Archive-Date: Mon, 01 Feb 1993 00:04:19 CST Sender: list-mgr@WKUVX1.BITNET Date: Mon, 01 Feb 1993 00:04:09 CST From: "Hunter Goatley, WKU" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <0096774C.DBE77F80.5682@WKUVX1.BITNET> Subject: MX-LIST Administrivia: Monthly Post Last modified: 2-JUL-1992 00:40 01:26 (Created) Welcome to MX-List@WKUVX1.BITNET, an electronic mailing list established for the discussion of the Message Exchange mail software. This is a routine posting you will see from time to time on MX-List. The MX-List archives are maintained at ARCHIVES@WKUVX1.BITNET. To get a copy of any month's postings, send an e-mail message with the body SEND MX-List.yyyy-mm to ARCHIVES@WKUVX1.BITNET, where "yyyy" is the year and "mm" is the numeric representation of the month. For example, the message SENDME MX-List.1992-04 will send the archives for April 1992. To remove yourself from the mailing list, send the following command to LISTSERV@WKUVX1.BITNET: SIGNOFF MX-List LISTSERV supports a few other commands for your convenience. The following commands can be handled automatically by the list processor: SIGNOFF MX-List - to remove yourself from the list REVIEW MX-List - to get a list of subscribers QUERY MX-List - to get the status of your entry on the list SET MX-List NOMAIL - to remain on the list but not receive mail SET MX-List MAIL - to resume receiving mail from the list SET MX-List CONCEAL - to not report your address in a REVIEW SET MX-List NOCONCEAL - to report your address in a REVIEW SET MX-List REPRO - to receive posts you make to MX-List SET MX-List NOREPRO - to not receive posts you make to MX-List LIST - to get a list of mailing lists served by WKUVX1 HELP - to receive a help file By default, subscriptions are set to MAIL, REPRO, NOCONCEAL. If you have any questions, comments, or suggestions about MX-List, please contact the list owner at the address below. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Hunter Goatley, VAX Systems Programmer goathunter@WKUVX1.BITNET Western Kentucky University Academic Computing, STH 226 (502) 745-5251 Bowling Green, KY 42101 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- ================================================================================ Archive-Date: Mon, 01 Feb 1993 10:50:53 CST Sender: list-mgr@WKUVX1.BITNET Date: Mon, 01 Feb 1993 09:21:33 EST From: "Brian Tillman, Smiths Industries, VAX Support, x8425" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <0096779A.BA5E2780.11706@swdev.si.com> Subject: Re: MX and Alpha machines Dan Wing (dwing@uh01.colorado.edu) asks: >Has anyone VESTed MX? I tried. It won't VEST. Rather, one image in MX (I think it's MX_MAILSHR, but I don't remember) won't VEST because of operating system dependencies. -----------------------------+-------------------------------- Brian Tillman | Internet: tillman@swdev.si.com Smiths Industries, Inc. | tillman_brian@si.com 4141 Eastern Ave., MS129 | Hey, I said this stuff myself. Grand Rapids, MI 49518-8727 | My company has no part in it. -----------------------------+-------------------------------- ================================================================================ Archive-Date: Mon, 01 Feb 1993 10:51:14 CST Sender: list-mgr@WKUVX1.BITNET Date: Mon, 01 Feb 1993 09:27:36 EST From: "Brian Tillman, Smiths Industries, VAX Support, x8425" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <0096779B.92AAC620.11718@swdev.si.com> Subject: On VESTing MX I said in an earlier message that MX wouldn't VEST because of one image. That image is MX_MAILSHRP.EXE (not MX_MAILSHR, as I said before). Here's why: $ vest mx_mailshrp.exe %VEST-F-LNKSYS, Image is linked against OpenVMS VAX and references symbols in it -- not translatable %VEST-F-TRANSFATAL, Translation was impossible -----------------------------+-------------------------------- Brian Tillman | Internet: tillman@swdev.si.com Smiths Industries, Inc. | tillman_brian@si.com 4141 Eastern Ave., MS129 | Hey, I said this stuff myself. Grand Rapids, MI 49518-8727 | My company has no part in it. -----------------------------+-------------------------------- ================================================================================ Archive-Date: Mon, 01 Feb 1993 10:55:43 CST Sender: list-mgr@WKUVX1.BITNET Date: Mon, 01 Feb 1993 10:55:11 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <009677A7.CE5DDCA0.19840@WKUVX1.BITNET> Subject: RE: On VESTing MX "Brian Tillman, Smiths Industries, VAX Support, x8425" writes: > >I said in an earlier message that MX wouldn't VEST because of one image. That >image is MX_MAILSHRP.EXE (not MX_MAILSHR, as I said before). Here's why: > >$ vest mx_mailshrp.exe >%VEST-F-LNKSYS, Image is linked against OpenVMS VAX and references symbols in it > -- not translatable >%VEST-F-TRANSFATAL, Translation was impossible > Yes, MX_MAILSHRP is linked against the system symbol table. If you can give me just 2--3 more weeks, I hope to have MX compiled and running on Alpha.... Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Mon, 01 Feb 1993 14:05:23 CST Sender: list-mgr@WKUVX1.BITNET From: goathunter@WKUVX1.BITNET Reply-To: MX-List@WKUVX1.BITNET Subject: Re: MX and Alpha machines Message-ID: <1993Jan31.100658.5339@wkuvx1.bitnet> Date: 31 Jan 93 10:06:58 CDT To: MX-List@WKUVX1.BITNET In article <01GU3GO71QQA0000FK@VAXF.COLORADO.EDU>, Dan Wing writes: > Javier Henderson, jav@crash.cts.com, writes: > >>Has anyone tried MX on Alpha machines? We'll be getting our systems next >>week, and I'm considering installing this package. > > MX is written in Bliss, and DEC has said the Bliss compiler won't be ported to > Alpha. However, DEC is releasing Rdb on Alpha, and Rdb is written almost > entirely in Bliss (well, as least that's what I've heard). > > I've also heard a rumor that Bliss is going to be made available on Alpha, but > maybe not for customers.... > As far as I know, there is no native BLISS compiler for AXP, but there is (was?) an AXP BLISS compiler in the porting toolkit. That's what I'll be using to port MX. I've tried various avenues to try to convince DEC to *bundle* BLISS with VMS, but they've pretty much ignored me. > Has anyone VESTed MX? > It *will* compile under AXP---Matt Madison has already had parts of it running on an Alpha box. I don't think much had to be changed (all of the BLISS code I ported to Alpha for the DECUS AXP startup CD was easier to port than the C code I ported---just one reason of many that I wish DEC would bundle BLISS with VMS). It's primarily a matter of my finishing up the v3.2 changes/enhancements and then taking a few days to compile and test MX under Alpha (I think). Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 03 Feb 1993 10:24:56 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 03 Feb 1993 11:08:45 EST From: "Steve Thompson, Cheme System Mangler" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <0096793C.08466A73.9089@cheme.cornell.edu> Subject: MX on a disk with quotas? Has anybody installed MX (specifically the queue files) on a disk on which disk quotas are enabled? Any gotchas? The MX documentation appears to state, by apparently having no mention of this, that the only disk quota required is that of the owner of the MX files. (ie there is no possibility of users sending mail requiring quotas on the MX queue device). Steve ================================================================================ Archive-Date: Wed, 03 Feb 1993 13:01:10 CST Sender: list-mgr@WKUVX1.BITNET Subject: Re: MX on a disk with quotas? Message-ID: <1993Feb3.174842.1114@news.arc.nasa.gov> From: Date: Wed, 3 Feb 1993 17:48:42 GMT Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET In article <0096793C.08466A73.9089@cheme.cornell.edu>, "Steve Thompson, Cheme System Mangler" writes: >Has anybody installed MX (specifically the queue files) on a disk on >which disk quotas are enabled? Any gotchas? The MX documentation appears >to state, by apparently having no mention of this, that the only disk >quota required is that of the owner of the MX files. (ie there is no >possibility of users sending mail requiring quotas on the MX queue device). That's because everything that writes in the queue directory has SYSPRV turned on when it does, so all the files inherit the parent directory's ownership. Just make sure that you assign plenty of diskquota to [SYSTEM] (or whoever owns the directory) to handle all the messages that will be stored there (plus whatever else it owns on the disk). -Matt -- Matthew Madison | madison@tgv.com | +1 408 427 4366 TGV, Inc. | 603 Mission Street | Santa Cruz, CA 95060 USA ================================================================================ Archive-Date: Thu, 04 Feb 1993 03:36:54 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: Error-codes returned from SITE-Interface Message-ID: <1993Jan26.173010.1302@aragorn.unibe.ch> Reply-To: MX-List@WKUVX1.BITNET Date: Tue, 26 Jan 1993 17:30:10 GMT To: MX-List@WKUVX1.BITNET Hi all I'm using the SITE-Interface of MX to deliver mails to another mailsystem. If a message will be rejected from the SITE-interface, then it returns a FATAL-messagecode - which is included in the reject message. What i would prefer is that the reject message tells the exact reason, ie "No such user" or "ambigiuous username" - But MX seems to ignore my own defined messages (which should be valid and are loaded in the MX_STARTUP with a "SET MESSAGE messagefile") If I enable the MX_SITE_DEBUG logical I get the following trace: 26-JAN-1993 18:06:00.16 DELIVER_FILE: Subprocess says: $! 26-JAN-1993 18:06:00.16 DELIVER_FILE: Subprocess says: $ int_receive MX_ROOT :[SITE]SITE_MSG_FDE99F60_0096732C_21A08E7B.TMP;1 MX_ROOT:[SITE]SITE_ADR_FDE99F6 0_0096732C_21A08E7B.TMP;1 26-JAN-1993 18:06:01.88 DELIVER_FILE: Subprocess says: %DM2-F-AMBUSER, Ambig iuous Username 26-JAN-1993 18:06:01.97 DELIVER_FILE: Subprocess says: WRITE SYS$OUTPUT F$FA O("EXIT:!XL",F$INTEGER($STATUS)) 26-JAN-1993 18:06:01.98 DELIVER_FILE: Subprocess says: EXIT:18018034 which tells me that the message (%IRE-F-AMBUSER) is known to the process... but the reject just looks like this: ... Note: this message was generated automatically. The following error(s) occurred during local delivery of your message. Error delivering to user "": Message number 08018034 Message follows. ... which is useless for the sender. (Who should he know this number?) I would prefer getting something like Error delivering to user "": %DM2-F-AMBUSER, Ambigiuous Username Message follows. Do I need to relink the MX to include my Message-codes ? Greetings, Martin Winter ---------------------------------------------------------------------------- Radio Suisse Ltd, Internet: winter@vision.rs.ch Laupenstrasse 18a postmaster@vision.rs.ch 3008 Bern PSI-Mail: +22846431062::WINTER (public X.25) Switzerland X.400: C=CH; A=ARCOM; P=RS; O=DM; S=WINTER, G=MARTIN ---------------------------------------------------------------------------- ================================================================================ Archive-Date: Thu, 04 Feb 1993 03:36:59 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: Re: Invalid "To" header Message-ID: <1993Jan29.074744.23370@aragorn.unibe.ch> Reply-To: MX-List@WKUVX1.BITNET Date: Fri, 29 Jan 1993 07:47:44 GMT To: MX-List@WKUVX1.BITNET In article <1ka1m0INNphm@gap.caltech.edu>, carl@SOL1.GPS.CALTECH.EDU (Carl J Lydick) writes: >In article <4589@winnie.fit.edu>, karin@zach.fit.edu (Karin Nicholson) writes: > >>Another question I have: Is it at all possible to run MX on a non-UCX >>TCP/IP, e.g. Process Software's TCP/IP? > >MX also works with MULTInet and CMU-TEK TCP/IP. I don't know about Process >Software's product. Yes, it does run with Process TCPware. This product is supported by NETLIB. MX run together over TCPware without any problems. Martin ******************************************************************************* Martin Egger, Ph.D., Computing Services - Head of System/User Support Group University of Bern, P.O. Box, Laenggassstrasse 51, CH-3012 Bern, Switzerland Phone: ++41 (0)31 65 38 45, Fax: ++41 (0)31 65 38 65, Telex: 753 112 unib ch RFC: egger@id.unibe.ch, X.400: S=egger;OU=id;O=unibe;P=switch;A=arcom;C=ch; HEPNET/SPAN: 20579::49202::egger, DECnet (Switzerland): 49202::egger ******************************************************************************* ================================================================================ Archive-Date: Thu, 04 Feb 1993 06:51:39 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 04 Feb 1993 06:51:24 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: winter@vision.rs.ch Message-ID: <009679E1.3FAE7A00.20723@WKUVX1.BITNET> Subject: RE: Error-codes returned from SITE-Interface writes: > >I'm using the SITE-Interface of MX to deliver mails to another mailsystem. >If a message will be rejected from the SITE-interface, then it returns >a FATAL-messagecode - which is included in the reject message. >What i would prefer is that the reject message tells the exact reason, >ie "No such user" or "ambigiuous username" - But MX seems to ignore >my own defined messages (which should be valid and are loaded in >the MX_STARTUP with a "SET MESSAGE messagefile") > For the MX_SITE_DEBUG stuff, that's because the command that's sent to the SITE subprocess is: $ WRITE SYS$OUTPUT 'F$FAO("EXIT:!XL",F$INTEGER($STATUS))' >Note: this message was generated automatically. > >The following error(s) occurred during local delivery of your message. > >Error delivering to user "": Message number 08018034 >Message follows. >... > [...] >Do I need to relink the MX to include my Message-codes ? > I haven't tried this, but I think that it would work if you added the SET MESSAGE command to the SITE startup in MX_EXE:MX_START.COM. Just putting it in MX_STARTUP.COM isn't going to define the messages for the individual processes. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Thu, 04 Feb 1993 15:38:31 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 04 Feb 1993 16:26:14 GMT From: David Candlin Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00967A31.8D15CA00.3635@v2.ph.ed.ac.uk> Subject: Name completion Edinburgh, 4-FEB-1993 Four weeks ago I sent the appended query to the list, and had a few helpful replies, which did not, however, resolve the problem. I have now had time to investigate further, and produce a diagnosis, but not a cure. The module DOM_EXPANSION_CMU included in the kit uses the GTH_NAMADR function of NET$GTHST (of CMUIP) to return not only an address, which it ignores, but also an expanded name in the GH$NLNAMSTR field, which can be used for mail. This is fine, provided that the MX-record in the nameserver out there has a mail-domain name which points to an identical host name, because DOM_EXPANSION_CMU wants a host name, not a mail-domain name, if it is to work. But we have MX domain names which point to a particular host in a cluster, in general not identical. [Below I retain part of the data included in my first enquiry, which shows this] Unfortunately the GTH_RRLOOK function of NET$GTHST does NOT return the expanded mail-domain name, so we cannot just slightly modify DOM_EXPANSION_CMU to solve the problem. Any suggestions? David Candlin --------------------------------------------------------------------------- Edinburgh, 8-JAN-1993 I'm a newcomer, and have just installed MX 3.1c over CMUIP 6.6. Name completion for outgoing smtp isn't working; what am I doing wrong? ---------------------------------------------------------------------------- edphv4$ ipncp IPNCP> rr th/mx # of RRs = [00000005] MX: Pref = [0000000A] Name = [aida.ph.ed.ac.uk] MX: Pref = [0000000D] Name = [castle.ed.ac.uk] MX: Pref = [0000000E] Name = [festival.ed.ac.uk] MX: Pref = [00000014] Name = [nsfnet-relay.ac.uk] MX: Pref = [0000001E] Name = [sun2.nsfnet-relay.ac.uk] IPNCP> rr th.ph.ed.ac.uk/mx # of RRs = [00000005] MX: Pref = [0000000A] Name = [aida.ph.ed.ac.uk] MX: Pref = [0000000D] Name = [castle.ed.ac.uk] MX: Pref = [0000000E] Name = [festival.ed.ac.uk] MX: Pref = [00000014] Name = [nsfnet-relay.ac.uk] MX: Pref = [0000001E] Name = [sun2.nsfnet-relay.ac.uk] ---------------------------------------------------------------------------- ================================================================================ Archive-Date: Thu, 04 Feb 1993 19:58:34 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: Re: Name completion Message-ID: <1993Feb5.001534.19703@news.arc.nasa.gov> Reply-To: MX-List@WKUVX1.BITNET Date: Fri, 5 Feb 1993 00:15:34 GMT To: MX-List@WKUVX1.BITNET In article <00967A31.8D15CA00.3635@v2.ph.ed.ac.uk>, David Candlin writes: >Unfortunately the GTH_RRLOOK function of NET$GTHST does NOT return the >expanded mail-domain name, so we cannot just slightly modify >DOM_EXPANSION_CMU to solve the problem. The information is not easy to get, at least in the general case (various TCP/IP packages and various DNS setups at various sites), which is why the default domain_expansion module works the way it does. What you could do is something like: if the name does not contain a '.' then tack on '.ph.ed.ac.uk' (or whichever domain suffix) to the name check if there's an MX or A RR for that expanded name if so, return the expanded name otherwise, continue {keep the original code here} You could customize it further if you wish. -Matt -- Matthew Madison | madison@tgv.com | +1 408 427 4366 TGV, Inc. | 603 Mission Street | Santa Cruz, CA 95060 USA ================================================================================ Archive-Date: Thu, 04 Feb 1993 21:44:38 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: simple(?) rewrite rule Message-ID: <1993Feb4.111024.1893@elcsci.com> Date: 4 Feb 93 19:10:24 GMT To: MX-List@WKUVX1.BITNET I'm having trouble with what seems should be a simple rewrite rule. We have recently received a registered domain name and I'm trying to use MX to route incomming messages to our various internal nodes via SMTP. Many of these internal nodes have not been updated to recognize there own address as node.site.com. In the past we just used the node name since everything was internal. To save time in getting this going (and since I have no control over most of these machines), I thought I'd use a rewrite rule to change 'user@node.site.com' to just 'user@node'. However I haven't gotten the rewrite rule to be recognized. Below is a router log and copy of the rewrite rules I've tried. I seem to be missing something obvious here. Configuration file: MX_DEVICE:[MX031]MX_CONFIG.MXCFG;15 MX version id is: MX V3.1C Address-rewriting rules: Rewrite "<{USER}@{NODE}.ELCSCI.COM>" => "<{USER}@{NODE}>" Rewrite "<{user}@esipdx.elcsci.com>" => "<{user}@esipdx>" Rewrite "<{user}@xyzzy>" => "<{user}@tenor>" 4-FEB-1993 10:49:09.79 %PROCESS, Processing entry number 1113 4-FEB-1993 10:49:09.99 %PROCESS, Status from READ_INFO was 00000001 4-FEB-1993 10:49:09.99 %PROCESS, Message originated in VMS Mail. 4-FEB-1993 10:49:10.00 %PROCESS, will run domain expander on envelope addresses. 4-FEB-1993 10:49:10.00 %PROCESS, Processing address: 4-FEB-1993 10:49:10.00 %PROCESS, ... address now reads: 4-FEB-1993 10:49:10.00 %PROCESS, will run domain expander on message headers. 4-FEB-1993 10:49:10.05 %PROCESS, ... for expanded TENOR to tenor 4-FEB-1993 10:49:10.78 %PROCESS, Finished VMSmail-origin preprocessing. 4-FEB-1993 10:49:10.78 %PROCESS, Recipient #0: 4-FEB-1993 10:49:10.81 %REWRITE, No rewrite rules matched 4-FEB-1993 10:49:10.81 %FINDPATH, Site-spec expand on xyzzy err=00000870 4-FEB-1993 10:49:10.81 %FINDPATH, domain name XYZZY matched path pattern * 4-FEB-1993 10:49:10.81 %PROCESS, Rewrote as - next hop qiclab, path 4 4-FEB-1993 10:49:10.88 %PROCESS, Adding to UUCP path: . 4-FEB-1993 10:49:11.29 %PROCESS, Path UUCP gets 1 rcpts, entry number 1114 ================================================================================ Archive-Date: Fri, 05 Feb 1993 10:51:12 CST Sender: list-mgr@WKUVX1.BITNET Date: 05 Feb 1993 11:38:05 -0500 (EST) From: "Jonathan E. Hardis" Reply-To: MX-List@WKUVX1.BITNET Subject: RE: simple(?) rewrite rule To: MX-List@WKUVX1.BITNET CC: hardis@garnet.nist.gov Message-ID: <00967AD2.76D6AE20.9542@garnet.nist.gov> > I'm having trouble with what seems should be a simple rewrite rule. > Address-rewriting rules: > Rewrite "<{USER}@{NODE}.ELCSCI.COM>" => "<{USER}@{NODE}>" > Rewrite "<{user}@esipdx.elcsci.com>" => "<{user}@esipdx>" > Rewrite "<{user}@xyzzy>" => "<{user}@tenor>" The second rewrite rule is superfluous since anything that would match it already matched the first rule. Remember, the third rule won't be applied to any address that matches rule #1. ================================================================================ Archive-Date: Sun, 07 Feb 1993 15:56:06 CST Sender: list-mgr@WKUVX1.BITNET Date: Sun, 7 Feb 1993 15:34:48 -0500 (EST) From: J_CERNY@UNHH.UNH.EDU Reply-To: MX-List@WKUVX1.BITNET Message-ID: <930207153448.2240072f@UNHH.UNH.EDU> Subject: info on current status of MX needed X-MX-Warning: Warning -- Invalid "To" header. To: mx-list@wkuvx1.bitnet@mitvma.mit.edu <@mitvma.mit.edu:mx-list@wkuvx1.bitnet> Hi, Matthew Madison tells me that he doesn't do MX anymore now that he has moved to TGV. I used MX 3.1 with UCX 1.3 with great satisfaction and then tried running the new UCX 2.0 without MX. I know, chuckle, chuckle. After several months of farting around with the bugs in UCX's SMTP software, and a series of patches from DEC that don't fix it, I'm ready to go back to MX. So, does MX work with UCX 2.0 and, if so, where is the current source for the code? Still at RPI? Jim Cerny, Computing and Information Services, Univ. N.H. ================================================================================ Archive-Date: Sun, 07 Feb 1993 17:21:18 CST Sender: list-mgr@WKUVX1.BITNET Date: Sun, 07 Feb 1993 16:21:04 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: J_CERNY@UNHH.UNH.EDU Message-ID: <00967C8C.53A53500.21758@WKUVX1.BITNET> Subject: RE: info on current status of MX needed J_CERNY@UNHH.UNH.EDU writes: > >Matthew Madison tells me that he doesn't do MX anymore now that >he has moved to TGV. I used MX 3.1 with UCX 1.3 with great >satisfaction and then tried running the new UCX 2.0 without >MX. I know, chuckle, chuckle. After several months of farting >around with the bugs in UCX's SMTP software, and a series of >patches from DEC that don't fix it, I'm ready to go back to >MX. > >So, does MX work with UCX 2.0 and, if so, where is the current >source for the code? Still at RPI? > Yes, MX works with UCX 2.0. The current version (MX v3.1C) is available via anonymous ftp from ftp.spc.edu in [.MX]. You can also get it via e-mail by sending the following commands in the body of a mail message to FILESERV@WKUVX1.BITNET: SEND MX031 SEND FILESERV_TOOLS MX V3.2 should be going into field-test this week, assuming I have the time to work on it that I'm planning to. 8-) An announcement will be made here when it's ready. Hunter, MX maintainer ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Mon, 08 Feb 1993 05:23:18 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: RE: info on current status of MX needed Message-ID: <1993Feb8.033834.1@sejnet.sunet.se> Reply-To: MX-List@WKUVX1.BITNET Date: Mon, 8 Feb 1993 03:38:34 GMT To: MX-List@WKUVX1.BITNET As a reminder, BITNET users can obtain a copy of MX in VMSDUMP format from the VMS Store at SEARN, which mirrors the MX directories of FTP.SPC.EDU. The syntax would be: $ send listserv@searn get [mx.mx032]*.* (substitute MX031 for a non-empty directory ;-), and INDEX [MX...]*.* for a list of all MX-related files). The VMS Store has T1 connectivity to the US, and the traffic is strongly asymmetric: until the phone companies offer lines which are 3 times faster in one direction, we'll have huge amounts of free bandwidth in the Europe->US direction which you are welcome to use if your preferred US server is down or backlogged or whatever. Eric ================================================================================ Archive-Date: Mon, 08 Feb 1993 20:07:37 CST Sender: list-mgr@WKUVX1.BITNET Date: Mon, 08 Feb 1993 17:56:20 PST From: Ed Bates Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <00967D62.CD0F5700.18837@vms1.cc.uop.edu> Subject: %SYSTEM-F-REJECT I have recently (today) upgraded our CMU-OpenVMS-IP software to Multinet, and, in the process, have reinstalled MX. After installing Multinet (v3.2), I re-installed MX 3.1 and then re-installed REVC. Everything appears to be working fine now, except some older messages in the MX queue. These messages keep requeuing themselves with the message: %SYSTEM-F-REJECT, connect to network object rejected They are all LOCAL to SMTP entries. I have attempted to READY both the SMTP and LOCAL parts, but they remain in the queue displaying the message above. Has someone dealt with this problem before? I would be grateful for any suggestions. -- Ed - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Edwin J. (Ed) Bates VAX Administrator/Postmaster . . _ _ Technical Support Specialist Internet: ebates@uop.edu | | / \ | \ Office of Computing Services AppleLink: U1441 | | | | |_/ University of the Pacific Telephone: (209) 946-2251 | | | | | Stockton, CA 95211 Fax: (209) 946-2898 \_/ \_/ | %MAIL-W-NOFNCYQTE, no fancy quote or joke found in signature file ================================================================================ Archive-Date: Tue, 09 Feb 1993 13:10:29 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: Re: %SYSTEM-F-REJECT Message-ID: <1993Feb9.182154.9334@news.arc.nasa.gov> Reply-To: MX-List@WKUVX1.BITNET Date: Tue, 9 Feb 1993 18:21:54 GMT To: MX-List@WKUVX1.BITNET In article <00967D62.CD0F5700.18837@vms1.cc.uop.edu>, Ed Bates writes: >Everything appears to be working fine now, except some older >messages in the MX queue. These messages keep requeuing >themselves with the message: > > %SYSTEM-F-REJECT, connect to network object rejected > >They are all LOCAL to SMTP entries. I have attempted to >READY both the SMTP and LOCAL parts, but they remain in the >queue displaying the message above. It means that when MX tried to connect to the remote host, it got back a "connection refused" message, indicating that no SMTP server was running there. -Matt -- Matthew Madison | madison@tgv.com | +1 408 427 4366 TGV, Inc. | 603 Mission Street | Santa Cruz, CA 95060 USA ================================================================================ Archive-Date: Tue, 09 Feb 1993 18:38:36 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: RE: Error-codes returned from SITE-Interface Message-ID: <1993Feb9.153011.4303@aragorn.unibe.ch> Reply-To: MX-List@WKUVX1.BITNET Date: Tue, 9 Feb 1993 15:30:11 GMT To: MX-List@WKUVX1.BITNET In article <009679E1.3FAE7A00.20723@WKUVX1.BITNET>, Hunter Goatley writes: > writes: >> >>I'm using the SITE-Interface of MX to deliver mails to another mailsystem. >>If a message will be rejected from the SITE-interface, then it returns >>a FATAL-messagecode - which is included in the reject message. >>What i would prefer is that the reject message tells the exact reason, >>ie "No such user" or "ambigiuous username" - But MX seems to ignore >>my own defined messages (which should be valid and are loaded in >>the MX_STARTUP with a "SET MESSAGE messagefile") >> >For the MX_SITE_DEBUG stuff, that's because the command that's sent to >the SITE subprocess is: > > $ WRITE SYS$OUTPUT 'F$FAO("EXIT:!XL",F$INTEGER($STATUS))' > >>Note: this message was generated automatically. >> >>The following error(s) occurred during local delivery of your message. >> >>Error delivering to user "": Message number 08018034 >>Message follows. >>... >> >[...] >>Do I need to relink the MX to include my Message-codes ? >> >I haven't tried this, but I think that it would work if you added the >SET MESSAGE command to the SITE startup in MX_EXE:MX_START.COM. Just >putting it in MX_STARTUP.COM isn't going to define the messages for >the individual processes. > I finally got it. It seemed that I was on the right way - but the SET MESSAGE is process-specific and all the MX-processes are DETACHED processes. The MX_START.COM starts all this detached processes, so it won't help changing it there. But (my buig luck) the SITE-process is a Command-procedure and adding the SET MESSAGE there does the trick... Thanks for your help... It got me into the right direction. Martin Winter ---------------------------------------------------------------------------- Radio Suisse Ltd, Internet: winter@vision.rs.ch Laupenstrasse 18a postmaster@vision.rs.ch 3008 Bern PSI-Mail: +22846431062::WINTER (public X.25) Switzerland X.400: C=CH; A=ARCOM; P=RS; O=DM; S=WINTER, G=MARTIN ---------------------------------------------------------------------------- ================================================================================ Archive-Date: Tue, 09 Feb 1993 21:45:12 CST Sender: list-mgr@WKUVX1.BITNET Subject: Re: %SYSTEM-F-REJECT Message-ID: <18398@umd5.umd.edu> From: bleau@umdsp.umd.edu Date: 9 Feb 93 20:11:31 GMT Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET In article <00967D62.CD0F5700.18837@vms1.cc.uop.edu>, ebates@vms1.cc.uop.edu (Ed Bates) writes: >I have recently (today) upgraded our CMU-OpenVMS-IP software >to Multinet, and, in the process, have reinstalled MX. After >installing Multinet (v3.2), I re-installed MX 3.1 and then >re-installed REVC. > >Everything appears to be working fine now, except some older >messages in the MX queue. These messages keep requeuing >themselves with the message: > > %SYSTEM-F-REJECT, connect to network object rejected > >They are all LOCAL to SMTP entries. I have attempted to >READY both the SMTP and LOCAL parts, but they remain in the >queue displaying the message above. > >Has someone dealt with this problem before? I would be >grateful for any suggestions. > While I haven't dealt with the specific problem you describe (can't connect to SMTP), I have received an error message like that while trying to connect to the FTP object/service. Did you create separate acounts for each of the UCX services? If not, this answer probably won't apply. I asked UCX to create its accounts, which went nicely. The problem came when they were used. I only experienced the problem described below with ftp, but then I didn't use the UCX SMTP stuff. The ftp account/service creates a log file of some type in its login directory. The problem came about because I created the accounts on a device (the system device) that had disk quotas enabled. What the UCX installation failed to do - and I also, forgot to do - is to create a disk quota entry for account UCX$FTP. So, while UCX could start up a process under UCX$FTP, the login failed due to the inability to create a log file. This resulted in the message you reported being seen. If these conditions apply to you (multiple UCX accounts, quotas enabled on login device), then there's a good chance this is what is wrong. Another time I had the above message (also with ftp) is when I had rerun the UCX$CONFIG procedure. Not wanting to answer too many questions, I responded NO to the "Configure TCP/IP services?" prompt. I though it would keep the previous services intact. WRONG! It *disabled* all my services. To be more specific, UCX did not enable them as part of its startup (see UCX$INET_SET_INTERFACES.COM in SYS$SPECIFIC:[SYSMGR]). In this case rerun UCX$CONFIG and enable the services. It'll add several UCX ENABLE SERVICE commands to the file. If it's not one of these causes, I'm out of ideas. Do an ACC/FULL, though, and check the exit status of the errant processes. That may give you a clue. Also look at any log files that get created. >-- Ed > >- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > Edwin J. (Ed) Bates VAX Administrator/Postmaster . . _ _ > Technical Support Specialist Internet: ebates@uop.edu | | / \ | \ > Office of Computing Services AppleLink: U1441 | | | | |_/ > University of the Pacific Telephone: (209) 946-2251 | | | | | > Stockton, CA 95211 Fax: (209) 946-2898 \_/ \_/ | > > >%MAIL-W-NOFNCYQTE, no fancy quote or joke found in signature file Larry Bleau University of Maryland bleau@umdsp.umd.edu 301-405-6223 ================================================================================ Archive-Date: Tue, 09 Feb 1993 21:53:21 CST Sender: list-mgr@WKUVX1.BITNET Date: Tue, 09 Feb 1993 13:01:04 PST From: Ed Bates Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00967E02.B83C6AC0.19159@vms1.cc.uop.edu> Subject: RE: %SYSTEM-F-REJECT Thank you for your responses: carl@SOL1.GPS.CALTECH.EDU ake@dayton.saic.com (Earle Ake) Hunter Goatley madison@TGV.COM (Matt Madison) I failed to mention it, but I had looked at the SMTP logs, and they didn't tell me anything new. It appears that Matt had the right idea, so here's his explanation... - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Date: Tue, 9 Feb 93 10:22:01 PST From: madison@TGV.COM (Matt Madison) Message-ID: <930209102201.28800941@TGV.COM> Subject: Re: %SYSTEM-F-REJECT To: ebates@vms1.cc.uop.edu X-ST-Vmsmail-To: ST%"ebates@vms1.cc.uop.edu" In article <00967D62.CD0F5700.18837@vms1.cc.uop.edu>, Ed Bates writes: >Everything appears to be working fine now, except some older >messages in the MX queue. These messages keep requeuing >themselves with the message: > > %SYSTEM-F-REJECT, connect to network object rejected > >They are all LOCAL to SMTP entries. I have attempted to >READY both the SMTP and LOCAL parts, but they remain in the >queue displaying the message above. It means that when MX tried to connect to the remote host, it got back a "connection refused" message, indicating that no SMTP server was running there. -Matt -- Matthew Madison | madison@tgv.com | +1 408 427 4366 TGV, Inc. | 603 Mission Street | Santa Cruz, CA 95060 USA - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - As it turns out, the SMTP server on our router was non- functional. I received messages from local users last night about other programs experiencing this problem. When I spoke with router's administrator, he mentioned that SMTP was indeed down on that computer. -- Ed - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Edwin J. (Ed) Bates VAX Administrator/Postmaster . . _ _ Technical Support Specialist Internet: ebates@uop.edu | | / \ | \ Office of Computing Services AppleLink: U1441 | | | | |_/ University of the Pacific Telephone: (209) 946-2251 | | | | | Stockton, CA 95211 Fax: (209) 946-2898 \_/ \_/ | %MAIL-W-NOFNCYQTE, no fancy quote or joke found in signature file ================================================================================ Archive-Date: Wed, 10 Feb 1993 05:31:45 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 10 Feb 1993 05:29:13 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00967E8C.C28ADCC0.22575@WKUVX1.BITNET> Subject: RE: Error-codes returned from SITE-Interface writes: > >>I haven't tried this, but I think that it would work if you added the >>SET MESSAGE command to the SITE startup in MX_EXE:MX_START.COM. Just >>putting it in MX_STARTUP.COM isn't going to define the messages for >>the individual processes. >> >I finally got it. It seemed that I was on the right way - but the SET MESSAGE >is process-specific and all the MX-processes are DETACHED processes. The >MX_START.COM starts all this detached processes, so it won't help changing >it there. But (my buig luck) the SITE-process is a Command-procedure and >adding the SET MESSAGE there does the trick... > Duh. I see that now. When I first looked at it, it looked like the MX_START was the input file for the detached processes, which doesn't make sense now that I look at it again. Sorry. >Thanks for your help... It got me into the right direction. > Good! Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 10 Feb 1993 06:01:20 CST Sender: list-mgr@WKUVX1.BITNET Subject: SMTP problem with MX 3.1C and CMUTEK 6.6-5A Message-ID: <1993Feb10.081619.873@actrix.gen.nz> From: Clive.Nicolson@bbs.actrix.gen.nz Reply-To: MX-List@WKUVX1.BITNET Date: Wed, 10 Feb 1993 08:16:19 GMT To: MX-List@WKUVX1.BITNET I am using MX 3.1C and CMUTEK 6.6-5A with VMS 5.5 = I am trying to reply to a mail from: UserZ@nzomis.enet.dec.com SMTP fails to deliver the mail, and does not try any of the lower priority mail gateways for nzomis.enet.dec.com .Did it try any of them??. = I am also running a DNS on my system: $ domain/lookup Lookup> set/server=baby.bedroom.gen.nz/type=mx/full Lookup> lookup nzomis.enet.dec.com Response from BABY.BEDROOM.GEN.NZ: Query 1, response, standard query, recursion desired, recursion available, no error, 1 queries, 3 answers, 0 name servers, 3 additional Query: NZOMIS.ENET.DEC.COM MX IN Answer: NZOMIS.ENET.DEC.COM MX IN 7200 0 enet-gw.PA.DEC.COM NZOMIS.ENET.DEC.COM MX IN 7200 1 CRL.DEC.COM NZOMIS.ENET.DEC.COM MX IN 7200 2 inet-gw-2.PA.DEC.COM Additional: enet-gw.PA.DEC.COM A IN 86400 16.1.240.15 CRL.DEC.COM A IN 2592000 192.58.206.2 inet-gw-2.PA.DEC.COM A IN 86400 16.1.0.23 Response time: 5.140 seconds; Times sent: 2 = enet-gw.PA.DEC.COM is unreachable via IP from my system $ telnet enet-gw.PA.DEC.COM %TELNET-I-TRYING, Trying to connect to ENET-GW.PA.DEC.COM %TELNET-W-NO_OPEN, Can't open connection to ENET-GW.PA.DEC.COM -IPACP-E-URC, Destination Unreachable %TELNET-W-NO_OPEN, Can't open connection to !AS = I can reach both CRL.DEC.COM and inet-gw-2.PA.DEC.COM = $ run mx_root:[exe]mailqueue Entry: 773, Origin: [Local] Status: IN-PROGRESS SMTP entry #774, status: READY Waiting for retry until: 10-FEB-1993 19:54:22.28 Recipient #1: , Route=nzomis.enet.dec.com Error count=1 Last error: %SYSTEM-F-IVADDR, invalid media address (=%X00000134) = Turning on SMTP debugging gives: 10-FEB-1993 01:04:53.56 Processing queue entry number 697 on node BABY 10-FEB-1993 01:04:55.12 Recipient: , route=nzomis.enet.dec.com 10-FEB-1993 01:04:55.13 SMTP_SEND: looking up host name nzomis.enet.dec.com 10-FEB-1993 01:04:57.35 SMTP_SEND: Failed, sts=00000134 10-FEB-1993 01:04:57.36 SMTP send failed, sts=0C27804A, sts2=00000134 10-FEB-1993 01:04:57.36 Recipient status=0C27804A for 10-FEB-1993 01:04:59.09 1 rcpts need retry, next try 10-FEB-1993 01:34:59.09 10-FEB-1993 01:04:59.13 *** End of processing pass *** = By using the following form of address I can get a mail back to UserZ@... UserZ%nzomis.enet.dec.com@inet-gw-2.PA.DEC.COM = Not knowing what the division of labour is between MX and CMUTEK when SMTP is involved, I dont know where to start looking. Or is this just beyond the current versions of these packages?. Clive. ================================================================================ Archive-Date: Thu, 11 Feb 1993 02:08:14 CST Sender: list-mgr@WKUVX1.BITNET From: millerh@CCVAX4.CCS.CSUS.EDU Subject: Re: SMTP problem with MX 3.1C and CMUTEK 6.6-5A Message-ID: <1993Feb11.032202.6784@csus.edu> Reply-To: MX-List@WKUVX1.BITNET Date: Thu, 11 Feb 1993 03:22:02 GMT To: MX-List@WKUVX1.BITNET In article <1993Feb10.081619.873@actrix.gen.nz>, Clive.Nicolson@bbs.actrix.gen.nz writes: > >I am using MX 3.1C and CMUTEK 6.6-5A with VMS 5.5 >= It is not clear from your post - is this a new problem that has just started since the installation of the new software, or can it be confirmed that this exact scenerio worked using the older code versions? >I am trying to reply to a mail from: > >UserZ@nzomis.enet.dec.com > >SMTP fails to deliver the mail, and does not try any of the lower >priority mail gateways for nzomis.enet.dec.com .Did it try any of them??. >= >I am also running a DNS on my system: > >$ domain/lookup >Lookup> set/server=baby.bedroom.gen.nz/type=mx/full >Lookup> lookup nzomis.enet.dec.com >Response from BABY.BEDROOM.GEN.NZ: > Query 1, response, standard query, recursion desired, recursion available, > no error, 1 queries, 3 answers, 0 name servers, 3 additional > Query: > NZOMIS.ENET.DEC.COM MX IN > Answer: > NZOMIS.ENET.DEC.COM MX IN 7200 0 enet-gw.PA.DEC.COM > NZOMIS.ENET.DEC.COM MX IN 7200 1 CRL.DEC.COM > NZOMIS.ENET.DEC.COM MX IN 7200 2 inet-gw-2.PA.DEC.COM > Additional: > enet-gw.PA.DEC.COM A IN 86400 16.1.240.15 > CRL.DEC.COM A IN 2592000 192.58.206.2 > inet-gw-2.PA.DEC.COM A IN 86400 16.1.0.23 >Response time: 5.140 seconds; Times sent: 2 >= >enet-gw.PA.DEC.COM is unreachable via IP from my system > >$ telnet enet-gw.PA.DEC.COM >%TELNET-I-TRYING, Trying to connect to ENET-GW.PA.DEC.COM >%TELNET-W-NO_OPEN, Can't open connection to ENET-GW.PA.DEC.COM >-IPACP-E-URC, Destination Unreachable >%TELNET-W-NO_OPEN, Can't open connection to !AS >= >I can reach both > CRL.DEC.COM >and > inet-gw-2.PA.DEC.COM >= >$ run mx_root:[exe]mailqueue >Entry: 773, Origin: [Local] > Status: IN-PROGRESS > SMTP entry #774, status: READY > Waiting for retry until: 10-FEB-1993 19:54:22.28 > Recipient #1: , Route=nzomis.enet.dec.com > Error count=1 > Last error: %SYSTEM-F-IVADDR, invalid media address (=%X00000134) >= Don't know where the heck this is coming from. The error message indicates that this is the result of an invalid disk address for rean or write, or a hardware error. It does not say if this can be a generic hardware error, or only a disk error. Are you having any hardware problems? Due to the nature of the error, I can see my MX cashed it in. >Turning on SMTP debugging gives: > >10-FEB-1993 01:04:53.56 Processing queue entry number 697 on node BABY >10-FEB-1993 01:04:55.12 Recipient: , > route=nzomis.enet.dec.com >10-FEB-1993 01:04:55.13 SMTP_SEND: looking up host name nzomis.enet.dec.com >10-FEB-1993 01:04:57.35 SMTP_SEND: Failed, sts=00000134 >10-FEB-1993 01:04:57.36 SMTP send failed, sts=0C27804A, sts2=00000134 >10-FEB-1993 01:04:57.36 Recipient status=0C27804A for > >10-FEB-1993 01:04:59.09 1 rcpts need retry, next try 10-FEB-1993 01:34:59.09 >10-FEB-1993 01:04:59.13 *** End of processing pass *** >= >By using the following form of address I can get a mail back to UserZ@... > >UserZ%nzomis.enet.dec.com@inet-gw-2.PA.DEC.COM >= >Not knowing what the division of labour is between MX and CMUTEK when >SMTP is involved, I dont know where to start looking. Or is this just beyond >the current versions of these packages?. > > There should be no problem with the addressing working. The problem is probably elsewhere. Did you reboot the system after bringing 6.6-5A up? ================================================================================ Archive-Date: Fri, 12 Feb 1993 15:06:37 CST Sender: list-mgr@WKUVX1.BITNET From: meregalli@cesi.it Reply-To: MX-List@WKUVX1.BITNET Subject: posting test from cesi Message-ID: <1993Feb10.110200.52@cesi.it> Date: 10 Feb 93 11:02:00 +0100 To: MX-List@WKUVX1.BITNET this is a posting test ------------------------------------------------------------------ Alberto Meregalli, SIC tel. +39 2 2125 249 Centro Elettrotecnico Sperimentale Italiano fax +39 2 2125 520 Via Rubattino, 54 - I 21034 Milano Internet: meregalli@cesi.it UUCP: ...i2unix!cesi!meregalli ================================================================================ Archive-Date: Fri, 12 Feb 1993 20:30:56 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: MX with JNET and A1-mailbus? Message-ID: <1993Feb12.174139.16712@wega.rz.uni-ulm.de> Date: Fri, 12 Feb 1993 17:41:39 GMT To: MX-List@WKUVX1.BITNET Here comes another beginners question: My system manager ordered my to ask competent people (you! :-) ) about the interaction of MX, JNET and A1-mailbus. Our current mail configuration for VMS is: VAX-11 mail with an ultrix gateway for smtp mail via decnet It looks: From: NETWAY::"test@rzmain.rz.uni-ulm.de" To: rzmain::ORAKEL Subj: smtp where NETWAY:: is our ultrix smtp/decnet gateway. we receive bitnet-mail via jnet: From: Jnet%"TEST@DULRUU51" To: JNET%"orakel@dulruu51" Subj: bitnet Now, we want to migrate to x.400. We are using A1-mail/mailbus system. Our problem is, how to integrate bitnet mail into A1, because jnet will not cooperate with A1-mail. For smtp-mail we have found a hook in ultrix sendmail.cf I suggested to use MX (with VAX-mail), but my system manager will only look at MX, if it supports x.400/A1-mail What can I tell him? :-) (We are running VMS 5.5) thanx in advance \ Ulli 'Framstag' Horlacher \ psi%(0262)45050260103::orakel \ \ Student consultant VAX/VMS and networks \ orakel@dulruu51.bitnet \ \ Rechenzentrum Universitaet Ulm Germany \ orakel@rz.uni-ulm.de \ \ "With mainframes you just can do nonsense, with PCs not even that." \ ================================================================================ Archive-Date: Sat, 13 Feb 1993 07:23:16 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: mx memory display Message-ID: <8741@lib.tmc.edu> Date: 11 Feb 1993 00:26:11 GMT Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET I have been using MX for some time now and I recently installed MX 3.1C along with CMUIP 6.6-5A. This is on a 6510 and VMS 5.4-2. When I run IPNCP I find the following when only 2 connections plus the MX listen on port 25 are shown as active. Anyone have an idea of what to look for and what are these items? netstat/mem Item Alloc Free Overrun Max overrun QBlks 64 0 302 311 Uarg blks 20 16 0 0 Control segs 0 0 0 0 Max size segs 64 2 301 301 Net recv bufs 30 29 0 0 IPNCP> exi Also, in MX, using MCP queue show, I find a few zero length queue entries. Looking in [.queue] the .SRC_INFO for most of these have some IP address not in my domain. There is no other info in the file. The .HDR_INFO and .MSG_TEXT files are zero length. It doesn't happen much, as I have about 10 entries like this, and the message numbers are now over 5000. I suppose 10 out of 5000 is no big deal, but I am curious. _________________________________________________________________________ Lawrence Newton | Z999012@mdacc.mda.uth.tmc.edu Mail Box 167 | LNEWT@mdacc.mda.uth.tmc.edu University of Texas | BITNET Z999012@UTMDAH Z999012@UTHVM1 M.D. Anderson Cancer Center | THENET MDACC::Z999012 1515 Holcombe Houston, Texas 77030 | Voice (713) 792-6344 "Maybe we can eventually make language a complete impediment to understanding." Hobbes of Calvin and Hobbes ------------------------------------------------------------------------ ================================================================================ Archive-Date: Sun, 14 Feb 1993 06:29:39 CST Sender: list-mgr@WKUVX1.BITNET From: millerh@CCVAX4.CCS.CSUS.EDU Subject: Re: mx memory display Message-ID: <00968181.F129A140@CCVAX4.CCS.CSUS.EDU> Reply-To: MX-List@WKUVX1.BITNET Date: Sun, 14 Feb 1993 07:44:53 GMT To: MX-List@WKUVX1.BITNET In article <8741@lib.tmc.edu>, z999012@MDACC.MDA.UTH.TMC.EDU (Lawrence Newton) writes: > > I have been using MX for some time now and I recently installed MX 3.1C >along with CMUIP 6.6-5A. This is on a 6510 and VMS 5.4-2. > > When I run IPNCP I find the following when only 2 connections >plus the MX listen on port 25 are shown as active. Anyone have an idea >of what to look for and what are these items? > >netstat/mem >Item Alloc Free Overrun Max overrun >QBlks 64 0 302 311 >Uarg blks 20 16 0 0 >Control segs 0 0 0 0 >Max size segs 64 2 301 301 >Net recv bufs 30 29 0 0 >IPNCP> exi > What IPNCP is telling you here is about the number of times IPACP went to one of it's "look-aside" lists for a memory resource, and the pre-allocated list had no more free entries. Contrary to folklore, this is not the amount of overdraft. I would suggest that you increase the number of Queue Blocks and Max Sized Segments to about 96 each, and continue to monitor the situation. You may also have to increase the /BUFFER_LIMIT parameter in the IP_STARTUP file to a larger value, say 131072 (128K). However, if IPACP has not been crashing, nor have connections been refused, then you are probably OK for the time being. But you may want to adjust these parameters as a preventative measure. The memory allocation issues will be resolved in a fure release of IPACP. > Also, in MX, using MCP queue show, I find a few zero length queue >entries. Looking in [.queue] the .SRC_INFO for most of these have some >IP address not in my domain. There is no other info in the file. >The .HDR_INFO and .MSG_TEXT files are zero length. It doesn't >happen much, as I have about 10 entries like this, and the message numbers >are now over 5000. I suppose 10 out of 5000 is no big deal, but I am >curious. > _________________________________________________________________________ > Lawrence Newton | Z999012@mdacc.mda.uth.tmc.edu > Mail Box 167 | LNEWT@mdacc.mda.uth.tmc.edu > University of Texas | BITNET Z999012@UTMDAH Z999012@UTHVM1 > M.D. Anderson Cancer Center | THENET MDACC::Z999012 > 1515 Holcombe Houston, Texas 77030 | Voice (713) 792-6344 > "Maybe we can eventually make language a > complete impediment to understanding." Hobbes of Calvin and Hobbes > ------------------------------------------------------------------------ > ================================================================================ Archive-Date: Tue, 16 Feb 1993 01:23:59 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: Problem with MLists and queues Message-ID: <1993Feb16.162518.1@uqvax.cc.uq.oz.au> Date: Tue, 16 Feb 1993 06:25:18 GMT To: MX-List@WKUVX1.BITNET Greetings one and all, I've recently installed (for the first time) MX Mail 3.1C, on a VAXstation 4000/60 running VMS 5.5-2. I've struck a couple of problems, which I was hoping some kind soul might have the answers to. One pertains to mailing list management, and the other to queue management. (1) In the log below, I have a mailing list (NET_MGT) set up, owned by the postmaster, who also happens to be a system user. (For the purposes of what I'm doing, I've made the SYSTEM account the postmaster for the time being). Using the example name conversion module in the programming manual, I've been able to get username <--> alias conversion going with minimal effort, so that SYSTEM is translated to postmaster and vice versa. In the log, it can be seen that if SYSTEM attempts to ADD user J.Bloggs to the mailing list, the debug log (apparently) shows that SYSTEM/postmaster can do this, but the operation simply does not seem to be performed. On the other hand, SYSTEM/postmaster can quite happily SUBSCRIBE to the list. (In both cases access is granted as OWNER, but the MLIST_ADD operation only occurs for the SUBSCRIBE). I did have a browse of some source code, but never having seen BLISS before, I suspect that there is really minimal benefit in doing this unless I get up to speed on this language first. Is there anything I have left out or misconfigured? (Is there more relevent information I should supply?) $ mcp MCP> sho list net_mgt Mailing lists: Name: NET_MGT Owner: "Postmaster@CC.UQ.OZ.AU" Reply-to: List, Sender Archive: DSKC:[ARCHIVE.MLIST.NET_MGT] Add message: MX_MLIST_DIR:MLIST_ADD_MESSAGE.TXT Remove message: MX_MLIST_DIR:MLIST_REMOVE_MESSAGE.TXT Forwarded-to-owner message: MX_MLIST_DIR:MLIST_FORWARD_MESSAGE.TXT Description: Net Management Contact List Errors-to: Postmaster@CC.UQ.OZ.AU Protection: (SYSTEM:RWED,OWNER:RWED,GROUP:W,WORLD:W) MCP> review net_mgt MCP> sho system_users System (privileged) users: "Postmaster@CC.UQ.OZ.AU" MCP> ^Z $ mail MAIL> send To: mx%"net_mgt-request@CC.UQ.OZ.AU" Subj: ADD NET_MGT J.Bloggs@CC.UQ.OZ.AU Enter your message below. Press CTRL/Z when complete, or CTRL/C to quit: ADD NET_MGT J.Bloggs@CC.UQ.OZ.AU ^Z New mail on node YOBBO from YOBBO::SYSTEM "SYSTEM on YOBBO" MAIL> ^Z $ mcp MCP> que sho /all /full Entry: 349, Origin: [Local] Status: FINISHED, size: 35 bytes Created: 16-FEB-1993 14:58:49.57, expires 18-MAR-1993 14:58:49.57 Last modified 16-FEB-1993 14:59:23.76 Recipient #1: MCP> review net_mgt MCP> ^Z $ set def MX_ROOT:[MLF] $ dir *.log; Directory MX_ROOT:[MLF] FILESERV_LOG.LOG;1 0 11-FEB-1993 17:15:45.94 (RWED,RWED,RE,) MX_MLF_YOBBO.LOG;39 0 16-FEB-1993 13:47:03.42 (RWED,RWED,RE,) MX_MLF_LOG.LOG;1 2 16-FEB-1993 14:59:23.18 (RWED,RWED,RE,) Total of 3 files, 2 blocks. $ ty MX_MLF_LOG.LOG;1 16-FEB-1993 14:59:23.31 Processing queue entry number 350 16-FEB-1993 14:59:23.65 Checking local name: net_mgt-request 16-FEB-1993 14:59:23.65 This is a list control address. 16-FEB-1993 14:59:23.67 LIST_CTRL_REQ: Message is from: Postmaster@CC.UQ.OZ.AU 16-FEB-1993 14:59:23.75 LIST_CTRL_REQ: command is: ADD NET_MGT J.Bloggs@CC.UQ.OZ.AU 16-FEB-1993 14:59:23.75 CHECK_ACCESS: checking Postmaster@CC.UQ.OZ.AU for access mask=00000010 16-FEB-1993 14:59:23.75 CHECK_ACCESS: Found address on owner list. 16-FEB-1993 14:59:23.75 CHECK_ACCESS: -- access granted under OWNER class. 16-FEB-1993 14:59:23.80 All done with this entry. $ mail MAIL> send To: mx%"net_mgt-request@CC.UQ.OZ.AU" Subj: SUBSCRIBE NET_MGT Enter your message below. Press CTRL/Z when complete, or CTRL/C to quit: SUBSCRIBE NET_MGT ^Z New mail on node YOBBO from YOBBO::SYSTEM "SYSTEM on YOBBO" MAIL> ^Z $ dir *.log Directory MX_ROOT:[MLF] FILESERV_LOG.LOG;1 0 11-FEB-1993 17:15:45.94 (RWED,RWED,RE,) MX_MLF_YOBBO.LOG;39 0 16-FEB-1993 13:47:03.42 (RWED,RWED,RE,) MX_MLF_LOG.LOG;2 2 16-FEB-1993 15:01:40.67 (RWED,RWED,RE,) MX_MLF_LOG.LOG;1 2 16-FEB-1993 14:59:23.18 (RWED,RWED,RE,) Total of 4 files, 4 blocks. $ ty MX_MLF_LOG.LOG;2 16-FEB-1993 15:01:40.80 Processing queue entry number 352 16-FEB-1993 15:01:41.28 Checking local name: net_mgt-request 16-FEB-1993 15:01:41.28 This is a list control address. 16-FEB-1993 15:01:41.30 LIST_CTRL_REQ: Message is from: Postmaster@CC.UQ.OZ.AU 16-FEB-1993 15:01:41.36 LIST_CTRL_REQ: command is: SUBSCRIBE NET_MGT 16-FEB-1993 15:01:41.36 CHECK_ACCESS: checking Postmaster@CC.UQ.OZ.AU for access mask=00000004 16-FEB-1993 15:01:41.36 CHECK_ACCESS: Found address on owner list. 16-FEB-1993 15:01:41.36 CHECK_ACCESS: -- access granted under OWNER class. 16-FEB-1993 15:01:41.37 MLIST_ADD: Adding Postmaster@CC.UQ.OZ.AU (Name=SYSTEM on YOBBO) to list. 16-FEB-1993 15:01:41.37 MLIST_ADD: successfully added. 16-FEB-1993 15:01:43.10 All done with this entry. $ mcp MCP> review net_mgt SYSTEM on YOBBO MCP> ^Z (2) Quite often since this package was installed (about a week ago) situations seem to occur where a message seems to be "frozen" in an IN-PROGRESS state - that is, the message will stay on the queue indefinitely in this state, and hence effectively prevent any other subsequent messages from being sent. Typically, these "frozen" messages will be from the postmaster (as in the example below) - the destination has varied in each case. If I simply cancel the "frozen" message, subsequent messages (i.e. >= 354) will still not be sent, even though they are in the READY state. I have tried every MCP queue oriented command I can think of, and have reset the retry interval for all channels to one second (from the original one hour, as can be seen in the example below), all without success. Shutting down MX Mail and restarting it will still not do the trick - the problem is preserved. However, shutting down the entire machine and restarting it *will* get the messages delivered. Needless to say, however, this is not my preferred course of action in these situations! :-) I'm unaware of any way to manually trigger delivery, as is possible with various command files in packages like PMDF. I had thought that possibly sending a message when the queue was in this state might trigger delivery, but alas it doesn't. So...as in the first question, is there something I'm missing, or is there something I've misconfigured? YOBBO> mcp MCP> que sho /all /full Entry: 349, Origin: [Local] Status: FINISHED, size: 35 bytes Created: 16-FEB-1993 14:58:49.57, expires 18-MAR-1993 14:58:49.57 Last modified 16-FEB-1993 14:59:23.76 Recipient #1: Entry: 351, Origin: [Local] Status: FINISHED, size: 18 bytes Created: 16-FEB-1993 15:01:16.30, expires 18-MAR-1993 15:01:16.30 Last modified 16-FEB-1993 15:01:43.06 Recipient #1: Entry: 353, Origin: [Local] Status: IN-PROGRESS, size: 189 bytes Created: 16-FEB-1993 15:01:41.42, expires 18-MAR-1993 15:01:41.42 Last modified 16-FEB-1993 15:01:42.31 Recipient #1: MCP> sho local LOCAL agent settings: DECnet delivery retry interval: 0 00:00:01.00 Maximum number of retries: 96 Accounting enabled. Top headers: FROM,SENDER,TO,CC,BCC,SUBJECT,ENCRYPTED,DATE Bottom headers: (none) Thank you for your time and help. Cheers, Rob. ----------------------+------------------------------------------------------- Rob McMillan | E-mail: R.McMillan@cc.uq.oz.au Systems Programmer | Phone: +61 07 365 4290 FAX: +61 07 365 4477 The Prentice Centre | Post: University of Queensland, QLD 4072, AUSTRALIA ================================================================================ Archive-Date: Tue, 16 Feb 1993 08:03:30 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: Re: simple(?) rewrite rule Message-ID: <1993Feb16.045818.1223@dmc.com> Date: 16 Feb 93 04:58:18 EST To: MX-List@WKUVX1.BITNET In article <1993Feb4.111024.1893@elcsci.com>, lewisc@elcsci.com (Chuck Lewis) writes: > I'm having trouble with what seems should be a simple rewrite rule. > We have recently received a registered domain name and I'm trying to > use MX to route incomming messages to our various internal nodes via > SMTP. Many of these internal nodes have not been updated to recognize > there own address as node.site.com. In the past we just used the node > name since everything was internal. To save time in getting this going > (and since I have no control over most of these machines), I thought > I'd use a rewrite rule to change 'user@node.site.com' to just 'user@node'. > However I haven't gotten the rewrite rule to be recognized. Below is > a router log and copy of the rewrite rules I've tried. I seem to be > missing something obvious here. > Did you remember to: MCP RESET ROUTER or MCP RESET /CLUSTER after adding the rewrite rules? If so, try: $ MCP SHUT ROUTER $ @SYS$STARTUP:MX_STARTUP ROUTER Sometimes the reset doesn't "take". -- Dick Munroe Internet: munroe@dmc.com Doyle Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. ================================================================================ Archive-Date: Tue, 16 Feb 1993 12:24:20 CST Sender: list-mgr@WKUVX1.BITNET Subject: mlfake question ? Message-ID: From: Reply-To: MX-List@WKUVX1.BITNET Date: Tue, 16 Feb 1993 15:58:32 GMT Keywords: mlfake signoff global To: MX-List@WKUVX1.BITNET I want to use the MX utility MLFAKE to do a global listserv signoff when i remove a user. But the documentation is unclear about the operation of the utility. The syntax is MLFAKE listname hostname [command] But i want to send to a listserv with a command of "signoff * global". In a case like that where the command is replaced, it the listname parameter completely ignored ? I think that the command i want to use is this mlfake junk uicvm.uic.edu/listserv "signoff * global" John Andrea andrea@essex.stfx.ca ================================================================================ Archive-Date: Tue, 16 Feb 1993 12:27:14 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: Re: Problem with MLists and queues Message-ID: <1993Feb16.172803.1597@news.arc.nasa.gov> Reply-To: MX-List@WKUVX1.BITNET Date: Tue, 16 Feb 1993 17:28:03 GMT To: MX-List@WKUVX1.BITNET In article <1993Feb16.162518.1@uqvax.cc.uq.oz.au>, ccrob@uqvax.cc.uq.oz.au (Rob McMillan) writes: > In the log, it can be seen that if SYSTEM attempts to ADD user J.Bloggs > to the mailing list, the debug log (apparently) shows that > SYSTEM/postmaster can do this, but the operation simply does not seem > to be performed. On the other hand, SYSTEM/postmaster can quite happily > SUBSCRIBE to the list. (In both cases access is granted as OWNER, but > the MLIST_ADD operation only occurs for the SUBSCRIBE). Looks like a simple syntax problem. If you send to NET_MGT-Request, the ADD command should read ADD user@host You shouldn't be specifying the list name on the command. As for the "frozen" entries: >Entry: 353, Origin: [Local] > Status: IN-PROGRESS, size: 189 bytes > Created: 16-FEB-1993 15:01:41.42, expires 18-MAR-1993 15:01:41.42 > Last modified 16-FEB-1993 15:01:42.31 > Recipient #1: Try turning on debugging in the Router. Or at least check to make sure the Router process hasn't exited abnormally. To give an entry a manual "kick" to make it ready for processing again: MCP> QUEUE READY 353 -Matt -- Matthew Madison | madison@tgv.com | +1 408 427 4366 TGV, Inc. | 603 Mission Street | Santa Cruz, CA 95060 USA ================================================================================ Archive-Date: Tue, 16 Feb 1993 15:36:40 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: can't install the jnet interface Message-ID: <1993Feb16.173159.14267@ncsu.edu> Reply-To: MX-List@WKUVX1.BITNET Date: Tue, 16 Feb 1993 17:31:59 GMT To: MX-List@WKUVX1.BITNET -- I'm trying to install the mx/jnet interface and I'm running into a problem. I'm running vms 5.4-2, jnet 3.6, ucx 2.0 and mx 3.1c. During vmsinstal here's what happens: %MX-I-LINKING, Linking image MX_JNET... %LINK-W-MULDEF, symbol RECEIVE_DELVER multiply defined in module JNETMSG file $62$DUA0:[SYS0.SYSUPD.MX031]JNET.OLB;106 %LINK-W-MULDEF, symbol RECEIVE_REWIND multiply defined in module JNETMSG file $62$DUA0:[SYS0.SYSUPD.MX031]JNET.OLB;106 %LINK-W-MULDEF, symbol RECEIVE_VMERR multiply defined in module JNETMSG file $62$DUA0:[SYS0.SYSUPD.MX031]JNET.OLB;106 %LINK-W-MULDEF, symbol RECEIVE_CLOSE multiply defined in module JNETMSG file $62$DUA0:[SYS0.SYSUPD.MX031]JNET.OLB;106 %LINK-W-MULDEF, symbol RECEIVE_FORWERR multiply defined in module JNETMSG file $62$DUA0:[SYS0.SYSUPD.MX031]JNET.OLB;106 %LINK-W-MULDEF, symbol RECEIVE_DEST multiply defined in module JNETMSG file $62$DUA0:[SYS0.SYSUPD.MX031]JNET.OLB;106 %VMSINSTAL-E-INSFAIL, The installation of MX V3.1 has failed. Any advice/help/suggestions are quite appreciated... Harry -------------------------------------------------------------------------- Harry Nicholos KD4TCS |>>Not always correct, but never in doubt<< VMS Systems Programmer | Harry_Nicholos@ncsu.edu Computing Center | hmn@ncsuvax.bitnet North Carolina State University | (919) 515-5497 -------------------------------------------------------------------------- ================================================================================ Archive-Date: Wed, 17 Feb 1993 01:42:26 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 10:56 GMT From: ALI SHOKOUFANDEH Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Hi Friends, We are new in Mx enviromet, By the time we are connected to the EARN using Jnet ver 3.6 on a micro vax 3100 system, and now we are trying to install mx-Mailer for sending mail to Internet addresses. After it we started the installation of mx-mailer, we have selected the following components for installation: Base MX software Jnet interface support Mailing list/file server support Documentation example files user-contributed... After ward system have started the installation procedure , during providing Jnet support when system was about linking image mx_jnet... system prompts the following errors: %LINK-W-MULDEF, symbol RECEIVE_DELVER multiply defined in module JNETMSG file DKA300:[SYS0.SYSUPD.MX031]JNET.OLB;106 and also the same warning for the following symbols: RECEIVE_REWIND RECEIVE_VMERR RECEIVE_CLOSE RECEIVE_FORERR RECEIVE_DEST After it installation will be failed . Can anybody help us kindly in this regard? -Ali- ================================================================================ Archive-Date: Wed, 17 Feb 1993 08:16:39 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 08:16:12 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: ALI@IREARN.BITNET, hmn@ccvax1.cc.ncsu.edu Message-ID: <00968424.3F3C3D80.1182@WKUVX1.BITNET> Subject: MX Jnet installation problems ALI SHOKOUFANDEH writes: > >Hi Friends, >We are new in Mx enviromet, By the time we are connected to the >EARN using Jnet ver 3.6 on a micro vax 3100 system, and now [...] >After ward system have started the installation procedure >, during providing Jnet support when system was about >linking image mx_jnet... system prompts the following >errors: >%LINK-W-MULDEF, symbol RECEIVE_DELVER multiply defined > in module JNETMSG file DKA300:[SYS0.SYSUPD.MX031]JNET.OLB;106 > Two people reporting the same problem (sorry I didn't get back to you Harry---I was working on MX v3.2 and forgot to respond). I'm not really sure what's going. I've had no problems installing MX v3.2 with Jnet v3.6, but I have not tried installing V3.1. MX Jnet *does* include the Jnet messages (file JNETMSG.MSG from JAN_LIB:), and they're also defined in JANSHR in MX v3.6. However, I've had no problems linking the two. In fact, I just removed the JNETMSG module from JNET.OLB and successfully linked MX_JNET, so it looks like its no longer necessary. I don't have an older Jnet around anymore. Does anybody out there still have 3.4 or 3.5? If so, could you send me (via SEND/FILE/VMSDUMP) the JANSHR.EXE file from JAN_LIB:? Before leaving this out of MX v3.2, I'd like to make sure that it's safe to do so. I assume that, at one time, JANSHR did *not* include the message symbols. MX v3.2 should be ready for field testing today; if you want to give that a shot instead, let me know. Otherwise, I'd suggest the following: 1. Unpack the files in MX031.E (the MX Jnet saveset): $ BACKUP/LOG MX031.E/save [tmp]*.* 2. Delete the JNETMSG module from JNET.OLB: $ library/object/log/del=jnetmsg jnet 3. Put the kit back together again: $ backup/log [tmp]*.*; mx031.e/saveset Then try reinstalling it. That should do the trick. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 17 Feb 1993 09:29:08 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 18:19 GMT From: ALI SHOKOUFANDEH Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Hi , As Hunter told I have done the followins: 1 $backup /log mx031.e/save [tmp]*.* 2 library/object/log/del=jnetmsg jnet 3 backup/log [tmp]*.*; mx031.e/saveset Afterward I tried again the installation procedure, during the installation of mx_jnet image system promted the following message and installation failed again. here is the message: %LINK-W-NOSUCHMOD, module JNETMSG not found in library DKA300:[sys0.sysupd.mx031]JNET.OLB Then I have searched the DKA300:[sys0.sysupd] there was no [MX031] sub_dir. I realy don't know what shall I do. Is it a coflict between mx and jnet's version. -Ali- ================================================================================ Archive-Date: Wed, 17 Feb 1993 09:32:22 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 09:31:56 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: ALI@IREARN.BITNET Message-ID: <0096842E.D3DDF820.1210@WKUVX1.BITNET> Subject: MX/Jnet linking --- ooops ALI SHOKOUFANDEH writes: > >Hi , >As Hunter told I have done the followins: >1 $backup /log mx031.e/save [tmp]*.* >2 library/object/log/del=jnetmsg jnet >3 backup/log [tmp]*.*; mx031.e/saveset > >Afterward I tried again the installation procedure, during the installation >of mx_jnet image system promted the following message and installation failed >again. here is the message: >%LINK-W-NOSUCHMOD, module JNETMSG not found > in library DKA300:[sys0.sysupd.mx031]JNET.OLB > Ooops. I left out a step. Between 2 and 3, edit MX_JNET.OPT and change the line: VMI$KWD:JNET.OLB/INCLUDE=(MX_JNET,JNETMSG)/LIB to VMI$KWD:JNET.OLB/INCLUDE=(MX_JNET)/LIB Then repack. That *should* work. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 17 Feb 1993 09:51:48 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 09:51:18 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <00968431.88B62220.1214@WKUVX1.BITNET> Subject: MX V3.2 ready for field testing Well, finally, MX v3.2 is ready for field testing. If you are interested in field testing this version, please drop me a private e-mail message and I'll let you know how you can get it. I'd like to know which components you plan to install, so that we can try out everything. If you're using SMTP with a TCP package, please let me know which package you're using. I'm expecting a short test cycle---I've tested everything I can very thoroughly. Most of the v3.2 code has been running on WKUVX1 since early December, so it's been running "live" for a couple of months now. NOTE: this version of MX changes the format of the MX file queue FLQ_DIR:SYSTEM_QUEUE.FLQ_CTL. If you install MX v3.2 on a system with messages still in the READY state in the file queue, you'll be unable to go back to v3.1 after installing v3.2 (at least, not without reinstalling v3.1 from scratch). The MX installation automatically converts your queue file. Also, MX v3.2 uses 10 subdirectories under MX_FLQ_DIR: to hold the files. This was done in an attempt to help with performance problems caused when the QUEUE.DIR file grows larger than 128 blocks (which it won't with MX v3.2). There have been a lot of changes made, though many of them will not be noticable by end-users. A new utility, MXALIAS, has been added to let users maintain their own personal MX aliases for addresses. MXALIAS does *not* use logical names---it's integrated with MX_MAILSHR. MX v3.2 features: o FLQ routines rewritten in BLISS o MX_FLQ_DIR subdirectories o Local delivery error messages compatible with LISTSERV o Proper support for BSMTP continuation lines in MX Jnet. o A number of enhancements to the MLF agent. o X25_SMTP bug fixes. o MX SITE breaks the DCL command line into pieces for longer commands. o MX_MAILSHR correctly handles logicals defined as lists of logicals. o Lots, lots more! Once again, not that I'm expecting problems, *but*: if you can't afford for there to be some disruption of MX service, then field testing is probably not for you. I don't know of any problems, but that's the whole point of field testing. Don't volunteer to help test if you can't live with potential problems. Thanks for your help! Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 17 Feb 1993 10:37:55 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: Re: MX Jnet installation problems Date: 17 Feb 1993 10:27:48 -0500 Message-ID: <1ltlhkINN4oe@dayub.dayton.saic.com> To: MX-List@WKUVX1.BITNET Hunter Goatley (goathunter@WKUVX1.BITNET) wrote: : ALI SHOKOUFANDEH writes: : > : >Hi Friends, : >We are new in Mx enviromet, By the time we are connected to the : >EARN using Jnet ver 3.6 on a micro vax 3100 system, and now : [...] : >After ward system have started the installation procedure : >, during providing Jnet support when system was about : >linking image mx_jnet... system prompts the following : >errors: : >%LINK-W-MULDEF, symbol RECEIVE_DELVER multiply defined : > in module JNETMSG file DKA300:[SYS0.SYSUPD.MX031]JNET.OLB;106 : > : Two people reporting the same problem (sorry I didn't get back to you : Harry---I was working on MX v3.2 and forgot to respond). : I'm not really sure what's going. I've had no problems installing MX : v3.2 with Jnet v3.6, but I have not tried installing V3.1. Might also check any JNET* logical names. Might also edit the command that links the image and add in a /MAP qualifier. It might give you an idea where the extra symbols are coming from. Not having JNET, I can't help any further. -Earle -- Earle Ake Internet: NSI-DECnet (SPAN): 28276::ake ================================================================================ Archive-Date: Wed, 17 Feb 1993 12:25:20 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 12:11:11 EST From: Rick Westerman Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00968445.1386A740.17111@bchm1.aclcb.purdue.edu> Subject: RE: MX V3.2 ready for field testing I volunteer for the MX v3.2 test. I am running version 3.0 under VMS 5.5-1 using UCX 2.0. We have a small VAXcluster (11 machines) with all mail routed through 1 machine. Mail is solely TCP/IP based, except the local machines. In other words, a rather simple setup. -- Rick Rick Westerman System Manager of the AIDS Center Laboratory westerm@aclcb.purdue.edu for Computational Biochemistry (ACLCB), BCHM Phone: (317) 494-0505 bldg., Purdue University, W. Lafayette, IN 47907 FAX: (317) 494-7897 ================================================================================ Archive-Date: Wed, 17 Feb 1993 13:52:20 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 13:07:14 EST From: Alan Simon Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <0096844C.E7E93640.10939@shrsys.hslc.org> Subject: RE: MX V3.2 ready for field testing Hunter - I'm interested in field-testing MX 3.2. ------------------------------------------------------------------------------- Alan Simon simon@hslc.org Associate Director simon@shrsys.hslc.org Health Sciences Libraries Consortium VOICE: (215) 222-1532 3600 Market Street, Suite 550 FAX: (215) 222-0416 Philadelphia, PA 19104 ------------------------------------------------------------------------------- ================================================================================ Archive-Date: Wed, 17 Feb 1993 14:39:56 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 10:58:24 -0800 (PST) From: Phil Rand Reply-To: MX-List@WKUVX1.BITNET Subject: Port pine to VMS? Integrate with MX? User agent for MX? To: MX discussion CC: Pine Bugs Message-ID: Has anyone written a new user-agent for MX to replace the VMS mail utility on VMS. Has anyone ported pine to VMS? Or is such a porting project under way? We've been using MX (Mail Exchange) on our VMS systems for SMTP mail delivery to standard VMS mail, and have been pleased with it, but we're less pleased with the VMS mail utility, which is difficult for new users and clumsy for old hands. Simultaneously, we've been using Pine on our one Ultrix system, and love it. Its design goal of sacrificing functionality for ease of use and learning has paid off in a big way. It doesn't sacrifice that much functionality, anyway. How hard could it be to port Pine to VMS and make it the user interface through which users deal with MX? (Pretty hard, I suppose :-) Perhaps the VMS POSIX subsystem could be used? Has anybody done such a project? Is one under way? --Phil // Phil Rand prand@paul.spu.edu // Computer & Information Systems (206) 281-2428 // Seattle Pacific University, 3307 3rd Ave W, Seattle, WA 98119 ================================================================================ Archive-Date: Wed, 17 Feb 1993 14:50:21 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 17 Feb 1993 14:49:56 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: prand@paul.spu.edu Message-ID: <0096845B.403B0AE0.1569@WKUVX1.BITNET> Subject: RE: Port pine to VMS? Integrate with MX? User agent for MX? Phil Rand writes: > >Has anyone written a new user-agent for MX to replace the VMS mail utility >on VMS. > Not as far as I know. >Has anyone ported pine to VMS? Or is such a porting project under way? > I picked this up once with the intention of trying to port it (or at least determine how feasible it would be). I never got around to even unpacking the .tar file---I just have too many other things going on to really attempt it. >We've been using MX (Mail Exchange) on our VMS systems for SMTP mail >delivery to standard VMS mail, and have been pleased with it, but we're >less pleased with the VMS mail utility, which is difficult for new users >and clumsy for old hands. > >Simultaneously, we've been using Pine on our one Ultrix system, and love >it. Its design goal of sacrificing functionality for ease of use and >learning has paid off in a big way. It doesn't sacrifice that much >functionality, anyway. > >How hard could it be to port Pine to VMS and make it the user interface >through which users deal with MX? (Pretty hard, I suppose :-) > >Perhaps the VMS POSIX subsystem could be used? > >Has anybody done such a project? Is one under way? > If anybody out there wants to give it a try, I'm all for it. I'd appreciate hearing any experiences anyone may have had trying to port it.... Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 17 Feb 1993 20:47:24 CST Sender: list-mgr@WKUVX1.BITNET Subject: Re: mlfake question ? Message-ID: <1993Feb17.204528.1@sejnet.sunet.se> From: Date: Wed, 17 Feb 1993 20:45:28 GMT Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET In article , andrea@pX1.stfx.ca (John Andrea) writes: > I want to use the MX utility MLFAKE to do a global listserv signoff > when i remove a user. With MX V3.2 you won't have to do this any longer. MX will generate formatted delivery errors which LISTSERV understands and the users will be automatically signed off as the bounces come back. Whatever you do, PLEASE don't make a DCL program to MLFAKE signoff commands for 5,000 users as someone did a few years ago. LISTSERV administrators don't generally like to find their server working on a pile of thousands of deletion requests for a user which isn't subscribed to any of their lists :-) Eric ================================================================================ Archive-Date: Thu, 18 Feb 1993 21:13:33 CST Sender: list-mgr@WKUVX1.BITNET Date: 18 Feb 1993 16:09:28 -0700 (MST) From: ccvax@academic.cc.colorado.edu Subject: Distribution Lists To: MX-List@WKUVX1.BITNET Reply-To: MX-List@WKUVX1.BITNET Message-ID: <0096852F.8897B840.3919@academic.cc.colorado.edu> When I send a message to a distribution list the VMS To: header shows just the name of the list: To: @dumb However some of the users in the list have forwarding set to another machine with an MX address: MAIL> set forward To: MX%"""eric@admin""" In the messages they receive, each individual in the VMS distribution list is included in the To: header, with the fully qulified name: eric@academic.cc.colordao.edu, kmichels@academic.cc.colorado.edu, ..... for all the users on the list - around 100 names. Is there any way to get mx to report the simple VMS form - i.e. just the name of the distribution list? Thanks, Karen ********************************************************************* * Karen Michels * * Academic Computing ccvax@cc.colorado.edu (internet) * * The Colorado College ccvax%ccnode@colorado (bitnet) * * 14 E. Cache La Poudre * * Colorado Springs, CO 80903 719-389-6457 * ********************************************************************* ================================================================================ Archive-Date: Fri, 19 Feb 1993 07:10:48 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: Re: Distribution Lists Date: 19 Feb 1993 07:58:56 -0500 Message-ID: <1m2ligINNatd@dayub.dayton.saic.com> To: MX-List@WKUVX1.BITNET ccvax@academic.cc.colorado.edu wrote: : When I send a message to a distribution list the VMS To: header shows : just the name of the list: : To: @dumb : However some of the users in the list have forwarding set to another : machine with an MX address: : MAIL> set forward : To: MX%"""eric@admin""" : In the messages they receive, each individual in the VMS distribution : list is included in the To: header, with the fully qulified name: : eric@academic.cc.colordao.edu, kmichels@academic.cc.colorado.edu, : ..... for all the users on the list - around 100 names. : Is there any way to get mx to report the simple VMS form - i.e. : just the name of the distribution list? If you are using MX then why not use the mailing list idea? This way the entire list is not sent on the To: line. Only the mailing list name appears. -Earle -- Earle Ake Internet: NSI-DECnet (SPAN): 28276::ake ================================================================================ Archive-Date: Fri, 19 Feb 1993 09:45:27 CST Sender: list-mgr@WKUVX1.BITNET Date: Fri, 19 Feb 93 16:07-0500 Message-ID: <009685F8.64324C80.2661@merlin.berlin.ptb.de*> From: "Mario Meyer, Physikalisch-Technische Bundesanstalt, Inst. Berlin, Labor 10.42, Abbestr. 2-12, 1000 Berlin 10" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET CC: mmeyer@MERLIN Subject: subscribe subscribe ================================================================================ Archive-Date: Fri, 19 Feb 1993 17:28:20 CST Sender: list-mgr@WKUVX1.BITNET Date: Fri, 19 Feb 1993 15:50:23 EST From: "David B. Dierker" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET CC: dierker@wums.wustl.edu Message-ID: <009685F6.0761D720.13326@wums.wustl.edu> Subject: MX and UCX and UCX/LPD problem... I've come across a nasty(?) problem with MX and callable MAIL and UCX/LPD... I Think. Outline of problem: 1) user on unix box prints to a queue located on VMS box using LPD... 2) user wants to know when job is finished so they include the -m qualifier on the unix lp command (-m means send a mail message when done.) 3) Job queues up, prints, and when the job completes the server queue UCX$LPD_QUEUE falls over dead with an callable mail error about unable to find transport SMTP... 4) This problem is reported to system manager(ME) and then I setup the appropriate logicals such that use of the mail transport SMTP% will call the MX_MAILSHR.EXE. 5) User on unix box tries the print again for grins. 6) Job queues up, prints, and when the job completes the server queue UCX$LPD_QUEUE falls over dead with an callable mail error about unable to find file SYS$SCRATCH:[SYS0.UCX_LPD.POST_2UP]ETC.ETC* 7) This additional problem is reported to the system manager(ME) and after much fussing and fuming I define a system wide logical for SYS$SCRATCH: which points to a publicly writable directory. The problem goes away... My question? Does MX use the SYS$SCRATCH logical? If it does this is a problem since server symbionts don't run with the benefit of DCL and it looks like the SYS$SCRATCH logical isn't created for a non-DCL process. *) the name ETC.ETC isn't the real name, I just can't at this moment remember the name, and it's a temporary filename anyway. -----------------------------------------------+-------------------------------- David B. Dierker (dierker@wums.wustl.edu) |Brevity in signatures is Wash U. Med Ctr VAX Facility |a blessing. WU = +1 314 362 3354 / DSC = +1 314 432 8200 | ================================================================================ Archive-Date: Fri, 19 Feb 1993 18:26:01 CST Sender: list-mgr@WKUVX1.BITNET Subject: >mlfake? Message-ID: From: Reply-To: MX-List@WKUVX1.BITNET Date: Fri, 19 Feb 1993 20:07:46 GMT Keywords: mlfake works ? To: MX-List@WKUVX1.BITNET Thanks for the info about not sending 3000 listserv signoffs if only a fraction are needed. Good point, i'll keep it in mind. Now to answer my own question. Using mlfake for another user. The syntax is mlfake /for=username listname address/listserv "command" 'listname' is always (appearantly) appended to 'command' But does it really work. After trying one mlfake (mx v2.3) it was still running after 2 cpu minutes so i decided to abort it. Should i have just waited longer ? john andrea andrea@essex.stfx.ca ================================================================================ Archive-Date: Tue, 23 Feb 1993 11:38:30 CST Sender: list-mgr@WKUVX1.BITNET Date: Tue, 23 Feb 1993 09:22:22 -0800 (PST) From: Sheryl Rose Erez Reply-To: MX-List@WKUVX1.BITNET Subject: Re: Port pine to VMS? Integrate with MX? User agent for MX? To: Phil Rand CC: MX discussion , Pine Bugs Message-ID: Phil, A few people have talked about porting Pine to VMS, but we haven't heard of anybody actually doing the work yet. We'd be happy to provide the source code and whatever advice we can to the lucky one who wants to do the port. If somebody who reads this would like to take on the port effort, write to use at pine@cac.washington.edu. Sheryl Erez erez@cac.washington.edu UW Network Information Center On Wed, 17 Feb 1993, Phil Rand wrote: > Date: Wed, 17 Feb 1993 10:58:24 -0800 (PST) > From: Phil Rand > To: MX discussion > Cc: Pine Bugs > Subject: Port pine to VMS? Integrate with MX? User agent for MX? > > Has anyone written a new user-agent for MX to replace the VMS mail utility > on VMS. > > Has anyone ported pine to VMS? Or is such a porting project under way? > > We've been using MX (Mail Exchange) on our VMS systems for SMTP mail > delivery to standard VMS mail, and have been pleased with it, but we're > less pleased with the VMS mail utility, which is difficult for new users > and clumsy for old hands. > > Simultaneously, we've been using Pine on our one Ultrix system, and love > it. Its design goal of sacrificing functionality for ease of use and > learning has paid off in a big way. It doesn't sacrifice that much > functionality, anyway. > > How hard could it be to port Pine to VMS and make it the user interface > through which users deal with MX? (Pretty hard, I suppose :-) > > Perhaps the VMS POSIX subsystem could be used? > > Has anybody done such a project? Is one under way? > > --Phil > > // Phil Rand prand@paul.spu.edu > // Computer & Information Systems (206) 281-2428 > // Seattle Pacific University, 3307 3rd Ave W, Seattle, WA 98119 > > > ================================================================================ Archive-Date: Tue, 23 Feb 1993 13:03:27 CST Sender: list-mgr@WKUVX1.BITNET Subject: Re: simple(?) rewrite rule Message-ID: <1993Feb18.102312.1913@elcsci.com> From: Reply-To: MX-List@WKUVX1.BITNET Date: 18 Feb 93 18:23:12 GMT To: MX-List@WKUVX1.BITNET In article <1993Feb16.045818.1223@dmc.com>, munroe@dmc.com (Dick Munroe) writes: > In article <1993Feb4.111024.1893@elcsci.com>, lewisc@elcsci.com (Chuck Lewis) writes: >> I'm having trouble with what seems should be a simple rewrite rule. >> However I haven't gotten the rewrite rule to be recognized. Below is >> a router log and copy of the rewrite rules I've tried. I seem to be >> missing something obvious here. >> > > Did you remember to: > > MCP RESET ROUTER > > $ MCP SHUT ROUTER > $ @SYS$STARTUP:MX_STARTUP ROUTER > > Sometimes the reset doesn't "take". That did correct my problem! I had done the RESET command after defining the rewrite rule. It had no effect. As soon as I shutdown things and restarted them, my rewrite rules worked just fine. Thanks! ================================================================================ Archive-Date: Tue, 23 Feb 1993 13:42:16 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: MX -> MR Gateway? (was RE: HELP... uucp w/all in one) Message-ID: <1993Feb23.055314.1247@dmc.com> Date: 23 Feb 93 05:53:14 EST To: MX-List@WKUVX1.BITNET In article <00968798.B81EBF80.10261@kpy.com>, Bill Pedersen writes: > Flame on! > > And while I at it, do not believe that this communications mechanism is > the place for subtle marketing. I speak of the message from the fellow at > Innosoft, the publishers of PMDF, which he is recommending. > > While we can not keep this from happening I think that a PRIVATE message is > reasonable if you want to sell someone something but when someone is > looking for help, to suggest he buy something... > > Anyway, my thoughts. > > Flame Off... Actually, this is a good argument for putting such functionality into something that IS free, e.g., MX. The "I want to hook transport% to A1" question has come up again and again. Perhaps it's time for MX to route to MR? If there are any A1 mavins, running MX, who know BLISS want to volunteer? I'm not up on MR, but I have the last two qualifications and am more than willing to participate in a joint development venture. Dick Munroe -- Dick Munroe Internet: munroe@dmc.com Doyle Munroe Consultants, Inc. UUCP: ...uunet!thehulk!munroe 267 Cox St. Office: (508) 568-1618 Hudson, Ma. FAX: (508) 562-1133 GET CONNECTED!!! Send mail to info@dmc.com to find out about DMConnection. ================================================================================ Archive-Date: Tue, 23 Feb 1993 15:23:41 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: FAQ? or ftp sites for mx? Message-ID: <16B7EC2DC.DSTEIS01@ulkyvm.louisville.edu> Date: Tue, 23 Feb 1993 18:51:24 GMT To: MX-List@WKUVX1.BITNET I am trying to find the FAQ for this group, so that I might also be able to locate the LISTSERV or anonymous FTP site where the mx package is available. Thanks in advance ... ********************************************************************* ** Donald S. Teiser voice: (502) 588-7994 * ** Tech Support for VMS & Ultrix fax: (502) 588-8896 * ** University of Louisville Louisville, KY 40292 * ** dsteis01@ulkyvm.louisville.edu * ********************************************************************* ================================================================================ Archive-Date: Tue, 23 Feb 1993 15:54:30 CST Sender: list-mgr@WKUVX1.BITNET Date: Tue, 23 Feb 1993 13:40:29 PST From: Bill Pedersen Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: pedersen@kpy.com Message-ID: <00968908.8B1744A0.10955@kpy.com> Subject: RE: MX -> MR Gateway? (was RE: HELP... uucp w/all in one) Dick Munroe writes: >> Actually, this is a good argument for putting such functionality >> into something that IS free, e.g., MX. The "I want to hook >> transport% to A1" question has come up again and again. Perhaps >> it's time for MX to route to MR? If there are any A1 mavins, >> running MX, who know BLISS want to volunteer? I'm not up on MR, >> but I have the last two qualifications and am more than willing >> to participate in a joint development venture. Well, actually I do not think that you necessarily have to be a bliss hacker. You could get by with writing a Message Router interface, as PMDF has and then you are done... Well, maybe it is not all that easy but it can be done and there is documentation on how to do it. There is the issue of address maintenance and the ability of the of the DEC DDF to handle internet/uucp addresses but then that may be a matter of establishing a reasonable dialect. In the mean time, for those who only want a SIMPLE interface my previous message to Bob had the re-write rules for UUCP which work and therefore it should be possible to transcribe them to MX%. Anyway, I'm not volunteering at present, just commenting. __________________________________________________________________________ Bill Pedersen | KPY Network Partners (KPY Corporation) | 1037 North Fair Oaks Avneue, Sunnyvale, CA 94089-2101 | 1-408-734-2564 1-800-458-8930 FAX:1-408-734-2940 ================================================================================ Archive-Date: Tue, 23 Feb 1993 16:45:36 CST Sender: list-mgr@WKUVX1.BITNET Date: Tue, 23 Feb 1993 17:10:02 EST From: Paul Simons Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00968925.D13362A0.24479@nuconvex.com> Subject: RE: MX -> MR Gateway? > Actually, this is a good argument for putting such functionality > into something that IS free, e.g., MX. The "I want to hook > transport% to A1" question has come up again and again. Perhaps > it's time for MX to route to MR? If there are any A1 mavins, > running MX, who know BLISS want to volunteer? > Dick Munroe I am working on a gateway into MR using the site interface of MX. The design is done, the code is about 25% written. Obviously, I can only do this part time, so it's going to be a while before it's done. The time frame is sometime this year. I'm willing to share what I have with someone who can get it done faster, but that someone will have to have the Message Router Programmers Kit. By the way, I can't seem to get a signature at the end of my messages. I have the following logicals defined: (LNM$PROCESS_TABLE) "MX_AUTO_SIGNATURE" = "True" "MX_SIGNATURE" = "USER:[SIMONPG]SIGNATURE.TXT" And if I "type mx_signature", it works. I think I just figured out the problem. I'm using DECW$MAIL. Would those logicals get defined in its process? They're in login.com. ================================================================================ Archive-Date: Wed, 24 Feb 1993 05:46:12 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 93 06:38:45 -0500 Message-ID: <9302241138.AA08091@genrad.com> From: Reply-To: MX-List@WKUVX1.BITNET X-MX-Warning: Warning -- Invalid "To" header. To: "mx-list%wkuvx1.bitnet@ukcc.uky.edu" CC: Subject: RE: MX -> MR Gateway? Someone unknown wrote: >> Actually, this is a good argument for putting such functionality >> into something that IS free, e.g., MX. The "I want to hook >> transport% to A1" question has come up again and again. Perhaps >> it's time for MX to route to MR? If there are any A1 mavins, >> running MX, who know BLISS want to volunteer? > >> Dick Munroe > > I am working on a gateway into MR using the site interface of MX. The > design is done, the code is about 25% written. Obviously, I can only do > this part time, so it's going to be a while before it's done. The time > frame is sometime this year. I'm willing to share what I have with > someone who can get it done faster, but that someone will have to have the > Message Router Programmers Kit. By the way, I can't seem to get a > signature at the end of my messages. I have the following logicals > defined: > > (LNM$PROCESS_TABLE) > > "MX_AUTO_SIGNATURE" = "True" > "MX_SIGNATURE" = "USER:[SIMONPG]SIGNATURE.TXT" > > And if I "type mx_signature", it works. I think I just figured out the > problem. I'm using DECW$MAIL. Would those logicals get defined in its > process? They're in login.com. Knowing that your sig file was not being included, would it not have been a good idea to *manually* insert some indication of who/where you are? -------------------------------------------------------------------------------- Name :Derek Dongray, Systems Manager, GenRad Ltd. Phone :061 486 1511 InterNet : Dongray@GenRad.com UKnet : Derek.Dongray@GenRad.co.uk PSS : 234261600119::Dongray CompuServe : 70374,2745 Address : Monmouth House, Monmouth Road, Cheadle Hulme, Cheshire, SK8 7AY, UK. ================================================================================ Archive-Date: Wed, 24 Feb 1993 07:34:55 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 93 13:13 GMT From: "UK TeX Archive Manager " Reply-To: MX-List@WKUVX1.BITNET To: MX-LIST%BITNET.WKUVX1@NSFNET-RELAY.AC.UK Subject: RE: MX -> MR Gateway? In message <9302241138.AA08091@genrad.com> dated Wed, 24 Feb 93 06:38:45 -0500 dongray wrote: > Someone unknown wrote: [requoting of original messages omitted...] > Knowing that your sig file was not being included, would it not have been a > good idea to *manually* insert some indication of who/where you are? What sort of mailer are you using to read your MX-List traffic? I didn't have any trouble identifying the poster of the message to which you objected; here's a portion of that message's headers, just as received by me: > Date: Tue, 23 Feb 1993 17:10:02 EST > From: Paul Simons > Reply-To: MX-List@BITNET.WKUVX1 > To: MX-List@BITNET.WKUVX1 > Message-ID: <00968925.D13362A0.24479@nuconvex.com> > Subject: RE: MX -> MR Gateway? > Sender: list-mgr%BITNET.WKUVX1@EDU.UKY.UKCC So he appears to be Paul Simons. When I first saw the message to which I'm currently replying, I immediately noticed an anomaly amongst *its* headers (just as received, preceded by MX's annotation of its complaint): > X-MX-Warning: Warning -- Invalid "To" header. > To: "mx-list%wkuvx1.bitnet@ukcc.uky.edu"@uk.ac.nsfnet-relay This *must* be an artefact of the original mailing; I don't see anything like this on anyone else's mailings to the list (although I think this is the first UK one I've seen --- this current message should generate another:-) Judging by the information in *his* signature box, I would have expected Derek Dongray's mailing, since he's registered in the .com domain (albeit located physically in the UK) to have simply gone out To: mx-list%wkuvx1.bitnet@ukcc.uky.edu directly over the Internet (and presumably using MX:-). Similarly, I would have expected his incoming mail to arrive by the same route, and hence go nowhere near nsfnet-relay. He's also registered in the .co.uk domain, this time through UKnet (and hence using one of the UKC gateways, for example ben.uknet.ac.uk). This would again use direct (SMTP) Internet delivery. The intrusion of the uk.ac.nsfnet-relay above suggests that mailings are going to him via the default Internet/Janet gateway nsfnet-relay.ac.uk. But there's no DNS MX record that suggests mailings for either of Derek's addresses should go by that route. It may be that once upon a time, his mailings *had* to go via this route (prior to the introduction of the JIPS [Internet over Janet] facility in the UK). I've tried to REVIEW MX-List, but have insufficient privilege: could it be that Derek's registration therein has an explicit nsfnet-relay.ac.uk tacked onto the end? Brian {Hamilton Kelly} System Manager for the UK TeX Archive at Aston University ================================================================================ Archive-Date: Wed, 24 Feb 1993 07:49:37 CST Sender: list-mgr@WKUVX1.BITNET Message-ID: <9302240226.AA03634@monk.proteon.com> From: John_Woodward_at_CTC@relay.proteon.com Reply-To: MX-List@WKUVX1.BITNET Date: 23 Feb 93 16:59 EST Subject: Problem running MX 2.3-1 under VMS 5.5-1 To: MX-List@RPIECSVX.BITNET CC: Bill_Horsthemke@relay.proteon.com Ladies/Gentlemen, We ran into a problem running MX 2.3-1. MX 2.3 is working fine, so consider this a bug report rather than a cry for help. Our machine is a VAX 4000-500 running VMS 5.5-1, nodename REX. We are using CMU's TCP/IP v 6.6 and configuring MX for SMTP support only. When we start MX 2.3-1, the MX SMTP process dies the first time it tries to process a message. The log w/ source line #s follows. GET_REPLY's line 650 is calling TCP_GET_LINE from library NETLIB_CMU_TCP. Since it is linked w/ trace off, I don't have line numbers from it, but TCP_GET_LINE calls something else, probably a system service. You can see that it makes a call as there are two layers unaccounted for in the stack dump, TCP_GET_LINE and whatever it called. ------------------------------------------------------------------------------- MX_SMTP_LOG.LOG ------------------------------------------------------------------------------- %SYSTEM-F-ACCVIO, access violation, reason mask=04, virtual address=00000000, PC=00053189, PSL=03C00000 %TRACE-F-TRACEBACK, symbolic stack dump follows module name routine name line rel PC abs PC 00053189 00053189 0002C8E5 0002C8E5 SMTP_OUT GET_REPLY 650 00000089 00003AD8 SMTP_OUT SMTP_SEND 281 000002F5 00003395 PROCESS PROCESS 283 00000365 000025C8 MX_SMTP MX_SMTP 31 0000020D 00001C71 ------------------------------------------------------------------------------- MX_SMTP_REX.LOG ------------------------------------------------------------------------------- 23-FEB-1993 11:46:01.66 Processing queue entry number 1337 on node REX 23-FEB-1993 11:46:01.96 Recipient: 23-FEB-1993 11:46:01.96 Identified next hop as relay 23-FEB-1993 11:46:01.96 SMTP_SEND: looking up host name relay 23-FEB-1993 11:46:01.96 SMTP_SEND: Attempting to start session with relay 23-FEB-1993 11:46:01.98 SMTP_SEND: Connected 23-FEB-1993 11:46:02.31 SMTP_SEND: Rcvd: 220-relay.aar.com SMTP service ------------------------------------------------------------------------------- Please address any replies directly to me, as I do not read the news. /John Woodward John_Woodward_at_CTC@relay.proteon.com ================================================================================ Archive-Date: Wed, 24 Feb 1993 07:50:04 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 07:49:37 CST From: "Hunter Goatley" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <009689A0.B1BB1940.3814@WKUVX1.BITNET> Subject: RE: MX -> MR Gateway? "UK TeX Archive Manager " writes: > >In message <9302241138.AA08091@genrad.com> dated Wed, 24 Feb 93 06:38:45 -0500 >dongray wrote: > >> Someone unknown wrote: >[requoting of original messages omitted...] >> Knowing that your sig file was not being included, would it not have been a >> good idea to *manually* insert some indication of who/where you are? > >What sort of mailer are you using to read your MX-List traffic? I didn't have >any trouble identifying the poster of the message to which you objected; here's >a portion of that message's headers, just as received by me: > [...] >When I first saw the message to which I'm currently replying, I immediately >noticed an anomaly amongst *its* headers (just as received, preceded by MX's >annotation of its complaint): > >> X-MX-Warning: Warning -- Invalid "To" header. >> To: "mx-list%wkuvx1.bitnet@ukcc.uky.edu"@uk.ac.nsfnet-relay > And what I saw was: >X-MX-Warning: Warning -- Invalid "To" header. >To: "mx-list%wkuvx1.bitnet@ukcc.uky.edu" Which is invalid because of the quotes. I don't still have the message around and the "Received:" lines are stripped from the archive. But it looks like the @uk.ac.nsfnet-relay may have just been added to yours. Anyone else? >He's also registered in the .co.uk domain, this time through UKnet (and hence >using one of the UKC gateways, for example ben.uknet.ac.uk). This would again >use direct (SMTP) Internet delivery. The intrusion of the uk.ac.nsfnet-relay >above suggests that mailings are going to him via the default Internet/Janet >gateway nsfnet-relay.ac.uk. But there's no DNS MX record that suggests >mailings for either of Derek's addresses should go by that route. It may be >that once upon a time, his mailings *had* to go via this route (prior to the >introduction of the JIPS [Internet over Janet] facility in the UK). I've tried >to REVIEW MX-List, but have insufficient privilege: could it be that Derek's >registration therein has an explicit nsfnet-relay.ac.uk tacked onto the end? > Derek is registered as . Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 24 Feb 1993 07:53:25 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 07:53:09 CST From: "Hunter Goatley" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: John_Woodward_at_CTC@relay.proteon.com Message-ID: <009689A1.2FDC38E0.3821@WKUVX1.BITNET> Subject: RE: Problem running MX 2.3-1 under VMS 5.5-1 John_Woodward_at_CTC@relay.proteon.com writes: > >Ladies/Gentlemen, > We ran into a problem running MX 2.3-1. MX 2.3 is working fine, so >consider this a bug report rather than a cry for help. And I don't consider it a bug since you're several versions behind on MX and NETLIB. The latest version of MX is V3.1C, available via anonymous ftp from ftp.spc.edu in [.MX]. MX v3.2 is in the final stages of field test and will be released this week, I think. > Please address any replies directly to me, as I do not read the news. > I have to side with other people here. If you can post to the group with a question, you can be bothered to read it. I am cc'ing this to you via e-mail, but.... Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 24 Feb 1993 08:11:39 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 93 08:49:45 -0500 Message-ID: <9302241349.AA10713@genrad.com> From: Reply-To: MX-List@WKUVX1.BITNET X-MX-Warning: Warning -- Invalid "To" header. To: "mx-list%wkuvx1.bitnet@ukcc.uky.edu" CC: Subject: Mail headers revisited. In reply to Brian {Hamilton Kelly}, my mx-list mail goes through a gateway machine in Concord, Massachussetts, which is a Vax6000 running Ultrix. This machine is known asd genrad.com to internet or internally as either genrad or CO6301 (decnet name). It forwards all mail to username dongray to . CDCLU1 is a VMScluster in Manchester England. I read the mail using VMSmail. The are few meaningful headers. Preesumably, CO6301 strips them. This is why I regularly complain about people who do not put their addresses in the text of the message. E.g. > From: CDCLU1::CO6301::"@UKCC.UKY.EDU:list-mgr@WKUVX1.BITNET" 24-FEB-1993 13:35:19.58 > To: NSFNET-RELAY.AC.UK::MX-LIST%BITNET.WKUVX1 > CC: > Subj: RE: MX -> MR Gateway? > > In message <9302241138.AA08091@genrad.com> dated Wed, 24 Feb 93 06:38:45 -0500 > dongray wrote: > [rest deleted] > > Brian {Hamilton Kelly} > System Manager for the > UK TeX Archive at Aston University I notice that you didn't bother to put in your E-mail address either. BTW this is the header for Simon's mail >From: CDCLU1::CO6301::"@UKCC.UKY.EDU:list-mgr@WKUVX1.BITNET" 23-FEB-1993 23:03:17.19 >To: WKUVX1.BITNET::MX-List >CC: >Subj: RE: MX -> MR Gateway? > [rest deleted] My other mail routes are via PSS into a Vax here, or via a gateway Vax750 running Ultrix, genrad.co.uk/curlew/FH7504 which knows to forward mail to Derek.Dongray to me here. It similarly strips almost all headers. I think the Nsfnet-realy.ac.uk entry must have been added after it went through MX-list. -------------------------------------------------------------------------------- Name :Derek Dongray, Systems Manager, GenRad Ltd. Phone :061 486 1511 InterNet : Dongray@GenRad.com UKnet : Derek.Dongray@GenRad.co.uk PSS : 234261600119::Dongray CompuServe : 70374,2745 Address : Monmouth House, Monmouth Road, Cheadle Hulme, Cheshire, SK8 7AY, UK. ================================================================================ Archive-Date: Wed, 24 Feb 1993 09:34:23 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 09:50:31 EST From: Paul Simons Reply-To: MX-List@WKUVX1.BITNET To: MX-list@WKUVX1.BITNET Message-ID: <009689B1.95A55C70.24537@nuconvex.com> Subject: UUCP + MX = Ugly addresses I guess by now it's obvious that I am trying to get MX configured to met the nee ds of my site. I hope that my travails will help the next person. Is there a way to help MX convert UUCP headers such as: CS.YALE.EDU!harvard.UUCP!WKUVX1.BITNET!list-mgr to instead of: WKUVX1.BITNET!list-mgr%harvard.UUCP@CS.YALE.EDU ? ---------- Paul G. Simons simonpg@nuconvex.com CONnecticut Valley Electric eXchange (CONVEX) 203.276.6435 Southington, CT, USA {yale,uunet!hsi}!nuconvex!simonpg ================================================================================ Archive-Date: Wed, 24 Feb 1993 09:37:18 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 09:38:27 EST From: Paul Simons Reply-To: MX-List@WKUVX1.BITNET To: MX-list@WKUVX1.BITNET Message-ID: <009689AF.E59FEB20.24525@nuconvex.com> Subject: Sigs are fixed. (was RE: MX -> MR Gateway?) As an aside, MX did not complain when delivering <9302241138.AA08091@genrad.com> to my site: Message-ID: <9302241138.AA08091@genrad.com> From: genrad.com!dongray%harvard.UUCP@CS.YALE.EDU Reply-To: WKUVX1.BITNET!MX-List%harvard.UUCP@CS.YALE.EDU To: "mx-list%wkuvx1.bitnet@ukcc.uky.edu".?Unknown.bulldog.CS.YALE.EDU.user@CS.YALE.E DU >> And if I "type mx_signature", it works. I think I just figured out the >> problem. I'm using DECW$MAIL. Would those logicals get defined in its >> process? They're in login.com. >Knowing that your sig file was not being included, would it not have been a >good idea to *manually* insert some indication of who/where you are? Yes, it would have been a good idea. I defined the logicals in the system table and I get my signature file. Since I am using a workstation, I find that an exceptable solution. ---------- Paul G. Simons simonpg@nuconvex.com CONnecticut Valley Electric eXchange (CONVEX) 203.276.6435 Southington, CT, USA {yale,uunet!hsi}!nuconvex!simonpg ================================================================================ Archive-Date: Wed, 24 Feb 1993 10:00:36 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 10:43:53 EST From: "Tim McKee, 803-366-2620" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <009689B9.09FF9BE0.4870@cvse.cortex.prospect.com> Subject: RE: Sigs are fixed. (was RE: MX -> MR Gateway?) Signatures work fine for me from my VS (as far as I know). I use a DEFINE/JOB to define the signature logical and have no problem! (Indeed, I do all the defines in my login.com in /JOB mode just to be safe - as well as to reduce the time spent copying logicals when spawning a subprocess. This may seem trivial, but I sometimes use a software product that places HUNDREDS of logicals in the process table - definately not a trivial copying task!) tim ================================================================================ Timothy R. McKee | Senior Consultant - VAX/VMS Internals Cortex Corp. - SouthEast Field Office | mckeetr@cvse.cortex.prospect.com Phone: (803) 366-2620 | FAX: (803) 366-6517 ================================================================================ ================================================================================ Archive-Date: Wed, 24 Feb 1993 11:07:39 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 11:33:37 EST From: "Brian Tillman, Smiths Industries, VAX Support, x8425" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <009689BF.FD1D19A0.27577@swdev.si.com> Subject: RE: Sigs are fixed. Timothy R. McKee (mckeetr@cvse.cortex.prospect.com) writes: >I use a DEFINE/JOB to define the signature logical and have >no problem! > >(Indeed, I do all the defines in my login.com in /JOB mode >just to be safe... Just be aware that you are limited by the JTQUOTA value for your account in the UAF. Also be aware that, for DECwindows, too many job logical names can adversely affect your session manager. -----------------------------+-------------------------------- Brian Tillman | Internet: tillman@swdev.si.com Smiths Industries, Inc. | tillman_brian@si.com 4141 Eastern Ave., MS129 | Hey, I said this stuff myself. Grand Rapids, MI 49518-8727 | My company has no part in it. -----------------------------+-------------------------------- ================================================================================ Archive-Date: Wed, 24 Feb 1993 14:41:46 CST Sender: list-mgr@WKUVX1.BITNET Subject: MX SMTP and Wollongong - Working! Message-ID: <1993Feb24.133946.62@vtrm01.uucp> From: Reply-To: MX-List@WKUVX1.BITNET Date: 24 Feb 93 13:39:46 EST To: MX-List@WKUVX1.BITNET Wollongong TCP/IP aka WIN/TCP for VMS aka PATHWAY..... can now be used with MX. The lastest version, that is. The Wollongong PATHWAY Runtime component now supplies UCX compatibility. I installed the MX/NETLIB package and directed it to setup it's communication using UCX 1.3+ software. I had been using MX 3.1C with the Wollongong UCX compatibility feature as a beta site and now with the officially released product. SMTP mail works just fine. However, trying to gateway mail between UUCP,VMSmail,A1 and internal TCP/IP is really an experience. Someday I hope to have it working flawlessly. ;-) -- Fred LaForest UUCP: vtrm01!flaforest VMS/UNIX/(Whatever else we have in the building) System Programmer Vickers, Inc. GEnie: FALAFOREST Troy, MI 48007-0302 ================================================================================ Archive-Date: Wed, 24 Feb 1993 14:42:23 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 93 16:05 GMT From: "UK TeX Archive Manager " Reply-To: MX-List@WKUVX1.BITNET To: MX-LIST%BITNET.WKUVX1@NSFNET-RELAY.AC.UK Subject: RE: Mail headers revisited. In message <9302241349.AA10713@genrad.com> dated Wed, 24 Feb 93 08:49:45 -0500, dongray wrote: > In reply to Brian {Hamilton Kelly}, my mx-list mail goes through a gateway > machine in Concord, Massachussetts, which is a Vax6000 running Ultrix. This > machine is known asd genrad.com to internet or internally as either genrad or > CO6301 (decnet name). It forwards all mail to username dongray to > . CDCLU1 is a VMScluster in Manchester England. I read the mai l > using VMSmail. The are few meaningful headers. Preesumably, CO6301 strips them . »»» I'd even go so far as to say that there's only one: the subject line! Seriously though, I can't see why you put up with this. If I was a user on a system where some jumped-up little administrator decided to strip out all RFC-822 header lines, I'd be going WAY over his/her head to get something done about it. After all, without the *original* date, and from line, how is one to know when a message was posted, and by whom? > This is why I regularly complain about people who do not put their addresses i n > the text of the message. Well, I can understand that now; personally, I've always felt that providing my e-mail address in a signature block is redundant, and only needlessly extending the message, to the general annoyance of 99% of the recipients. For let's face it, we *do* bitch about the miles of headers on messages, and *don't* need to look at them most of the time. But when something goes wrong, one even wants to see Received and Return-Path lines. So MX provides one with the ideal facility: shove the uninteresting header lines at the *end* of the message, and only put the really useful ones (that permit one readily to compose replies) above the message body (and following the VMSmail headers). > I notice that you didn't bother to put in your E-mail address either. OK, here goes: Brian {Hamilton Kelly} System Manager for the UK TeX Archive at Aston University Tel: +44 793 785252 P.S. That signature points up another reason for *not* putting e-mail addresses in signature blocks. You'll note that my host appears in the UK's big-endian ordering: that's because we're connected to Janet. So people in the rest of the world have to be intelligent enough to know that, if they want to mail me directly, they have to reorder the mailbox into system@tex.ac.uk. But this re-ordering takes place *automatically* when my message passes through the Janet/Internet gateway uk.ac.nsfnet-relay: all addresses are reversed by the gateway into the format appropriate to the target network. But this only happens to mailboxes that appear in the RFC-822 (or Grey-Book Mail) headers of the message: it can't (and shouldn't) happen to anything else within the body of the message --- if you (or someone else) choses to throw away this invaluable information, by discarding the headers, then ... ================================================================================ Archive-Date: Wed, 24 Feb 1993 15:36:21 CST Sender: list-mgr@WKUVX1.BITNET From: osymh%msu.oscs.montana.edu@MTSUNIX1.BITNET Reply-To: MX-List@WKUVX1.BITNET Date: Wed, 24 Feb 1993 14:27:22 MST From: "Michael L. Hitch" To: MX-List@WKUVX1.BITNET Message-ID: <009689D8.428BD540.28838@msu.oscs.montana.edu> Subject: Problem with distribution file expansion MX 3.1C has a problem expanding a distribution file when the distribution file resides on a PC. It appears that the PCSA$MAIL process is attempting to use the distribution file on VMS and will get an access violation if the file doesn't exist. When I create a dummy .DIS file on VMS, then the mail works fine (except that the information in the dummy file is used, which may not be correct). MX_MAILSHR should handle a missing distribution file a little better. ----- NETSERVER.LOG file for case where distribution file does not exist on VMS: 24-FEB-1993 11:44:44.38 To:_TREX::OSYMH From:_OSYMH::_PC_0D3E Logged on TREX as PCSA$MAIL_95 at 24-FEB-1993 11:44:47.43 $ exit $ ! Copyright (c) 1987, 1990 Digital Equipment Corporation. All rights reserved. $ IF "" .NES. "" THEN NETSERVER$COMMAND $ IF "" .EQS. "" THEN NETSERVER$VERIFY = 0 $ V = F$VERIFY(NETSERVER$VERIFY) -------------------------------------------------------- Connect request received at 24-FEB-1993 11:44:49.86 from remote process OSYMH::"0=_PC_0D3E" for object "SYS$SYSTEM:PCSA$MAIL_SERVER.EXE" -------------------------------------------------------- %SYSTEM-F-ACCVIO, access violation, reason mask=04, virtual address=00008600, PC Improperly handled condition, image exit forced. Signal arguments Stack contents Number = 00000005 00000000 Name = 0000000C 207C0000 00000004 7FE84CE8 00008600 7FE84CCC 000745ED 00008BC1 0BC00004 0000003F 7FE80041 00000000 0000000F 00000000 Register dump R0 = 04E80029 R1 = 7FE84AC8 R2 = 00000041 R3 = 000085E8 R4 = 00000000 R5 = 00000000 R6 = FFFFFFFF R7 = 7FE84FC8 R8 = 00000004 R9 = 7FE84FC0 R10= 00179694 R11= 00000000 AP = 7FE84A58 FP = 7FE84A18 SP = 7FE84A94 PC = 000745ED PSL= 0BC00004 OSYMH job terminated at 24-FEB-1993 11:46:26.85 Accounting information: Buffered I/O count: 262 Peak working set size: 1473 Direct I/O count: 228 Peak page file size: 7013 Page faults: 1852 Mounted volumes: 0 Charged CPU time: 0 00:00:03.82 Elapsed time: 0 00:01:44.36 --- Michael L. Hitch osymh@msu.oscs.montana.edu Computer Consultant OSYMH@MTSUNIX1.BITNET Office of Systems and Computing Services Montana State University Bozeman, MT USA ================================================================================ Archive-Date: Wed, 24 Feb 1993 15:56:57 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 15:56:27 CST From: "Hunter Goatley" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <009689E4.B43ACE60.4138@WKUVX1.BITNET> Subject: RE: Problem with distribution file expansion osymh%msu.oscs.montana.edu@MTSUNIX1.BITNET writes: > > MX 3.1C has a problem expanding a distribution file when the distribution >file resides on a PC. It appears that the PCSA$MAIL process is attempting to >use the distribution file on VMS and will get an access violation if the file >doesn't exist. When I create a dummy .DIS file on VMS, then the mail works >fine (except that the information in the dummy file is used, which may not be >correct). MX_MAILSHR should handle a missing distribution file a little >better. > Agreed. I'll fix this in MX v3.2. Thanks for reporting the problem. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 24 Feb 1993 15:57:46 CST Sender: list-mgr@WKUVX1.BITNET Date: Wed, 24 Feb 1993 15:57:16 CST From: "Hunter Goatley" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <009689E4.D16B0900.4142@WKUVX1.BITNET> Subject: RE: MX SMTP and Wollongong - Working! writes: > >Wollongong TCP/IP aka WIN/TCP for VMS aka PATHWAY..... can now be used >with MX. The lastest version, that is. > >The Wollongong PATHWAY Runtime component now supplies UCX compatibility. I >installed the MX/NETLIB package and directed it to setup it's communication >using UCX 1.3+ software. > >I had been using MX 3.1C with the Wollongong UCX compatibility feature as a >beta site and now with the officially released product. SMTP mail works just >fine. > Good deal! I'll add this to the MX documentation. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Wed, 24 Feb 1993 17:40:31 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: Re: MX -> MR Gateway? (was RE: HELP... uucp w/all in one) Message-ID: <1993Feb24.222539.4350@netcom.com> Date: Wed, 24 Feb 1993 22:25:39 GMT To: MX-List@WKUVX1.BITNET munroe@dmc.com (Dick Munroe) writes: >Actually, this is a good argument for putting such functionality >into something that IS free, e.g., MX. The "I want to hook >transport% to A1" question has come up again and again. Perhaps >it's time for MX to route to MR? If there are any A1 mavins, >running MX, who know BLISS want to volunteer? I'm not up on MR, >but I have the last two qualifications and am more than willing >to participate in a joint development venture. i think it is not as difficult as you suggest, dick. what i did not know is that AI1 requires an "_" in the to: address to route to xport%. ken p., who rightly railed against the blatent PDMF marking folks, was kind enough to tell me about oa$lib:special.com, which is where the secret stuff lies. mere publication of these two factoids (in the FAQ and in the UUCP DELIVER and MX documentation) might go along way to edifying us all-in-1 bozos. f you do want to do this, i can do bliss (and have a compilier :-) and _mailshr hacking, too, so count me in. -- -- -- bob pasker -- rbp@netcom.com -- ================================================================================ Archive-Date: Wed, 24 Feb 1993 18:01:52 CST Sender: list-mgr@WKUVX1.BITNET Date: 24 Feb 1993 17:20:32 -0500 (EST) From: "Jonathan E. Hardis" Reply-To: MX-List@WKUVX1.BITNET Subject: RE: FAQ? or ftp sites for mx? To: MX-List@WKUVX1.BITNET CC: DSTEIS01@ulkyvm.louisville.edu, hardis@garnet.nist.gov Message-ID: <009689F0.732D3FA0.10164@garnet.nist.gov> > in message > <16B7EC2DC.DSTEIS01@ulkyvm.louisville.edu> really wants to know: > I am trying to find the FAQ for this group, so that I might also be able > to locate the LISTSERV or anonymous FTP site where the mx package is > available. In the words of its author: Message Exchange V3.1 Message Exchange (MX) is electronic mail distribution and routing software for VMS V5.0 or later. It supports Internet mail over TCP/IP using CMU-Tek, DEC VMS/ULTRIX Connection, TGV MultiNet, and Process Software Corporation's TCPware, as well as SMTP over DECnet and X.25; BITNET mail over Jnet; also interfaces with DECUS UUCP. MX uses VMS Mail for local message entry and delivery, and includes support for mailing lists and mail-based file servers. Support is included for interfacing MX with a custom mail transport and for providing custom address handling routines. V3.1 includes the following new features over previous releases: * SMTP over X.25 is now supported, using VAX P.S.I. * The Jnet interface no long requires a special account to run. * Mailing list processor supports NOCASE, CONCEAL, and NOREPRO flags for subscriber entries. * A NAME_CONVERSION module is included in the examples save set that handles address conversions for both DECnet and MRGATE users. * MX processes now log their startups and exits to log files, and will also log to an operator class, if desired. * New STATUS command added to MCP that lists all running MX processes. * Several other minor improvements and bug fixes. MX V3.1, including installation kit, docs, and sources, is available via anonymous FTP on ftp.spc.edu, in directory [.MX]. If you do not have FTP access, you can obtain MX by requesting it via E-mail. Send the following commands in the body of a mail message to MXSERVER@WKUVX1.BITNET: SEND MX031 SEND FILESERV_TOOLS It's a huge package and, because of BITNET delays, make take a few days for all the pieces to arrive. There is no FAQ compilation, though we would be pleased if someone would start one. ================================================================================ Archive-Date: Thu, 25 Feb 1993 04:36:26 CST Sender: list-mgr@WKUVX1.BITNET Subject: Strange SET FORWARD behaviour Message-ID: <1993Feb25.090134.170@logica.co.uk> From: gerry@pluto.logica.co.uk Date: 25 Feb 93 09:01:33 GMT Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET This is a bit obscure but here goes... I'm using a feature of DEC Pathworks to get mail notification on my PC of new mail delivered to my VMS Mail account on our Pathworks server. This involves doing a SET FORWARD on my account GERRY to a forwarded address of PCSA%LI274$GERRY which will broadcast a mail notification to PC node LI274 and deliver then deliver the mail into my VMS Mail account GERRY as normal. All is well for standard mail coming and going locally and through remote DECnet connections. We are also very happy MX3.1C users, however incoming mail delivered from MX to my account ends up with the horrible To: PCSA%LI274$GERRY recipient address instead of To: GERRY that I get for local VMS Mail (and MX delivery without the SET FORWARD), or To: PLUTO::GERRY that I get for remote DECnet mail. Is there a way to have the SET FORWARD PCSA%LI274$GERRY setup *and* still have the mail recipient field generated by MX appear as just GERRY, or s this just a feature of how MX does it's LOCAL delivery? -Gerry ------------------------------------------------------------------------------- Gerard Magennis | Logica Industry Ltd, London UK | Internet: gerry@logica.co.uk "Born to drive, learned to walk later" - CA bumper sticker ================================================================================ Archive-Date: Thu, 25 Feb 1993 06:03:05 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 1993 06:02:54 CST From: "Hunter Goatley" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00968A5A.F3D969C0.4341@WKUVX1.BITNET> Subject: RE: Strange SET FORWARD behaviour gerry@pluto.logica.co.uk writes: > >I'm using a feature of DEC Pathworks to get mail notification on my PC of new >mail delivered to my VMS Mail account on our Pathworks server. > >This involves doing a SET FORWARD on my account GERRY to a forwarded address of >PCSA%LI274$GERRY which will broadcast a mail notification to PC node LI274 and >deliver then deliver the mail into my VMS Mail account GERRY as normal. > >All is well for standard mail coming and going locally and through remote >DECnet connections. We are also very happy MX3.1C users, however incoming >mail delivered from MX to my account ends up with the horrible > >To: PCSA%LI274$GERRY > [...] >Is there a way to have the SET FORWARD PCSA%LI274$GERRY setup *and* still have >the mail recipient field generated by MX appear as just GERRY, or s this just >a feature of how MX does it's LOCAL delivery? > This is apparently a problem with DELIVER too. I'll look into it for the next release of MX (the one after the upcoming v3.2). Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Thu, 25 Feb 1993 07:29:32 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 93 13:15 GMT From: "UK TeX Archive Manager " Reply-To: MX-List@WKUVX1.BITNET To: MX-LIST%BITNET.WKUVX1@NSFNET-RELAY.AC.UK Subject: RE: Strange SET FORWARD behaviour In message <1993Feb25.090134.170@logica.co.uk> dated 25 Feb 93 09:01:33 GMT gerry@pluto.logica.co.uk wrote: > This is a bit obscure but here goes... > > I'm using a feature of DEC Pathworks to get mail notification on my PC of new > mail delivered to my VMS Mail account on our Pathworks server. > > This involves doing a SET FORWARD on my account GERRY to a forwarded address o f > PCSA%LI274$GERRY which will broadcast a mail notification to PC node LI274 and > deliver then deliver the mail into my VMS Mail account GERRY as normal. > > All is well for standard mail coming and going locally and through remote > DECnet connections. We are also very happy MX3.1C users, however incoming > mail delivered from MX to my account ends up with the horrible > > To: PCSA%LI274$GERRY > > recipient address instead of > > To: GERRY > > that I get for local VMS Mail (and MX delivery without the SET FORWARD), or > > To: PLUTO::GERRY > > that I get for remote DECnet mail. > > Is there a way to have the SET FORWARD PCSA%LI274$GERRY setup *and* still have > the mail recipient field generated by MX appear as just GERRY, or s this just > a feature of how MX does it's LOCAL delivery? BINGO! I've reported an almost identical problem whilst field testing V3.2 (but never having used any earlier version of MX). I use Ned Freed's Deliver_Mailshr to post me a notification (at my home machine) of arrival of mail at this account through which I remotely (from 100 miles away) manage the Archive (in its SYSTEM aspects). I also use the "V" option of Deliver to effect privileged delivery (hence preserving whatever VMSmail headers would have been attached without the intervention of Deliver). To activate this delivery, the appropriate commands are in my SYS$LOGIN:MAIL.DELIVERY, and I have done a SET FORWARD DELIVER%SYSTEM. Incoming mail arriving locally, over DECnet from machines in Aston's setup, or over Janet (using Colour Book Software) has perfectly normal To lines; that arriving by SMTP/UUCP through MX ends up with the To address as DELIVER%SYSTEM. It would appear that MX is writing the To line as determined AFTER the forwarding, and not the "original". This is going to cause absolute and total havoc if I start to receive mail for the various mailing lists run here (which use home-grown DCL and *not* MX's MLF): at present, they work quite happily with all the lists being forwarded to DELIVER%LIST_SERVER; the latter can still tell for which list they are intended, because Deliver's ..FROM variables are set to the original addressee. So *please* Hunter, can you fix this problem, tout de suite!! Brian {Hamilton Kelly} System Manager for the UK TeX Archive at Aston University ================================================================================ Archive-Date: Thu, 25 Feb 1993 07:34:07 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 1993 07:29:04 EST From: "David B. Dierker" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET Message-ID: <00968A66.FCFC7860.15921@wums.wustl.edu> Subject: Re: MX -> MR Gateway? (was RE: HELP... uucp w/all in one) bob pasker (rbp@netcom.com) writes: >>Dick Munroe writes: >>Actually, this is a good argument for putting such functionality >>into something that IS free, e.g., MX. The "I want to hook >>transport% to A1" question has come up again and again. Perhaps >>it's time for MX to route to MR? If there are any A1 mavins, >>running MX, who know BLISS want to volunteer? I'm not up on MR, >>but I have the last two qualifications and am more than willing >>to participate in a joint development venture. > >i think it is not as difficult as you suggest, dick. > >what i did not know is that AI1 requires an "_" in the to: address to >route to xport%. ken p., who rightly railed against the blatent PDMF >marking folks, was kind enough to tell me about oa$lib:special.com, >which is where the secret stuff lies. Keep in mind that OA$LIB:special.com is only half the picture. While it is fairly easy to get ALLIN1 to allow the sending of mail just about anywhere using the "_" special.com hack. The reverse, getting mail delivered to ALLIN1 mailboxes with easy to understand, and easy to reply to addresses is another thing alltogether(sp). The only clean way to do this looks like a MR delivery agent for MX similar to MRGATE and PMDF's implementation. Not trivial. -----------------------------------------------+-------------------------------- David B. Dierker (dierker@wums.wustl.edu) |Brevity in signatures is Wash U. Med Ctr VAX Facility |a blessing. WU = +1 314 362 3354 / DSC = +1 314 432 8200 | ================================================================================ Archive-Date: Thu, 25 Feb 1993 10:57:34 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 1993 09:23:25 MST From: "Russ Wilton, Telecom Mgr" Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: wilton@hg.uleth.ca Message-ID: <00968A76.F740EE00.9503@hg.uleth.ca> Subject: RE: Strange SET FORWARD behaviour Hi: > It would appear that MX is writing the To line as determined AFTER the > forwarding, and not the "original". This may or may not be related to the above problem, but I have noticed an annoying feature of the NAME_CONVERSION facility. I do simple name conversion such that username SMITH gets translated to alias JOE_SMITH on the From: line of outgoing mail and incoming mail addresses to JOE_SMITH gets delivered to SMITH. I also do accounting for LOCAL and SMTP and the problem is that the name stored in the accounting records is always the name after translation. So the SMTP accounting file shows JOE_SMITH sending mail and the LOCAL accounting file shows SMITH receiving mail. It makes it difficult to relate the two. I would like to request that the actual username be used in both cases, so the outgoing delivery processes would have to write the accounting record before translation. This is obviously not as high a priority as the 'forwarding' problem originally reported, but if you are in there poking around anyway... :-) Russ #===============================================================# # Russell D Wilton E Mail: WILTON@HG.ULeth.CA # # Telecommunications Manager Voice: (403) 329-2525 # # Computing Services FAX: (403) 382-7108 # # University of Lethbridge # # 4401 University Drive Lethbridge, Alberta, CANADA T1K 3M4 # #===============================================================# ================================================================================ Archive-Date: Thu, 25 Feb 1993 12:31:05 CST Sender: list-mgr@WKUVX1.BITNET From: Subject: Re: UUCP + MX = Ugly addresses Message-ID: <1993Feb25.175835.1@sejnet.sunet.se> Reply-To: MX-List@WKUVX1.BITNET Date: Thu, 25 Feb 1993 17:58:35 GMT To: MX-List@WKUVX1.BITNET In article <009689B1.95A55C70.24537@nuconvex.com>, Paul Simons writes: > help MX convert UUCP headers such as: > > CS.YALE.EDU!harvard.UUCP!WKUVX1.BITNET!list-mgr > > to instead of: > > WKUVX1.BITNET!list-mgr%harvard.UUCP@CS.YALE.EDU ? And why should MX do that? Routing information has been placed in the address and, while it is usually redundant, there is no way for MX to know for a fact that trashing everything but the last token is going to result in a working address. Eric ================================================================================ Archive-Date: Thu, 25 Feb 1993 12:33:47 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 1993 12:08:59 EST From: "Brian Tillman, Smiths Industries, VAX Support, x8425" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <00968A8E.18364F80.28413@swdev.si.com> Subject: Will the mail go through if the recipient list contains an error? Suppose I send mail to two people using the "fake" .decnet domain, like this: MAIL> send To: mx%"smith@nodea.decnet",mx%"jones@nodea.decnet" Now, assume that SMITH does exist on NODEA, but JONES does not. I get an error back from Postmaster about the invalid user JONES, but does the mail still get delivered to SMITH, or does MX just cancel the whole thing? -----------------------------+-------------------------------- Brian Tillman | Internet: tillman@swdev.si.com Smiths Industries, Inc. | tillman_brian@si.com 4141 Eastern Ave., MS129 | Hey, I said this stuff myself. Grand Rapids, MI 49518-8727 | My company has no part in it. -----------------------------+-------------------------------- ================================================================================ Archive-Date: Thu, 25 Feb 1993 12:33:58 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 1993 12:23:39 EST From: "Brian Tillman, Smiths Industries, VAX Support, x8425" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <00968A90.24726980.28423@swdev.si.com> Subject: Postmaster doesn't show up in MCP QUEUE SHOW When a delivery error occurs, the sender receives a message from Postmaster (the MX delivery agent). The occurrence of the Postmaster notification appears in the message queue, but the source field is empty; that is, the source appears as <>, like this: $ mcp que show/all Entry Status Size Source Agent Entry Status Size ----- ------ ------ ------ ------- ----- ------ ------ 28411 FINISH 1489 LOCAL <> I'd like a wish added to have "Postmaster@mydomain" or simply "Postmaster" appear in the last field. Thanks. -----------------------------+-------------------------------- Brian Tillman | Internet: tillman@swdev.si.com Smiths Industries, Inc. | tillman_brian@si.com 4141 Eastern Ave., MS129 | Hey, I said this stuff myself. Grand Rapids, MI 49518-8727 | My company has no part in it. -----------------------------+-------------------------------- ================================================================================ Archive-Date: Thu, 25 Feb 1993 14:33:08 CST Sender: list-mgr@WKUVX1.BITNET Date: 25 Feb 1993 12:54:01 -0700 (MST) From: KMM@CCNODE.Colorado.EDU Reply-To: MX-List@WKUVX1.BITNET Subject: MX not delivering mail. To: MX-List@WKUVX1.BITNET Message-ID: <01GV511ICNQA003UCE@VAXF.COLORADO.EDU> Help ! I know the answer is to upgrade to MX3.1c (I was planning to this weekend). Meanwhile, running MX 2.1 on VMS 5.4-2, UCX 3.1, I am having problems getting three days of mail delivered. There are several possible problems, and I'm hoping someone can help me isolate and correct them. The problem seemed to start at the time one user pushed 300 messages (6000 bytes each) into the que at 2am. By the time I saw the problem in the morning, MX was not delivering anything and the queue file and directory were HUGE. [mx]queue.dir 580 blocks [mx.queue]system_queue.fql_ctl 13899 blocks (The system process has a diskquota of 100000 blocks on the que disk.) The SMTP server aborts within minutes of starting with the following in smpt_server.log ----------------------------------------------------------------------------- %SYSTEM-F-ACCVIO, access violation, reason mask=04, virtual address=00000072, PC =0000E1F0, PSL=03C00021 %TRACE-F-TRACEBACK, symbolic stack dump follows module name routine name line rel PC abs PC 0000E1F0 0000E1F0 00034F6B 00034F6B SMTP_SERVER PROCESS_CMD 1028 00000F57 000030A3 SMTP_SERVER SMTP_SERVER 315 0000032F 00001D93 ----------------------------------------------------------------------------- I tried CONVERT to see if the que file could be condensed - and it stayed the same size. I am now deleting the 300 messages (I can get the user to use another method), and purging them with FLQU. Questions: Is the above abort caused by the size of the queue file? Is there some quota on the SYSTEM account that may be causing the problem? (All of the processes are run by system, not a separate account.) Should I use CONVERT or CONVERT/RECLAIM to reduce the size of the queue file? Is there any way I can have MX deliver what is in the que without accepting any more messages till the mess is gone? When the SMTP server isn't running, MX_SMTP and MX_LOCAL don't appear to be delivering messages (i.e. FLQU> SHO doesn't move anything to FIN status). This may be just a timing issue, (when I look, and when MX delivers) - but I'm trying to clean this all up in the most efficient way. Thanks for any help, pointers, and suggestions (I know -- I need to upgrade to 3.1, but I have to get the mail out first.) Karen Michels kmichels@ccnode.colorado.edu (This address avoids MX - the other one won't get here) ================================================================================ Archive-Date: Thu, 25 Feb 1993 15:00:05 CST Sender: list-mgr@WKUVX1.BITNET Date: 25 Feb 1993 15:49:59 -0500 (EST) From: "Jonathan E. Hardis" Reply-To: MX-List@WKUVX1.BITNET Subject: RE: MX not delivering mail. To: MX-List@WKUVX1.BITNET CC: kmichels@ccnode.colorado.edu, hardis@garnet.nist.gov Message-ID: <00968AAC.F7AFCCE0.10242@garnet.nist.gov> > The problem seemed to start at the time one user pushed 300 messages > (6000 bytes each) into the que at 2am. By the time I saw the problem in > the morning, MX was not delivering anything and the queue file and > directory were HUGE. I had a problem like this, once upon a time. I forgot the details, but the root of the problem turned out to be that there wasn't wasn't enough contiguous disk space left to enlarge the file directory. I think I ended up backing up and restoring the disk to defragment it. In my case, I gave up trying to get MX to resend the old messages. I renamed the queue directory and let MX create a new one. Then, I dealt with the stuck messages at my leisure. - Jonathan ================================================================================ Archive-Date: Thu, 25 Feb 1993 21:21:29 CST Sender: list-mgr@WKUVX1.BITNET Date: 25 Feb 1993 20:10:58 -0700 (MST) From: KMM@CCNODE.Colorado.EDU Reply-To: MX-List@WKUVX1.BITNET Subject: Re: MX not delivering mail. To: MX-List@WKUVX1.BITNET Message-ID: <01GV5G9YI3420017PR@VAXF.COLORADO.EDU> This afternoon I wrote: > The problem seemed to start at the time one user pushed 300 messages > (6000 bytes each) into the que at 2am. By the time I saw the problem in > the morning, MX was not delivering anything and the queue file and > directory were HUGE. ...... and more Thanks to all who helped. By copying all of the files to an unfragmentted disk, I was able to get things going again (and get complaints "What - 100 mail messges in an hour?"). It is slowly recovering from it's indigestion. Since our usage has quadrupled in the last 2 months (and continuing to climb), I'll just have to be more vigilant (and update the software). Thanks much, Karen ================================================================================ Archive-Date: Thu, 25 Feb 1993 22:18:53 CST Sender: list-mgr@WKUVX1.BITNET Date: Thu, 25 Feb 1993 13:56:09 EST From: "Brian Tillman, Smiths Industries, VAX Support, x8425" Reply-To: MX-List@WKUVX1.BITNET To: munroe@dmc.com CC: mx-list@WKUVX1.BITNET Message-ID: <00968A9D.10567B00.28519@swdev.si.com> Subject: RE: Update to cmuip066? Dick Munroe (munroe@dmc.com) writes: >I noticed csus.edu:/pub/cmuip/contrib/tekip0665a.save_z. Is this >a patch or update to csus.edu:/pub/cmuip/vms-v5/*? It's an update. It converts CMUIP V6.6-5 (CMUIP066) to V6.6-5A. You'll need to restore the TEKIP0665A saveset and copy the images it contains to the necessary directories. The destinations should be obvious from the names in the saveset, but they won't match exactly, since CMUIP066 has all references to "Tek" removed, while TEKIP066A does not. -----------------------------+-------------------------------- Brian Tillman | Internet: tillman@swdev.si.com Smiths Industries, Inc. | tillman_brian@si.com 4141 Eastern Ave., MS129 | Hey, I said this stuff myself. Grand Rapids, MI 49518-8727 | My company has no part in it. -----------------------------+-------------------------------- ================================================================================ Archive-Date: Fri, 26 Feb 1993 14:50:33 CST Sender: list-mgr@WKUVX1.BITNET Subject: a1mail address garbling Message-ID: <1993Feb26.124137.186@condor> From: newsmgr@logica.co.uk Reply-To: MX-List@WKUVX1.BITNET Date: 26 Feb 93 12:41:37 GMT To: MX-List@WKUVX1.BITNET I can't work out how to send mail through MX from all-in-1 mail, using the command line - can anyone help? From a1mail, I can send for example to mx%"user@host"@mrgate and everything works fine. But what I want to do is to include a line in a batch file, something like a1mail foobar.txt "mx%""user@host""@mrgate" but every combination of quotes and double quotes I can think of fails - mrgate seems determined to open up 'user@host', change it into 'host::user' and then tell me that it's not a valid address. The same problem appears to prevent me from including MX gateway addresses in the DDS - which is another problem I'd really like to solve. Thanks William Lees leesw@logica.co.uk ================================================================================ Archive-Date: Fri, 26 Feb 1993 18:46:45 CST Sender: list-mgr@WKUVX1.BITNET From: yale!nuconvex!nuconvex.com!simonpg@HARVARD.BITNET Reply-To: MX-List@WKUVX1.BITNET Date: Fri, 26 Feb 1993 17:48:30 EST From: Paul Simons To: MX-List@WKUVX1.BITNET Message-ID: <00968B86.B085C3F0.24658@nuconvex.com> Subject: RE: a1mail address garbling In <1993Feb26.124137.186@condor> William Lees says: >The same problem appears to prevent me from including MX gateway addresses in >the DDS - which is another problem I'd really like to solve. This doesn't directly have to do with MX, but I have been trying to deal with that problem as well. Is DDS an ALLIN1 product or a part of MessageBus? Where is any documentation for DDS? ---------- Paul G. Simons simonpg@nuconvex.com CONnecticut Valley Electric eXchange (CONVEX) 203.276.6435 Southington, CT, USA {yale,uunet!hsi}!nuconvex!simonpg ================================================================================ Archive-Date: Fri, 26 Feb 1993 19:01:09 CST Sender: list-mgr@WKUVX1.BITNET From: Reply-To: MX-List@WKUVX1.BITNET Subject: Re: UUCP + MX = Ugly addresses Date: 24 Feb 1993 12:47:12 -0500 Message-ID: <1mgcb0INNt0o@dayub.dayton.saic.com> To: MX-List@WKUVX1.BITNET Paul Simons (yale!nuconvex.com!simonpg%harvard@HARVUNXW.BITNET) wrote: : I guess by now it's obvious that I am trying to get MX configured to met the nee : ds : of my site. I hope that my travails will help the next person. Is there a way : to : help MX convert UUCP headers such as: : CS.YALE.EDU!harvard.UUCP!WKUVX1.BITNET!list-mgr : to instead of: : WKUVX1.BITNET!list-mgr%harvard.UUCP@CS.YALE.EDU ? MX as of 3.1 or so will now use the UUCP rewrite rules if you want it to. You can then rewrite the messy addresses as you want them to show up. I have the docs for 3.0 in front of me so it must have been 3.1 where that feature was added. Check out the current release notes and documentation for more information. -Earle -- Earle Ake Internet: NSI-DECnet (SPAN): 28276::ake ================================================================================ Archive-Date: Sat, 27 Feb 1993 17:10:32 CST Sender: list-mgr@WKUVX1.BITNET Date: Sat, 27 Feb 1993 15:01:08 PST From: "W. Todd Wipke" Reply-To: MX-List@WKUVX1.BITNET To: mx-list@RPIECSVX.BITNET Message-ID: <00968C38.79789040.2490@SECS.UCSC.EDU> Subject: electronic voting Our campus uses MX for Academic Senate (faculty) communication and fileserv and listserv things to make academic plans, budget plans avail campus wide without consuming paper. We are now embarking on a plan to eliminate the $600 that it costs to have a mail ballot by creating a scheme for confidential voting by email, using a program to tally the votes and check authenticity of voters as well as assuring only one vote per person. I am sure some other universities must already be doing this and would like to learn or borrow from them. ¶qWþªì= ¸5Zö×ü I envision having a validation list of username and unique assigned identifier. The outgoing ballot will include the substance to be voted upon, plus the name and id. To be valid, the user must send a msg to a special address and include commands such as name Jake Green id 20a745f vote item 1 yes vote item 2 no vote item 3 Jerome Street vote item 3 Sally Cornwall vote item 4 3 The system would check that the id is the one assigned and mailed out to Jake Green and that no vote has been yet recorded. It would record the votes and send a confirming msg to the voter that the vote was recorded successfully in a timely fashion. Voter fraud would be immediately apparent. For audit purposes, all messages probably should be recorded in a log file that would be deleted if the election is not contested. One could also create an artificial list called "vote" that had no distribution list, a log would be created that could be processed by any program to do the validation and tally, the problem with this approach is that there is no immediate feedback about validation and vote recording. I am not concerned about network security or encryption, winning an election here is a commitment to a lot of work. I also envision being able to run a reminder program that would remind people that have not voted yet to get their votes in promptly. Look forward to your help. Thanks in advance -Todd Wipke University of California, Santa Cruz (yes, city of TGV) ================================================================================ Archive-Date: Sat, 27 Feb 1993 17:21:32 CST Sender: list-mgr@WKUVX1.BITNET Date: Sat, 27 Feb 1993 17:21:12 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: wipke@secs.ucsc.edu Message-ID: <00968C4C.0A04FAA0.5408@WKUVX1.BITNET> Subject: RE: electronic voting "W. Todd Wipke" writes: > >Our campus uses MX for Academic Senate (faculty) communication and >fileserv and listserv things to make academic plans, budget plans avail [...] >I envision having a validation list of username and unique >assigned identifier. The outgoing ballot will include the substance >to be voted upon, plus the name and id. To be valid, the user must >send a msg to a special address and include commands such as [...] I'm not 100% sure what you're looking for here, but it sounds like a perfect example of a SITE interface to MX. What I would do, if this is what you're looking for: o Create an address, say VOTE@SITE, that gets delivered to the MX Site agent. o From within SITE_DELIVER.COM, run whatever program to tally the votes and send mail acknowledging receipt of the vote. If you wanted to make sure that only certain addresses could vote, one way would be to create a mailing list that contains all valid addresses (set to NOMAIL) and the SITE address set to MAIL. If you set the list protection to (G:RW,W), I think, only those people on the list can send mail to the list address. Any such mail will be delivered to the mailing list subscribers, but since they're all set to NOMAIL, only the SITE address would actually get the message. With this approach, you could also automatically get an archive of all messages posted to the list. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Sat, 27 Feb 1993 17:36:14 CST Sender: list-mgr@WKUVX1.BITNET Date: Sat, 27 Feb 1993 17:35:59 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: mx-list@WKUVX1.BITNET Message-ID: <00968C4E.1AC2A520.5420@WKUVX1.BITNET> Subject: Implementing multiple SITE delivery agents "W. Todd Wipke" writes: > >I also run BULLETIN which uses SITE deliver. I don't understand how >to sort out stuff sent to @bulletin from those sent to the >voting address. I guess I don't understand the site stuff very well, >the site_deliver.com I have is specific to BULLETIN. >-Todd With MX v3.1, you can specify /ROUTE on the DEFINE PATH xxxx SITE command. The ROUTE value gets passed to the SITE_DELIVER.COM as P1, allowing SITE_DELIVER.COM to check the value and branch to the appropriate code or call another procedure or whatever. For example: MCP> DEFINE PATH BULLETIN SITE/ROUTE=BULLETIN MCP> DEFINE PATH VOTE SITE/ROUTE=VOTE Then in SITE_DELIVER.COM: $ if p1.eqs."BULLETIN" then goto bulletin_message $ if p1.eqs."VOTE" then goto vote_message $ ..... I'm going to forward this to the MX-List too.... Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU) ================================================================================ Archive-Date: Sun, 28 Feb 1993 08:32:05 CST Sender: list-mgr@WKUVX1.BITNET Date: Sun, 28 Feb 1993 9:27:40 -0500 (EST) From: J_CERNY@UNHH.UNH.EDU Reply-To: MX-List@WKUVX1.BITNET Message-ID: <930228092740.21631e7e@UNHH.UNH.EDU> Subject: any insight on files of size zero blocks? To: mx-list%wkuvx1.bitnet@mitvma.mit.edu Hi, First I want to thank several folks on this list who responded to my query a few weeks ago (before I became a subscriber) ... about using MX with UCX 2.0. Works great. Despite applying every patch that DEC sent us (it was no joke to be a member of the patch-of-the-week club), we still had the UCX SMTP queue keep dying on ACCVIO errors. MX has cured that. But a question. We notice that sometimes we get files in the MX_QUEUE directory of size zero. In particular the MSG_TEXT and HDR_INFO files will be empty, but the SRC_INFO file will have an address in it. These don't seem to be deliverable and we've been manually deleting them. Any comments or insights as to what is going on? Jim Cerny, Computing and Information Services, Univ. N.H. ================================================================================ Archive-Date: Sun, 28 Feb 1993 09:23:20 CST Sender: list-mgr@WKUVX1.BITNET Date: Sun, 28 Feb 1993 09:23:00 CST From: Hunter Goatley Reply-To: MX-List@WKUVX1.BITNET To: MX-List@WKUVX1.BITNET CC: j_cerny@unhh.unh.edu Message-ID: <00968CD2.66B72DA0.5663@WKUVX1.BITNET> Subject: RE: any insight on files of size zero blocks? J_CERNY@UNHH.UNH.EDU writes: > >But a question. We notice that sometimes we get files in the MX_QUEUE >directory of size zero. In particular the MSG_TEXT and HDR_INFO files >will be empty, but the SRC_INFO file will have an address in it. These >don't seem to be deliverable and we've been manually deleting them. >Any comments or insights as to what is going on? > They're left over sometimes when users are sending mail and get disconnected or killed before completing the send. I think what's going on is that, when the process is killed (or user hangs up or whatever), the normal exit handlers from Mail aren't invoked and the MX message doesn't get cancelled. You can cancel them using MCP QUEUE CANCEL---that'll cause the MX Router to delete the files on its next purge pass. In fact, that's a better way to get rid of them, as the queue entries get deleted too. Hunter ------ Hunter Goatley, VMS Systems Programmer, Western Kentucky University goathunter@WKUVX1.BITNET (or goathunter%wkuvx1.bitnet@UKCC.UKY.EDU)