: 67 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug23.164044.23@meteor>, itopsjc@meteor.syscon.hii.com writes: >We've just recently switched Internet mail handlers from SMTP, provided with >MultiNet, to the freeware package, MX. Our users are used to sending mail out >over the Internet using SMTP. Does anyone have any recommendations on how we >can make the transition easier for them? > >As a temporary solution, I've copied MX_EXE:MX_MAILSHR.EXE to >SYS$LIBRARY:SMTP_MAILSHR.EXE and INSTALLed it /OPEN/SHARED/HEADER_RES. This >appears to work okay for our purposes, but is there a better way? From MX_DOC:MGMT_GUIDE.TXT: 5.1.1 VMS MAIL Protocol Prefix MX by default uses the foreign protocol prefix MX% when interfacing with VMS Mail. You can define alternate foreign protocol prefixes for use with MX, to provide a migration path for users from other mail systems to MX. MX will correctly handle the following prefixes: SMTP%, WINS%, IN%, JNET%, IHMF%, VN%, ST%, INET%, and UUCP%.[1] To set up one of these alternate prefixes in VMS Mail, define the logical name MAIL$PROTOCOL_prefix: $ DEFINE/SYSTEM/EXEC MAIL$PROTOCOL_prefix MX_MAILSHR where prefix is one of the above-mentioned prefixes, without the trailing percent sign. Note that incoming mail from MX will always bear the MX% prefix. If you wish to use another prefix for incoming mail, you can define the logical name MX_ PROTOCOL_PREFIX: $ DEFINE/SYSTEM/EXEC MX_PROTOCOL_PREFIX prefix% where prefix is one of the above-mentioned prefixes, with the trailing percent sign. The default prefix MX% is the recommended prefix. ________________ [1] You should not re-direct the UUCP% prefix to MX if you are using MX with UUCP. Doing so will prevent messages from being delivered to UUCP from MX, since MX uses the UUCP_MAILSHR interface (the same as UUCP% does). If you want, you could add the logicals to MX's startup by carefully editing MX_ROOT:[000000]MX_LOGICALS.DAT. Personally, I'd leave incoming mail so MX% is used, if for nothing more than to gently remind people that things have changed. >Also, where are MX_MAILSHR and SMTP_MAILSHR normally kept? Neither are in >SYS$LIBRARY, where MAIL appears to be looking. Of course, there's no >documentation on foreign mail transports, so I can't just look it up... $ SHOW LOGICAL/FULL MX_MAILSHR "MX_MAILSHR" [exec] = "MX_EXE:MX_MAILSHR" (LNM$SYSTEM_TABLE) In the absence of a logical, Mail will look in SYS$SHARE (SYS$LIBRARY). [followups to vmsnet.mail.mx.] -Dan Wing, Systems Administrator, University Hospital, Denver dwing@uh01.colorado.edu or wing_d@ucolmcc.bitnet X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 17:31:17 PDT From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: name server ================================================================================ Archive-Date: xxx, 23 Aug 1993 23:33:00 GMT Date: 23 Aug 1993 23:33:00 GMT Organization: HST Wide Field/Planetary Camera Lines: 28 Distribution: world Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930823082035.40a005b9@TGV.COM>, adelman@TGV.COM (Kenneth Adelman) writes: => Actually, MultiNet SHOULD do neither. During the SMTP transaction, when => MultiNet's SMTP server sees an invalid From: or To: field, it should reject the => message. Disposition of the message then falls to the machine that tried => sending the message with the bogus address. = = Actually, RFC1123 says that a receiving SMTP *MAY* validate destination =addresses as they are received and reject the destination, and says nothing =about validating the source address. That's true. If I'd meant to say that the RFCs required Multinet to validate source addresses, I'd've phrased it that way. The spirit of the RFCs seems to be that SMTP is designed to try to ensure that either the message is delivered or that the sender receive a notification of failure to deliver. If a host accepts a message with an unreplyable From: field, then there's no way that a failure to deliver notification can be returned to the sender (unless, of course, some or all the mailers en route parse the Received: lines to generate a return path, or the host that finds the From: field to be unreplyable mungs that field to include explicit routing back through the host that handed it the bad From: field. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from panix.com ([198.7.0.2]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 19:58:35 PDT Received: by panix.com id AA08114 (5.65c/IDA-1.4.4 for info-multinet@tgv.com); Mon, 23 Aug 1993 22:58:33 -0400 From: "Karen G. Schneider" Message-Id: <199308240258.AA08114@panix.com> Subject: Vague Misogyny To: info-multinet@tgv.com ================================================================================ Archive-Date: Mon, 23 Aug 1993 22:58:33 -0400 Date: Mon, 23 Aug 1993 22:58:33 -0400 (EDT) Cc: kgs@panix.com (Karen G. Schneider) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Length: 1270 People have different ways to ask questions. It concerns me that when a *woman* posts a general question to this board, she is suddenly a "moron." That insult is intended to silence the poster. If you don't like a question, why not try to write a *personal* note to the poster explaining how, and why, the question could have been phrased better? What do you accomplish by insulting people? The Internet is no longer an ol'-white-boys' network. If you can't function comfortably in an environment increasingly informed by a plurality of backgrounds and values, then perhaps you need to evaluate the efficacy of your participation on the 'Net. In the past, belittling newcomers may have sufficed to keep your kingdom free of us riffraff. But today, however, such behavior only sets you apart as an anachronism, expelling your asocial effluvia into the winds of change. Remember, "you're either on the bus, or you're off the bus." If you want to stay on the bus, make nice with the taxpaying "morons" who control your future. -- Karen G. Schneider kgs@panix.com karens@queens.lib.ny.us * * * "Revolutions wouldn't be so bad if you could eat at home every * * day with your family." -- _Like Water for Chocolate_ * X-ST-Status: N Return-path: Received: from VESTA.SUNQUEST.COM ([149.138.1.42]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 20:53:21 PDT ================================================================================ Archive-Date: Mon, 23 Aug 1993 20:53:19 -0700 Date: Mon, 23 Aug 1993 20:53:19 -0700 (MST) From: Ehud Gavron To: INFO-MULTINET@TGV.COM CC: GAVRON@VESTA.SUNQUEST.COM Subject: People who yell "discrimination" Date sent: 23-AUG-1993 20:49:48 >Subj: Vague Misogyny >From: kgs@panix.com (Karen G. Schneider) > >People have different ways to ask questions. It concerns me that >when a *woman* posts a general question to this board, she is suddenly >a "moron." It concerns me that when a stupid question is posted and this is pointed out, too many people are eager to find HIDDEN NEFARIOUS EVIL reasons for the response instead of the simple truth. She asked a stupid vague unanswerable question; she was told to quit being a moron. The same response would have been given a man. However, he would then not have come back to whine about discrimination. I SHUDDER TO THINK THAT ONE DAY I will be accused of being a jew-hating black-hating woman-hating nazi pig, just for telling some person that their question is moronic. Catch a clue and keep it. >Remember, "you're either on the bus, or you're off the bus." You're out in the weeds. Go visit Cluemasters. Clues prepared in... about an hour. >Karen G. Schneider kgs@panix.com karens@queens.lib.ny.us * * * Ehud -- Ehud Gavron (EG76) gavron@vesta.sunquest.com "The world bores you when you're cool." X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 23 Aug 1993 21:43:44 PDT Date: Mon, 23 Aug 93 21:43:44 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: name server To: carl @ SOL1.GPS.CALTECH.EDU cc: info-multinet @ TGV.COM > In article <930823082035.40a005b9@TGV.COM>, adelman@TGV.COM (Kenneth Adelman) writes: > => Actually, MultiNet SHOULD do neither. During the SMTP transaction, when > => MultiNet's SMTP server sees an invalid From: or To: field, it should reject the > => message. Disposition of the message then falls to the machine that tried > => sending the message with the bogus address. > = > = Actually, RFC1123 says that a receiving SMTP *MAY* validate destination > =addresses as they are received and reject the destination, and says nothing > =about validating the source address. > That's true. If I'd meant to say that the RFCs required Multinet to validate > source addresses, I'd've phrased it that way. The spirit of the RFCs seems to > be that SMTP is designed to try to ensure that either the message is delivered > or that the sender receive a notification of failure to deliver. If a host > accepts a message with an unreplyable From: field, then there's no way that a > failure to deliver notification can be returned to the sender (unless, of > course, some or all the mailers en route parse the Received: lines to generate > a return path, or the host that finds the From: field to be unreplyable mungs > that field to include explicit routing back through the host that handed it the > bad From: field. I think it is reasonable for a receiver SMTP to assume that the sender SMTP is sending a valid return address. Why? Because the overhead of checking to see if the return address is valid is prohibitive (one could argue that you actually need to send a message to it for a full verification). I've actually considered hacking TGV.COM's SMTP server to reject mail with bogus return host names because we get a lot of mail to SERVICE@TGV.COM that we can't reply to (rarely from misconfigured MultiNet machines; usually some brain-dead UUCP gateway). Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 23 Aug 93 23:02:45 PDT From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: RE: general question re multinet name serving capabilities ================================================================================ Archive-Date: xxx, 24 Aug 1993 05:37:38 GMT Date: 24 Aug 1993 05:37:38 GMT Organization: HST Wide Field/Planetary Camera Lines: 16 Distribution: world Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <0097173F.F8773260.23343@sacto.mp.usbr.gov>, jimww@sacto.mp.usbr.gov (Jim Wheeler # 916-978-5089) writes: =Don't hesitate to post it, but, Carl is just as abusive to anyone that asks a =question he disaproves of. Why does he even expend the effort if he thinks =someone is beneath him ? In the hopes that idiots like yourself who are impervious to gentle persuasion, might, eventually, learn to think before posting. I admit that you bear the signs ofg being a hopeless case, though -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from uvs1.orl.mmc.com ([129.243.52.3]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 03:06:55 PDT Received: by uvs1.orl.mmc.com (5.57/Ultrix3.0-C) id AA23975; Tue, 24 Aug 93 06:09:52 -0400 Received: from taurus (taurus.orl.mmc.com) by polaris.orl.mmc.com (4.1/SMI-4.1) id AA29714; Tue, 24 Aug 93 06:09:50 EDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 06:09:50 EDT Date: Tue, 24 Aug 93 06:09:50 EDT From: chitty@polaris.orl.mmc.com (Tom Chitty) To: info-multinet@tgv.com Subject: RE: People who yell "discrimination" Cc: chitty@polaris.orl.mmc.com Well, maybe to you it was a 'stupid' question, but to those of us less genius-like than you appear to be, it may be a very good question. If you don't like the question, or the fact that she is a woman, most of the rest of us could really care less. At least try to be tolerant of us 'stupid' people, both male and female, and pray that you never ask what one of use considers a 'stupid' question. Tom C. ]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_] ] Tom Chitty Martin Marietta Technologies ] ] chitty@polaris.orl.mmc.com PO Box 555837 ] ] (407)356-9194 - Voicemail MP - 1170 ] ] (407)356-9302 - FAX Orlando, FL 32855-5837 ] ] (407)291-0521 - Digital Pager ] ] ] ]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_] >Subj: Vague Misogyny >From: kgs@panix.com (Karen G. Schneider) > >People have different ways to ask questions. It concerns me that >when a *woman* posts a general question to this board, she is suddenly >a "moron." It concerns me that when a stupid question is posted and this is pointed out, too many people are eager to find HIDDEN NEFARIOUS EVIL reasons for the response instead of the simple truth. She asked a stupid vague unanswerable question; she was told to quit being a moron. The same response would have been given a man. However, he would then not have come back to whine about discrimination. I SHUDDER TO THINK THAT ONE DAY I will be accused of being a jew-hating black-hating woman-hating nazi pig, just for telling some person that their question is moronic. Catch a clue and keep it. >Remember, "you're either on the bus, or you're off the bus." You're out in the weeds. Go visit Cluemasters. Clues prepared in... about an hour. >Karen G. Schneider kgs@panix.com karens@queens.lib.ny.us * * * Ehud -- Ehud Gavron (EG76) gavron@vesta.sunquest.com "The world bores you when you're cool." ----- End Included Message ----- X-ST-Status: N Return-path: Received: from relay.iunet.it ([192.106.1.2]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 03:18:16 PDT Received: from cesi.UUCP by relay.iunet.it with UUCP id AA25344 (5.65c8/IDA-1.4.4 for TGV.COM!info-multinet); Tue, 24 Aug 1993 12:20:35 +0200 Received: by cesi.it (DECUS UUCP /2.0/2.0/2.0/); Tue, 24 Aug 93 12:17:38 +0200 Received: by cesi.it (MX V3.3 VAX) id 26071; Tue, 24 Aug 1993 12:17:27 +0200 ================================================================================ Archive-Date: Tue, 24 Aug 1993 12:17:26 +0200 Date: Tue, 24 Aug 1993 12:17:26 +0200 From: Sacchi Annamaria To: Adelman@TGV.COM Cc: info-multinet@TGV.COM, sacchi@cesi.it Subject: Customizing USER_LPD_CLIENT for MULTINET_LPD_SYMBIONT Hello! I mail you about my problems on Print-Symbiont and then I took holidays. But now I'm here again (sigh) with the same problems. To make you easer, let me make again the question. >> I modified the VMS print symbiont adding my PSM$K_INPUT_FILTER routine. >> Because My_Symbiont is multithreaded and i need to record some informations >> about each stream, more particularly about the current print job, i use a >> workarea as it's defined in Print Symbiont Modification manual. > >> Now, i would also modify the Symbiont to make it work with MULTINET. >> To do this I must customize MULTINET_LPD_SYMBIONT with my USER_LPD_CLIENT.C, >> as TGV suggests. >> My question is this: >> How can i define in USER_LPD_CLIENT the workarea, (its size is normally >> passed to Symbiont when PSM$PRINT is called while USER_LPD_CLIENT doesn't call >> psm$print) ? >> If it isn't possible, how can i do in order to save informations about each >> stream ? > Hmm. I can't see any easy way out of this. We currently use the >data area for our own data, and anything you put in it is going to >screw up our replacement output routines. About all I could think of is >using a static area to store a list of (buffer, address of data area) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >pairs, and search it each time you're called to fine your own private ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ >data area. Do you mean that i can use the address of the buffer containing the data to process as access key to my own private data area? There is a one to one corrispondence between this address and the stream? Even if it is so, i cannot identify the stream when the task starts. In fact the address of buffer is passed to PSM$K_INPUT_FILTER only when the function requested by Symbiont is PSM$K_FORMAT. Have you any suggest? We would suggest TGV to modify MULTINET_LPD_SYMBIONT in order to allow the customer to use the workarea, also maintaining a portion reserved to TGV itself. It would be also usefull if the customer could define the umber of threads. >> Another question: in USER_LPD_CLIENT is defined the function that replaces >> ROUTINE CODE of psm-routines with user-routines. Now, in the example i found, >> psm$replace use a FUNCTION CODE not a ROUTINE CODE, is it an error or don't I >> understand what there is behind ? >> I include the source of the function: >> int psm_init_v32(PSM$REPLACE) >> int (*PSM$REPLACE) (); >> { >> #ifdef notdef >> Code = PSM$K_READ; **** >> (*PSM$REPLACE) (&Code, Someroutine); **** >> #endif /*notdef*/ >> } > I think it is right. We replace the output routine using PSM$K_OUTPUT. I didn't explain sufficiently clear. I would know if this source is only a template (and the function-code, note it's not a routine-code, PSM$K_READ was used for error, in such case i can cancel the statements marked with **** and substitute them with code = PSM$K_INPUT_FILTER (*PSM$REPLACE) (&Code, MY_routine); or add these two statements and keep ****. In this case is it right to use the function-code PSM$K_READ? Don't i must use only routine-code for PSM$REPLACE? I hope to be more clear now and excuse me if I am not again. Many thanks. --------------------------------------------------------------------------- Annamaria Sacchi, SIC tel. +39 2 2125 344 CESI, Centro Elettrotecnico Sperimentale Italiano fax +39 2 2125 520 Via Rubattino, 54 - I 21034 Milano E-mail: sacchi@cesi.it X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 08:53:30 PDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 10:52:00 -0500 Date: Tue, 24 Aug 1993 10:52:00 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: general question Thanks to those who took the time to try to answer my vague question and to those who took the time to sympathesize/explain/answer during the ensuing exchange. I repeat at the bottom the question, which I later amended by replacing the word 'weird' with 'unusual.' In reading it over I can see why people like Carl could get frustrated, because it DOES sound as though I had in my mind some specific difficulty or an example which I wasn't divulging because I said 'which appears to me to be weird'. I meant to say 'which appear to be weird' or 'which are said to be weird.' It was triggered by some almost-forgotten mail about some multinet incompatibilities in the past which had been attributed to some unusual addresses generated on campus. If I had it to do over again, I would have phrased it better. I should have done that rather than give in and try to find an example (which I didn't have, and which turned out to have been about a mail server problem instead, in any event). This is not to say that I think that the question should have gotten the reaction that it did from a few people. Personal attacks are not an appropriate response. Like someone said, if it isn't understood/understandable, let someone ask, or ignore it. Perhaps some of you have no idea how little time some of us have to read/compose mail. Also, some of us like to try to get as broad a range of responses and participation as possible, and like to cast things broadly. The mailing lists and newsgroups are not limited to those who want themselves considered the world geniuses in something or another. People should not be discouraged from posting because they fear they will be trounced (men or women)! Anyway, I later asked a simpler question which boiled down to rating multinet's name serving performance in a mixed vms/unix environment. I have received a number of answers which have been reassuring, and I have appreciated them. I have also heard from women who had heard of similar studies about the experience of women posting to the internet. Apparently one study has indicated that the posts of women are examined more critically than those of men. I don't know anything about the differences in communication between the sexes, but I am aware that email and newsgroups present difficult communication problems. A little politeness works a lot better than bullying in making truly democratic communications. But I will leave that to the people who understand the process better than I do. Peace. Original ill-fated post: I would like to end our reliance on campus nameservers for a number of reasons. But this campus appears to use/promote all sorts of addressing which appears to me to be weird. Does anyone there have a feeling for how well the multinet name servers would respond to unusual addresses? Ginny Metze ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + Virginia Metze, not speaking for anyone but herself. + + metze@vmetze.mrl.uiuc.edu + + "'When I use a word,' Humpty Dumpty said in a rather + + scornful tone, 'it means just what I choose it to + + mean,--neither more nor less.'" Alice in Wonderland + ======================================================== | I remember you, Prairie Bayou! I knew you would | | die before you 'threw a clunker.' | ======================================================== X-ST-Status: N Return-path: Received: from CVAX.IPFW.INDIANA.EDU ([149.164.187.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 12:02:59 PDT Received: by CVAX.IPFW.INDIANA.EDU (MX V3.1C) id 27738; Tue, 24 Aug 1993 14:06:56 EST ================================================================================ Archive-Date: Tue, 24 Aug 1993 14:06:11 EST Date: Tue, 24 Aug 1993 14:06:11 EST From: "Joe Dillon, IU - PU at Fort Wayne, DP Dept." To: INFO-MULTINET@TGV.COM Subject: communication, my two cents There are several good books on gender communication differences by noted author, Deb Tannen, that might interest many of you. Two titles that I can present are: _You Just Don't Understand_, ISBN 0-345-37205-0 _That's Not What I Meant_, ISBN 0-345-34090-6. What you say and the words you choose to say it give clues to your gender, signatures anonymous. One poster wrote how males seem to ignore offensive statements, I think this is predominant but not categorical. Another poster wrote that oftentimes the signature line was ignored and that the signature is not conclusive of gender anyway, I agree. /====================================================================\ || Joseph H. Dillon /\ Manager of User Support || || Computing and Data Processing || || || Indiana-Purdue University || Dillon@Cvax.IPFW.indiana.edu || || 2101 Coliseum Boulevard East || (219) 481-6197 || || Fort Wayne, IN 46805-1499 \/ || \====================================================================/ X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 12:05:00 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Restricting DNS Info Message-ID: <1993Aug24.181800.16079@picker.com> From: SYSNET@engvax.picker.com (Leonard A. Visconti) ================================================================================ Archive-Date: Tue, 24 Aug 1993 18:18:00 GMT Date: Tue, 24 Aug 1993 18:18:00 GMT Sender: news@picker.com Organization: PICKER INTERNATIONAL Nntp-Posting-Host: engvax.picker.com X-News-Reader: VMS NEWS 1.23 Lines: 26 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I am working to make our DNS structure more vertical, but currently I still have about 30 systems or so in our top level 'picker.com' domain. I would like to break this into two separate zones delegated by two, distinct nameservers (NS). One NS (which will act as our external NS) will contain a zone file with just a few systems to be advertised to the Internet. The other NS (our internal NS) will have a zone file which contains the rest of these systems, and will share this information internally only. I want each NS to be an authoritative server for different parts of the same domain --- picker.com. And, although I want them to share this information with each other, I do not want the external NS to advertise to the Internet any node information it learns from our internal NS. Configuring both as secondaries to each other allows them to exchange host information with each other, but also causes the external NS to share this learned data with the outside. I would appreciate any help with this. Thank you. ------------------------------------------------------------------------------- | Leonard A. Visconti Picker International | | Network Analyst 595 Miner Road | | visconti@picker.com Highland Heights, Ohio 44143 | | (216)473-4801 | ------------------------------------------------------------------------------- X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 12:24:44 PDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 14:23:16 -0500 Date: Tue, 24 Aug 1993 14:23:16 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: lantastic I would like to trade experiences with someone else who has used lantastic for peer-to-peer PC networking. I am sure that this is out of the range of the interests of this group, so please send me email at metze@vmetze.mrl.uiuc.edu. Thank you. Ginny X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 12:37:23 PDT From: Mark.Berryman@Mvb.Saic.Com (MUMPS Wizard) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Restricting DNS Info Message-ID: <15091792@MVB.SAIC.COM> ================================================================================ Archive-Date: Tue, 24 Aug 1993 19:23:23 GMT Date: Tue, 24 Aug 1993 19:23:23 GMT Organization: Science Applications Int'l, Corp. - Computer systems operation Lines: 36 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET SYSNET@engvax.picker.com (Leonard A. Visconti) writes: >I am working to make our DNS structure more vertical, but currently I still >have about 30 systems or so in our top level 'picker.com' domain. I would >like to break this into two separate zones delegated by two, distinct >nameservers (NS). One NS (which will act as our external NS) will contain a >zone file with just a few systems to be advertised to the Internet. The other >NS (our internal NS) will have a zone file which contains the rest of these >systems, and will share this information internally only. > >I want each NS to be an authoritative server for different parts of the same >domain --- picker.com. And, although I want them to share this information >with each other, I do not want the external NS to advertise to the Internet >any node information it learns from our internal NS. > >Configuring both as secondaries to each other allows them to exchange host >information with each other, but also causes the external NS to share this >learned data with the outside. (I will use EX to refer to the system acting as a nameserver to the outside world and IN to refer to the internal only nameserver.) Create a file containing the zone information for all hosts you wish to advertise to the outside world as well as the SOA record for the zone. Place this file in a place where it can be read by both EX and IN. This file will contain a $INCLUDE directive. On EX the $INCLUDE needs to resolve to an empty file. On IN the $INCLUDE needs to resolve to the file that contains the hosts within this same zone that are to be resolved internally only (and must NOT contain an SOA record). Configure both IN and EX as PRIMARY for the zone in question and have them read the first file. Mark Berryman Mark.Berryman@Mvb.Saic.Com X-ST-Status: N Return-path: Received: from PFC.MIT.EDU ([18.77.0.113]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 12:58:06 PDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 15:58:05 -0400 Date: Tue, 24 Aug 1993 15:58:05 -0400 (EDT) From: Mark London To: info-multinet@tgv.com Subject: Info rpc? What is the best source for info on RPC, and what does MULTINET support with regard to RPC? Thanks. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 13:08:32 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: CLOSING state in MULTINET SHOW Message-ID: <1993Aug24.110932.1697@muvms6.wvnet.edu> From: boag@muvms6.wvnet.edu (Bob Boag) ================================================================================ Archive-Date: xxx, 24 Aug 1993 11:09:32 -0500 Date: 24 Aug 93 11:09:32 -0500 Organization: Marshall University Computer Center Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET This is probably a real simple question, but I'm new to MultiNet so please forgive me. I've noticed quite a few (somtimes 10 or 15) states which are listed as CLOSING when I enter a MULTINET SHOW command. It seems like they hang around for quite some time. What are the most common causes of this, and should I worry about them? -- Bob Boag Software Systems Analyst BITNET: boag@muvms6 Marshall University Internet: boag@muvms6.wvnet.edu University Computer Center Phone: 304/696-2624 Huntington, WV 25755-5320 FAX: 304/696-3601 X-ST-Status: N Return-path: Received: from medac3.med.utoronto.ca ([128.100.83.3]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 13:09:47 PDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 16:13:05 -0400 Date: Tue, 24 Aug 1993 16:13:05 -0400 (EDT) From: "Nancy Worth, Computing Support Div." To: info-multinet@tgv.com CC: WORTH@medac.med.utoronto.ca Subject: Irix 4.0.x sending mail to Multinet server Dear Multinetters, Has anyone had problems with Iris 4.0.x not being able to send mail to the DNS which is running Multinet 3.2C under VMS 5.4? The sendmail version is 5.65. The iris sendmail doesn't allow the mail server to be off of the domain. Mail can be bounced off the Multinet server to other sites around the world and even within the same domain, but will not send mail to any person on the server machine. It times out after the HELO. A telnet to port 25 is able to send a message. Anybody have any suggestions? Nancy. -- Nancy E. Worth worth@medac.med.utoronto.ca & postmaster (416)978-8504 Computing Support Div., Faculty of Medicine, Univ. of Toronto, ON Canada M5S 1A8 (Formerly Division of Medical Computing) Standard disclaimers... "I'm for Sail" X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 24 Aug 1993 13:30:33 PDT Date: Tue, 24 Aug 93 13:30:33 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: CLOSING state in MULTINET SHOW To: boag @ muvms6.wvnet.edu cc: info-multinet @ TGV.COM > This is probably a real simple question, but I'm new to MultiNet so please > forgive me. > I've noticed quite a few (somtimes 10 or 15) states which are listed as CLOSING > when I enter a MULTINET SHOW command. It seems like they hang around for quite > some time. What are the most common causes of this, and should I worry about > them? TCP's CLOSING state is entered when we send a FIN to close a connection (eg, in response to an application exiting or otherwise closing a connection) and receive a FIN from the other side before receiving an ACK of our FIN. We'll sit in this state retransmitting our FIN until acknowledged. I suspect the only way to spend significant amounts of time in this state is to have a misbehaving machine on the other end, perhaps because it is powered off before sending the ACK. Ken X-ST-Status: N Return-path: Received: from lobby.ti.com ([192.94.94.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 14:15:51 PDT Received: from mk1501.dseg.ti.com ([128.247.206.175]) by lobby.ti.com with SMTP (5.65c/LAI-3.2) id AA20747; Tue, 24 Aug 1993 16:19:11 -0500 Received: from mordor (mordor.dseg.ti.com) by mk1501.dseg.ti.com with SMTP (5.65c/LAI-3.2) id AA28420; Tue, 24 Aug 1993 16:10:32 -0500 Received: by mordor (4.1/SMI-4.1) id AA05543; Tue, 24 Aug 93 16:15:35 CDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 16:15:35 CDT Date: Tue, 24 Aug 93 16:15:35 CDT From: rbeck@mordor.dseg.ti.com (Ronald Beck) To: info-multinet@TGV.COM, METZE@VMETZE.MRL.UIUC.EDU Subject: Re: general question > From info-multinet-relay@TGV.COM Tue Aug 24 11:33:50 1993 > Date: Tue, 24 Aug 1993 10:52:00 -0500 (CDT) > From: METZE@VMETZE.MRL.UIUC.EDU > To: info-multinet@TGV.COM > Cc: METZE@VMETZE.MRL.UIUC.EDU > Subject: general question > Content-Length: 3417 > > ... much stuff about posting & patience deleted, > > Peace. > > Original ill-fated post: > I would like to end our reliance on campus nameservers for a number > of reasons. But this campus appears to use/promote all sorts of > addressing which appears to me to be weird. Does anyone there have > a feeling for how well the multinet name servers would respond > to unusual addresses? Do you have an example of the "unusual addresses"? What type of addressing does the campus promote and would they like to go to some standard. For example, do the addresses look like: person@system.domain.com or uunet!system!person or smtp%"user@domain.com" or user%user@domain.com It's hard to say how Multinet name servers would respond unless there's some idea of what types of address format will be encountered. In general, I have faith that Multinet can handle a lot of diverse systems. At least their track record so far supports my faith in their abilities. It would be easier to respond to the question though if more specifics were available. Hope this helps, Ron > > Ginny Metze > > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > + Virginia Metze, not speaking for anyone but herself. + > + metze@vmetze.mrl.uiuc.edu + > + "'When I use a word,' Humpty Dumpty said in a rather + > + scornful tone, 'it means just what I choose it to + > + mean,--neither more nor less.'" Alice in Wonderland + > ======================================================== > | I remember you, Prairie Bayou! I knew you would | > | die before you 'threw a clunker.' | > ======================================================== > X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 14:19:19 PDT From: owen@astro.washington.edu (Russell E. Owen) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: People who yell "discrimination" Followup-To: vmsnet.networks.tcp-ip.multinet ================================================================================ Archive-Date: Tue, 24 Aug 1993 13:39:58 -0700 Date: Tue, 24 Aug 1993 13:39:58 -0700 Organization: University of Washington Lines: 5 NNTP-Posting-Host: rowen.astro.washington.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Ehud is poisonous to everybody on the net. If my news reader had kill file capability, he would be the only person in it. Just be glad he's not a coworker. -- Russell X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 16:29:40 PDT From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Vague Misogyny ================================================================================ Archive-Date: xxx, 24 Aug 1993 22:58:55 GMT Date: 24 Aug 1993 22:58:55 GMT Organization: HST Wide Field/Planetary Camera Lines: 52 Distribution: world Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <199308240258.AA08114@panix.com>, kgs@panix.com (Karen G. Schneider) writes: =People have different ways to ask questions. It concerns me that =when a *woman* posts a general question to this board, she is suddenly =a "moron." That insult is intended to silence the poster. Would you have preferred that I use the term "shit-for-brains" as I have at times in the past in describing the sort of person who posts such vague questions, regardless of gender? Or is it that you're too feeble-minded to have read the rest of the posts in the thread before jumping in with you own ill-founded paranoid claims? =If you don't like a question, why not try to write a *personal* note =to the poster explaining how, and why, the question could have been =phrased better? When someone posts to the group complaining about network problems, there's no guarantee that any e-mail addressed to them will be received. =What do you accomplish by insulting people? In this case, it resulted in the person in question actually stating what it was that she REALLY wanted to know. =The Internet is no longer =an ol'-white-boys' network. If you can't function comfortably in an =environment increasingly informed by a plurality of backgrounds and =values, then perhaps you need to evaluate the efficacy of your =participation on the 'Net. Gee. You know, the same argument could be made for you. You're the one who apparently can't function comfortably in the environment. =In the past, belittling newcomers may have =sufficed to keep your kingdom free of us riffraff. I had no intention, despite your paranoid ravings, to keep anything free of any "riffraff." My intention was to get someone to actually THINK about the question being posted. =But today, however, =such behavior only sets you apart as an anachronism, expelling your =asocial effluvia into the winds of change. And you idiocy only sets you apart as someone who's too paranoid to think. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 16:29:42 PDT From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: name server ================================================================================ Archive-Date: xxx, 24 Aug 1993 23:02:16 GMT Date: 24 Aug 1993 23:02:16 GMT Organization: HST Wide Field/Planetary Camera Lines: 24 Reply-To: carl@SOL1.GPS.CALTECH.EDU NNTP-Posting-Host: sol1.gps.caltech.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930823214344.20800705@TGV.COM>, adelman@TGV.COM (Kenneth Adelman) writes: = I think it is reasonable for a receiver SMTP to assume that the =sender SMTP is sending a valid return address. Why? Because the =overhead of checking to see if the return address is valid is =prohibitive (one could argue that you actually need to send a message =to it for a full verification). I've actually considered hacking =TGV.COM's SMTP server to reject mail with bogus return host names =because we get a lot of mail to SERVICE@TGV.COM that we can't reply to =(rarely from misconfigured MultiNet machines; usually some brain-dead =UUCP gateway). It's a tradeoff between how much time the system spends validating the hosts on incoming mail and how much time the postmaster of the system ends up spending trying to figure out a return address from the headers once mail bounces and can't be returned to the sender. Perhaps a configurable option would be the best solution. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 16:59:38 PDT From: mparisi@ripple.jpl.nasa.gov (Mark Parisi) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Vague Misogyny ================================================================================ Archive-Date: xxx, 24 Aug 1993 23:35:11 GMT Date: 24 Aug 1993 23:35:11 GMT Organization: Jet Propulsion Laboratory Lines: 21 Distribution: world Reply-To: mparisi@ripple.jpl.nasa.gov (Mark Parisi) NNTP-Posting-Host: 137.79.114.102 X-Newsreader: mxrn 6.18-4 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET When I read this newsgroup, I tend to skip over headers and start reading the body of the article to see if it is of interest to me. When I saw this particular discussion start, my thoughts were: 1. I wish I knew what the poster was talking about -- what kind of strange addresses? Do I have to worry about this? 2. Somebody who doesn't know what is coming at them is going to get seriously flamed. It wasn't until cires of discrimination were heard that I noticed that the poster was a woman. While I can't claim to understand what goes on in the minds of those who flame, I think a male poster would have gotten the same answer. Peace. -- -- Mark Parisi (mparisi@ripple.jpl.nasa.gov) X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 17:14:23 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: RE: People who yell "discrimination" Message-ID: <25e70aINNl2v@gap.caltech.edu> From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) ================================================================================ Archive-Date: xxx, 24 Aug 1993 23:07:54 GMT Date: 24 Aug 1993 23:07:54 GMT Reply-To: carl@SOL1.GPS.CALTECH.EDU Distribution: world Organization: HST Wide Field/Planetary Camera NNTP-Posting-Host: sol1.gps.caltech.edu Lines: 32 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <9308241009.AA29714@polaris.orl.mmc.com>, chitty@polaris.orl.mmc.com (Tom Chitty) writes: =Well, maybe to you it was a 'stupid' question, but to those of us less =genius-like than you appear to be, it may be a very good question. Our educational system at work, I guess. Now it takes a genius to figure out that a question of the form: I've heard rumors that a piece of software had a bug (or was lacking a feature; I'm not going to tell you whether it was a bug or the lack of a feature) at some point (no, I'm not going to tell you when, either), processing certain types of data (and no, of course I won't tell you what kind of data caused the problem, nor will I even give you a list of the types of data that were being used). Does it still have this problem? is a stupid question. So, tell me, Tom, just how stupid do you have to be to think that the above question isn't stupid? If you just tell us how stupid YOU are, that'll give us a lower limit. =If you don't like the question, or the fact that she is a woman, The gener issue was a straw-person argument introduced by the person who asked the question in the first place. Your failure to notice that might give us some sort of upper limit on your intelligence. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 18:16:54 PDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 18:16:53 -0700 Date: Tue, 24 Aug 1993 18:16:53 -0700 (PDT) From: "L. Stuart Vance" Subject: Recent flamage To: info-multinet@TGV.COM Mail-System-Version: Organization: TGV, Inc. X-Phone: 408/457-5200 (work); 408/457-5205 (fax) X-Address: 101 Cooper Street; Santa Cruz, CA 95060 (work) Okay folks, enough is enough. info-multinet (and it's corresponding newsgroup) are not moderated. We don't want them to be moderated. Given these facts, we rely on the discretion of the subscribers to show proper etiquette in asking and answering questions. For the last week, alas, there has been more flamage on the list than actual technical information. This is bad for a number of reasons: o People's feelings get hurt (who really likes getting stepped on?) o Because of the low technical content, people who might benefit from the information presented on the list unsubscribe because they don't want to waste their time having to read through 10 or 15 flames to read the 2 or 3 messages that might be useful PLEASE keep the content of messages posted to this list/newsgroup technically oriented. If you can't stand seeing naive or beginner questions fly by the list occasionally, then please consider unsubscribing. If you feel you have to flame someone, please do it in a private mail message; there is no reason to CC info-multinet. In the next few weeks, we will be setting up a new mailing list, tgv-announce@tgv.com, where new release announcements, new product announcements, training announcements, etc. will be posted. This way, those of you who wish to receive information from TGV via the Internet will be able to without having to deal with the volume of info-multinet. If any of you have any questions about the above, please feel free to drop me a note or give me a call (although I may not be able to respond immediately - we're currently in the thick of Interop). Regards, -----Stuart X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 18:21:30 PDT ================================================================================ Archive-Date: Tue, 24 Aug 1993 20:19:51 -0500 Date: Tue, 24 Aug 1993 20:19:51 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: carl@SOL1.GPS.CALTECH.EDU CC: info-multinet@tgv.com, METZE@VMETZE.MRL.UIUC.EDU Subject: RE: People who yell "discrimination" I thought I carefully explained, so that even all of you genuises could understand, that I was not claiming 'discrimination' as such. I said that studies have shown that women posters are apparently intimidated by the treatment given them and other women when they post; treatment which men are apparently more willing to ignore. If you want more details, I can give you the name of the book that a couple of U of I women recently wrote/edited. What seems to you as normal give and take does not come across that way to us. Also, I submit that the question was vague. So what? Why does that deserve the kind of treatment it has been given. As to the apparent parody of the question below, it is far more ridiculous than the question. Many people immediately realized that I was simply trying to get a handle on how likely multinet name servers were to handle a wide variety of unusual addresses. As I have said about 10 times, I didn't have specific information to post because the excellent networking staff at UofI and TGV didn't give me such specific information. But that is all right; they were all men. I didn't think there was anything that odd about their not giving specifics even when warning of problems. I guess I should have really gotten on them, huh? I don't know why some people get their kicks from being internet bullies but it is not creating an atmosphere worthy of the millions being spent on it -- mostly our tax dollars. Maybe TGV could give us all an exam before we are allowed to post? >In article <9308241009.AA29714@polaris.orl.mmc.com>, chitty@polaris.orl.mmc.com >(Tom Chitty) writes: >=Well, maybe to you it was a 'stupid' question, but to those of us less >=genius-like than you appear to be, it may be a very good question. >... gross parody of my question deleted >The gener issue was a straw-person argument introduced by the person who asked >the question in the first place. Your failure to notice that might give us >some sort of upper limit on your intelligence. Now this is a really interesting approach. Because I objected to the WAY you objected to my vague question, anything I had to say about the kind of atmosphere that is conducive to discussion is to be ignored. That is really too much. I did not mind the criticism; I am one of the few people left on earth who seems perfectly willing to admit that they are sometimes wrong and sometimes don't say things well and sometimes don't understand. But I don't see anything useful in going into full attack mode at the first lapse. Or for that matter, for the nth lapse. It is not necessary to reduce everything to the level of a personal attack, especially one done with such ill humor. Ginny >-------------------------------------------------------------------------------- >Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL> >Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My >understanding of astronomy is purely at the amateur level (or below). So >unless what I'm saying is directly related to VAX/VMS, don't hold me or my >organization responsible for it. If it IS related to VAX/VMS, you can try to >hold me responsible for it, but my organization had nothing to do with it. X-ST-Status: N Return-path: Received: from whistler.sfu.ca ([142.58.103.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 21:28:35 PDT Received: from nickel.ucs.sfu.ca by whistler.sfu.ca (5.65/SFU-2.0) id AA12841; Tue, 24 Aug 93 21:28:33 -0700 Received: by nickel.ucs.sfu.ca (NX5.67c/NX3.0S) id AA10689; Tue, 24 Aug 93 21:27:48 -0700 From: Bill Baines Message-Id: <9308250427.AA10689@nickel.ucs.sfu.ca> Subject: Re: Info rpc? To: MRL@PFC.MIT.EDU ================================================================================ Archive-Date: Tue, 24 Aug 1993 21:27:47 PDT Date: Tue, 24 Aug 93 21:27:47 PDT Cc: info-multinet@TGV.COM X-Mailer: ELM [version 2.3 PL11] > > What is the best source for info on RPC, and what does MULTINET support with > regard to RPC? Thanks. > I can't speak for Multinet, but if you want a pretty good RPC intro look at "Power Programming with RPC", John Bloomer, O'Reilly & Associates, ISBN 0-937175-77-3. It talks about some different implementations, (SUN ONC, NCS, etc.) and has lots of example code. -- Bill Baines, Operations and Tech. Support, Simon Fraser University, Vancouver, British Columbia, Canada bill@sfu.ca, (604) 291-3955, (fax 291-4242), VE7FML X-ST-Status: N Return-path: Received: from whistler.sfu.ca ([142.58.103.1]) by TGV.COM via INTERNET ; Tue, 24 Aug 93 22:18:46 PDT Received: from nickel.ucs.sfu.ca by whistler.sfu.ca (5.65/SFU-2.0) id AA14084; Tue, 24 Aug 93 22:18:43 -0700 Received: by nickel.ucs.sfu.ca (NX5.67c/NX3.0S) id AA10731; Tue, 24 Aug 93 22:17:58 -0700 From: Bill Baines Message-Id: <9308250517.AA10731@nickel.ucs.sfu.ca> Subject: tcpdump question To: info-multinet@TGV.COM ================================================================================ Archive-Date: Tue, 24 Aug 1993 22:17:57 PDT Date: Tue, 24 Aug 93 22:17:57 PDT X-Mailer: ELM [version 2.3 PL11] When I use the command "$MU TCPD PORT NFS" I see descriptions of the NFS traffic on an isolated NFS network -- same network the Multinet host is on, but the VAX is not playing NFS. Some of the NFS packet descriptions from the server have the following information appended to the record: reply ok OK 1472 (frag 6013:1480@0+) reply ok OK 1472 (frag 6014:1480@0+) . . . reply ok Errno 28261 1472 (frag 5ff3:1480@0+) reply ok Errno 28261 1472 (frag 5ff4:1480@0+) The hex after "frag" is incrementing. Can someone comment? Is this badness? -- Bill Baines, Operations and Tech. Support, Simon Fraser University, Vancouver, British Columbia, Canada bill@sfu.ca, (604) 291-3955, (fax 291-4242), VE7FML X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 01:30:20 PDT From: zhangj@igd.fhg.de (Jian Zhang) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: ftp programming interface ================================================================================ Archive-Date: xxx, 24 Aug 1993 07:30:33 GMT Date: 24 Aug 93 07:30:33 GMT Organization: IGD Darmstadt Lines: 14 NNTP-Posting-Host: batida.igd.fhg.de Summary: is there lower level programming interface to ftp Keywords: ftp programming interface To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hello, using command or script can man use ftp to transport files among computers connected to network. Is there other lower level programming interface to ftp, so that one can develop himself a client program on a local machine to communicate with ftp server on any machine through network? thank you? Jian P.S. email: zhangj@igd.fhg.de X-ST-Status: N Return-path: Received: from uvs1.orl.mmc.com ([129.243.52.3]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 03:34:43 PDT Received: by uvs1.orl.mmc.com (5.57/Ultrix3.0-C) id AA26367; Wed, 25 Aug 93 06:37:40 -0400 Received: from taurus (taurus.orl.mmc.com) by polaris.orl.mmc.com (4.1/SMI-4.1) id AA11460; Wed, 25 Aug 93 06:37:37 EDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 06:37:37 EDT Date: Wed, 25 Aug 93 06:37:37 EDT From: chitty@polaris.orl.mmc.com (Tom Chitty) To: info-multinet@tgv.com Subject: RE: People who yell "discrimination" Cc: chitty@polaris.orl.mmc.com I think you (easily) confuse a stupid question, with one that is not clear (in fact the term vague was used in one e-mail). Why on earth did the kind sole decide to flame the sender, rather than simply saying "Hey, your e-mail is not straight forward, can you state your question clearly?" Gee, what a novel idea (at least to this area), being _KIND_ to someone instead of an obnoxious, flaming poster. I would gladly tell you how stupid I am, but my boss is much better at it than I, and he is currently too busy to write back to you ;) I 'jumped' into the fray because I find it hard to believe that people like ehund actually exist, and that he considers flaming someone who posts a vague question to be the correct thing to do. Gee, I guess I learned my lesson, huh? It appears as though ehund (and yourself) consider yourselves the guardians of the Free World, preventing 'stupid' questions by simply flaming back at the sender. I will sleep soundly tonight, knowing that you two are watching my e-mail, and protecting the internet. Why not simply respond to e-mail that is vague with the old stand-by, 'what did you mean?' Tom C. ]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_] ] Tom Chitty Martin Marietta Technologies ] ] chitty@polaris.orl.mmc.com PO Box 555837 ] ] (407)356-9194 - Voicemail MP - 1170 ] ] (407)356-9302 - FAX Orlando, FL 32855-5837 ] ] (407)291-0521 - Digital Pager ] ] ] ] SMILEY of the week: #:o+= - Betty Boop ] ] ] ]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_]_] ----- Begin Included Message ----- From info-multinet-relay@tgv.com Tue Aug 24 20:44:03 1993 X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: RE: People who yell "discrimination" From: carl@sol1.gps.caltech.edu (Carl J Lydick) ================================================================================ Archive-Date: xxx, 24 Aug 1993 23:07:54 GMT Date: 24 Aug 1993 23:07:54 GMT Reply-To: carl@sol1.gps.caltech.edu Distribution: world Organization: HST Wide Field/Planetary Camera Nntp-Posting-Host: sol1.gps.caltech.edu Lines: 32 To: INFO-MULTINET@tgv.com X-Gateway-Source-Info: USENET Content-Length: 1792 In article <9308241009.AA29714@polaris.orl.mmc.com>, chitty@polaris.orl.mmc.com (Tom Chitty) writes: =Well, maybe to you it was a 'stupid' question, but to those of us less =genius-like than you appear to be, it may be a very good question. Our educational system at work, I guess. Now it takes a genius to figure out that a question of the form: I've heard rumors that a piece of software had a bug (or was lacking a feature; I'm not going to tell you whether it was a bug or the lack of a feature) at some point (no, I'm not going to tell you when, either), processing certain types of data (and no, of course I won't tell you what kind of data caused the problem, nor will I even give you a list of the types of data that were being used). Does it still have this problem? is a stupid question. So, tell me, Tom, just how stupid do you have to be to think that the above question isn't stupid? If you just tell us how stupid YOU are, that'll give us a lower limit. =If you don't like the question, or the fact that she is a woman, The gener issue was a straw-person argument introduced by the person who asked the question in the first place. Your failure to notice that might give us some sort of upper limit on your intelligence. -------------------------------------------------------------------------------- Carl J Lydick | INTERnet: CARL@SOL1.GPS.CALTECH.EDU | NSI/HEPnet: SOL1::CARL Disclaimer: Hey, I understand VAXen and VMS. That's what I get paid for. My understanding of astronomy is purely at the amateur level (or below). So unless what I'm saying is directly related to VAX/VMS, don't hold me or my organization responsible for it. If it IS related to VAX/VMS, you can try to hold me responsible for it, but my organization had nothing to do with it. ----- End Included Message ----- X-ST-Status: N Return-path: Received: from ACC.HAVERFORD.EDU ([165.82.1.30]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 06:03:48 PDT Received: from ACC.HAVERFORD.EDU by ACC.HAVERFORD.EDU (PMDF #12293) id <01H25NVFNTV895NXHB@ACC.HAVERFORD.EDU>; Wed, 25 Aug 1993 09:06 EDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 09:06 EDT Date: Wed, 25 Aug 1993 09:06 EDT From: DBESTIDE@ACC.HAVERFORD.EDU Subject: flame wars To: info-multinet@TGV.COM X-Envelope-to: info-multinet@TGV.COM X-VMS-To: IN%"info-multinet@TGV.COM" X-VMS-Cc: DBESTIDE hey guys/gals. this is getting ridiculous! i am on this list to receive technical information, nothing else. it is getting me po'd that in reviewing the list daily, there is more *garbage* to sift through than what is appropriate. lists are not for this purpose. why don't you take your comments offline, one on one, with the appropriate individual? diane bestider haverford college X-ST-Status: N Return-path: Received: from deneb.mc.duke.edu ([152.3.78.66]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 06:10:08 PDT Received: by deneb.mc.duke.edu (5.65/DEC-Ultrix/4.3) id AA11111; Wed, 25 Aug 1993 09:06:51 -0400 ================================================================================ Archive-Date: Wed, 25 Aug 1993 09:01:10 -0400 Date: Wed, 25 Aug 1993 09:01:10 -0400 (EDT) From: "James D. Dryfoos" Subject: Re: ftp programming interface To: Jian Zhang Cc: INFO-MULTINET@TGV.COM Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 24 Aug 1993, Jian Zhang wrote: > Date: Tue, 24 Aug 1993 07:30:33 +0000 (GMT) > From: Jian Zhang > To: INFO-MULTINET@TGV.COM > Subject: ftp programming interface > > Hello, > > using command or script can man use ftp to transport files among computers > connected to network. Is there other lower level programming interface > to ftp, so that one can develop himself a client program on a local > machine to communicate with ftp server on any machine through network? > > thank you? > > Jian > > P.S. email: zhangj@igd.fhg.de > > A callable interface to FTP would be one of the greatest things in the world. Can't tell you how many ciritical applications I have doen where I end of spawn to VMS to execute a com file with some FTP commands in it. This is very crude and error handling is almost nill. Anyway, I just started looking for this myself (although I am not using Multinet). I did find on VMSSERV something called LIBFTP. Here is the readme: LIBFTP is a program interface to FTP servers. In effect it allows you to run FTP from batch procedures. I've done a quick port from U*x to Multinet. The performance is terrible as is the usual case for simple minded ports, but it DOES work. ---- If you have any questions or suggestions, I'll be glad to listen to them. If you find any bugs, be a sport and let me know what you've fixed. I'ld LOVE to see version of this rewritten in terms of NETLIB. It would have a lot better performance and it could be used by anybodies TCP/IP stack. Maybe I'll do it when I get some spare time. Dick Munroe Doyle, Munroe Consultants. 267 Cox St. Hudson, Ma. 01749 munroe@dmc.com (508) 568-1618 FAX: (508) 562-1133 I would like to here more on this. -- Jim ======================================================================= James D. Dryfoos | dryfo001@mc.duke.edu Duke University Medical Center | dryfo001@dukemc.bitnet 2200 West Main Street | Suite 450 Room 36A | (919) 286-6391 - office Durham, NC 27710 Earth | (919) 286-6369 - fax ======================================================================= X-ST-Status: N Return-path: ================================================================================ Archive-Date: Wed, 25 Aug 1993 06:38:53 PDT Date: Wed, 25 Aug 93 06:38:53 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: tcpdump question To: bill @ nickel.ucs.sfu.ca cc: info-multinet @ TGV.COM > When I use the command "$MU TCPD PORT NFS" I see descriptions of the > NFS traffic on an isolated NFS network -- same network the Multinet host > is on, but the VAX is not playing NFS. > Some of the NFS packet descriptions from the server have the following > information appended to the record: > reply ok OK 1472 (frag 6013:1480@0+) > reply ok OK 1472 (frag 6014:1480@0+) > . > . > . > reply ok Errno 28261 1472 (frag 5ff3:1480@0+) > reply ok Errno 28261 1472 (frag 5ff4:1480@0+) > The hex after "frag" is incrementing. Can someone comment? Is this > badness? This is TCPDUMP's way of telling you that you are looking at the first fragment of a fragmented IP datagram. The syntax is: (frag ID:SIZE@OFFSET) (The + means that there are more fragments) So what you're seeing is the first 1480 bytes of fragments with the IP ID (sequence count) of 5ff3 and 5ff4. If you TCPDUMP all IP packets between those two machines, you'll see the other fragments as well. ken X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 06:47:22 PDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 8:45:47 -0500 Date: Wed, 25 Aug 1993 8:45:47 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: carl@SOL1.GPS.CALTECH.EDU CC: info-multinet@tgv.com, METZE@VMETZE.MRL.UIUC.EDU Subject: Re: Vague Misogyny > >From: carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) >In article <199308240258.AA08114@panix.com>, kgs@panix.com (Karen G. Schneider) writes: >=What do you accomplish by insulting people? >In this case, it resulted in the person in question actually stating what it >was that she REALLY wanted to know. > I hate to be the one to break it to you, but YOUR flaming me was not the reason, but the fact that several people more politely had obvious difficulty figuring out what the hell I wanted. However, I have had conversations with you via email and I have concluded that you are indeed human. This probably really is your way of trying to make me think about what I say. But with all due respect, it does not necessarily have the effect of making one resort to reasoning that you hoped. Perhaps you have the same things going on all day, just as I do: People coming in that 'the printer doesn't work,' 'the machine gave wrong answers,' 'you changed the library,' etc., etc. (Specific? heck, I haven't seen specific for so long I don't remember what it looks like...) I have to be polite to all of these people and try to figure out what they really want. I can't insult them, and actually, since they obviously have SOME problem I have to worry about, I don't even want to. But it is different and more tempting on the internet. What a great way to vent one's frustrations. But if bullying, insulting, etc., is going to carry the day on the 'net' more people than you ever know will be turned off. As usual as you are, there seem to be lots of people who have you in their 'kill' file. I probably shouldn't have written this before reading my mail. Ginny X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 07:20:31 PDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 9:19:03 -0500 Date: Wed, 25 Aug 1993 9:19:03 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: clarification of comment about tgv In a previous post I made an offhand comment which it has been pointed out to me can be taken in a way that I certainly never intended. The comment was: > >Maybe TGV could give us all an exam before we are allowed to post? > I meant only that if the posters (NOT TGV) expected so much of those posting, perhaps we should take an exam to make sure we meet the standards before entering the 'fray.' I certainly did not intend, as could be implied, that TGV's conduct in any way made that necessary. TGV has always been extremely polite and responsible in ANY message that I have seen from them on this mailing list, and I am sincerely grieved that I seemed to impugn them. Ginny X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 07:38:29 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: People who yell "discrimination" Message-ID: From: Stafford@Vax2.Winona.MSUS.edu (John Stafford) ================================================================================ Archive-Date: xxx, 25 Aug 1993 09:24:00 -0600 Date: 25 Aug 93 09:24:00 -0600 Followup-To: vmsnet.networks.tcp-ip.multinet Distribution: world Organization: Minnesota State University @ Winona Nntp-Posting-Host: 134.29.65.55 Lines: 25 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <9308251037.AA11460@polaris.orl.mmc.com>, chitty@polaris.orl.mmc.com (Tom Chitty) wrote: > [...] > I 'jumped' into the fray because I find it hard to believe that people like > ehund actually exist, [...] It is interesting that those who offer some of the most concise, accurate and informed technical advice have such difficult personalities _and we tolerate it_. Why? We recognize aberation and simply cope. If any of you have been fortunate enough to read Lydick's posting in other forums, you might find that he has a sharply critical mind which is highly focused upon rational discourse - a very discursive orientation. What makes him a good programmer and logician (and skeptic) unfortunately dominates his expectations of human discourse. Lydic and company are excellent models of personalities that simply don't communicate well in the human pardigm. --- John Stafford Minnesota State University @ Winona X-ST-Status: N Return-path: Received: from SACWMS.MP.USBR.GOV ([140.214.12.9]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 07:44:42 PDT Received: by sacto.mp.usbr.gov (MX V3.3 VAX) id 24313; Wed, 25 Aug 1993 07:44:14 PDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 07:44:03 PDT Date: Wed, 25 Aug 1993 07:44:03 PDT From: Jim Wheeler # 916-978-5089 To: zhangj@igd.fhg.de CC: info-multinet@tgv.com, jimww@sacto.mp.usbr.gov Subject: RE: ftp programming interface >From: MX%"zhangj@igd.fhg.de" 25-AUG-1993 01:56:12.79 > >Hello, > >using command or script can man use ftp to transport files among computers >connected to network. Is there other lower level programming interface >to ftp, so that one can develop himself a client program on a local >machine to communicate with ftp server on any machine through network? > There are two solutions we use here. One has the originating side on the VAX using Multinet, but the destination is a CYBER. The code is a VAX queue that sends and receives ftp to the CYBERS. The second solution is more generic, it uses LPD to shift files around our network using filters to put the files somewhere. Some security issues have been addressed, but not all. Contact Henry Miller - henrym@sacto.mp.usbr.gov for more info. >thank you? > >Jian > >P.S. email: zhangj@igd.fhg.de jimww # James W. Wheeler # # # Regional Systems & WAN Manager # voice: 916 978-5089 fax: 916-978-5284 # # Mid Pacific Region # # # U.S. Bureau of Reclamation # inet: jimww@sacto.mp.usbr.gov # # 2800 Cottage Way MP1130 # # # Sacramento, CA 95825 # # X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 07:56:25 PDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 9:54:57 -0500 Date: Wed, 25 Aug 1993 9:54:57 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: use of DECNET<=>TCP/IP gateway instead of some uses of ftp This comment addresses only a few of the cases in which a callable interface to ftp would be useful, but if you have a DECNET<=>TCP/IP gateway that you can use, one can reference a file on a unix machine using the general filename format of VMS. That is, copy uimrl7"root@mrlaxm password"::"/etc/rc.tcpip" rc_tcpip.mrlaxm X-ST-Status: N Return-path: Received: from primus.cstp.umkc.edu ([134.193.2.53]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 08:21:38 PDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 10:21:27 -0500 Date: Wed, 25 Aug 1993 10:21:27 -0500 (CDT) From: Brian McKeever To: mparisi@ripple.jpl.nasa.gov CC: info-multinet@tgv.com Subject: RE: Vague Misogyny > When I read this newsgroup, I tend to skip over headers and start > reading the body of the article to see if it is of interest to me. > > When I saw this particular discussion start, my thoughts were: > > 1. I wish I knew what the poster was talking about -- what kind > of strange addresses? Do I have to worry about this? > > 2. Somebody who doesn't know what is coming at them is going to > get seriously flamed. > > It wasn't until cires of discrimination were heard that I noticed that > the poster was a woman. > > While I can't claim to understand what goes on in the minds of those > who flame, I think a male poster would have gotten the same answer. > > Peace. I agree. Boy! Errr... Person! Do I agree! I enjoy INFO-MULTINET. I have received very useful information/programs from a lot of individuals out there, including (and especially) Ehud, and so have others. In his defense, I kind-of feel that using a kill file to keep from seeing anything from him is sort-of like cutting off your nose to spite your face. IMHO, technically, he knows what he's talking about. But it IS, of course, your call. I do wish he (a few others, and myself at times) would be a little more tolerant of nascent system management questions. But EVERYONE knows about Ehud. As Mark pointed out in 1 & 2 above the question was more than a bit vague, and I too "expected" something like what happened. And if he wasn't on vacation, Ehud would have been a good bet to have started the ball rolling. Come on, laugh at and then delete what you don't like. We're all adults after all. Although I know nobody on this list personally, I don't think that anyone meant to flame anyone else because of gender. I've seen it happen to LOTS of others over the months (e'gad, that's years!) on this and other lists. God knows I try to gather as much information as possible before I bother hundreds of other system & network managers. Call it professional courtesy. Being a straight white male means that I've not been subjected to anything I'd classify as discrimination so I'm not sensitized to it, but I do think that some individuals I've run across should re-orient the order of their introspection queues though. Did I provide enough information? Were the sentences clear and concise as possible? Did I demonstrate that I knew what I was talking about? And then maybe, only maybe, consider descrimination. Discrimination is the easy out. It requires no personal changes at all. We all make mistakes and I've asked more than my fair share of stupid questions (and been flamed) and haven't once thought it was because someone didn't like men or WASPs. However, I can also see how a female working in our profession would have to have a healthy ego to survive. Unfortunately, a healthy ego sometimes makes us jump to the wrong conclusions. Anyway, my 2 cents worth. Sorry for using your list for this TGV. -=Brian X-ST-Status: N Return-path: Received: from FEDC06.FED.ORNL.GOV ([128.219.48.6]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 08:48:59 PDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 11:48:57 -0400 Date: Wed, 25 Aug 1993 11:48:57 -0400 (EDT) From: Dave Greenwood To: INFO-MULTINET@TGV.COM Subject: Lpd problem after IBM upgrade One of our IBM RS6000 workstations was upgraded from AIX 3.2 to AIX 3.2.4. Before the upgrade we had no problems printing from the IBM to our VAX printers, which are served by lpd to the IBMs. Since the upgrade nothing gets printed. Obviously, this looks like an IBM problem but I'm hoping for some help on how to identify the problem. There are no error messages displayed on the IBM and none in the MULTINET:LPD_SERVER.LOG or MULTINET:_LPD.LOG files. The files are copied to our VAX and are in MULTINET_SPOOL:.LPD__0 files. I assume the is a pid. The _LPD.LOG file contains an entry for each file such as the following: Job morris.txt from gilesca@fedu55, queue ln3, local user gilesca - Aug 25 10:57 But the entry doesn't contain any of the additional info such as for the following entry from another IBM node: Job STDIN.15028 from root@fedu54, queue ln3 - Aug 25 11:30 /queue=ln3/name=STDIN.15028 file: _VAR_SPOOL_QDAEMON_TORQBPO.;1 (8 blocks) Job STDIN.15028 (queue LN3, entry 318) started on LN3 Jobs which are printed on a PC which serves a queue to the IBM are printed just fine - it's only the jobs which are queued to the VAX printers that don't print. The IBM system person says no changes were made to the /etc/printcap file during the upgrade (at least intentionally anyway). We're running VMS A5.5-2 and MultiNet 3.2C. I restarted the server process but that didn't help. Any suggestions on what the problem is or how to track the problem down? Thanks, Dave -------------- Dave Greenwood Internet: Greenwoodde@ORNL.GOV Oak Ridge National Lab %STD-W-DISCLAIMER, I don't speak for nobody else X-ST-Status: N Return-path: Received: from AC.GRIN.EDU ([132.161.10.8]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 11:13:37 PDT Received: from DREAMS.GRIN.EDU by VMSMAIL.GRIN.EDU (PMDF V4.2-11 #4381) id <01H25WH43FDC8Y5TTJ@VMSMAIL.GRIN.EDU>; Wed, 25 Aug 1993 13:12:19 CDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 13:12:15 -0500 Date: Wed, 25 Aug 1993 13:12:15 -0500 (cdt) From: Brian D McMahon Subject: Re: flame wars To: info-multinet@TGV.COM DBESTIDE@ACC.HAVERFORD.EDU writes: > hey guys/gals. this is getting ridiculous! NO KIDDING. Folks, let's all please SHUT UP. Take it to the talk.* groups, or go find something PRODUCTIVE to do. Thank you for your cooperation. Brian McMahon Postmaster / Acad. Software Support Grinnell College Computer Services Grinnell, Iowa 50112 USA Voice: +1 515 269 4901 Fax: +1 515 269 4936 Motto on Innosoft t-shirt: "Making the world a safer place for e-mail." X-ST-Status: N Return-path: Received: from baylor.ccis.baylor.edu ([129.62.1.2]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 13:17:04 PDT Received: from baylor.edu by baylor.edu (PMDF V4.2-11 #3376) id <01H260S2POPWA733AL@baylor.edu>; Wed, 25 Aug 1993 15:16:48 CDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 15:16:48 -0500 Date: Wed, 25 Aug 1993 15:16:48 -0500 (CDT) From: David Severson Subject: Re: People who yell "discrimination" To: INFO-MULTINET@TGV.COM X-VMS-To: IN%"INFO-MULTINET@TGV.COM" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Gentleman, Please take this outside. The rest of us would like to work on solving our computer problems. David Severson Baylor University Seversond@baylor X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 13:43:24 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: flame wars Message-ID: <930825095133@reedlb.ho.dupont.com> From: reedlb@[138.196.90.13] ("Boyd Reed, ReedLB@hoser.dnet.dupont.com") ================================================================================ Archive-Date: Wed, 25 Aug 1993 14:51:33 GMT Date: Wed, 25 Aug 1993 14:51:33 GMT Sender: news@es.dupont.com (USENET News System) Organization: Du Pont Experimental Station Nntp-Software: PC/TCP NNTP Nntp-Posting-Host: reedlb.ho.dupont.com Lines: 26 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <01H25NVFNTV895NXHB@ACC.HAVERFORD.EDU> DBESTIDE@ACC.HAVERFORD.EDU writes: >hey guys/gals. this is getting ridiculous! > >i am on this list to receive technical information, nothing else. it is getting me po'd that in reviewing the list daily, there is more *garbage* to sift >through than what is appropriate. lists are not for this purpose. > >why don't you take your comments offline, one on one, with the appropriate >individual? > >diane bestider >haverford college AMEN! -- ------------------------------------------------------------------------------ L. Boyd Reed, II |DISCLAIMER: The opinions are my own, Associate Systems Analyst | not those of Conoco/DuPont Distributed Computing Division |-------------------------------------- Conoco Information Systems | Conoco, Inc. | "Save a tree, eat a gopher" Houston, TX | | -- Unknown Internet: ReedLB@hoser.dnet.DuPont.com| ----------------------------------------------------------------------------- X-ST-Status: N Return-path: Received: from ingate.adc.com ([155.226.10.200]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 16:54:17 PDT Received: by ingate.adc.com (5.65/DEC-Ultrix/4.3) id AA02263; Wed, 25 Aug 1993 18:55:07 -0500 Received: from valandil.YP.sun386i by saruman (4.0/SMI-4.0) id AA03541; Wed, 25 Aug 93 18:54:04 CDT Received: by valandil.YP.sun386i (4.1/SMI-4.1) id AA07396; Wed, 25 Aug 93 18:53:33 CDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 18:53:33 CDT Date: Wed, 25 Aug 93 18:53:33 CDT From: mal@valandil.adc.com (Michael A Leo) To: info-multinet@tgv.com Subject: SunOS and Ultrix equivalents of TRACEROUTE Hi, What is the SunOS and Ultrix equivalents of the MULTINET TRACEROUTE command? X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 17:27:29 PDT Received: from psulias.psu.edu by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H26BFH4MVK8ZFV1V@psulias.psu.edu>; Wed, 25 Aug 1993 20:27:25 EDT ================================================================================ Archive-Date: Wed, 25 Aug 1993 20:27:25 -0400 Date: Wed, 25 Aug 1993 20:27:25 -0400 (EDT) From: "J.Lance Wilkinson, 814-865-1818" Subject: Limit number of concurrent outgoing TELNET connections? To: INFO-MULTINET@psulias.psu.edu Organization: Penn State University / University Libraries X-VMS-To: IMULTINET X-VMS-Cc: JLW MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Is there any way we can cause the multinet telnet client to limit the number of concurrent outgoing connections a user has? We tend to lock up our users pretty well, but we have the situation where users hit the escape sequence and connect to another host, and then another, and then another. We'd like to be able to limit these to one or two at a time, per client. +-"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+ | J.Lance Wilkinson ("Lance") BitNet: JLW@PSULIAS.BITNET | | Systems Design Specialist - Lead InterNet: jlw@psulias.psu.edu | | Library Computing Services AT&T:(814) 865-1818 FAX:(814)863-3560 | | E8 Pattee Library "I'd rather be dancing..." | | Penn State University A host is a host from coast to coast, | | University Park, PA 16802 And no one will talk to a host that's close | | Unless the host that isn't close | | Is busy, hung or dead. | +------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+ X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 20:35:30 PDT From: ewilts@vmsmail.gov.bc.ca (Ed Wilts) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: TGV-announce (was: Recent flamage) Message-ID: <1993Aug25.084842.2979@vmsmail.gov.bc.ca> ================================================================================ Archive-Date: xxx, 25 Aug 1993 08:48:42 PST Date: 25 Aug 93 08:48:42 PST Reply-To: EWILTS@GALAXY.GOV.BC.CA Organization: BC Systems Corporation Lines: 14 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <746241413.537723.VANCE@TGV.COM>, VANCE@TGV.COM (L. Stuart Vance) writes: > In the next few weeks, we will be setting up a new mailing list, > tgv-announce@tgv.com, where new release announcements, new product > announcements, training announcements, etc. will be posted. This way, those > of you who wish to receive information from TGV via the Internet will be able > to without having to deal with the volume of info-multinet. Stuart, I'd like to suggest that the tgv-announce list gets gatewayed to a new newsgroup, biz.tgv. This would be along the lines of biz.dec where we see all the Digital press releases. -- Ed Wilts, BC Systems, 4000 Seymour Place, Victoria, B.C., Canada, V8X 4S8 EWilts@Galaxy.Gov.BC.CA Office: (604) 389-3430 Fax: (604) 389-3412 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 20:35:31 PDT From: ewilts@vmsmail.gov.bc.ca (Ed Wilts) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: DECnet over IP - address remapping Message-ID: <1993Aug25.085243.2980@vmsmail.gov.bc.ca> ================================================================================ Archive-Date: xxx, 25 Aug 1993 08:52:42 PST Date: 25 Aug 93 08:52:42 PST Reply-To: EWILTS@GALAXY.GOV.BC.CA Organization: BC Systems Corporation Lines: 12 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In configuring DECnet over IP using MultiNet, there appears to be no way to do any DECnet area/number re-mapping. This functionality is currently served by a few sites that use Cisco routers, but I don't know how to solve this problem using MultiNet. Are there any plans to allow name/number remapping in a future release of MultiNet? How about filtering? This would make it a *lot* easier to join two large networks... -- Ed Wilts, BC Systems, 4000 Seymour Place, Victoria, B.C., Canada, V8X 4S8 EWilts@Galaxy.Gov.BC.CA Office: (604) 389-3430 Fax: (604) 389-3412 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Wed, 25 Aug 93 21:03:53 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: henrym@sacto.mp.usbr.gov Subject: RE: ftp programming interface Sender: news@sacsa3.mp.usbr.gov ( News Login Account) Message-ID: <0097190F.5F580C90@sacto.mp.usbr.gov> ================================================================================ Archive-Date: Thu, 26 Aug 1993 03:32:09 GMT Date: Thu, 26 Aug 1993 03:32:09 GMT Reply-To: henrym@sacto.mp.usbr.gov Organization: United States Bureau of Reclamation, Mid Pacific Region, Sacramento, CA Lines: 66 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <009718A4.11FDF740.24313@sacto.mp.usbr.gov>, jimww@sacto.mp.usbr.gov (Jim Wheeler # 916-978-5089) writes: >>From: MX%"zhangj@igd.fhg.de" 25-AUG-1993 01:56:12.79 >> >>Hello, >> >>using command or script can man use ftp to transport files among computers >>connected to network. Is there other lower level programming interface >>to ftp, so that one can develop himself a client program on a local >>machine to communicate with ftp server on any machine through network? >> There are two portions of this code, the server and the client. Both run as VMS symbionts. The client symbiont queues files to the remote system, in this case, a CYBER and shoves the data into a temporary queue file, then creates a command file that submits the first file. The server symbiont wakes up periodically and connects to the remote system and attempts to pull back any files in a specific queue. Both the client and server use pretty standard FTP commands, with the exception that the file specs on the CYBER side for accessing the queues are a little off the wall! It operates with a low-level I/O package that supports MultiNet and CMU-TEK-TCP. I wrote this package about the same time Matt Madison was writing NETLIB. Had I known about Matt's package at the time... >There are two solutions we use here. One has the originating side on the VAX >using Multinet, but the destination is a CYBER. The code is a VAX queue that >sends and receives ftp to the CYBERS. > >The second solution is more generic, it uses LPD to shift files around our >network using filters to put the files somewhere. Some security issues have >been addressed, but not all. > Since MultiNet does not support filters in the same way that UNIX does, this solution is probably not usuable for non-UNIX systems. This filter code is HEAVILY dependent upon the UNIX filesystem and system calls. >Contact Henry Miller - henrym@sacto.mp.usbr.gov for more info. > > >>thank you? >> >>Jian >> >>P.S. email: zhangj@igd.fhg.de > >jimww > >====================================================================== ># James W. Wheeler # # ># Regional Systems & WAN Manager # voice: 916 978-5089 fax: 916-978-5284 # ># Mid Pacific Region # # ># U.S. Bureau of Reclamation # inet: jimww@sacto.mp.usbr.gov # ># 2800 Cottage Way MP1130 # # ># Sacramento, CA 95825 # # >====================================================================== > -HWM ---------- Henry W. Miller Assistant Systems and Network Manager U.S. Bureau of Reclamation, Mid Pacific Region 2800 Cottage Way MP1130 Sacramento, CA 95825 (916) 978-5108 X-ST-Status: N Return-path: Received: from PFC.MIT.EDU ([18.77.0.113]) by TGV.COM via INTERNET ; Thu, 26 Aug 93 07:05:35 PDT ================================================================================ Archive-Date: Thu, 26 Aug 1993 10:05:29 -0400 Date: Thu, 26 Aug 1993 10:05:29 -0400 (EDT) From: Mark London To: info-multinet@tgv.com Subject: rpc, again? Can anyone at TGV respond to my request for info as to what capabilities MULTINET has with regard to RPC? There is little info in the docs but scant references to them, and I see that there are include files for RPC, but I'm not sure what that means. Thanks. X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 26 Aug 1993 08:36:19 PDT Date: Thu, 26 Aug 93 08:36:19 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: DECnet over IP - address remapping To: ewilts @ vmsmail.gov.bc.ca cc: info-multinet @ TGV.COM > In configuring DECnet over IP using MultiNet, there appears to be no way to do > any DECnet area/number re-mapping. This functionality is currently served by a > few sites that use Cisco routers, but I don't know how to solve this problem > using MultiNet. > Are there any plans to allow name/number remapping in a future release of > MultiNet? How about filtering? This would make it a *lot* easier to join two > large networks... Absolutely none; unless DEC implements this in VMS DECnet. Cisco wrote their own DECnet implementation that allows them to do this; we just wrote a device driver which lets you take the DECnet already running under VMS and send the packets over TCP or UDP. We'd have to write a DECnet implementation to do this... Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 26 Aug 93 10:38:03 PDT From: rgoldstone@OAVAX.CSUCHICO.EDU (Robin, Girl Wonder) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Multinet finger server ================================================================================ Archive-Date: xxx, 26 Aug 1993 15:29:14 GMT Date: 26 Aug 1993 15:29:14 GMT Organization: California State University, Chico Lines: 42 Reply-To: rgoldstone@OAVAX.CSUCHICO.EDU NNTP-Posting-Host: oavax.csuchico.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I would love it if Multinet's finger server were a little more generous in the information it returned. Here is an example: $ finger jmensching@oavax JMENSCHI Mensching, Jim (MIN JMENSCHING not logged in Last login Wed 25-Aug-93 9:11AM-PDT Would it be possible for finger to return the complete 12 character userid and 31 character owner field, instead of 8 and 19 characters, respectively? Why is Multinet's finger server so stingy? That is my minimum request. Now for the wishful thinking dept: I would like to run a finger server under Multinet that can respond to lastname or random string queries by searching a specified file or files of user information. We currently run a finger server on a unix host, that is pointed to by the top-level "A" record for our campus, which does the above. To see it in action, try: finger goldstone@csuchico.edu. We would like to change our "A" record to point to our VAX but we really want to provide this extended finger service. Has anyone hacked the Multinet finger server in this way? Ideally I would like it to work as follows: 1) search Authorize database for exact username match. If found, print relevant info and plan. file if there is one. 2) search other "user info" files, as designated by a logical name FINGER_INFO or something. As I said, we already have this implemented under unix. I don't know if it would be possible to port the unix finger server to VMS or if it would be better to try to modify Multinet's finger server code. Either way, we don't have any programmers on our staff with TCP/IP or even C programming experience. I am wondering if anyone has undertaken this project or would be interested in doing it, for me and for you and for other Multinet users who might be seeking this capability. Thanks and go easy on me 'cuz I'm a girl. 8^) -Robin ************************************************************************ Robin Goldstone, Systems Software Specialist ____ California State University, Chico Computing Services \ / rgoldstone@oavax.csuchico.edu \/ Closets are for clothes X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 26 Aug 1993 11:12:02 PDT Date: Thu, 26 Aug 93 11:12:02 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: SunOS and Ultrix equivalents of TRACEROUTE To: mal @ valandil.adc.com cc: info-multinet @ TGV.COM > What is the SunOS and Ultrix equivalents of the MULTINET TRACEROUTE > command? "traceroute". It is available from FTP.EE.LBL.GOV, get the file traceroute.tar.Z Ken X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Thu, 26 Aug 93 11:16:01 PDT Received: from TGV.COM (HQ.TGV.COM) by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H27CZBXD4G8ZG4DX@psulias.psu.edu>; Thu, 26 Aug 1993 14:15:56 EDT ================================================================================ Archive-Date: Thu, 26 Aug 1993 11:15:51 -0700 Date: Thu, 26 Aug 1993 11:15:51 -0700 (PDT) From: adelman@TGV.COM (Kenneth Adelman) Subject: Re: Limit number of concurrent outgoing TELNET connections? To: JLW@psulias.psu.edu Cc: INFO-MULTINET@psulias.psu.edu Reply-to: Adelman@TGV.COM (Kenneth Adelman) > Is there any way we can cause the multinet telnet client to limit the number of > concurrent outgoing connections a user has? We tend to lock up our users > pretty well, but we have the situation where users hit the escape sequence and > connect to another host, and then another, and then another. We'd like to be > able to limit these to one or two at a time, per client. You could do this by customizing the USER_ACCESS.C module to check to see if it was invoked from TELNET, and if it is, then try to acquire a VMS lock by the name of "TELNET_username" where username is the username of the person. If the lock is already held, then refuse the connection. See the example code in [MULTINET.EXAMPLES]USER_ACCESS.C. Ken X-ST-Status: N Return-path: Received: from psulias.psu.edu ([128.118.88.200]) by TGV.COM via INTERNET ; Thu, 26 Aug 93 11:22:26 PDT Received: from psulias.psu.edu by psulias.psu.edu (PMDF V4.2-13 #4134) id <01H27D6WFAX8HXIRGK@psulias.psu.edu>; Thu, 26 Aug 1993 14:22:04 EDT ================================================================================ Archive-Date: Thu, 26 Aug 1993 14:22:04 -0400 Date: Thu, 26 Aug 1993 14:22:04 -0400 (EDT) From: "J.Lance Wilkinson, 814-865-1818" Subject: Re: Limit number of concurrent outgoing TELNET connections? To: Adelman@TGV.COM, INFO-MULTINET@psulias.psu.edu Organization: Penn State University / University Libraries X-VMS-To: IN%"Adelman@TGV.COM",IN%"INFO-MULTINET@psulias.psu.edu" X-VMS-Cc: JLW, EGF MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII >> Is there any way we can cause the multinet telnet client to limit the number of >> concurrent outgoing connections a user has? We tend to lock up our users >> pretty well, but we have the situation where users hit the escape sequence and >> connect to another host, and then another, and then another. We'd like to be >> able to limit these to one or two at a time, per client. > > You could do this by customizing the USER_ACCESS.C module to check >to see if it was invoked from TELNET, and if it is, then try to >acquire a VMS lock by the name of "TELNET_username" where username is >the username of the person. If the lock is already held, then refuse >the connection. See the example code in [MULTINET.EXAMPLES]USER_ACCESS.C. This would be good if it weren't for the fact that the users we need to control in this way all use the same user id, and there could be hundreds of them logged on at the same time. We don't mind 20% of them doing outgoing telnet at the same time, but we do mind if any one of them has more than one telnet session open at the same time; it's individual copies of the client that we need to limit in the number of connections. We were hoping for some kind of logical we could define, for example: $ define/sys/exec MULTINET_TELNET_CONNECTIONS_PER_CLIENT 2 which would cause the Telnet client to refuse to open more than two concurrent sessions. +-"Never Underestimate the bandwidth of a station wagon full of mag tapes"--+ | J.Lance Wilkinson ("Lance") BitNet: JLW@PSULIAS.BITNET | | Systems Design Specialist - Lead InterNet: jlw@psulias.psu.edu | | Library Computing Services AT&T:(814) 865-1818 FAX:(814)863-3560 | | E8 Pattee Library "I'd rather be dancing..." | | Penn State University A host is a host from coast to coast, | | University Park, PA 16802 And no one will talk to a host that's close | | Unless the host that isn't close | | Is busy, hung or dead. | +------"He's dead, Jim. I'll get his tricorder. You take his wallet."-------+ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 26 Aug 1993 11:28:15 PDT Date: Thu, 26 Aug 93 11:28:15 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Limit number of concurrent outgoing TELNET connections? To: JLW @ psulias.psu.edu cc: info-multinet @ TGV.COM > >> Is there any way we can cause the multinet telnet client to limit the number of > >> concurrent outgoing connections a user has? We tend to lock up our users > >> pretty well, but we have the situation where users hit the escape sequence and > >> connect to another host, and then another, and then another. We'd like to be > >> able to limit these to one or two at a time, per client. > > >> You could do this by customizing the USER_ACCESS.C module to check > >to see if it was invoked from TELNET, and if it is, then try to > >acquire a VMS lock by the name of "TELNET_username" where username is > >the username of the person. If the lock is already held, then refuse > >the connection. See the example code in [MULTINET.EXAMPLES]USER_ACCESS.C. > This would be good if it weren't for the fact that the users we need to control > in this way all use the same user id, and there could be hundreds of them > logged on at the same time. We don't mind 20% of them doing outgoing telnet at > the same time, but we do mind if any one of them has more than one telnet > session open at the same time; it's individual copies of the client that we > need to limit in the number of connections. We were hoping for some kind of > logical we could define, for example: > $ define/sys/exec MULTINET_TELNET_CONNECTIONS_PER_CLIENT 2 > which would cause the Telnet client to refuse to open more than two concurrent > sessions. The hook is certainly there to inplement this yourself, but you'd need to add your own mechanism to keep count. For example, one crude way of doing it is to loop through the locks by the name: TELNET_username_sequence util you found a "sequence" that wasn't in-use. Then you would hold that lock until the image exists. If you check all valid sequence counts and can't get a lock, then you report a failure to the user. Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Thu, 26 Aug 1993 11:54:27 PDT Date: Thu, 26 Aug 93 11:54:27 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Multinet finger server To: rgoldstone @ OAVAX.CSUCHICO.EDU cc: info-multinet @ TGV.COM > I would love it if Multinet's finger server were a little more generous in > the information it returned. Here is an example: > $ finger jmensching@oavax > JMENSCHI Mensching, Jim (MIN JMENSCHING not logged in > Last login Wed 25-Aug-93 9:11AM-PDT > Would it be possible for finger to return the complete 12 character userid > and 31 character owner field, instead of 8 and 19 characters, respectively? > Why is Multinet's finger server so stingy? Trying to fit all of the output in 80 columns. > That is my minimum request. Now for the wishful thinking dept: > I would like to run a finger server under Multinet that can respond to > lastname or random string queries by searching a specified file or files > of user information. We currently run a finger server on a unix host, that > is pointed to by the top-level "A" record for our campus, which does the > above. To see it in action, try: finger goldstone@csuchico.edu. We would > like to change our "A" record to point to our VAX but we really want to provide > this extended finger service. > Has anyone hacked the Multinet finger server in this way? Ideally I would like > it to work as follows: > 1) search Authorize database for exact username match. If found, print relevant > info and plan. file if there is one. > 2) search other "user info" files, as designated by a logical name FINGER_INFO > or something. > As I said, we already have this implemented under unix. I don't know if it > would be possible to port the unix finger server to VMS or if it would be > better to try to modify Multinet's finger server code. Either way, we don't > have any programmers on our staff with TCP/IP or even C programming experience. > I am wondering if anyone has undertaken this project or would be interested in > doing it, for me and for you and for other Multinet users who might be seeking > this capability. In a future release of MultiNet, we plan on providing our FINGER in source code form as an example that you could customize. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Thu, 26 Aug 93 17:40:16 PDT From: bleau@umdsp.umd.edu X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Limit number of concurrent outgoing TELNET connections? ================================================================================ Archive-Date: xxx, 26 Aug 1993 23:49:39 GMT Date: 26 Aug 1993 23:49:39 GMT Organization: University of Maryland Physics Dept. Lines: 38 Reply-To: bleau@umdsp.umd.edu NNTP-Posting-Host: umdsp.umd.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930826112638.400402@TGV.COM>, adelman@TGV.COM (Kenneth Adelman) writes: [stuff about original idea - limiting number of outgoing telnet sessions per TELNET client (ie, per user) deleted] > > The hook is certainly there to inplement this yourself, but you'd >need to add your own mechanism to keep count. For example, one crude >way of doing it is to loop through the locks by the name: > > TELNET_username_sequence > >util you found a "sequence" that wasn't in-use. Then you would hold >that lock until the image exists. If you check all valid sequence >counts and can't get a lock, then you report a failure to the user. > > Ken Another idea that came to mind to keep count is to acquire a lock named TELNET_username If it doesn't exist create it and give it a value of "1" (as I recall you can associate a string value with a lock name); you're done, release the lock. If the lock already exists get its value and hold onto the lock; now translate the /SYSTEM/EXEC logical name MULTINET_TELNET_CLIENT_OUTGOING_LIMIT or any other name you want to use, and compare them. If you new lock value is greater than or equal to the logical name value, reject the request and release the lock. If it isn't, increment the lock value and modify it, releasing the lock. Be sure to reverse these operations when closing down a connection. This would actually limit outgoing sessions on a per-user basis, not a per-login or a per-process basis, which you may actually want to implement. Larry Bleau University of Maryland bleau@umdsp.umd.edu 301-405-6223 X-ST-Status: N Return-path: Received: from gatekeeper.es.dupont.com ([192.26.233.2]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 00:48:18 PDT Received: by gatekeeper.es.dupont.com (5.65/ULTRIX-mjr-062991); id AA15763; Fri, 27 Aug 93 03:48:14 -0400 Received: by esds01.es.dupont.com (5.65/ULTRIX-mjr-063191); id AA25157; Fri, 27 Aug 93 03:48:01 -0400 ================================================================================ Archive-Date: Fri, 27 Aug 1993 03:47:55 -0400 Date: Fri, 27 Aug 93 03:47:55 -0400 From: eng_mb@nbdtf1.dnet.dupont.com (Michael B. Benga) To: "info-multinet@TGV.COM"@esds01.dnet.dupont.com Cc: "Adelman@TGV.COM"@esds01.dnet.dupont.com Wmvx: :FRISICEJ Subject: MIB II Extensions for DECnet IV. I would like to manage the DECnet IV VMS systems that run TGV/Multinet using SNMP. However, it seems that the only MIB II extensions provided in the MULTINET::SNMP_MIB.TXT are TGV related. How can I include the MIB II extension specified in RFC 1289 into TGV's SNMP? Did anybody do this ? IS TGV planning to adopt the RFC 1289 and when? Thank you in advance. Regards, Mike B. X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 02:37:04 PDT Via: uk.ac.tex; Fri, 27 Aug 1993 10:36:40 +0100 ================================================================================ Archive-Date: Fri, 27 Aug 1993 9:36 GMT Date: Fri, 27 Aug 93 9:36 GMT From: "UK TeX Archive Manager " To: JLW , INFO-MULTINET Subject: Re: Limit number of concurrent outgoing TELNET connections? In message <01H27D6WJUVYHXIRGK@psulias.psu.edu> dated Thu, 26 Aug 1993 14:22:04 -0400 (EDT) JLW@edu.psu.psulias (J.Lance Wilkinson, 814-865-1818) wrote: > This would be good if it weren't for the fact that the users we need to control > in this way all use the same user id, and there could be hundreds of them > logged on at the same time. We don't mind 20% of them doing outgoing telnet at > the same time, but we do mind if any one of them has more than one telnet > session open at the same time; it's individual copies of the client that we > need to limit in the number of connections. We were hoping for some kind of > logical we could define, for example: > > $ define/sys/exec MULTINET_TELNET_CONNECTIONS_PER_CLIENT 2 > > which would cause the Telnet client to refuse to open more than two concurrent > sessions. So why not, instead of having a lock of the form TELNET_username, have them called TELNET_parent_pid? Combine this with the use of a counter as the lock *content*, as suggested in someone else's response, and compare with your logical. Brian {Hamilton Kelly} (BH29) System Manager for the (VMS) UK TeX Archive at Aston University X-ST-Status: N Return-path: Received: from sun2.nsfnet-relay.ac.uk ([128.86.8.45]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 04:32:22 PDT Via: uk.ac.tex; Fri, 27 Aug 1993 12:29:46 +0100 ================================================================================ Archive-Date: Fri, 27 Aug 1993 11:28 GMT Date: Fri, 27 Aug 93 11:28 GMT From: "UK TeX Archive Manager " To: INFO-MULTINET Subject: ftp introducing an arbitrary cut-off, before end of file Following the appearance of an announcement of some new TeX-support software, I attempted to fetch the relevant files from the announcer's site. One of the two files came over quite happily, but the second ``jammed up''. (I always perform transfers of big files with hash on --- in this case, I got 70 hashes and then things went quiet: eventually, the remote ftp server timed me out and kicked me off). At this stage, I'd got 1638 VMS disk blocks on my machine. The account under which the transfer was running has about 500,000 blocks of free quota, and the disk itself has more than 400,000 blocks free. The file in question is 1,581,010 bytes long, so ought to fit into about 3087 blocks --- the file is binary, and that's been specified for the transfer. My first assumption was that network connectivity had been lost, so I repeated the exercise. Same thing happened: only 1638 blocks of the file transferred (although this time, 76 hashes were printed --- I'd always suspected that they represented a random quantum of transfer:-) To absolve the remote end of responsibility, I logged into a local U**x box, and fetched the file without any problem. I then attempted the transfer to this machine (once more running the ftp client on TeX.ac.uk and using the server on the U**x box). Yet again the transfer stops at 1638 blocks. It's not quota or diskspace related: apart from the aforementioned ample quantities of each, a repeated transfer, to a different local name, results in TWO copies, each truncated to 1638 blocks. I've now attempted the transfer (onto TeX.ac.uk), both from the Stateside machine and from the local U**x box, a total of eight times. Every time, the resultant file has occupied 1638 blocks of diskspace, although the number of hashes displayed has varied between 69 and 76. I'd be grateful if someone Stateside could attempt the transfer and report whether they are able to fetch the whole file. The parent site is ftp.cs.umb.edu; the file is pub/tex/psfonts.tar.gz -rw-rw-r-- 1 karl wheel 1581010 Aug 4 15:02 psfonts.tar.gz Incidentally, the file that I was successful in fetching was this: -rw-rw-r-- 1 karl wheel 245460 Aug 26 13:17 xdvik-1.1.tar.gz Just as one additional bit of information, I spawned out of the editor whilst creating this message and moved to a different disk; once again, the transfer stopped after 70 hashes, but this time the resultant file was only 1512 blocks long --- consistently in three attempts (two remote and one local). Does the ftp client throttle itself back depending upon the space left on the disk?! I'm doing all this from the SYSTEM account, with all privileges turned on (including BYPASS, because the original directory into which I was attempting the transfer has an ACL which prevented me from creating the new directory). $ multinet show/version TGV MultiNet V3.2 Rev B, VAX-11/785, VAX/VMS V5.4-2 The 1638 block transfers were taking place to this directory: $ dir/sec [-]berry.dir Directory DISK$TEX:[TEX-ARCHIVE.DVIWARE.XDVI] BERRY.DIR;1 AUI$TEX (RE,RWE,RE,RE) (IDENTIFIER=AUI$TEX,ACCESS=READ+WRITE+EXECUTE+CONTROL) (IDENTIFIER=AUI$TEX,OPTIONS=DEFAULT,ACCESS=READ+WRITE+EXECUTE+DELETE+CONTROL) (IDENTIFIER=[*,*],ACCESS=READ+EXECUTE) (IDENTIFIER=[*,*],OPTIONS=DEFAULT,ACCESS=READ+EXECUTE) (DEFAULT_PROTECTION,SYSTEM:RE,OWNER:RWED,GROUP:RE,WORLD:RE) Total of 1 file. The AUI$TEX resource identifier has ample quota, as has SYSTEM on this disk: $ show quota/user=aui$tex User AUI$TEX has 2025107 blocks used, 534893 available, of 2560000 authorized and permitted overdraft of 512000 blocks on DISK$TEX $ show quota/user=system User [SYSTEM] has 90693 blocks used, 159307 available, of 250000 authorized and permitted overdraft of 1000 blocks on DISK$TEX Finally, although the system disk is fairly full (and has no quotas, of course), DISK$TEX is a volume set, with plenty of room: $ show device sys$sysdevice Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DUA0: Mounted 0 TEXSYS 8667 313 1 $ show device disk$tex Device Device Error Volume Free Trans Mnt Name Status Count Label Blocks Count Cnt DUB6: Mounted 0 ARCHIVE3 139869 1 1 DUB18: Mounted 0 ARCHIVE1 139215 1 1 DUB19: Mounted 5 ARCHIVE2 139119 2 1 (I'd be happiest if someone from TGV could tell me `` oh dear, there's a bug''. But I think it more likely that I'm doing something wrong, although I cannot see what! Horrible thought: would all this be fixed had I bothered to upgrade to V3.2C?) Brian {Hamilton Kelly} (BH29) System Manager for the (VMS) UK TeX Archive at Aston University X-ST-Status: N Return-path: Received: from cis.saultc.on.ca ([192.197.90.3]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 08:38:36 PDT Received: from CIS.SAULTC.ON.CA by CIS.SAULTC.ON.CA (PMDF V4.2-10 #4035) id <01H28LUIPEAS000111@CIS.SAULTC.ON.CA>; Fri, 27 Aug 1993 11:40:56 EDT ================================================================================ Archive-Date: Fri, 27 Aug 1993 11:40:56 -0400 Date: Fri, 27 Aug 1993 11:40:56 -0400 (EDT) From: Mike Cliffe Subject: Routing between Netware and Multinet To: info-multinet@tgv.com X-VMS-To: in%"info-multinet@tgv.com" X-VMS-Cc: MIKE MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII (I probably could have picked a better subject to describe this, however...) Problem: The routing table (we are not running gated) is being updated but some of the routes are not what I would really like to have. Background: We have two class C addreses: one for our VAXen/Routers (not sub-netted) I'll call this one ADDRESS 1, and one for our Netware PC lans (this one is sub-netted ff.ff.ff.e0) I'll call this one ADDRESS 2 Our routing table originally looks like this: Destination Gateway Flags Refcnt Use Interface ----------- ------- ----- ------ --- --------- LOCALHOST LOCALHOST Up,Host 1 16 lo0 IT-NET MARTINI.SAULTC. Up,Gateway,H 0 0 se0 EWING-NET MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 DEFAULT-GATEWAY 192.197.90.224 Up,Gateway 2 213 se0 SAULTC-NET ACS.SAULTC.ON.C Up 1 54 se0 it-net and ewing-net were added vis mu set/route... The gateways for it-net and ewing-net are file servers that are using the ip routing capability of Netware. What happens is this. I can ping each of the file servers using their ADDRESS 1 address. I can also ping one of the file server's ADDRESS 2 address. However, once I have done that I can no longer ping the other file server's ADDRESS 2 address. i.e. After doing this the routing table looks like this: Destination Gateway Flags Refcnt Use Interface ----------- ------- ----- ------ --- --------- LOCALHOST LOCALHOST Up,Host 1 16 lo0 IT-NET MARTINI.SAULTC. Up,Gateway,H 0 0 se0 EWING-NET MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 WINE.SAULTC1.ON MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 STDL0.SAULTC1.O MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 DEFAULT-GATEWAY 192.197.90.224 Up,Gateway 2 213 se0 SAULTC-NET ACS.SAULTC.ON.C Up 1 54 se0 The `far-side' of the file servers/routers are being routed through the same gateway "MUSTANG". When in reality MUSTANG should be the gateway for stdl0 and ewing-net and MARTINI should be the gateway for it-net. In this example the ewing-net via mustang works fine but we can no longer get to WINE. Any suggestions as to why this might be happening. If I've left out some important info I apologize. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Mike Cliffe, Technical Services Voice: (705) 759-2554 x645 Sault College of Applied Fax: (705) 256-5944 Arts & Technology <> Internet: mike@cis.saultc.on.ca Sault Ste. Marie, ON CANADA <><> X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 27 Aug 1993 09:13:03 PDT Date: Fri, 27 Aug 93 09:13:03 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Routing between Netware and Multinet To: mike @ cis.saultc.on.ca cc: info-multinet @ TGV.COM > We have two class C addreses: one for our VAXen/Routers (not sub-netted) > I'll call this one ADDRESS 1, and one for our Netware PC lans (this one is > sub-netted ff.ff.ff.e0) I'll call this one ADDRESS 2 > Our routing table originally looks like this: > Destination Gateway Flags Refcnt Use Interface > ----------- ------- ----- ------ --- --------- > LOCALHOST LOCALHOST Up,Host 1 16 lo0 > IT-NET MARTINI.SAULTC. Up,Gateway,H 0 0 se0 > EWING-NET MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 > DEFAULT-GATEWAY 192.197.90.224 Up,Gateway 2 213 se0 > SAULTC-NET ACS.SAULTC.ON.C Up 1 54 se0 > it-net and ewing-net were added vis mu set/route... > The gateways for it-net and ewing-net are file servers that are using the > ip routing capability of Netware. I note that the route to IT-NET is listed as a "Host" route (the "H" under Flags). Try adding it with the /FORCE_NETWORK qualifier to SET/ROUTE. > What happens is this. I can ping each of the file servers using their > ADDRESS 1 address. I can also ping one of the file server's ADDRESS 2 > address. However, once I have done that I can no longer ping the other > file server's ADDRESS 2 address. i.e. After doing this the routing table > looks like this: > Destination Gateway Flags Refcnt Use Interface > ----------- ------- ----- ------ --- --------- > LOCALHOST LOCALHOST Up,Host 1 16 lo0 > IT-NET MARTINI.SAULTC. Up,Gateway,H 0 0 se0 > EWING-NET MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 > WINE.SAULTC1.ON MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 > STDL0.SAULTC1.O MUSTANG.SAULTC. Up,Gateway,H 0 0 se0 > DEFAULT-GATEWAY 192.197.90.224 Up,Gateway 2 213 se0 > SAULTC-NET ACS.SAULTC.ON.C Up 1 54 se0 > The `far-side' of the file servers/routers are being routed through the > same gateway "MUSTANG". When in reality MUSTANG should be the gateway for > stdl0 and ewing-net and MARTINI should be the gateway for it-net. In this > example the ewing-net via mustang works fine but we can no longer get to > WINE. Any suggestions as to why this might be happening. If I've left out > some important info I apologize. My guess is that because those routes are int he table wrong, your default gateway is getting the packets and sending ICMP redirects to you, which end up getting cached as these routes. Ken X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 09:32:25 PDT ================================================================================ Archive-Date: Fri, 27 Aug 1993 9:32:21 -0700 Date: Fri, 27 Aug 1993 9:32:21 -0700 (PDT) From: "JEFFREY D. BUSMA" To: eng_mb@nbdtf1.dnet.dupont.com CC: info-multinet@TGV.COM Subject: RE: MIB II Extensions for DECnet IV. > I would like to manage the DECnet IV VMS systems that >run TGV/Multinet using SNMP. However, it seems that the >only MIB II extensions provided in the >MULTINET::SNMP_MIB.TXT are TGV related. How can I >include the MIB II extension specified in RFC 1289 into >TGV's SNMP? > >Did anybody do this ? >IS TGV planning to adopt the RFC 1289 and when? > >Thank you in advance. > >Regards, > > Mike B. > > MultiNet's 3.2 SNMP agent is not currently extensible. It will be in the release _after_ the 3.3 release. We'll look at the DECnet IV MIB then as well, either including it in our agebt or providing hooks for a sub-agent. Jeff X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 09:40:19 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: rdb connection via tcp/ip with multinet Message-ID: <20217@pitt.UUCP> From: skyking@cs.pitt.edu (John Candee) ================================================================================ Archive-Date: xxx, 27 Aug 1993 15:32:00 GMT Date: 27 Aug 93 15:32:00 GMT Sender: news@cs.pitt.edu Organization: Univ. of Pittsburgh Computer Science Lines: 55 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I am having a problem using rdb over a tcp/ip network. I am using a client program (written in C) on a DECstation 5000, that connects to a system running VMS and accesses a database residing on the VMS machine. I am using the API that comes from DEC for an Ultrix machine. Environment: VMS ver 5.5 RDB ver 4.2 Client API running on Ultrix on a DECstation 5000. VMS machine running multinet software. The database has a table that has 62 fields. When I try to select from this table, I get an error stating invalid SQLDA at varying times in the retrieval. As a test program, I connect to the machine, attach to the database, prepare a select statement, allocate_sqlda_data, open a cursor, close the cursor, (in this test program, I do not even do a fetch), release the statement, then go back and prepare the select statement and do it again. At some iteration, anywhere from 1 to about 20, I will get the error. I can do this successfully if I use a table that has fewer fields, but I do not know the number of fields that will start to cause this problem. I should also point out that I have a similar problem with updating this big table although I don't have any problem with tables consisting of 5 to 10 fields. The connection has read/write buffers that are allocated by the API. The buffer is supposed to be 1300 bytes, and it looks like the data being sent is around 500 or so bytes, so I don't think the buffers are being overloaded. Can anyone shed some light on this problem? In the process of looking into this, I tried to look at the sqlda structure. I had trouble trying to get the size of the sqlda from the SQLDABC field, since it appears this number is not stored in the machine normal way for an integer. Does anyone know how this info is stored? Big-endian, Little, endian, BCD, ... ? Thanks in advance. John C. skyking@speedy.cs.pitt.edu X-ST-Status: N Return-path: Received: from po5.andrew.cmu.edu ([128.2.10.105]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 11:56:20 PDT Received: from localhost (postman@localhost) by po5.andrew.cmu.edu (8.5/8.5) id OAA18829; Fri, 27 Aug 1993 14:56:15 -0400 Received: via switchmail; Fri, 27 Aug 1993 14:56:14 -0400 (EDT) Received: from unix4.andrew.cmu.edu via qmail ID ; Fri, 27 Aug 1993 14:55:34 -0400 (EDT) Received: from unix4.andrew.cmu.edu via qmail ID ; Fri, 27 Aug 1993 14:55:16 -0400 (EDT) Received: from mms.0.1.23.EzMail.2.0.CUILIB.3.45.SNAP.NOT.LINKED.unix4.andrew.cmu.edu.pmax.ul4 via MS.5.6.unix4.andrew.cmu.edu.pmax_ul4; Fri, 27 Aug 1993 14:55:16 -0400 (EDT) Message-ID: <0gTZWIC00WB55U71Us@andrew.cmu.edu> ================================================================================ Archive-Date: Fri, 27 Aug 1993 14:55:16 -0400 Date: Fri, 27 Aug 1993 14:55:16 -0400 (EDT) From: "Thomas J. Joyce" To: info-multinet@tgv.com Subject: Pathworks for VMS (DOS) and Multinet Cc: Does Pathworks for VMS (DOS) require any special layered product in order to work with Multinet. This question applies to the server and the client. I was a little confused about a reference to "Leverage for Pathworks" being needed for Pathworks to function with Multinet. I found only one instance of this in the SSA for Pathworks for VMS V4.1-1. I didn't see an SPD number for Leverage for Pathworks PAS. Could someone clear this up for me? thanks ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ + Thomas Joyce Thomas.Joyce@cmu.edu Carnegie-Mellon ACIS + + + + "Mors aut Honorabilis Vita" + ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 27 Aug 1993 12:03:16 PDT Date: Fri, 27 Aug 93 12:03:16 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Pathworks for VMS (DOS) and Multinet To: Thomas.Joyce @ cmu.edu cc: info-multinet @ TGV.COM > Does Pathworks for VMS (DOS) require any special layered product > in order to work with Multinet. This question applies to the server > and the client. > I was a little confused about a reference to "Leverage for Pathworks" > being needed for Pathworks to function with Multinet. I found only > one instance of this in the SSA for Pathworks for VMS V4.1-1. I didn't > see an SPD number for Leverage for Pathworks PAS. > Could someone clear this up for me? You do not need PAS to run Pathworks over MultiNet; what you need to do is to configure Pathworks as if it was running over UCX. PAS won't even work over contemporary versions of MultiNet. Do not listen to what anyone else tells you about PAS. Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Fri, 27 Aug 1993 12:05:46 PDT Date: Fri, 27 Aug 93 12:05:46 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: ftp introducing an arbitrary cut-off, before end of file To: system @ uk.ac.tex cc: info-multinet @ TGV.COM > Following the appearance of an announcement of some new TeX-support software, I > attempted to fetch the relevant files from the announcer's site. One of the > two files came over quite happily, but the second ``jammed up''. (I always > perform transfers of big files with hash on --- in this case, I got 70 hashes > and then things went quiet: eventually, the remote ftp server timed me out and > kicked me off). > At this stage, I'd got 1638 VMS disk blocks on my machine. The account under > which the transfer was running has about 500,000 blocks of free quota, and the > disk itself has more than 400,000 blocks free. The file in question is > 1,581,010 bytes long, so ought to fit into about 3087 blocks --- the file is > binary, and that's been specified for the transfer. > My first assumption was that network connectivity had been lost, so I repeated > the exercise. Same thing happened: only 1638 blocks of the file transferred > (although this time, 76 hashes were printed --- I'd always suspected that they > represented a random quantum of transfer:-) This sounds like either a network connectivity problem (it could even be transient) OR a problem with a pattern-sensitive link somewhere in between you and the other side (a link that is not letting a particular pattern of data through). The "#" printing in FTP is a random quantum -- it represents a read or write operation completing, and reads can complete in different sizes. Think of it as an "I`m awake" more than anything else. > Does the > ftp client throttle itself back depending upon the space left on the disk?! No. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 12:51:23 PDT From: Don.Rainwater@UC.Edu (Don Rainwater, Systems Manager) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: MSDOS / Windows equiv of Eudora Followup-To: vmsnet.networks.tcp-ip.multinet ================================================================================ Archive-Date: Fri, 27 Aug 1993 12:05:48 -0500 Date: Fri, 27 Aug 1993 12:05:48 -0500 Organization: Univ of Cincinnati Academic IT Services Lines: 30 NNTP-Posting-Host: ucrain.san.uc.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug17.074828.341@eisner.decus.org>, roseberry@eisner.decus.org wrote: > > The guys with Macs really like how Eudora works the the Multinet POP Server. > > Is there some software for the IBM PCs that is just as good as Eudora and > will work with Multinet's POP Server ? > > - Bert Roseberry roseberry@eisner.decus.org - or - > US Coast Guard bert@mailstorm.dot.gov Eudora will soon be available for the MS Windows. I think it uses Winsock. Eudora is going commercial, with version 2.0 for the Macintosh to be released at Interop, and the Windows version to be released in... November(?). The price is around $50 a copy in single quantities. Quantity and edu discounts are available. Contact Qualcomm at eudora-sales@qualcomm.com for information about either version. There is still going to be a free version of Eudora for the Macintosh (version 1.4), but I don't know about the PC version. A PC POP3 client that people around here seem to like is NuPOP from NWU. It provides access to POP3 and NNTP client services, and is available from ftp.acns.nwu.edu. I have no connection with Eudora, Qualcomm, NuPOP, or NWU. I *have* used the Macintosh version of Eudora, and I'm very satisfied with it. I'll probably try to buy the commercial version when it becomes available. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Fri, 27 Aug 93 22:43:52 PDT X-Newsgroups: comp.os.vms,vmsnet.networks.tcp-ip.multinet Subject: PC-VAX-LaserJet printing problem Message-ID: <27AUG199322113432@ariel.lerc.nasa.gov> From: uugblum@ariel.lerc.nasa.gov (Greg Blumers [3-6777]) ================================================================================ Archive-Date: xxx, 27 Aug 1993 22:11 EST Date: 27 Aug 1993 22:11 EST Distribution: world Organization: NASA Lewis Research Center NNTP-Posting-Host: ariel.lerc.nasa.gov News-Software: VAX/VMS VNEWS 1.41 Lines: 61 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Has anyone else had problems printing under Multinet V3.2 Rev. C to HP laserJet printers? I am experiencing a problem printing HP-PCL files from a PC to a LaserJet printer attached to a Milan print server. If I bypass the VAX, it printers fine. Unfortunately, the majority of our PCs don't have FTP's TCP/IP software. In this case, Ungermann-Bass XNS PC software talks to the VAX Interconnections software which can submit the job for printing in a VAX queue. The VAX queue and form definitions are: Printer queue B142B2_JET, idle, on LIMS02::NLP69:, mounted form PASSALL (stock=DEFAULT) /BASE_PRIORITY=4 /DEFAULT=(FORM=PASSALL (stock=DEFAULT)) /LIBRARY=HPLJ4_DEVCTL Lowercase /OWNER=[SYSTEM,MANAGER] /PROCESSOR=MULTINET_STREAM_SYMBIONT /PROTECTION=(S:E,O:D,G:R,W:RW) /RETAIN=ERROR /SEPARATE=(RESET=(RESET)) Form name Number Description --------- ------ ----------- PASSALL (stock=DEFAULT) 21 TCP/IP Non-Ansi Printing /LENGTH=255 /STOCK=DEFAULT /WIDTH=65535 I put the LaserJet printer in DISPLAY mode which produces a graphical picture of the data received. I'd prefer a hex dump mode, but even the LaserJet 4 doesn't offer that. I've summarized the finding below. CASE #1 - LPR from PC to Sun with Milan driver to Milan MIL-3000. Prints fine LPR Milan Stream PC ----------> Sun Server -----------> Milan MIL-3000 ----> HP LaserJet CASE #2 - LPR through VAX. EOT (Hex 04) character generates a carriage return character. LPR Multinet Stream Queue PC ----------> VAXcluster -----------> Milan MIL-3000 ----> HP LaserJet CASE #3 - FTP file to VAX. Print from VAX. Extract spaces in document. I couldn't find any relation to other characters in the file. FTP-BINARY Multinet Stream Queue PC ----------> VAXcluster -----------> Milan MIL-3000 ----> HP LaserJet Also, if I print using the /PASSALL qualifier, the data gets really munged. Any ideas? -------------------------------------------------------------------------------- Greg Blumers Sverdrup Technology, Inc. Systems Programmer c/o NASA Lewis Research Center (216)433-6777 Mail Stop 142-2 21000 Brookpark Road uugblum@scivax.lerc.nasa.gov Cleveland, OH 44135 -------------------------------------------------------------------------------- X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sat, 28 Aug 93 18:22:06 PDT From: rankin@eql.caltech.edu (Pat Rankin) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: ftp introducing an arbitrary cut-off, before end of file ================================================================================ Archive-Date: xxx, 28 Aug 1993 18:08 PDT Date: 28 Aug 1993 18:08 PDT Organization: California Institute of Technology Lines: 34 Distribution: world NNTP-Posting-Host: eql.caltech.edu News-Software: VAX/VMS VNEWS 1.41x4 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <15172517@MVB.SAIC.COM>,\ SYSTEM@TEX.AC.UK (UK TeX Archive Manager ) writes... [...] > I'd be grateful if someone Stateside could attempt the transfer and report > whether they are able to fetch the whole file. The parent site is > ftp.cs.umb.edu; the file is pub/tex/psfonts.tar.gz > -rw-rw-r-- 1 karl wheel 1581010 Aug 4 15:02 psfonts.tar.gz I didn't have any problem, although it wasn't particularly speedy. We're running 3.2C. An earlier version--I don't recall which--did have a problem with MGET hanging while transfering lots of small files, but I don't think GET ever had any such problem. $ show symbol aftp AFTP == "FTP/USERN="anonymous"/PASSW="rankin@eql.caltech.edu"" $ aftp ftp.cs.umb.edu /bin "get pub/tex/psfonts.tar.gz tmp:psfonts.tar-gz" EQL.Caltech.EDU MultiNet FTP user process 3.2(106) Connection opened (Assuming 8-bit connections) Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sat, 28 Aug 93 20:29:48 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: david@hmcvax.claremont.edu Subject: Problem with mail and NFS mounted disks. Message-ID: <1993Aug27.201509.1@hmcvax.claremont.edu> Lines: 22 Sender: news@news.claremont.edu (The News System) Organization: Harvey Mudd College ================================================================================ Archive-Date: Sat, 28 Aug 1993 04:15:09 GMT Date: Sat, 28 Aug 1993 04:15:09 GMT To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I need some advice with a multinet NFS problem. Multinet NFS translates filenames with characters that are illegal in VMS filenames into very strange names with a liberal dose of dollar signs. An example would be $U$SERS.DIR;1 for the Un*x directory "/Users". The problem occurs when you create an account whose home directory is on an NFS mounted disk. Mail creates files with names like MAIL$047705D500050097.MAI;1, which then get translated into bizzare filenames on the NFS side with greek letters and other strange characters. Sometimes, these characters include extra periods or other characters that our NFS server (a newtware beast) doesn't seem to like. Basically, mail gets really badly hosed, as it can't create some of the files it spews externally to the mail.mai file. Has anyone out there ever set up a system with NFS mounted home disks? And is there a way to create files with dollar signs that aren't translated into other characters? (BTW, nfsmounting with the qulaifier /semantics=vms_filenames doesn't help a lot, since we would like to have these directories available from multiple platforms.) Ideas, thoughts, comments? Ack! -David X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Sun, 29 Aug 93 02:04:22 PDT ================================================================================ Archive-Date: Sun, 29 Aug 1993 02:04:17 -0700 Date: Sun, 29 Aug 1993 02:04:17 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: TGV-announce (was: Recent flamage) To: EWILTS@GALAXY.GOV.BC.CA Cc: INFO-MULTINET@TGV.COM In-Reply-To: <1993Aug25.084842.2979@vmsmail.gov.bc.ca> Mail-System-Version: Organization: TGV, Inc. X-Phone: 408/457-5200 (work); 408/457-5205 (fax) X-Address: 101 Cooper Street; Santa Cruz, CA 95060 (work) >> In the next few weeks, we will be setting up a new mailing list, >> tgv-announce@tgv.com, where new release announcements, new product >> announcements, training announcements, etc. will be posted. This way, those >> of you who wish to receive information from TGV via the Internet will be able >> to without having to deal with the volume of info-multinet. > >Stuart, I'd like to suggest that the tgv-announce list gets gatewayed to a new >newsgroup, biz.tgv. This would be along the lines of biz.dec where we see all >the Digital press releases. Good idea. I sincerely doubt that we'll be generating as many press releases as Digital, however. I'll look into this. Regards! -----Stuart X-ST-Status: N Return-path: Received: from THOR.INNOSOFT.COM ([192.160.253.66]) by TGV.COM via INTERNET ; Sun, 29 Aug 93 10:07:29 PDT Received: from INNOSOFT.COM by INNOSOFT.COM (PMDF V4.2-14 #1336) id <01H2A1SK5S4G984I58@INNOSOFT.COM>; Sun, 29 Aug 1993 10:07:22 PST ================================================================================ Archive-Date: Sun, 29 Aug 1993 09:57:09 -0800 Date: Sun, 29 Aug 1993 09:57:09 -0800 (PST) From: Ned Freed Subject: Re: Problem with mail and NFS mounted disks. To: david@hmcvax.claremont.edu Cc: INFO-MULTINET@TGV.COM MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII > Multinet NFS translates filenames with characters that are illegal in VMS > filenames into very strange names with a liberal dose of dollar signs. An > example would be $U$SERS.DIR;1 for the Un*x directory "/Users". The problem > occurs when you create an account whose home directory is on an NFS mounted > disk. Mail creates files with names like MAIL$047705D500050097.MAI;1, which > then get translated into bizzare filenames on the NFS side with greek letters > and other strange characters. Sometimes, these characters include extra > periods or other characters that our NFS server (a newtware beast) doesn't seem > to like. Basically, mail gets really badly hosed, as it can't create some of > the files it spews externally to the mail.mai file. Has anyone out there ever > set up a system with NFS mounted home disks? And is there a way to create > files with dollar signs that aren't translated into other characters? > (BTW, nfsmounting with the qulaifier /semantics=vms_filenames doesn't help a > lot, since we would like to have these directories available from multiple > platforms.) You absolutely, positively, must NOT do this. VMS MAIL makes extensive use of RMS indexed files, which in turn are utterly dependent on VMS locking mechanisms. (RMS global buffers are also involved, which makes the problem even worse.) NFS operates at the XQP level and knows nothing whatsoever about any of these RMS locks, so when you mount the same disk on multiple platforms in different VMS clusters the locks from one cluster are not visible to the other. The inevitable result will be file corruption and total chaos. I should emphasize that this is a limitation of NFS itself, which does not provide the underlying locking model that would be necessary to insure consistent VMS-style record-level locks. And as a purely practical matter it would be exceptionally difficult to implement without enhancing the VMS lock manager in several significant ways. I cannot comment on your problems with Netware, but I will say that if Netware's NFS server cannot support a reasonable repetiore of file names it should be dismissed as an inferior implementation. Ned X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Sun, 29 Aug 93 10:25:56 PDT From: smith@mcclb0.med.nyu.edu X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: TGV-announce (was: Recent flamage) Message-ID: <1993Aug29.113352.7263@mcclb0> ================================================================================ Archive-Date: xxx, 29 Aug 1993 16:33:52 GMT Date: 29 Aug 93 16:33:52 GMT Organization: NYU Medical Center, New York, NY 10016, USA Lines: 30 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <746615057.808723.VANCE@TGV.COM>, VANCE@TGV.COM (L. Stuart Vance) writes: >>> In the next few weeks, we will be setting up a new mailing list, >>> tgv-announce@tgv.com, where new release announcements, new product >>> announcements, training announcements, etc. will be posted. This way, those >>> of you who wish to receive information from TGV via the Internet will be able >>> to without having to deal with the volume of info-multinet. >> >>Stuart, I'd like to suggest that the tgv-announce list gets gatewayed to a new >>newsgroup, biz.tgv. This would be along the lines of biz.dec where we see all >>the Digital press releases. > > Good idea. I sincerely doubt that we'll be generating as many press releases > as Digital, however. I'll look into this. I have been as disgusted as anyone about the recent misuse of this list to conduct a mindless little flame-war. But PLEASE DON'T use this as a reason to create YET ANOTHER LIST. I have a job to do. And reading NEWS and mail from mailing lists is meant to be a tiny part of that job, not an all consuming avocation. On top of that if you create a newsgroup in the 'biz' section, I have to get 'biz' from our central site, which is a huge waste of time, not to mention additional disk space, etc. So PLEASE just use this list so that info about Multinet and similar issues can be found in JUST ONE PLACE. If you want to control access, then moderate the discussions. +----------------------------------------------------------------------------+ |Ross Smith, Research Computing Resource, Department of Cell Biology, NYU-MC| |E-Mail: SMITH@NYUMED.BITNET (BITNET), SMITH@MCCLB0.MED.NYU.EDU (Internet)| +----------------------------------------------------------------------------+ X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 03:29:53 PDT From: doelz@urz.unibas.ch X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Need Advice to $Creprc with non-0 UIC Message-ID: <1993Aug30.120718.43104@urz.unibas.ch> ================================================================================ Archive-Date: xxx, 30 Aug 1993 12:07:18 MET Date: 30 Aug 93 12:07:18 MET Organization: University of Basel, Switzerland Lines: 94 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I would like to start a command file from Multinet's inetd server functionality: 1) The server runs under SYSTEM 2) The server has no DCL 3) The command file shall run under a different UIC than SYSTEM. Thei means that I first get the UIC from a getuai call and then launch a creprc call which launches LOGINOUT.EXE under this ID. I have prepared an example which I hope can be read clearly enough to show this procedure. The program runs without error but starts the detached process with SYSTEM rather than the desired UIC. As an example I call LOGINOUT.EXE with a command called XXXX.COM: > $ open/write xxx hallo.hassle > $ write xxx "Hello, this is ",F$GetJPI("","USERNAME") > $ write SYS$OUTPUT "Hello, this is ",F$GetJPI("","USERNAME") > $ Close xxx > $ Exit The C program uses the user HASSLE but it could be another ID as well; it does a getuai call and a creprc afterwards, and it reads: > #include > #include > #include > #include > #include > main() > { > > /* Declarations for getuai */ > int my_gid, my_uid; > $DESCRIPTOR(usrnam,"HASSLE"); > unsigned long uic=0, uic_len; > struct item_list > { unsigned short buffer_length; > unsigned short item_code; > void *buffer_address; > unsigned long *returned_length; > } item[4] = > {{ 4, UAI$_UIC, &uic, &uic_len}, > { 0, 0, 0, 0}}; > > /* declarations for creprc */ > $DESCRIPTOR(Standard_Output,"XXXX.OUT"); > $DESCRIPTOR(Standard_Input,"XXXX.COM"); > $DESCRIPTOR(Proc_name,"HASSLE"); > $DESCRIPTOR(Image,"SYS$SYSTEM:LOGINOUT.EXE"); > > int status; > int pri; > int pid; > > > pri = 3; > > /* Getuai call first */ > > status = sys$getuai(0,0,&usrnam,item,0,0,0); > printf("Status Code from SYS$GETUAI: %d (hex: %x)\n", > status, status); > > my_gid = uic >> 16; > my_uid = uic & 0xFFFF; > printf( "%d -> UID %d GID %d \n", uic, my_uid, my_gid); > printf( "%o -> UID %o GID %o \n", uic, my_uid, my_gid); > > /* > * Create the sub-process > */ > > status = sys$creprc(&pid,&Image,&Standard_Input,&Standard_Output, > &Standard_Output,0,0,&Proc_name,pri,uic,0,0); > printf("Status code returned from $CREPRC: %d (hex: %X)\n", > status, status); > > } > > > Frustrating enough, the getuai and creprc calls both succeed, and the created process _still_belongs_to_the_creator (SYSTEM) rather than the desired uic (HASSLE). I have put xxxx.com in both sys$manager and the login directory of the HASSLE user. I would appreciate any input on this problem. Regards Reinhard X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 05:28:31 PDT From: jbasara@ssdc.SSDC.Sterling.COM (Jim Basara) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: TGV-announce (was: Recent flamage) ================================================================================ Archive-Date: xxx, 30 Aug 1993 00:48:32 -0400 Date: 30 Aug 1993 00:48:32 -0400 Organization: Sterling Software IMD. (Vienna,Va) Lines: 22 NNTP-Posting-Host: ssdc.ssdc.sterling.com X-Newsreader: TIN [version 1.1 PL6] To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET smith@mcclb0.med.nyu.edu wrote: : : I have been as disgusted as anyone about the recent misuse of this list to : conduct a mindless little flame-war. But PLEASE DON'T use this as a reason : to create YET ANOTHER LIST. I have a job to do. And reading NEWS and mail : from mailing lists is meant to be a tiny part of that job, not an all : consuming avocation. On top of that if you create a newsgroup in the 'biz' : section, I have to get 'biz' from our central site, which is a huge waste of : time, not to mention additional disk space, etc. : : So PLEASE just use this list so that info about Multinet and similar issues : can be found in JUST ONE PLACE. If you want to control access, then moderate : the discussions. : I agree completely. Please consider this recommendation. -- ------------------------------------------------------------------------------- Jim Basara jbasara@ssdc.SSDC.Sterling.COM X-ST-Status: N Return-path: Received: from VMETZE.MRL.UIUC.EDU ([128.174.119.119]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 07:27:53 PDT ================================================================================ Archive-Date: Mon, 30 Aug 1993 9:26:13 -0500 Date: Mon, 30 Aug 1993 9:26:13 -0500 (CDT) From: METZE@VMETZE.MRL.UIUC.EDU To: info-multinet@tgv.com CC: METZE@VMETZE.MRL.UIUC.EDU Subject: another mailing list >From: SMTP%"jbasara@ssdc.SSDC.Sterling.COM" 30-AUG-1993 08:04:01.37 >To: METZE >: >: I have been as disgusted as anyone about the recent misuse of this list to >: conduct a mindless little flame-war. But PLEASE DON'T use this as a reason >: to create YET ANOTHER LIST. I have a job to do. And reading NEWS and mail >: from mailing lists is meant to be a tiny part of that job, not an all >: consuming avocation. On top of that if you create a newsgroup in the 'biz' I agree, too. I don't like mail piling up on my disk, but I just don't have time for newsgroups and use only the most essential mailing lists. Multinet is a very small part of my responsibilities, and it is very handy to have everything important in one list. >: section, I have to get 'biz' from our central site, which is a huge waste of >: time, not to mention additional disk space, etc. >: >: So PLEASE just use this list so that info about Multinet and similar issues >: can be found in JUST ONE PLACE. If you want to control access, then moderate >: the discussions. >: I hate flamage as you can tell from my stuck pig reactions. If this is the motivation I vote for moderating. Would it have been more appropriate to send this to tgv rather than post it? > >I agree completely. Please consider this recommendation. > >-- >------------------------------------------------------------------------------- >Jim Basara > jbasara@ssdc.SSDC.Sterling.COM X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 30 Aug 1993 08:21:43 PDT Date: Mon, 30 Aug 93 08:21:43 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Need Advice to $Creprc with non-0 UIC To: doelz @ urz.unibas.ch cc: info-multinet @ TGV.COM > I would like to start a command file from Multinet's inetd server > functionality: > 1) The server runs under SYSTEM > 2) The server has no DCL > 3) The command file shall run under a different UIC than SYSTEM. > [...] > Frustrating enough, the getuai and creprc calls both succeed, and > the created process _still_belongs_to_the_creator (SYSTEM) rather than > the desired uic (HASSLE). I have put xxxx.com in both sys$manager and > the login directory of the HASSLE user. > I would appreciate any input on this problem. This trick works for creating a subprocess, but in this case LOGINOUT notices that you're creating a detached process and acts a little differently. Basically, it looks at the USERNAME of the process and logs in as that username. The username comes from the username in your JIB which is cloned to the newly created process by $CREPRC. You'd need to write some code to go into kernel mode, poke the username in YOUR JIB, create the process, and poke the username in your JIB BACK to the original value. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 08:37:48 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: TGV-announce (was: Recent flamage) Message-ID: <1993Aug30.150924.8554@colorado.edu> From: dwing@uh01.Colorado.EDU (Dan Wing) ================================================================================ Archive-Date: Mon, 30 Aug 1993 15:09:24 GMT Date: Mon, 30 Aug 1993 15:09:24 GMT Reply-To: dwing@uh01.Colorado.EDU Sender: news@colorado.edu (The Daily Planet) Organization: University of Colorado Hospital Authority, Denver Nntp-Posting-Host: buckie.hsc.colorado.edu Lines: 13 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <25s0r0INN9ri@ssdc.SSDC.Sterling.COM>, jbasara@ssdc.SSDC.Sterling.COM (Jim Basara) writes: >I agree completely. Please consider this recommendation. I don't understand. Lets say that TGV has four "product announcements" they'd like to send out. How is it any more work, or any more time consuming, to receive those four mail message with a return address of TGV-ANNOUNCE or INFO-MULTINET? -Dan Wing, Systems Administrator, University Hospital, Denver dwing@uh01.colorado.edu or wing_d@ucolmcc.bitnet X-ST-Status: N Return-path: Received: from AC.GRIN.EDU ([132.161.10.8]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 09:25:43 PDT Received: from DREAMS.GRIN.EDU by VMSMAIL.GRIN.EDU (PMDF V4.2-11 #4381) id <01H2CS579JPC8Y59GF@VMSMAIL.GRIN.EDU>; Mon, 30 Aug 1993 11:23:44 CDT ================================================================================ Archive-Date: Mon, 30 Aug 1993 11:23:30 -0500 Date: Mon, 30 Aug 1993 11:23:30 -0500 (cdt) From: Brian D McMahon Subject: Re: another mailing list To: info-multinet@tgv.com OK, folks, if we're voting, let me register a strong vote *AGAINST* moderating this list. I would far rather see TGV devote staff resources to development and to maintaining their excellent product support. The answer is simple: Keep the blinking flames (and meta-flames, and meta-meta-flames, and soul-searching discussions about meta-flames) off the list. If you enjoy that sort of thing, go back to the Info-VAX archives from last January. We thank you for your cooperation. This, by the way, is not a flame. ;-) Brian McMahon (*) Postmaster / Acad. Software Support Grinnell College Computer Services Grinnell, Iowa 50112 USA Voice: +1 515 269 4901 Fax: +1 515 269 4936 (*) No relation. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 10:46:08 PDT From: dumoulin@titan.ksc.nasa.gov (Jim Dumoulin) X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Too many system-wide file locks via Multinet NFS Message-ID: <1993Aug30.131402.7252@titan.ksc.nasa.gov> ================================================================================ Archive-Date: xxx, 30 Aug 1993 13:14:02 EST Date: 30 Aug 93 13:14:02 EST Organization: NASA, Kennedy Space Center Lines: 403 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I've got a configuration where we're attempting to do compilation of Ada programs between a Harris Nighthawk Unix platform (M88K HAPSE on CX/UX Unix) and a VMS VAX running Multinet Version 3.2a. Ada compiles run fine between Harris Nighthawk's using their own NFS but fails with an error "Too many system-wide file locks" when compiling files via a Multinet NFS mounted file system. I can manually read/write the file system via NFS so I don't think file permissions are a problem. I have also seen requests from the Unix side requesting access to the STATD deamon. Anyone know if STATD has been ported for use by Multinet NFS? I've include a Sniffer packet trace of the last few network packets before the error. Titan is the Multinet host and Station1 is the Unix NFS side. Sniffer Network Analyzer data from 26-Aug-93 at 17:43:52, Page 1 NFS: ----- SUN NFS ----- NFS: NFS: Proc = 4 (Look up file name) NFS: File handle = 7A221C000000550E4E0100004449534B NFS: 245348415245000000000000550E0000 NFS: File name = .hapse_locks NFS: NFS: [Normal end of "SUN NFS".] NFS: ADDR HEX ASCII 0000 AA 00 04 00 82 1C AA 00 04 00 69 1D 08 00 45 00 ..........i...E. 0010 00 94 FB 5B 00 00 FD 11 20 B3 80 9F 9F 0A 80 9F ...[.... ....... 0020 01 01 03 FE 08 01 00 80 45 57 00 04 5B 6A 00 00 ........EW..[j.. 0030 00 00 00 00 00 02 00 01 86 A3 00 00 00 02 00 00 ................ 0040 00 04 00 00 00 01 00 00 00 20 2C 7D 2C B4 00 00 ......... ,},... 0050 00 08 73 74 61 74 69 6F 6E 31 00 00 02 10 00 00 ..station1...... 0060 00 01 00 00 00 01 00 00 00 01 00 00 00 00 00 00 ................ 0070 00 00 7A 22 1C 00 00 00 55 0E 4E 01 00 00 44 49 ..z"....U.N...DI 0080 53 4B 24 53 48 41 52 45 00 00 00 00 00 00 55 0E SK$SHARE......U. 0090 00 00 00 00 00 0C 2E 68 61 70 73 65 5F 6C 6F 63 .......hapse_loc 00A0 6B 73 ks - - - - - - - - - - - - - - - - Frame 83 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 83 0.0012 station1 titan.KSC.NAS.. NFS R OK F=485A NFS: ----- SUN NFS ----- NFS: Sniffer Network Analyzer data from 26-Aug-93 at 17:43:52, unsaved capture data, Page 3 NFS: Proc = 4 (Look up file name) NFS: Status = 0 (OK) NFS: File handle = 1D234F0000007A221C0000004449534B NFS: 245348415245000000000000550E0000 NFS: File type = 1 (Regular file) NFS: Mode = 0100644 NFS: Type = Regular file NFS: Owner's permissions = rw- NFS: Group's permissions = r-- NFS: Others' permissions = r-- NFS: Link count = 1, UID = 528, GID = 1 NFS: File size = 5, Block size = 1024, No. of blocks = 2 NFS: File system id = 0, File id = 4026540829 NFS: Access time = 25-Aug-93 23:29:11.026272 GMT NFS: Modification time = 25-Aug-93 23:29:11.026272 GMT NFS: Inode change time = 25-Aug-93 23:29:11.038416 GMT NFS: NFS: [Normal end of "SUN NFS".] NFS: ADDR HEX ASCII 0000 AA 00 04 00 69 1D AA 00 04 00 82 1C 08 00 45 00 ....i.........E. 0010 00 9C 42 E9 00 00 FF 11 D7 1D 80 9F 01 01 80 9F ..B............. 0020 9F 0A 08 01 03 FE 00 88 E3 45 00 04 5B 6A 00 00 .........E..[j.. 0030 00 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 0040 00 00 00 00 00 00 1D 23 4F 00 00 00 7A 22 1C 00 .......#O...z".. 0050 00 00 44 49 53 4B 24 53 48 41 52 45 00 00 00 00 ..DISK$SHARE.... 0060 00 00 55 0E 00 00 00 00 00 01 00 00 81 A4 00 00 ..U............. 0070 00 01 00 00 02 10 00 00 00 01 00 00 00 05 00 00 ................ 0080 04 00 00 00 00 00 00 00 00 02 00 00 00 00 F0 00 ................ 0090 23 1D 2C 7B F5 C7 00 0E 66 A0 2C 7B F5 C7 00 0E #.,{....f.,{.... 00A0 66 A0 2C 7B F5 C7 00 0A 96 10 f.,{...... - - - - - - - - - - - - - - - - Frame 84 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 84 0.0086 titan.KSC.NAS.. station1 NFS C Create .hapse_locks in F=0176 NFS: ----- SUN NFS ----- NFS: NFS: Proc = 9 (Create file) NFS: File handle = 7A221C000000550E4E0100004449534B NFS: 245348415245000000000000550E0000 NFS: File name = .hapse_locks NFS: Mode = 0644 NFS: Type = ? NFS: Owner's permissions = rw- NFS: Group's permissions = r-- NFS: Others' permissions = r-- NFS: UID = -1 NFS: GID = 1 NFS: Size = -1 NFS: Access time = -1 NFS: Modification time = -1 NFS: NFS: [Normal end of "SUN NFS".] NFS: ADDR HEX ASCII 0000 AA 00 04 00 82 1C AA 00 04 00 69 1D 08 00 45 00 ..........i...E. 0010 00 B4 FB 5C 00 00 FD 11 20 92 80 9F 9F 0A 80 9F ...\.... ....... 0020 01 01 03 FE 08 01 00 A0 43 6C 00 04 5B 6B 00 00 ........Cl..[k.. 0030 00 00 00 00 00 02 00 01 86 A3 00 00 00 02 00 00 ................ 0040 00 09 00 00 00 01 00 00 00 20 2C 7D 2C B4 00 00 ......... ,},... 0050 00 08 73 74 61 74 69 6F 6E 31 00 00 02 10 00 00 ..station1...... 0060 00 01 00 00 00 01 00 00 00 01 00 00 00 00 00 00 ................ 0070 00 00 7A 22 1C 00 00 00 55 0E 4E 01 00 00 44 49 ..z"....U.N...DI 0080 53 4B 24 53 48 41 52 45 00 00 00 00 00 00 55 0E SK$SHARE......U. 0090 00 00 00 00 00 0C 2E 68 61 70 73 65 5F 6C 6F 63 .......hapse_loc 00A0 6B 73 00 00 01 A4 FF FF FF FF 00 00 00 01 FF FF ks.............. 00B0 FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF FF ................ 00C0 FF FF .. - - - - - - - - - - - - - - - - Frame 85 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 85 0.0331 station1 titan.KSC.NAS.. NFS R OK F=485A NFS: ----- SUN NFS ----- NFS: NFS: Proc = 9 (Create file) NFS: Status = 0 (OK) NFS: File handle = 1D234F0000007A221C0000004449534B NFS: 245348415245000000000000550E0000 NFS: File type = 1 (Regular file) NFS: Mode = 0100644 NFS: Type = Regular file NFS: Owner's permissions = rw- NFS: Group's permissions = r-- NFS: Others' permissions = r-- NFS: Link count = 1, UID = 528, GID = 1 NFS: File size = 5, Block size = 1024, No. of blocks = 2 NFS: File system id = 0, File id = 4026540829 NFS: Access time = 25-Aug-93 23:29:11.026272 GMT NFS: Modification time = 25-Aug-93 23:29:11.026272 GMT NFS: Inode change time = 25-Aug-93 23:29:11.038416 GMT NFS: NFS: [Normal end of "SUN NFS".] NFS: ADDR HEX ASCII 0000 AA 00 04 00 69 1D AA 00 04 00 82 1C 08 00 45 00 ....i.........E. 0010 00 9C 43 E9 00 00 FF 11 D6 1D 80 9F 01 01 80 9F ..C............. 0020 9F 0A 08 01 03 FE 00 88 E3 44 00 04 5B 6B 00 00 .........D..[k.. 0030 00 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 0040 00 00 00 00 00 00 1D 23 4F 00 00 00 7A 22 1C 00 .......#O...z".. 0050 00 00 44 49 53 4B 24 53 48 41 52 45 00 00 00 00 ..DISK$SHARE.... 0060 00 00 55 0E 00 00 00 00 00 01 00 00 81 A4 00 00 ..U............. 0070 00 01 00 00 02 10 00 00 00 01 00 00 00 05 00 00 ................ 0080 04 00 00 00 00 00 00 00 00 02 00 00 00 00 F0 00 ................ 0090 23 1D 2C 7B F5 C7 00 0E 66 A0 2C 7B F5 C7 00 0E #.,{....f.,{.... 00A0 66 A0 2C 7B F5 C7 00 0A 96 10 f.,{...... - - - - - - - - - - - - - - - - Frame 86 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 86 0.0080 titan.KSC.NAS.. station1 NFS C F=485A Get attributes NFS: ----- SUN NFS ----- NFS: NFS: Proc = 1 (Get file attributes) NFS: File handle = 1D234F0000007A221C0000004449534B NFS: 245348415245000000000000550E0000 NFS: NFS: [Normal end of "SUN NFS".] NFS: ADDR HEX ASCII 0000 AA 00 04 00 82 1C AA 00 04 00 69 1D 08 00 45 00 ..........i...E. 0010 00 84 FB 5D 00 00 FD 11 20 C1 80 9F 9F 0A 80 9F ...].... ....... 0020 01 01 03 FE 08 01 00 70 B9 F1 00 04 5B 6C 00 00 .......p....[l.. 0030 00 00 00 00 00 02 00 01 86 A3 00 00 00 02 00 00 ................ 0040 00 01 00 00 00 01 00 00 00 20 2C 7D 2C B4 00 00 ......... ,},... 0050 00 08 73 74 61 74 69 6F 6E 31 00 00 02 10 00 00 ..station1...... 0060 00 01 00 00 00 01 00 00 00 01 00 00 00 00 00 00 ................ 0070 00 00 1D 23 4F 00 00 00 7A 22 1C 00 00 00 44 49 ...#O...z"....DI 0080 53 4B 24 53 48 41 52 45 00 00 00 00 00 00 55 0E SK$SHARE......U. 0090 00 00 .. - - - - - - - - - - - - - - - - Frame 87 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 87 0.0037 station1 titan.KSC.NAS.. NFS R OK NFS: ----- SUN NFS ----- NFS: NFS: Proc = 1 (Get file attributes) NFS: Status = 0 (OK) NFS: File type = 1 (Regular file) NFS: Mode = 0100644 NFS: Type = Regular file NFS: Owner's permissions = rw- NFS: Group's permissions = r-- NFS: Others' permissions = r-- NFS: Link count = 1, UID = 528, GID = 1 NFS: File size = 5, Block size = 1024, No. of blocks = 2 NFS: File system id = 0, File id = 4026540829 NFS: Access time = 25-Aug-93 23:29:11.026272 GMT NFS: Modification time = 25-Aug-93 23:29:11.026272 GMT NFS: Inode change time = 25-Aug-93 23:29:11.038416 GMT NFS: NFS: [Normal end of "SUN NFS".] NFS: ADDR HEX ASCII 0000 AA 00 04 00 69 1D AA 00 04 00 82 1C 08 00 45 00 ....i.........E. 0010 00 7C 44 E9 00 00 FF 11 D5 3D 80 9F 01 01 80 9F .|D......=...... 0020 9F 0A 08 01 03 FE 00 68 91 46 00 04 5B 6C 00 00 .......h.F..[l.. 0030 00 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 0040 00 00 00 00 00 00 00 00 00 01 00 00 81 A4 00 00 ................ 0050 00 01 00 00 02 10 00 00 00 01 00 00 00 05 00 00 ................ 0060 04 00 00 00 00 00 00 00 00 02 00 00 00 00 F0 00 ................ 0070 23 1D 2C 7B F5 C7 00 0E 66 A0 2C 7B F5 C7 00 0E #.,{....f.,{.... 0080 66 A0 2C 7B F5 C7 00 0A 96 10 f.,{...... - - - - - - - - - - - - - - - - Frame 89 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 89 0.0096 titan.KSC.NAS.. station1 PMAP C Get port PROG=100021, VERS=1, IP=17 PMAP: ----- SUN Port Map ----- PMAP: PMAP: Proc = 3 (Look up a mapping) PMAP: Program = 100021, version = 1 PMAP: Protocol = 17 PMAP: PMAP: [Normal end of "SUN Port Map".] PMAP: ADDR HEX ASCII 0000 AA 00 04 00 82 1C AA 00 04 00 69 1D 08 00 45 00 ..........i...E. 0010 00 54 FB 5F 00 00 1C 11 01 F0 80 9F 9F 0A 80 9F .T._............ 0020 01 01 02 C9 00 6F 00 40 F1 C6 2C 7E 2F 36 00 00 .....o.@..,~/6.. 0030 00 00 00 00 00 02 00 01 86 A0 00 00 00 02 00 00 ................ 0040 00 03 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 0050 00 00 00 01 86 B5 00 00 00 01 00 00 00 11 00 00 ................ 0060 00 00 .. - - - - - - - - - - - - - - - - Frame 90 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 90 0.0233 station1 titan.KSC.NAS.. PMAP R PORT=0 PMAP: ----- SUN Port Map ----- PMAP: PMAP: Proc = 3 (Look up a mapping) PMAP: Port = 0 PMAP: PMAP: [Normal end of "SUN Port Map".] PMAP: ADDR HEX ASCII 0000 AA 00 04 00 69 1D AA 00 04 00 82 1C 08 00 45 00 ....i.........E. 0010 00 38 E9 45 00 00 3C 11 F4 25 80 9F 01 01 80 9F .8.E..<..%...... 0020 9F 0A 00 6F 02 C9 00 24 FF 6E 2C 7E 2F 36 00 00 ...o...$.n,~/6.. 0030 00 01 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 0040 00 00 00 00 00 00 ...... - - - - - - - - - - - - - - - - Frame 91 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 91 0.0098 titan.KSC.NAS.. station1 Telnet R PORT=2854 /usr Telnet:----- Telnet data ----- Telnet: Telnet:/usr Telnet: ADDR HEX ASCII 0000 AA 00 04 00 82 1C AA 00 04 00 69 1D 08 00 45 00 ..........i...E. 0010 00 2C FB 61 00 00 1C 06 02 21 80 9F 9F 0A 80 9F .,.a.....!...... 0020 01 01 00 17 0B 26 05 C9 DE CF 19 9D 9E AE 50 18 .....&........P. 0030 18 00 AB 75 00 00 2F 75 73 72 80 9F ...u../usr.. - - - - - - - - - - - - - - - - Frame 92 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 92 0.0069 [128.159.6.28] titan.KSC.NAS.. Telnet R PORT=9718 /usr Telnet:----- Telnet data ----- Telnet: Telnet:/usr Telnet: ADDR HEX ASCII 0000 08 00 05 07 D8 0D AA 00 04 00 82 1C 08 00 45 00 ..............E. 0010 00 2C E9 46 00 00 3C 06 8D 2A 80 9F 01 01 80 9F .,.F..<..*...... 0020 06 1C 00 17 25 F6 09 66 BC B3 95 FC 3D 41 50 18 ....%..f....=AP. 0030 10 00 35 21 00 00 2F 75 73 72 00 40 ..5!../usr.@ - - - - - - - - - - - - - - - - Frame 97 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 97 0.0334 station1 titan.KSC.NAS.. TCP D=23 S=2854 ACK=97115859 WIN=6144 TCP: ----- TCP header ----- TCP: TCP: Source port = 2854 TCP: Destination port = 23 (Telnet) TCP: Sequence number = 429760174 TCP: Acknowledgment number = 97115859 TCP: Data offset = 20 bytes TCP: Flags = 10 TCP: ..0. .... = (No urgent pointer) TCP: ...1 .... = Acknowledgment TCP: .... 0... = (No push) TCP: .... .0.. = (No reset) TCP: .... ..0. = (No SYN) TCP: .... ...0 = (No FIN) TCP: Window = 6144 TCP: Checksum = 4E65 (correct) TCP: No TCP options TCP: ADDR HEX ASCII 0000 AA 00 04 00 69 1D AA 00 04 00 82 1C 08 00 45 00 ....i.........E. 0010 00 28 E9 47 00 00 3C 06 F4 3E 80 9F 01 01 80 9F .(.G..<..>...... 0020 9F 0A 0B 26 00 17 19 9D 9E AE 05 C9 DE D3 50 10 ...&..........P. 0030 18 00 4E 65 00 00 00 40 2D 80 B1 06 ..Ne...@-... - - - - - - - - - - - - - - - - Frame 98 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 98 0.0067 titan.KSC.NAS.. station1 Telnet R PORT=2854 2/user/spftest/tst2/fixed.a, line 0, char 0:fa... Telnet:----- Telnet data ----- Telnet: Telnet:2/user/spftest/tst2/fixed.a, line 0, char 0:fatal: Could not lock library /u... Telnet: ADDR HEX ASCII 0000 AA 00 04 00 82 1C AA 00 04 00 69 1D 08 00 45 00 ..........i...E. 0010 00 B6 FB 64 00 00 1C 06 01 94 80 9F 9F 0A 80 9F ...d............ 0020 01 01 00 17 0B 26 05 C9 DE D3 19 9D 9E AE 50 18 .....&........P. 0030 18 00 D8 52 00 00 32 2F 75 73 65 72 2F 73 70 66 ...R..2/user/spf 0040 74 65 73 74 2F 74 73 74 32 2F 66 69 78 65 64 2E test/tst2/fixed. 0050 61 2C 20 6C 69 6E 65 20 30 2C 20 63 68 61 72 20 a, line 0, char 0060 30 3A 66 61 74 61 6C 3A 20 43 6F 75 6C 64 20 6E 0:fatal: Could n 0070 6F 74 20 6C 6F 63 6B 20 6C 69 62 72 61 72 79 20 ot lock library 0080 2F 75 73 72 32 2F 75 73 65 72 2F 73 70 66 74 65 /usr2/user/spfte 0090 73 74 2F 74 73 74 32 3B 20 74 6F 6F 20 6D 61 6E st/tst2; too man 00A0 79 20 73 79 73 74 65 6D 2D 77 69 64 65 20 66 69 y system-wide fi 00B0 6C 65 20 6C 6F 63 6B 73 0D 0A 73 74 61 74 69 6F le locks..statio 00C0 6E 31 3E 20 n1> - - - - - - - - - - - - - - - - Frame 100 - - - - - - - - - - - - - - - - - SUMMARY Delta T Destination Source Summary 100 0.0014 [128.159.6.28] titan.KSC.NAS.. Telnet R PORT=9718 2/user/spftest/tst2/fixed.a, line 0, char 0:fa... Telnet:----- Telnet data ----- Telnet: Telnet:2/user/spftest/tst2/fixed.a, line 0, char 0:fatal: Could not lock library /usr2 Telnet: ADDR HEX ASCII 0000 08 00 05 07 D8 0D AA 00 04 00 82 1C 08 00 45 00 ..............E. 0010 00 78 E9 48 00 00 3C 06 8C DC 80 9F 01 01 80 9F .x.H..<......... 0020 06 1C 00 17 25 F6 09 66 BC B7 95 FC 3D 41 50 18 ....%..f....=AP. 0030 10 00 32 F6 00 00 32 2F 75 73 65 72 2F 73 70 66 ..2...2/user/spf 0040 74 65 73 74 2F 74 73 74 32 2F 66 69 78 65 64 2E test/tst2/fixed. 0050 61 2C 20 6C 69 6E 65 20 30 2C 20 63 68 61 72 20 a, line 0, char 0060 30 3A 66 61 74 61 6C 3A 20 43 6F 75 6C 64 20 6E 0:fatal: Could n 0070 6F 74 20 6C 6F 63 6B 20 6C 69 62 72 61 72 79 20 ot lock library 0080 2F 75 73 72 32 2F /usr2/ -- -------------------------------------------------------------------------- Jim Dumoulin INTERNET: DUMOULIN@TITAN.KSC.NASA.GOV NASA / Payload Operations SPAN/HEPnet: KSCP00::DUMOULIN Kennedy Space Center Florida, USA 32899 "America needs SPACE to grow" X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 30 Aug 1993 11:15:19 PDT Date: Mon, 30 Aug 93 11:15:19 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Too many system-wide file locks via Multinet NFS To: DUMOULIN @ TITAN.KSC.NASA.GOV cc: service @ TGV.COM, info-multinet @ TGV.COM > I've got a configuration where we're attempting to do compilation of Ada > programs between a Harris Nighthawk Unix platform (M88K HAPSE on > CX/UX Unix) and a VMS VAX running Multinet Version 3.2a. Ada > compiles run fine between Harris Nighthawk's using their own NFS but > fails with an error "Too many system-wide file locks" when compiling > files via a Multinet NFS mounted file system. I can manually read/write > the file system via NFS so I don't think file permissions are a problem. Ask Harris what that error means. > I have also seen requests from the Unix side requesting access to the > STATD deamon. Anyone know if STATD has been ported for use by > Multinet NFS? I've include a Sniffer packet trace of the last few network > packets before the error. Titan is the Multinet host and Station1 is the > Unix NFS side. We've implemented our own STATD in MultiNet, not using the SUN code. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 11:15:54 PDT From: william@fractl.tn.cornell.edu X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Simple Question. ================================================================================ Archive-Date: xxx, 30 Aug 1993 17:56:44 GMT Date: 30 Aug 1993 17:56:44 GMT Organization: Dept of TAM Cornell Univ. Ithaca, NY Lines: 37 Reply-To: william@fractl.tn.cornell.edu NNTP-Posting-Host: fractl.tn.cornell.edu To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Hi This is a simple question. Does anyone have a program that will determine if a host is available and : a: define a symbol that can be checked by DCL. -or- b: set the vms status word. Like ping except I would like to set a symbol or have the status indicate wether or not the host can be reached. This DCL code does what I want but it is cumbersome and I would like somthing a little simpler. $ define sys$output foo.txt $ multinet ping host /number_on_packets=1 $ deassign sys$output $ search foo.txt "1 packets received" $ status=$status $ delete foo.txt; $ if .not. status $ then $ write sys$output "Host not available" $ endif $ There has to be an easier way. Thanks Bill X-ST-Status: N Return-path: ================================================================================ Archive-Date: Mon, 30 Aug 1993 11:17:06 PDT Date: Mon, 30 Aug 93 11:17:06 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Simple Question. To: william @ fractl.tn.cornell.edu cc: info-multinet @ TGV.COM > This is a simple question. Does anyone have a program that will determine > if a host is available and : > a: define a symbol that can be checked by DCL. > -or- > b: set the vms status word. > Like ping except I would like to set a symbol or have the status indicate > wether or not the host can be reached. > This DCL code does what I want but it is cumbersome and I would like somthing > a little simpler. > $ define sys$output foo.txt > $ multinet ping host /number_on_packets=1 > $ deassign sys$output > $ search foo.txt "1 packets received" > $ status=$status > $ delete foo.txt; > $ if .not. status > $ then > $ write sys$output "Host not available" > $ endif > $ > There has to be an easier way. MultiNet V3.3's PING will define $STATUS depending on the success or failure of the PING. I believe there is a public-domain PING available for VMS/MultiNet which will also do this. Ken X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 12:08:09 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet From: hughes@logos.ucs.indiana.edu (larry j. hughes jr.) Subject: Re: Simple Question. Message-ID: Sender: news@usenet.ucs.indiana.edu (USENET News System) Nntp-Posting-Host: logos.ucs.indiana.edu Organization: University Computing Services News ================================================================================ Archive-Date: Mon, 30 Aug 1993 18:28:27 GMT Date: Mon, 30 Aug 1993 18:28:27 GMT Lines: 29 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <25tf0s$51e@fitz.TC.Cornell.EDU>, william@fractl.tn.cornell.edu writes: |> a: define a symbol that can be checked by DCL. |> -or- |> b: set the vms status word. |> |> Like ping except I would like to set a symbol or have the status indicate |> wether or not the host can be reached. Look on ftp.indiana.edu in /pub/vms/misc for query.c. Usage: $ query host port|service [-s symbolname] Example: $ query localhost telnet -s connected $ show symbol connected CONNECTED == "1" It's not quite the same as ping, as far testing for IP availability, but it checks for the existence of a given TCP service like telnet. I use it in .COM files to check for existence of a remote FTP servers, before attempting a batch FTP. //==================================================================\\ || Larry J. Hughes, Jr. | hughes@indiana.edu || || Indiana University | "The person who knows everything || || University Computing Services | has a lot to learn." || \\===================================================================// X-ST-Status: N Return-path: Received: from MAXIMILLION.MCC.COM ([128.62.2.77]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 14:35:04 PDT Received: from MAXIMILLION.MCC.COM by MAXIMILLION.MCC.COM (PMDF #2318 ) id <01H2D3ERU16O001LS0@MAXIMILLION.MCC.COM>; Mon, 30 Aug 1993 16:46:48 CDT ================================================================================ Archive-Date: xxx, 30 Aug 1993 16:46:47 -0500 Date: 30 Aug 1993 16:46:47 -0500 (CDT) From: Bob Paver Subject: Re: Simple Question. To: william@fractl.tn.cornell.edu Cc: INFO-MULTINET@TGV.COM Organization: MCC, MIS/Telecomm Mail-System-Version: X-Phone: (512) 338-3316 X-FAX: (512) 338-3888 > This is a simple question. Does anyone have a program that will determine > if a host is available and : > > a: define a symbol that can be checked by DCL. > > -or- > > b: set the vms status word. > Here is the 'readme' file from a PD version of ping that I picked up earlier this year. I don't remember where I got it; probably the ftp server at spcvxa.spc.edu. -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- Bob Paver ----> paver@mcc.com Microelectronics and Computer Technology Corp (MCC) 3500 West Balcones Center Drive Austin, Texas 78759-6509 (512) 338-3316 +++++++++++++++ O-PING-0.README +++++++++++++++ [3-DEC-1992] Various updates and enhancements: o Now allows command line redirection, piping, and backgrounding o Now returns useful status on exit: NOSUCHNODE - The node does not exist (no Type A DNS record) UNREACHABLE - The node cannot be reached (all packets lost) NORMAL - The node is reachable (no packets lost) DATALOST - The node is reachable (some packets lost) IVBUFLEN - The buffer size requested is invalid INSFARG - Insufficient arguments PROTOCOL - Error creating socket All except NORMAL and DATALOST are returned as FATAL class errors. NORMAL is returned as SUCCESS and DATALOST as WARNING. Message display is inhib- ited, so if you are testing for an explicit return value, be sure to mask the %x10000000 bit first! Above work done by Terry Kennedy, terry@spcvxa.spc.edu. [8-FEB-1991] PING for VMS/UCX PING sends ICMP ECHO_REQUEST packets to network hosts. The purpose of "PINGing" is to check for connectivity between two cooperating hosts. The difference between PING and the UCX "LOOP " command is that PING reports statistics about round-trip times and packet loss rates. To build PING execute MAKE.COM (included in the distribution). Ping has only been tested under VMS 5.4/UCX 1.3a. In any case I can't guarantee that it will work on any particular configuration. I would be interrested in any bugs and/or modifications. -------------------------------------------------------------------------------- William P. Bame | sneakernet: William P. Bame internet: bill@bill.ab.umd.edu | M.I.E.M.S.S. bbame@achi1.ab.umd.edu | 22 South Greene Street AT&T : [Work] (301) 328-3062 | Baltimore, MD 21201 CIS : 71620,425 | Room: T1R55 -------------------------------------------------------------------------------- X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Mon, 30 Aug 93 16:24:37 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: One FINAL gated question... Message-ID: <1993Aug30.183633.2159@dmc.com> From: munroe@dmc.com (Dick Munroe) ================================================================================ Archive-Date: xxx, 30 Aug 1993 18:36:33 EDT Date: 30 Aug 93 18:36:33 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 26 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET Do I need to have explicit static routes declared in the GATEWAY-DAEMON.CONFIGURATION file for each slip interface and destination? or will gated detect the transition from down to up of the interface and broadcast the new route? I.e, does: Device sl0: flags=4031 IP Address = 140.186.89.1 IP Sub-Net Mask = 255.255.255.0 IP Point to Point Destination = 198.175.187.3 require: static { 198.175.187.3 gateway 140.186.89.1 ; } ; And if it does, did I get the syntax right??? -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 02:04:59 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Simple Question. Message-ID: <1993Aug31.042826.1@spcvxb.spc.edu> From: terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr.) ================================================================================ Archive-Date: Tue, 31 Aug 1993 08:28:26 GMT Date: Tue, 31 Aug 1993 08:28:26 GMT Sender: news@spcuna.spc.edu (Network News) Organization: St. Peter's College, US Nntp-Posting-Host: spcvxa.spc.edu Lines: 12 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <930830111706.2020bd3c@TGV.COM>, adelman@TGV.COM (Kenneth Adelman) writes: > I believe there is a public-domain > PING available for VMS/MultiNet which will also do this. Yup - I added exit status support to my VMS port of the BSD ping. You can get it from ftp.spc.edu in [.ucx]ping.bck. Despite being in a UCX directory, it works fine under MultiNet. I don't know if the status codes it returns will be the same as the ones under the MultiNet release that Ken mentioned. Terry Kennedy Operations Manager, Academic Computing terry@spcvxa.bitnet St. Peter's College, Jersey City, NJ USA terry@spcvxa.spc.edu +1 201 915 9381 X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 05:44:07 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Re: Too many system-wide file locks via Multinet NFS Message-ID: <1993Aug31.082156.7261@titan.ksc.nasa.gov> From: dumoulin@titan.ksc.nasa.gov (Jim Dumoulin) ================================================================================ Archive-Date: xxx, 31 Aug 1993 08:21:56 EST Date: 31 Aug 93 08:21:56 EST Organization: NASA, Kennedy Space Center Lines: 34 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET In article <1993Aug30.131402.7252@titan.ksc.nasa.gov>, dumoulin@titan.ksc.nasa.gov (Jim Dumoulin) writes: > > I've got a configuration where we're attempting to do compilation of Ada > programs between a Harris Nighthawk Unix platform (M88K HAPSE on > CX/UX Unix) and a VMS VAX running Multinet Version 3.2a. Ada > compiles run fine between Harris Nighthawk's using their own NFS but > fails with an error "Too many system-wide file locks" when compiling > files via a Multinet NFS mounted file system. I can manually read/write > the file system via NFS so I don't think file permissions are a problem. > > I have also seen requests from the Unix side requesting access to the > STATD deamon. Anyone know if STATD has been ported for use by > Multinet NFS? I've include a Sniffer packet trace of the last few network > packets before the error. Titan is the Multinet host and Station1 is the > Unix NFS side. > > The response from TGV was to say that Multinet has implemented a STATD deamon for VMS and to enable the RPCSTATUS daemon via MULTINET CONFIG. I enabled RPCSTATUS and that cleared some of my problems but I still get the "Too many system-wide file locks" error. Packet traces from our network sniffer indicate that the Unix side did a Sun Port Map request to port 100021 just before it failed. I've looked up Port 100021 and it's reserved for the NLOCKMGR daemon. I've enabled the RPCLOCKMGR program via MULTINET CONFIG/SERVERS but am still getting errors. -- -------------------------------------------------------------------------- Jim Dumoulin INTERNET: DUMOULIN@TITAN.KSC.NASA.GOV NASA / Payload Operations SPAN/HEPnet: KSCP00::DUMOULIN Kennedy Space Center Florida, USA 32899 "America needs SPACE to grow" X-ST-Status: N Return-path: Received: from ti.com ([192.94.94.5]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 06:42:31 PDT Received: from mk1501.dseg.ti.com ([128.247.206.175]) by ti.com with SMTP (5.65c/LAI-3.2) id AA08631; Tue, 31 Aug 1993 08:45:10 -0500 Received: from flopn1.dseg.ti.com by mk1501.dseg.ti.com with SMTP (5.65c/LAI-3.2) id AA18675; Tue, 31 Aug 1993 08:40:59 -0500 ================================================================================ Archive-Date: Tue, 31 Aug 1993 8:39:09 -0500 Date: Tue, 31 Aug 1993 8:39:09 -0500 (CDT) From: Carolyn Cointment 480-2278 To: info-multinet@tgv.com Cc: CAROLYN@FLOPN1.DSEG.TI.COM Subject: crash dump with NFS_SERVER as the current process We have had our vax 8550 crash twice within the last two weeks. The first time, we did not receive any messages in with sysdump or the error log. However, the second time the current process was NFS_SERVER. The bugcheck was SSRVEXCEPT, Unexpected system service exception. We are running VMS 5.4-2 . We are wondering if there are any known bugs that could be causing this problem. (DEC dialed in to look at the crash, and they seem to be pointing to multinet as a problem.) Please respond with any information that might be helpful to us. Thank you, Carolyn Cointment carolyn@flopn2.dseg.ti.com 214-480-2278 X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 31 Aug 1993 08:21:50 PDT Date: Tue, 31 Aug 93 08:21:50 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: Too many system-wide file locks via Multinet NFS To: DUMOULIN @ TITAN.KSC.NASA.GOV cc: info-multinet @ TGV.COM > In article <1993Aug30.131402.7252@titan.ksc.nasa.gov>, dumoulin@titan.ksc.nasa.gov (Jim Dumoulin) writes: >> >> I've got a configuration where we're attempting to do compilation of Ada >> programs between a Harris Nighthawk Unix platform (M88K HAPSE on >> CX/UX Unix) and a VMS VAX running Multinet Version 3.2a. Ada >> compiles run fine between Harris Nighthawk's using their own NFS but >> fails with an error "Too many system-wide file locks" when compiling >> files via a Multinet NFS mounted file system. I can manually read/write >> the file system via NFS so I don't think file permissions are a problem. >> >> I have also seen requests from the Unix side requesting access to the >> STATD deamon. Anyone know if STATD has been ported for use by >> Multinet NFS? I've include a Sniffer packet trace of the last few network >> packets before the error. Titan is the Multinet host and Station1 is the >> Unix NFS side. >> > > > Ken, > In your last response you mentioned that Multinet has implemented a STATD > deamon for VMS and to enable the RPCSTATUS daemon via MULTINET CONFIG. > I enabled RPCSTATUS and that cleared some of my problems but I still > get the "Too many system-wide file locks" error. Packet traces > from our network sniffer indicate that the Unix side did a Sun > Port Map request to port 100021 just before it failed. I've looked > up Port 100021 and it's reserved for the NLOCKMGR daemon. I've > enabled the RPCLOCKMGR program via MULTINET CONFIG/SERVERS but am > still getting errors. Do you know if Multinet has an equilivent for > NLOCKMGR? RPCLOCKMGR is NLOCKMGR. Can I see a packet trace taken with $ MULTINET TCPDUMP/SNAP=1500/RPC=ALL immediately before the failure? Ken X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 31 Aug 1993 08:23:08 PDT Date: Tue, 31 Aug 93 08:23:08 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: crash dump with NFS_SERVER as the current process To: CAROLYN @ FLOPN1.DSEG.TI.COM cc: info-multinet @ TGV.COM > We have had our vax 8550 crash twice within the last two weeks. The first > time, we did not receive any messages in with sysdump or the error log. > However, the second time the current process was NFS_SERVER. The bugcheck was > SSRVEXCEPT, Unexpected system service exception. We are running VMS 5.4-2 > . We are wondering if there are any known bugs that could be causing this > problem. (DEC dialed in to look at the crash, and they seem to be pointing > to multinet as a problem.) Please respond with any information that might be > helpful to us. I highly recommend you open a support call with us; either at SERVICE@TGV.COM or 408-457-5201. ken X-ST-Status: N Return-path: Received: from nuhub.dac.neu.edu ([129.10.1.6]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 08:57:19 PDT Received: from neu.edu by neu.edu (PMDF V4.2-13 #4985) id <01H2E7FU007K8WW0I5@neu.edu>; Tue, 31 Aug 1993 11:57:45 EST ================================================================================ Archive-Date: Tue, 31 Aug 1993 11:57:45 -0500 Date: Tue, 31 Aug 1993 11:57:45 -0500 (EST) From: "I am only an egg." Subject: tftp puts To: info-multinet@tgv.com X-Envelope-to: info-multinet@tgv.com X-VMS-To: IN%"info-multinet@tgv.com" X-VMS-Cc: JOHNSON MIME-version: 1.0 I'm having a bad time trying to get tftp put to work. I do get all the time and it works fine. I now need to do puts. Anyone get this to work? I'd like to see an example tftp.filename-translations file and any other information you have on getting it to work. Does the file need to exist? I tried that and it didn't help. Does the directory need to be world write? What's the catch? Maunal doesn't say a whole lot. Thanks much. Cheers, Chris Johnson. ============================================================================ Chris Johnson Internet: johnson@nuhub.dac.neu.edu Assistant Director, Systems BITNET: johnson@nuhub Division of Academic Computing Voice: 617.373.3300 Northeastern University FAX: 617.373.8600 ============================================================================ X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 31 Aug 1993 10:08:32 PDT Date: Tue, 31 Aug 93 10:08:32 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: tftp puts To: JOHNSON @ neu.edu cc: info-multinet @ TGV.COM > I'm having a bad time trying to get tftp put to work. I do get all > the time and it works fine. I now need to do puts. > Anyone get this to work? I'd like to see an example > tftp.filename-translations file and any other information you have on > getting it to work. Does the file need to exist? I tried that and it > didn't help. Does the directory need to be world write? What's the > catch? Maunal doesn't say a whole lot. The *file* needs to pre-exist and be world-writable. ken X-ST-Status: N Return-path: Received: from Fang.TGV.COM ([161.44.224.2]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 10:15:35 PDT ================================================================================ Archive-Date: Tue, 31 Aug 1993 10:15:33 PDT Date: Tue, 31 Aug 93 10:15:33 PDT From: moreau (Sylvie J. Moreau-Fox) @ Sales.TGV.COM Subject: New distributor in Germany To: info-multinet @ tgv.com TGV, Inc. announces that it has signed an agreement with Tekelec Airtronic GmbH of Munich under which Tekelec is now the official reseller of MultiNet, TGV's TCP/IP software for DEC VMS users, and MultiWare, the company's newest connectivity software that allows NetWare users to access OpenVMS resources. Consequently all supports agreements signed between Debis Systemhaus, TGV's former distributor, and end-users are being transferred to Tekelec. Tekelec has the technical expertise and the superior reputation for service that TGV requires for its international resellers. Just as TGV has built its international reputation on service, Tekelec has built its reputation among its work group and networking customers by providing the best technology available and backing those products with superior technical support. For any support or sales related issues, please contact Tekelec Airtronic Kapuzinerstrasse 9 8000- Munich 2 Telephone: 49-89-5164 0 Fax: 49-89-516 4110 or TGV at sales@TGV.COM or support@TGV.COM. If you have any questions or issue that you wish to discuss, please feel free to call TGV directly at +1 408 457 5200 and ask for the International Sales Department. Sylvie Moreau-Fox TGV, Inc. X-ST-Status: N Return-path: Received: from srs.gov ([192.33.240.5]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 11:22:06 PDT Received: by srs.gov id AA03458 (InterLock SMTP Gateway 1.1 for info-multinet@TGV.COM); Tue, 31 Aug 1993 14:14:36 -0400 Received: by srs.gov (Internal Mail Agent-1); Tue, 31 Aug 1993 14:14:36 -0400 ================================================================================ Archive-Date: Tue, 31 Aug 1993 14:22:16 -0400 Date: Tue, 31 Aug 1993 14:22:16 -0400 (EDT) From: "F. Arthur Cochrane" Subject: How to set up Multinet for SQL*Net for Oracle To: info-multinet@TGV.COM X-Envelope-To: info-multinet@TGV.COM Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII I am wondering of anyone is using SQL*Net for Oracle with Multinet. The SQL*Net TCP/IP User's Guide only talks about setting up for Wollongong and Excelan. Looks like it wants a TCP port on 1525 setup but I do not know the program it wants to run. For Multinet I think I would to a CONFIGURE/SERVERS and do an ADD SQLNET (?) for TCP and port 1525 but what program to I tell it to execute. This is so PCs can access Oracle via TCP/IP. I hope someone out there is doing this. Arthur Cochrane (803) 725-2627 cochrane@srs.gov X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 16:01:49 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: static route questions. Message-ID: <1993Aug31.172049.2165@dmc.com> From: munroe@dmc.com (Dick Munroe) ================================================================================ Archive-Date: xxx, 31 Aug 1993 17:20:49 EDT Date: 31 Aug 93 17:20:49 EDT Organization: Doyle, Munroe Consultants, Inc., Hudson, MA Lines: 42 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I believe I have to define static routes to destinations on my SLIP links using GATED. For interface: Device sl1: flags=4070 IP Address = 140.186.89.2 IP Sub-Net Mask = 255.255.255.0 IP Point to Point Destination = 140.186.89.254 I attempt to do: ... static { default gateway 140.186.88.6 ; 140.186.89.254 gateway 140.186.89.2 ; } ; ... But when I restart the server, I get: QUICK::LOG$ROOT:[DMCONNECTION$ %%%%%%%%%%% OPCOM 31-AUG-1993 17:20:02.79 %%%%%%%%%%% (from node HULK at 31-AUG-1993 17:19:56.71) Message from user SYSTEM on HULK MultiNet Server: GATED: parse: MULTINET:GATEWAY-DAEMON.CONFIGURATION:19 gateway not a host address on an attached netwo QUICK::LOG$ROOT:[DMCONNECTION$ %%%%%%%%%%% OPCOM 31-AUG-1993 17:20:02.82 %%%%%%%%%%% (from node HULK at 31-AUG-1993 17:19:56.74) Message from user SYSTEM on HULK rk: '140.186.89.2' QUICK::LOG$ROOT:[DMCONNECTION$ %%%%%%%%%%% OPCOM 31-AUG-1993 17:20:02.84 %%%%%%%%%%% (from node HULK at 31-AUG-1993 17:19:56.76) Message from user SYSTEM on HULK MultiNet Server: GATED: parse_parse: 1 parse error What am I doing wrong here??? -- Dick Munroe Internet: munroe@dmc.com Doyle, Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. 01749 USA FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. X-ST-Status: N Return-path: Received: from PFC.MIT.EDU ([18.77.0.113]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 16:41:02 PDT ================================================================================ Archive-Date: Tue, 31 Aug 1993 19:41:01 -0400 Date: Tue, 31 Aug 1993 19:41:01 -0400 (EDT) From: Mark London To: info-multinet@tgv.com Subject: BOOTP for HP 1200C I recently received a HP1200 desktop printer with ethernet interface. I planned to set up a tcp print queue for it. However, unlike the HP4M, there are no front panel controls to set the IP info! It's all done using BOOTP. All the documentation is for Unix, and HP themselves were of little help. However, having set up BOOTP for a NCD X-terminal, I put an entry in the BOOTP config file and it worked! Luckily, unlike the terminal, the printer did not seem to require any download file, but it seems like I had to specify one anyway, which I did (a non-existant file). Mark London X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 19:59:38 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Searching the Info-Multinet archives? Message-ID: <93243.153000RBYAML@rohvm1.rohmhaas.com> From: Aengus Lawlor ================================================================================ Archive-Date: Tue, 31 Aug 1993 15:30:00 EDT Date: Tue, 31 Aug 1993 15:30:00 EDT Organization: Rohm and Haas Company Lines: 21 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I wanted to check the archives for this group, but I found I couldn't figure out how to do it, short of downloading all the archive files, and searching them locally. Is Info-Multinet run by a Listserv, and if so, can I send database commands to it? What's the Listserv address? If it's not, is there another way to search the archives? (Actually, what I origibally wanted to know is where I could find a good NNTP news-reader for VMS w/ Multinet. Any pointers to that would also be appreciated. Pre compiled, like the Gopher client at niord.shsu.edu, would make it even nicer :-) Aengus -- rbyaml@rohmhaas.com Aengus Lawlor alawlor@ejovax.enet.dec.com (who used to be alawlor@dit.ie) An bhfuil cead agam dul amok? (For the record, neither Rohm & Haas nor DEC necessarily agree with me). X-ST-Status: N Return-path: Received: from MVB.SAIC.COM ([139.121.19.1]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 19:59:39 PDT X-Newsgroups: vmsnet.networks.tcp-ip.multinet Subject: Need to send text files to a SMTP Host Message-ID: <260t10$hiv@elroy.jpl.nasa.gov> From: rfowler@ins-server.jpl.nasa.gov ================================================================================ Archive-Date: xxx, 1 Sep 1993 01:14:08 GMT Date: 1 Sep 1993 01:14:08 GMT Distribution: usa Organization: JPL/NASA Keywords: smtp NNTP-Posting-Host: 137.79.114.180 Lines: 38 To: INFO-MULTINET@TGV.COM X-Gateway-Source-Info: USENET I need a WinSock utility that will take a text file (containing appropriate header info) and send it to a "smart" smtp host. I have a dBase programer who needs to be able to semi-automatically send ~500 messages once a week. She can write the report script to generate 500 text files with the appropriate "from" and "to" lines but no one here has experience with automating sending mail from the PC or with writing winsock/network software. I am looking for something that will work like this: smtpsend c:\outmail\*.* smtp.host.com Where the program is smtpsend.exe, there are many text files in c:\outmail\, and smtp.host.com is a "smart" smtp host, with it's ip address defined in the hosts file. This program will be used with Microsoft W4Wrkg & MS TCP/IP connection, (for now) and WinNT (starting soon, and for the rest of its useful life). So a WinNT console app would be just fine. If you know of anything like this or would be interested in hacking some ones smtp sample code in your spare time, please let me know. Alternatively I have access to a VAX 6310 w/Multinet, if there is a way to do this with the vax and multinet (copy the files to a spool directory on the vax and let multinet send them?) I would like to hear about it. ---- "Insisting on perfect safety is for people who don't have the balls to live in the real world." -- Mary Shafer, NASA Ames Dryden If mail to RFowler@ins-server.jpl.nasa.gov bounces try using RFowler@kelvin.jpl.nasa.gov X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 20:48:55 PDT ================================================================================ Archive-Date: Tue, 31 Aug 1993 20:48:52 -0700 Date: Tue, 31 Aug 1993 20:48:52 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: Searching the Info-Multinet archives? To: Aengus Lawlor Cc: INFO-MULTINET@TGV.COM In-Reply-To: <93243.153000RBYAML@rohvm1.rohmhaas.com> Mail-System-Version: Organization: TGV, Inc. X-Phone: 408/457-5200 (work); 408/457-5205 (fax) X-Address: 101 Cooper Street; Santa Cruz, CA 95060 (work) >I wanted to check the archives for this group, but I found I couldn't >figure out how to do it, short of downloading all the archive files, >and searching them locally. The info-multinet archives are available in the [.INFO-MULTINET] under anonymous on TGV.COM. They are packaged there on a monthly basis. >Is Info-Multinet run by a Listserv, and if so, can I send database commands >to it? What's the Listserv address? Not a Listserv, but rather a Treyserv (Trey Garlough is the current keeper of the info-multinet flame). >(Actually, what I origibally wanted to know is where I could find a good >NNTP news-reader for VMS w/ Multinet. Any pointers to that would also be >appreciated. Pre compiled, like the Gopher client at niord.shsu.edu, would >make it even nicer :-) Here's info on a variety of Newsreaders that work with VMS and MultiNet. Regards! -----Stuart While MultiNet does not come with either an NNTP server or client, there is one freely available NNTP server and many free NNTP clients available around the Internet. Here is a list of the ones that we know of: ANU-News - Not just a news reader, but a full-blown news system for VMS, including both an NNTP client and server. Available by anonymous FTP from kuhub.cc.ukans.edu and sao.aarnet.edu.au. VMS style user interface. Contact Geoff Huston (geoff@sao.aarnet.edu.au) for more information. NEWSRDR - A news reader (NNTP client). Requires access to a system running an NNTP server. Available by anonymous FTP from ftp.spc.edu in directory [.MADISON.NEWSRDR]. It is also available through e-mail distribution from the BITNET mail server address FILESERV@WKUVX1 (send a one-line e-mail message with the text "HELP" in the message body for more information). MultiNet customers who get CDROM distribution can also find a recent version in the conributed software section on the CDROM. Contact Matt Madison (madison@tgv.com) for more information. VNEWS - Another NNTP client. Available by anonymous FTP from arizona.edu in directory [SOFTWARE.VMS.VNEWS]. UNIX style user interface. Contact Joel Snyder (jms@arizona.edu) for more information. VMS NEWS - Yet another NNTP client. Available by anonymous FTP from ftp.rz.uni-ulm.de, directory /pub/VMS/communications, and on ftp.spc.edu, directory [.ucx]. Contact Bernd Onasch (bernd@rzsun2.rz.uni-karlsruhe.de) for more information. DXRN/MXRN - DECwindows and Motif implementations of xrn, an X Windows based NNTP client. Available by anonymous FTP from decuac.dec.com, file /pub/DEC/dxrn.share. Contact Rick Murphy (murphy@burfle.dco.dec.com) for more information. BULLETIN - Includes a USENET news reading mode. Send mail to BULLETIN@ORYANA.PFC.MIT.EDU with the word "INFO" in the message body for details on what BULLETIN is and how to get it. FNEWS - Basically a mixture of NEWSRDR and ANU-NEWS, providing a somewhat different full-screen interface and quick response to all 1800 groups. It can be found in the pub/fnews/vms directory on zephyr.grace.cri.nz. Contact Chrisp@grace.cri.nz for more information. Many of these programs are also available from DECUS, either from the DECUS Library or on recent VAX SIG tapes. [last update: 19 Jul 1993] X-ST-Status: N Return-path: Received: from Octavia.TGV.COM ([161.44.128.87]) by TGV.COM via INTERNET ; Tue, 31 Aug 93 20:57:32 PDT ================================================================================ Archive-Date: Tue, 31 Aug 1993 20:57:28 -0700 Date: Tue, 31 Aug 1993 20:57:28 -0700 (PDT) From: "L. Stuart Vance" Subject: Re: How to set up Multinet for SQL*Net for Oracle To: "F. Arthur Cochrane" Cc: info-multinet@TGV.COM In-Reply-To: <01H2ECBETCJM000HJ0@SAEMS.SRS.GOV> Mail-System-Version: Organization: TGV, Inc. X-Phone: 408/457-5200 (work); 408/457-5205 (fax) X-Address: 101 Cooper Street; Santa Cruz, CA 95060 (work) > I am wondering of anyone is using SQL*Net for Oracle with Multinet. The > SQL*Net TCP/IP User's Guide only talks about setting up for Wollongong > and Excelan. Looks like it wants a TCP port on 1525 setup but I do not > know the program it wants to run. For Multinet I think I would to a > CONFIGURE/SERVERS and do an ADD SQLNET (?) for TCP and port 1525 but > what program to I tell it to execute. This is so PCs can access Oracle > via TCP/IP. > > I hope someone out there is doing this. Sounds like you have a seriously old version of SQL*Net. SQL*Net has supported MultiNet for 2-3 years, and the currently shipping version (V1.2, I believe) certainly does. I'd give Oracle a call about getting an update. If you do want to run it with WIN/TCP, try doing the following: o Enable WIN/TCP emulation mode in MULTINET CONFIGURE: NET-CONFIG>SET WINS-COMPATIBILITY TRUE NET-CONFIG>EXIT Reboot your system. o Fire up SQL*Net. SQL*Net runs as a standalone process, and does not need to be entered into the Server Configuration Utility. If the above steps don't do it for you, you may need to also: o Create a file called MULTINET:HOSTS. (it should be a UNIX style hosts file) o Create a file called MULTINET:SERVICES. (it should be a UNIX style services file) Alternatively, if you are actually running SQL*Net V2, you should follow the instructions at the end of this message. Oracle, alas, has not ported SQL*Net V2 to run over MultiNet. If you need this functionality, please harass your Oracle sales rep... Regards! -----Stuart ============================== Some few weeks ago I put in a call to TGV technical support about how to make Oracle SQL*net version 2.0 (actually now 2.0.13) work with MULTINET. Oracle, at present, only supports Wollongong with this version. Anyway, I didn't get a clear answer from TGV or from Oracle on how to make 'trick' sql*net into thinking MULTINET is Wollongong. After fiddling around (dumping Oracle executable files & the like) with the problem for a while I found out how to make things work. The steps are: 1) Re-link Oracle NETCONFIG & enable the sql*net version 2 wollongong driver using ora_root:[install]oracleins.com. Remember to shutdown your Oracle instance & to stop all of the listener process before you do this re-link. 2) Examine an Oracle image to determine what files/paths it is looking along $set def ora_root:[sqlplus] $search sqlplus.exe twg$tcp .....Lots of binary gibberish with a few plain text strings including: TCP/IPdriverTWG$TCP:[NETDIST.ETC]HOSTS.TWG$TCP:[NETDIST.ETC]SERVICES. 3) Enable WINS-COMPATIBILITY $ multinet config/net NET-CONFIG> set WINS-COMPATIBILITY true NET-CONFIG> exit then reboot the VAX to make the change active. For example: $@sys$update:autogen reboot I'm not really sure if the re-boot is necessary. 3.5) Add the TCPSRV as a service in the MULTINET:HOSTS.LOCAL. Also, just for the heck of it add the version 1.x sql*net service. SERVICE : TCP : 1521 : TCPSRV : SERVICE : TCP : 1525 : ORASRV : If you are doing this for the first time remember that you also need to add your host to the HOST.LOCAL file as sql*net is real stupid (doesn't use bind). HOST : your.ip.address.xxx : yourhost : VAX-4000-300 : VMS : : remember to compile the host table: $MULTINET HOST_TABLE COMPILE $@multinet:install_databases $@multinet:start_server 4) Examine what logicals setting WINS-COMPATIBILITY enables $show logical twg$tcp "TWG$TCP" = "DSA0:[SYS2.MULTINET.]" (LNM$SYSTEM_TABLE) = "DSA0:[SYS2.SYSCOMMON.MULTINET.]" 5) As sql*net version 2 is looking for the HOSTS. & SERVICES. file off of twg$tcp:[netdist.etc] create these directories & files (with a Wollongong compatable structure) with the contents shown below. You want to be make these files on the [SYS2.MULTINET.] path if you are on a VAX cluster with a shared system disk. $set def DSA0:[SYS2.MULTINET] $create/dir [.netdist] $set def [.netdist] $create/dir [.etc] $set def [.etc] $edit hosts. $edit services. . . . $type services. tcpsrv 1521/tcp $type hosts. 192.197.123.123 yournode yournode.yourdomain where, of course, the IP address is yours not the above 6) Create some Oracle configuration files in the ora_root:[network.admin] and ora_root:[netconfig] directory as is outlined (sort of) in the pitiful sql*net version 2.0 reference guide (which is wrong). As an aid to others I will include the contents of my configuration which includes configuration information of DECnet & TCP/IP sql*net drivers. Note that this configuration is valid for 2.0.13 of sql*net. It may not work with other sql*net 2.x versions as Oracle sometimes changes the format of these files. Look at the contents of the ora_root:[netconfig]*.sample files for hints on what the contents should look like. There is a sql*net configuration tool called NCFW. At his writing I would not bother trying to figure out how to use it (I can't & there is NO documentation on the tools use). $type ora_root:[netconfig]config.ora REM The line(s) below identify all the aliases for the databases REM setup on current node, and are served through SQL*net TCP/IP REM REM note: ora_root= $1$dua74:[oracle7.] on my system SQLNET SIDMAP V7 = $1$dua74:[oracle7.netconfig]dora7.com $type ora_root:[netconfig]dora7.com $! the line below is commented out & only the critical logicals are assigned $! to speed startup of the server process. $! $!@ora_root:[util]orauser.com $ASS/NOLOG V7 ORA_SID $ASS/NOLOG V7 ORA_CLU $DEFINE/NOLOG/TRANS=CONCEALED ORA_ROOT SHR$DISK74:[ORACLE7.] $@ORA_ROOT:[RDBMS]RDBMSUSER.COM $ORASRV:=$ORA_SYSTEM:SRV.EXE $ORASRV "(LOCAL=NO)" $EXIT $type ora_root:[network.admin]listener.ora listener =(description = (address_list = (address = (protocol = decnet) (object = ora7t) (node = yournode)) (address = (protocol =tcp) (port = 1521) (host = yournode)) ) ) stop_listener = yes trace_directory_listener = $1$dua74:[oracle7.network.trace] trace_file_listener = listener trace_level_listener = 1 sid_list_listener = (sid_list= (sid_desc=(sid_name=v7) (PROGRAM='$1$dua74:[oracle7.netconfig]dora7.com')) ) $type ora_root:[network.admin]tnsnames.ora ora7dn =(description = (address = (protocol = decnet) (node = aspen) (object = ora7t)) (connect_data = (sid= v7)) ) ora7tcp =(description = (address = (protocol = tcp) (port = 1521) (host = aspen)) (connect_data = (sid = v7)) ) $type ora_root:[network.admin]sqlnet.ora TRACE_DIRECTORY_CLIENT = ORA_ROOT:[NETWORK.TRACE] TRACE_DIRECTORY_SERVER = ORA_ROOT:[NETWORK.TRACE] TRACE_FILE_CLIENT = CLIENT TRACE_FILE_SERVER = SERVER TRACE_LEVEL_CLIENT = 1 TRACE_LEVEL_SERVER = 1 AUTOMATIC_IPC = OFF 7) For grins I'm including the commands used to set up the DECnet object refered to in the above *.ora configuration files. Note: if your decnet service is mis-configured your tcp/ip service will also not work (the reverse is also true). $mc ncp $def object ora7t number 0 $set object ora7t number 0 8) Start up the sql*net server(s?) $lsn:==@ora_root:[netconfig]lsnrctl.com That is all there is to it. I hope that this bit will be of some help to others. X-ST-Status: N Return-path: ================================================================================ Archive-Date: Tue, 31 Aug 1993 21:31:56 PDT Date: Tue, 31 Aug 93 21:31:56 PDT From: adelman (Kenneth Adelman) @ TGV.COM Reply-To: Adelman (Kenneth Adelman) @ TGV.COM Subject: Re: static route questions. To: munroe @ dmc.com cc: info-multinet @ TGV.COM > I believe I have to define static routes to destinations on my SLIP links using > GATED. For interface: > Device sl1: flags=4070 > IP Address = 140.186.89.2 > IP Sub-Net Mask = 255.255.255.0 > IP Point to Point Destination = 140.186.89.254 > I attempt to do: > ... > static { > default gateway 140.186.88.6 ; > 140.186.89.254 gateway 140.186.89.2 ; > } ; > ... > But when I restart the server, I get: > MultiNet Server: GATED: parse: > MULTINET:GATEWAY-DAEMON.CONFIGURATION:19 gateway not a host address on > an attached network: '140.186.89.2' You don't need any such static route; it is illegal because you list the destination of the route as your local IP adress. Destinations of routes need to be machines that you are directly connected to. Ken ================================================================================