Archive-Date: Fri, 1 Aug 1997 00:02:09 -0500 Sender: Date: Fri, 1 Aug 1997 00:01:49 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8196.927D9275.181@wku.edu> Subject: MX-LIST Administrivia: Monthly Post Posting statistics for list MX-LIST during July 1997 Total number of posts: 115 Total number of posters: 40 Total number of subscribers: 0 Last modified: 28-SEP-1995 13:33 (Updated digest info) Welcome to MX-List@LISTS.WKU.EDU, 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. MX-List postings are also available in a daily digest format. To subscribe to the digest, send the following command in the body of a mail message to MXserver@LISTS.WKU.EDU: SUBSCRIBE MX-List-Digest "Your real name here" The MX-List archives are maintained at ARCHIVES@LISTS.WKU.EDU. To get a copy of any month's postings, send an e-mail message with the body SEND MX-List.yyyy-mm to ARCHIVES@LISTS.WKU.EDU, 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. MX itself is available via anonymous ftp from ftp.spc.edu in [.MX.MX041]. You can also get it via e-mail by sending the commands SEND MX and SEND FILESERV_TOOLS on separate lines in the body of a mail message to FILESERV@LISTS.WKU.EDU. To remove yourself from the mailing list, send the following command to MXserver@LISTS.WKU.EDU: SIGNOFF MX-List MXserver 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 DIGEST - to switch to digest mode SET MX-List NODIGEST - to switch to non-digest mode 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 WKU 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, Sr. OpenVMS Systems Programmer goathunter@MadGoat.com Process Software P.O. Box 51745 Bowling Green, KY 42102-6745 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- ================================================================================ Archive-Date: Fri, 1 Aug 1997 09:10:24 -0500 Sender: Date: Fri, 01 Aug 1997 16:10:08 MET_DST From: "Ruth Thieme, GSI Darmstadt,Tel. 06159 712556" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: thieme@axp602.gsi.de Message-ID: <009B821D.D8662A89.332@axp602.gsi.de> Subject: SMTP-Port Hi, I want to see if MX is working port 25 when using TELENET. Where can I check it? Thanks for e very answer Ruth | | | Name: Ruth Thieme | | Institute: Gesellschaft fuer Schwerionenforschung (GSI) | | - DV&EE - | | Planckstrasse 1 | | 62491 Darmstadt | | Office: Room 1.251 | | Office Phone: +49 (6159) 71-2556 | | FAX: 2986 | | E-mail: R.Thieme@gsi.de | |_______________________________________________________________________| ================================================================================ Archive-Date: Fri, 1 Aug 1997 09:40:31 -0500 Sender: Date: Fri, 1 Aug 1997 07:40:15 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B81D6.9D48F021.25@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: SMTP-Port > From: MX%"MX-List@MadGoat.com" 1-AUG-1997 07:25:03.86 > Subj: SMTP-Port > Hi, > I want to see if MX is working port 25 when using TELENET. Where can I check > it? > Thanks for e very answer > Ruth > $ TELNET/PORT=25 mxhost.your.org > > > > | | > | Name: Ruth Thieme | > | Institute: Gesellschaft fuer Schwerionenforschung (GSI) | > | - DV&EE - | > | Planckstrasse 1 | > | 62491 Darmstadt | > | Office: Room 1.251 | > | Office Phone: +49 (6159) 71-2556 | > | FAX: 2986 | > | E-mail: R.Thieme@gsi.de | > |_______________________________________________________________________| > -HWM ================================================================================ Archive-Date: Fri, 1 Aug 1997 10:08:15 -0500 Sender: From: "Nic CLEWS" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <802564E6.0052D743.00@aejacob.jacobs.co.uk> Date: Fri, 1 Aug 1997 16:12:13 +0100 Subject: Re: SMTP-Port MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII >Hi, >I want to see if MX is working port 25 when using TELENET. Where can I check >it? >Thanks for e very answer If you TELNET using whatever client you have, select port 25. For example on Multinet on OpenVMS, use TELNET/PORT:25 node.yoyodyne.com (or whatever the node name is). in NT or AS400 go to the OPTIONS and change the port from 23 (normal) to 25. You should get the SMTP server to identify itself, just type 'quit' in lower case to end the session. If this works then the SMTP server is responding to SMTP connections. Nic Clews -- Bolton, UK website http://www.bolton.ac.uk/bolton/ homepage http://www.python.demon.co.uk/ ================================================================================ Archive-Date: Fri, 1 Aug 1997 10:14:26 -0500 Sender: Date: Fri, 1 Aug 1997 11:13:45 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B81F4.708947C0.20@swdev.si.com> Subject: Re: SMTP-Port Nic Clews (Nic_CLEWS@jacobs.co.uk) writes: >You should get the SMTP server to identify itself, just type 'quit' in >lower case to end the session. I would use upper case as a general rule. While MX may recognize lower case, many SMTP servers do not. -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Fri, 1 Aug 1997 11:05:53 -0500 Sender: Date: Fri, 1 Aug 1997 09:05:43 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B81E2.8DB652D7.79@CVOBKU.CVO.MP.USBR.GOV> Subject: Re: SMTP-Port > From: MX%"MX-List@MadGoat.com" 1-AUG-1997 08:35:18.52 > Subj: Re: SMTP-Port Sorry about that previous null posting - testing out a new system... > > Nic Clews (Nic_CLEWS@jacobs.co.uk) writes: > > >You should get the SMTP server to identify itself, just type 'quit' in > >lower case to end the session. > > I would use upper case as a general rule. While MX may recognize lower case, > many SMTP servers do not. Brian, Can you elaborate - that's a protocol violation. Maybe there shoule be an RFC on SMTP server non-compliance. I wonder how many brownbag points sendmail would rack up? > -- > Brian Tillman Internet: tillman_brian at si.com > Smiths Industries, Inc. tillman at swdev.si.com > 4141 Eastern Ave., MS239 Addresses modified to prevent > Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" > This opinion doesn't represent that of my company -HWM ================================================================================ Archive-Date: Fri, 1 Aug 1997 12:19:23 -0500 Sender: Date: Fri, 1 Aug 1997 13:18:40 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8205.E46010A0.1@swdev.si.com> Subject: Re: SMTP-Port >Can you elaborate - that's a protocol violation. I'm not sure what more I can say. I just know that I've used telnet to connect to some SMTP servers (I don't know which ones; I didn't pay any attention to it) that would accept upper case only. That's all I know. -- Brian ================================================================================ Archive-Date: Mon, 4 Aug 1997 05:47:53 -0500 Sender: From: "Nic CLEWS" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <802564E9.003A0C03.00@aejacob.jacobs.co.uk> Date: Mon, 4 Aug 1997 11:51:59 +0100 Subject: Re: SMTP-Port MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII >>Can you elaborate - that's a protocol violation. >I'm not sure what more I can say. I just know that I've used telnet to connect to some >SMTP servers (I don't know which ones; I didn't pay any attention to it) >that would accept upper case only. That's all I know. That's odd. In my original posting I stated use lower case, because I have come across an SMTP server that accepted lower case only !!! If memory serves me correctly, it was an IBM system, and I don't recall having any problems using lower case in any SMTP conversations. nic ================================================================================ Archive-Date: Mon, 4 Aug 1997 10:59:35 -0500 Sender: From: "Robert H. McClanahan" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Date: Mon, 4 Aug 1997 10:59:21 -0500 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: MX SMTP Error Greetings, I'm running MX V4.2 over UCX V2.0 on VMS 5.5-2 on a Vaxstation Model 90. I've got an entry in my MX queue that looks like this: Entry: 1, Origin: [SMTP] Status: IN-PROGRESS, size: 80 bytes Created: 4-AUG-1997 07:20:34.98, expires 3-SEP-1997 07:20:34.98 Last modified 4-AUG-1997 10:34:29.69 SMTP entry #4, status: READY, size: 80 bytes, waiting for retry until 4-AUG-1 997 11:04:40.62 Created: 4-AUG-1997 07:20:35.90, expires 3-SEP-1997 07:20:34.98 Last modified 4-AUG-1997 10:34:40.64 Recipient #1: , Route=netstep.net Error count=7 Last error: %MX_SMTP-W-INSUFF_STORAGE, action not taken: insufficient system storage Can someone tell me what the "Last error" message means? What type of system storage? Thanks, RHM +--+ Robert H. McClanahan, Mgr, Tech Info Systems, rmcclanahan@tis.aecc.com <[]>< Arkansas Electric Coop Corp, PO Box 194208, Little Rock, AR 72219-4208 USA "If I find in myself a desire which no experience in this world can satisfy, the most probable explanation is that I was made for another world." C.S. Lewis ================================================================================ Archive-Date: Mon, 4 Aug 1997 11:08:49 -0500 Sender: Date: Mon, 4 Aug 1997 11:08:44 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B844F.3C69AFF2.16@ALPHA.WKU.EDU> Subject: RE: MX SMTP Error "Robert H. McClanahan" writes: > > Last error: %MX_SMTP-W-INSUFF_STORAGE, action not taken: insufficient >system storage > >Can someone tell me what the "Last error" message means? What type of system >storage? > That error is being returned from the remote server and may mean just about anything. It probably means there's no spool space available, or the user is over quota. In any case, there's nothing you can really do about it, short of contacting the postmaster at that node. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Mon, 4 Aug 1997 11:17:16 -0500 Sender: Date: Mon, 04 Aug 1997 11:17:07 CST From: Ruth Dodson 612-896-6143 Reply-To: MX-List@MadGoat.com To: mx-list@madgoat.com CC: dodson@vmsmail.ncent.attws.com Message-ID: <009B8450.68618EFA.9@vmsmail.ncent.mccaw.com> Subject: A non-MX question... This is not a question regarding MX, so please forgive me for that to start with. It's just that I know you folks are mostly the types that are literate in both VMS and Unix and therefore might be able to point me to an answer: I'm looking for PERL for VMS. Does anyone know of a good PERL for VMS? (shareware or purchased product - either one as long as it's stable.) We would prefer a version that ran OpenVMS for Alpha v6.2-1H3, but could also use OpenVMS for VAX v6.2. Thanks for any possible leads. Ruth Dodson --- Ruth Dodson AT&T Wireless Services System Manager 7900 S. Xerxes Ave, Suite 301 Scott Consulting Corporation Minneapolis, MN 55431 dodson@vmsmail.ncent.mccaw.com (612)896-6143 ================================================================================ Archive-Date: Mon, 4 Aug 1997 12:27:07 -0500 Sender: Date: Mon, 4 Aug 1997 13:26:17 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: dodson@vmsmail.ncent.mccaw.com Message-ID: <009B8462.73E06960.3@swdev.si.com> Subject: RE: A non-MX question... Ruth Dodson (dodson@vmsmail.ncent.mccaw.com) writes: >I'm looking for PERL for VMS. ftp://genetics.upenn.edu/perl5/ -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Mon, 4 Aug 1997 13:00:25 -0500 Sender: Message-ID: <33E5B5A9.3680653@tmc.naecker.com> Date: Mon, 04 Aug 1997 10:57:45 +0000 From: Brad Hughes Reply-To: MX-List@MadGoat.com MIME-Version: 1.0 To: MX-List@MadGoat.com Subject: Re: A non-MX question... References: <009B8450.68618EFA.9@vmsmail.ncent.mccaw.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Ruth Dodson 612-896-6143 wrote: > > This is not a question regarding MX, so please forgive > me for that to start with. It's just that I know you > folks are mostly the types that are literate in both > VMS and Unix and therefore might be able to point me > to an answer: > > I'm looking for PERL for VMS. Does anyone know of a good > PERL for VMS? (shareware or purchased product - either one > as long as it's stable.) We would prefer a version that > ran OpenVMS for Alpha v6.2-1H3, but could also use OpenVMS > for VAX v6.2. > > Thanks for any possible leads. > > Ruth Dodson > > --- > Ruth Dodson AT&T Wireless Services > System Manager 7900 S. Xerxes Ave, Suite 301 > Scott Consulting Corporation Minneapolis, MN 55431 > dodson@vmsmail.ncent.mccaw.com (612)896-6143 http://www.sidhe.org/vmsperl/ ================================================================================ Archive-Date: Mon, 4 Aug 1997 16:13:24 -0500 Sender: Date: Mon, 04 Aug 1997 22:42:54 CET-DST From: "Rok Vidmar, NUK Ljubljana" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B84B0.3645956A.15@NUK.Uni-Lj.Si> Subject: RE: MX SMTP Error > > Last error: %MX_SMTP-W-INSUFF_STORAGE, action not taken: insufficient > >system storage Hunter, it is not obvious from the above message that in fact it originates from the remote server; I believe a clear statement in this regard should prevent misunderstandings. Regards, Rok Vidmar Internet: rok.vidmar@uni-lj.si National and University Library Phone: +386 61 125 4218 Turjaska 1, 1000 Ljubljana Fax: +386 61 125 5007 Slovenia ================================================================================ Archive-Date: Mon, 4 Aug 1997 19:09:08 -0500 Sender: Date: Mon, 4 Aug 1997 19:09:00 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8492.54990684.5@ALPHA.WKU.EDU> Subject: RE: MX SMTP Error "Rok Vidmar, NUK Ljubljana" writes: > >> > Last error: %MX_SMTP-W-INSUFF_STORAGE, action not taken: insufficient >> >system storage > > Hunter, it is not obvious from the above message that in fact it originates >from the remote server; I believe a clear statement in this regard should >prevent misunderstandings. > I agree. The text above matches that given in RFC 821.... Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Tue, 5 Aug 1997 02:51:31 -0500 Sender: Date: Tue, 05 Aug 1997 08:51:11 GMT From: System Manager Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8505.2FB7570F.5@spotimage.fr> Subject: how to edit messages in queue hi all, I have this problem: I use a site gateway and sometimes, some messages are rejected by the gateway because of ill-formed RFC822 header lines (ex.: multiples address in the 'From:' line) or RFC822 header line that I do not support for now (ex.: list name in the 'To:' line instead of a real recipient). before I have time to modify my site gateway, is there a way to locate and edit the bad message (with its queue entry for example) ? thanks in advance, Pierre Bru System & Network Mgr. mailto:system@spotimage.fr ================================================================================ Archive-Date: Tue, 5 Aug 1997 06:56:45 -0500 Sender: Date: Tue, 5 Aug 1997 06:56:38 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B84F5.2F019644.11@ALPHA.WKU.EDU> Subject: RE: how to edit messages in queue System Manager writes: > >I use a site gateway and sometimes, some messages are rejected >by the gateway because of ill-formed RFC822 header lines >(ex.: multiples address in the 'From:' line) or RFC822 header line >that I do not support for now (ex.: list name in the 'To:' line >instead of a real recipient). > >before I have time to modify my site gateway, is there a way to >locate and edit the bad message (with its queue entry for example) ? > Not really. If you know the format of the .HDR_INFO file, you *could* edit it by hand, but I wouldn't recommend it unless you're confident that you understand the format of that file. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Tue, 5 Aug 1997 07:56:41 -0500 Sender: Date: Tue, 5 Aug 1997 5:56:33 -0700 From: Matt Madison Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <970805055633.2030da73@Cisco.COM> Subject: RE: MX SMTP Error Hunter wrote: >"Rok Vidmar, NUK Ljubljana" writes: >> >>> > Last error: %MX_SMTP-W-INSUFF_STORAGE, action not taken: insufficient >>> >system storage >> >> Hunter, it is not obvious from the above message that in fact it originates >>from the remote server; I believe a clear statement in this regard should >>prevent misunderstandings. >> >I agree. The text above matches that given in RFC 821.... The other tip-off (admittedly non-obvious) is "%MX_SMTP-..." facility name in the message code. If it begins with "MX_SMTP", it's an SMTP status code returned from the remote server. -Matt -- Matthew Madison | | madison@cisco.com Cisco Systems | 101 Cooper St. | Santa Cruz, CA 95060 USA | +1 408 457 5390 ================================================================================ Archive-Date: Tue, 5 Aug 1997 08:10:15 -0500 Sender: From: goathunter@madgoat.com (Hunter Goatley) Subject: Re: BASE64 decode Date: 5 Aug 1997 12:19:01 GMT Message-ID: <5s75nl$lb6$1@news.wku.edu> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU In article <5rsijd$m5n$1@paperboy.ids.net>, mike@kronos.egr-ri.ids.net (Mike Umbricht) writes: >Is there a utility that can be used with MX / VMS mail that will decode: > > Content-Type: application/zip; name="401aug97.zip" > Content-Transfer-Encoding: base64 > Content-Disposition: inline; filename="401aug97.zip" > >It appears that mx_decode can only process "application/VMS-RMS" > While it was designed to process those, you can also have it process the file you describe by cutting out the headers (leaving only the encoded text) and using: $ mxd :== $mx_exe:mx_decode.exe $ mxd/image/noheaders file.in file.out Hunter ------ Hunter Goatley, Process Software, http://www.process.com/ TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Tue, 5 Aug 1997 10:40:47 -0500 Sender: Date: Tue, 5 Aug 1997 11:34:09 -0400 Message-ID: <97080511340913@funyet.mro.dec.com> From: anderson@funyet.mro.dec.com (Paul Anderson) Reply-To: MX-List@MadGoat.com To: MX-LIST@madgoat.com Subject: NETLIB_STARTUP looks for NETLIB_MULTINET_SHR I'm running MX V4.2 and recently installed the new version of NETLIB, V2.1. Today I noticed the SMTP_SERVER was dying with a MX SMTP Server (pid 00007D6D) exiting, status = 10018294 error and that during system startup, NETLIB_STARTUP gave the error %NETLIB-E-NOSHR, could not find NETLIB_MULTINET_SHR library to install I did not choose Multinet support when installing either MX or NETLIB. Apparently, NETLIB_STARTUP.COM sees device INET0 on the system and tries to install the Multinet image, which of course doesn't exist. I've commented out the IF/END IF statement beginning with the F$GETDVI call to check INET0 in NETLIB_STARTUP.COM and now there's no error returned at startup and the SMTP_SERVER runs properly. Is this expected behavior? Paul __________________________________________________________________ Paul Anderson Digital Equipment Corporation Printing Systems technical support DEC:.mro.FUNYET::ANDERSON DTN 297.8927 anderson@funyet.mro.dec.com 508.467.8927 MRO1-2/J25 508.467.1310 fax ================================================================================ Archive-Date: Tue, 5 Aug 1997 13:49:51 -0500 Sender: Message-ID: <01BCA1E0.C0C2F640@ntdose.schumi.ch> From: "Kurt A. Schumacher" Reply-To: MX-List@MadGoat.com To: "'MX-List@MadGoat.com'" Subject: Forwarded mails containing embedded documents - not recognized Date: Tue, 5 Aug 1997 20:47:12 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hello We have several customers here on the DECUS Switzerland machine using a OpenVMS-Mail forwarding to other sites, like C*Serve, A*L... Now, they strated using the funny and colorful tools, like the provider ones or Bills Exchange and Outlook (as I do). As long there is no forwarding, all of them are mostly able to recognize any kind of embedded documents (like Word, Excel... data). If the mail is forwarded (by MX/VMS) they can't anymore. What happens with the SMTP headers - they dont lock wrong after forwarding. Thanks for any hint. Kurt +------------------------------------------------------------------------+ | Kurt A. Schumacher Web: http://www.schumi.ch | | E-Mail: Kurt.Schumacher@schumi.ch | | PSI-Mail: PSI%022847911312::K_SCHUMACHER | | Auenstrasse 61 Voice: +41 1 881 37 80 | | CH-8302 Kloten FAX: +41 1 881 37 88 | +------------------------------------------------------------------------+ ================================================================================ Archive-Date: Wed, 6 Aug 1997 07:43:42 -0500 Sender: Date: Wed, 6 Aug 1997 5:43:34 -0700 From: Matt Madison Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: anderson@funyet.mro.dec.com Message-ID: <970806054334.202e4d72@Cisco.COM> Subject: RE: NETLIB_STARTUP looks for NETLIB_MULTINET_SHR >I'm running MX V4.2 and recently installed the new version of NETLIB, V2.1. >Today I noticed the SMTP_SERVER was dying with a > > MX SMTP Server (pid 00007D6D) exiting, status = 10018294 > >error and that during system startup, NETLIB_STARTUP gave the error > > %NETLIB-E-NOSHR, could not find NETLIB_MULTINET_SHR library to install > >I did not choose Multinet support when installing either MX or NETLIB. >Apparently, NETLIB_STARTUP.COM sees device INET0 on the system and tries to >install the Multinet image, which of course doesn't exist. > >I've commented out the IF/END IF statement beginning with the F$GETDVI call to >check INET0 in NETLIB_STARTUP.COM and now there's no error returned at startup >and the SMTP_SERVER runs properly. > >Is this expected behavior? Sort of. The NETLIB startup procedure tries to be "smart" about figuring out which TCP/IP package you have installed, and automatically selecting the correct NETLIB_xxx_SHR for that package. Which TCP/IP package are you running on that system? If it's MultiNet, then it's the expected behavior (and you should have installed NETLIB's MultiNet support). Otherwise, it sounds like I need to smarten up the automatic detection code a bit more. -Matt -- Matthew Madison | | madison@cisco.com Cisco Systems | 101 Cooper St. | Santa Cruz, CA 95060 USA | +1 408 457 5390 ================================================================================ Archive-Date: Wed, 6 Aug 1997 07:48:44 -0500 Sender: Date: Wed, 6 Aug 1997 5:48:36 -0700 From: Matt Madison Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <970806054836.202e4d72@Cisco.COM> Subject: RE: Forwarded mails containing embedded documents - not recognized >We have several customers here on the DECUS Switzerland machine using a >OpenVMS-Mail forwarding to other sites, like C*Serve, A*L... > >Now, they strated using the funny and colorful tools, like the provider >ones or Bills Exchange and Outlook (as I do). As long there is no >forwarding, all of them are mostly able to recognize any kind of embedded >documents (like Word, Excel... data). If the mail is forwarded (by MX/VMS) >they can't anymore. What happens with the SMTP headers - they dont lock >wrong after forwarding. Do you mean forwarded automatically (through a VMS MAIL SET FORWARD entry or an MX alias), or forwarded manually (i.e., the user reading the message uses the FORWARD command to re-send the message elsewhere)? If the messages are forwarded automatically, then the headers should have been preserved intact, including the MIME headers. For a manual forwarding, however, the new message would not include any MIME headers in the wrapper message around the message being forwarded, which would prevent it from being decoded. -Matt -- Matthew Madison | | madison@cisco.com Cisco Systems | 101 Cooper St. | Santa Cruz, CA 95060 USA | +1 408 457 5390 ================================================================================ Archive-Date: Wed, 6 Aug 1997 08:57:38 -0500 Sender: Message-ID: <3.0.32.19970806085725.00b0da10@bible.acu.edu> Date: Wed, 06 Aug 1997 08:57:26 -0500 To: MX-List@MadGoat.com From: Tom Dolan Reply-To: MX-List@MadGoat.com Subject: RE: Forwarded mails containing embedded documents - not recognized MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Matt Said: >If the messages are forwarded automatically, then the headers should have >been preserved intact, including the MIME headers. I'm rather slow understanding all of this - but I think I just learned something important: If I set forwarding up using the VMS MAIL "SET FORWARD /USER=" command, then the MIME headers are left intact allowing the receiving persons email client to properly decode file attachments. If forwarding is via the MX Alias feature, MIME headers may not be left intact. This is why the documentation says to use VMS MAIL forwarding? (And also why attachments for some of my users don't work! - I had them forwarded via an MX Alias - I just changed them all to VMS MAIL forward.) What is the most common use of MX Alias (besides this incorrect one!)? Thanks! Tom Dolan Dolan@Bible.acu.edu 202 Bible Building Systems Manager ACU Box 29454 College of Biblical Studies Abilene, TX 79699 Abilene Christian University 915.674.3706 ================================================================================ Archive-Date: Wed, 6 Aug 1997 09:39:13 -0500 Sender: Date: Wed, 6 Aug 1997 09:39:06 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B85D5.0BAF7B99.16@ALPHA.WKU.EDU> Subject: RE: Forwarded mails containing embedded documents - not recognized Tom Dolan writes: > >If I set forwarding up using the VMS MAIL "SET FORWARD /USER=" >command, then the MIME headers are left intact allowing the >receiving persons email client to properly decode file attachments. > If forwarding is via the MX Alias feature, MIME headers may not be >left intact. > No, they're left intact in either case. The only time they're not is when the user uses the FORWARD command inside VMS MAIL to forward the message back out. When that happens, the original RFC822 headers are treated as part of the message text and new headers are added (just as if the user had typed SEND). With DEFINE ALIAS and SET FORWARD, MX recognizes the forwarding info and forwards the message with the original headers intact (simply adding a Received: line to the original headers. Another way that can cause two sets of headers is a situation like this: GOATEST is forwarded to GOATEST1 GOATEST1 is forwared to MX%"goathunter@MadGoat.com" In this case, mail sent to GOATEST1 is properly forwarded without new headers because MX Router sees the forward and re-routes the message without involving MX Local. HOWEVER, if the message is sent to GOATEST, MX Router sees the forward to GOATEST1 and queues the message for delivery to GOATEST1 by MX Local. MX Local goes to deliver the message (at which point the headers are considered part of the message text, since VMS Mail doesn't support RFC822 headers) and callable MAIL ends up re-queueing the message via MX, causing a new set of RFC822 headers to be created for the message. It's that dual-set of headers that confuses MIME e-mail clients. >This is why the documentation says to use VMS MAIL forwarding? > No, it recommends VMS MAIL forwarding if you have lots of aliases because MX Router scans the MX aliases sequentially. Storing hundreds of addresses as MX aliases would slow down MX Router process because it would search sequentially through all the aliases looking for a match on every incoming address. (This is why the ADDRESS_REWRITER should be used if you have more than a few dozen forwarding addresses---it does a binary search on the list of aliases.) >(And also why attachments for some of my users don't work! - I had >them forwarded via an MX Alias - I just changed them all to VMS >MAIL forward.) > Both should work A-OK. You might enable MX debugging to see exactly how those messages are being processed. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Wed, 6 Aug 1997 09:49:40 -0500 Sender: Date: Wed, 6 Aug 1997 09:49:33 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B85D6.81B575BD.41@ALPHA.WKU.EDU> Subject: RE: Forwarded mails containing embedded documents - not recognized Hunter Goatley writes: > >Another way that can cause two sets of headers is a situation like >this: > > GOATEST is forwarded to GOATEST1 > GOATEST1 is forwared to MX%"goathunter@MadGoat.com" > >In this case, mail sent to GOATEST1 is properly forwarded without new >headers because MX Router sees the forward and re-routes the message >without involving MX Local. A slight correction. MX Local gets the message either way, but in this case, it recognizes that the GOATEST1 is forwarded back through MX and forwards the message without involving callable VMS MAIL. Here's the debug log for a message to sent GOATEST1: Processing queue entry number 16 Checking local name: GOATEST1 LOCAL_USER: Found fwdg address MX%"goathunter@madgoat.com" (VMS) = (MX) Forwarding to: FORWARD_MESSAGE: Normal forward - including Resent hdrs. FORWARD_MESSAGE: Resent-From=GOATEST1 (VMS), (MX) FORWARD_MESSAGE: Forwarding queue entry number=17 FORWARD_MESSAGE: Forwarding to: All done with this entry. >HOWEVER, if the message is sent to >GOATEST, MX Router sees the forward to GOATEST1 and queues the message >for delivery to GOATEST1 by MX Local. MX Local goes to deliver the >message (at which point the headers are considered part of the message >text, since VMS Mail doesn't support RFC822 headers) and callable MAIL >ends up re-queueing the message via MX, causing a new set of RFC822 >headers to be created for the message. > And here's the log of a message sent to GOATEST: Processing queue entry number 20 Checking local name: GOATEST LOCAL_USER: Non-MX fwdg address GOATEST1; treat as local. This is a regular delivery. DELIVER: mime_headers = 0 DELIVER: fdlstr = "" DELIVER: Using MX%"goathunter@goat.process.com" "Hunter Goatley" as VMS MAIL From address. DELIVER: Using MX%"GOATEST@KID.MADGOAT.COM" as VMS MAIL To address. DELIVER: Using as VMS MAIL CC address. DELIVER: Using "test" as subject. DELIVER: Delivering to GOATEST1 DELIVER: Status=00000001 from MAIL$ routines All done with this entry. Does that help explain the difference?? Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Wed, 6 Aug 1997 23:00:01 -0500 Sender: Date: Wed, 06 Aug 1997 21:59:40 MDT From: Mark Tarka Reply-To: MX-List@MadGoat.com To: mx-list@wkuvx1.wku.edu Message-ID: <009B863C.806977A9.3@earth.oscs.montana.edu> Subject: Avoiding spam/commercial ads Has anyone written a document about how a simple user of a system can avoid spam/commercial ads? Is there an appropriate news/discussion group for this topic? In the last year or less, more and more addresses containing land-mines designed to thwart the "spam-bots" and e-mail address harvesters have crossed my screen. Well, not really "land-mines" ('though that would be a neat concept) but more like "nimbys" (Not In My BackYard... remember?). Say...can't code be written to automatically delete all those different "nimby" mail address disfigurements? Sure, I can merely delete the offending posts, often after just reading the subject line...but it is junk-mail, and I am spending a few seconds thinking about it, and expending some joules of energy pressing the keys that will delete the material from our campus system's storage space. How can a simple user find protection from the spam/commercial assualts occurring daily throught the email system? Yours for good cooking... Mark ================================================================================ Archive-Date: Thu, 7 Aug 1997 02:03:33 -0500 Sender: Date: Thu, 7 Aug 1997 09:02:04 +0200 (MET DST) From: PASZTOR Miklos Reply-To: MX-List@MadGoat.com To: Mark Tarka CC: mx-list@wkuvx1.wku.edu Subject: Re: Avoiding spam/commercial ads Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE On Wed, 6 Aug 1997, Mark Tarka wrote: >=20 > Has anyone written a document about how a simple user of a system > can avoid spam/commercial ads? Is there an appropriate > news/discussion group for this topic? Yes. Please have a look at the web page http://spam.abuse.net/spam, and newsgroups alt.spam, news.admin.net-abuse.*. =20 Regards, Mikl=F3s /------------------------------------------------------\ Pa'sztor Miklo's | E-mail: pasztor@sztaki.hu MTA SZTAKI/ASZI Budapest 1132 V. Hugo u. 18-22 | Tel: (36)-(1)-149-75-32 Institute for Computation and Automation, Hungarian Academy of Sciences \------------------------------------------------------/ =20 ================================================================================ Archive-Date: Thu, 7 Aug 1997 08:04:17 -0500 Sender: Date: Thu, 07 Aug 1997 15:04:06 +0200 From: "Bengt H. Nilsson, KTH Stockholm" Reply-To: MX-List@MadGoat.com To: mx-list@madgoat.com CC: nilsson@plasma.kth.se Message-ID: <009B86CB.9D5818E3.15@plasma.kth.se> Subject: QUOTED-PRINTABLE in Subject line Hello, We receive a lot of mail using "Content-Transfer-Encoding: quoted-printable". The text is decoded by the Local agent, but the header lines, including the Subject line, are not, making these unreadable if they contain national characters. Can something be done about this? MX 4.2 VMS 6.2 Bengt H. Nilsson Dep. of Plasma Physics Alfvén Laboratory The Royal Institute of Technology Stockholm Sweden ================================================================================ Archive-Date: Thu, 7 Aug 1997 08:47:59 -0500 Sender: Date: Thu, 7 Aug 1997 08:47:51 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: NILSSON@PLASMA.KTH.SE Message-ID: <009B8697.0D95B172.28@ALPHA.WKU.EDU> Subject: RE: QUOTED-PRINTABLE in Subject line "Bengt H. Nilsson, KTH Stockholm" writes: > >We receive a lot of mail using >"Content-Transfer-Encoding: quoted-printable". The text >is decoded by the Local agent, but the header lines, >including the Subject line, are not, making these >unreadable if they contain national characters. Can >something be done about this? > That's covered by a different RFC from the MIME RFC. We have not yet added support for that to MX, though it is on the wish list. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Thu, 7 Aug 1997 12:55:32 -0500 Sender: Date: Thu, 7 Aug 1997 13:45:37 -0400 Message-ID: <97080713453704@funyet.mro.dec.com> From: anderson@funyet.mro.dec.com (Paul Anderson) Reply-To: MX-List@MadGoat.com To: MX-List@madgoat.com Subject: Re: NETLIB_STARTUP looks for NETLIB_MULTINET_SHR Matt, Concerning NETLIB_STARTUP.COM that sees device INET0 and tries to install NETLIB_MULTINET_SHR: > The NETLIB startup procedure tries to be "smart" about figuring out which > TCP/IP package you have installed, and automatically selecting the correct > NETLIB_xxx_SHR for that package. Which TCP/IP package are you running on > that system? TPFKAUCX, or The Product Formerly Known as UCX. I've never run Multinet on this system, and never had this problem before upgrading to the new NETLIB. I have an INET0: device on all my Alpha and VAX systems. Once NETLIB_STARTUP.COM determines PACKAGE to be UCX, it overwrites it with MULTINET by checking combinations of logical names; it defaults to MULTINET because I don't have any of the non-Digital TCP/IP packages installed. Paul __________________________________________________________________ Paul Anderson Digital Equipment Corporation Printing Systems technical support DEC:.mro.FUNYET::ANDERSON DTN 297.8927 anderson@funyet.mro.dec.com 508.467.8927 MRO1-2/J25 508.467.1310 fax ================================================================================ Archive-Date: Thu, 7 Aug 1997 13:08:20 -0500 Sender: Date: Thu, 7 Aug 1997 14:07:15 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: anderson@funyet.mro.dec.com Message-ID: <009B86C3.ABE7EDC0.5@swdev.si.com> Subject: Re: NETLIB_STARTUP looks for NETLIB_MULTINET_SHR Paul Anderson (anderson@funyet.mro.dec.com) writes: >Concerning NETLIB_STARTUP.COM that sees device INET0 and tries to install >NETLIB_MULTINET_SHR: ... >TPFKAUCX, or The Product Formerly Known as UCX. I've never run Multinet on >this system, and never had this problem before upgrading to the new NETLIB. ... >I have an INET0: device on all my Alpha and VAX systems. Interesting! I have TPFKAUCX V4.0 on many VAX systems here (30 of them) and not one of them has device INET0. What version of TPFKAUCX? -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Mon, 11 Aug 1997 05:35:04 -0500 Sender: Date: Mon, 11 Aug 1997 12:35:42 MET_DST From: "Ruth Thieme, GSI Darmstadt,Tel. 06159 712556" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: thieme@axp602.gsi.de Message-ID: <009B89DB.8B771307.1@axp602.gsi.de> Subject: Umlauts Hi, I have a MX conversion problem for german umlauts. I send a file containing german umlauts from a PC to a VMS-Alpha-system which uses MX. VMS mail finds a forward to second PC. When reading the file on this destination, all german umlauts are no longer correct. Can I tell MX that there should be no conversion for those 8-bit -codes. Thanks for every answer Ruth ================================================================================ Archive-Date: Wed, 13 Aug 1997 08:16:51 -0500 Sender: Date: Wed, 13 Aug 1997 09:16:41 EDT From: Irv Eisen Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8B52.133B8A80.27@ccstat.mc.duke.edu> Subject: Problem with one domain in MLF list We use MLF to serve a list of over 1100 addresses. All of the addressees at one particular institution, domains OSU.EDU and MEDCTR.OSU.EDU, claim that they have never received any of our broadcasts even though we can send email directly to these people. MLF has never returned any errors for these domains, even though it is appropriately returning errors for many other bad addresses. I turned on debugging for ROUTER, SMTP and MLF and can see no indication of problems with the broadcasts. It's as if MLF broadcasts, but not regular mail, are being handled at OSU but not redistributed. Is there something else we can do on this end to debug this? Does this sound like something MLF could be doing to the addresses? Any other suggestions would be appreciated. Thanks for your help, ================================================================ Irv Eisen, Systems Manager Cancer Center Information Systems Email: IRV@CCSTAT.MC.DUKE.EDU Duke University Medical Center DUMC Box 3958 Voice: (919) 681-5405 Durham, NC 27710 Fax : (919) 681-8028 ================================================================ ================================================================================ Archive-Date: Wed, 13 Aug 1997 08:28:19 -0500 Sender: Date: Wed, 13 Aug 1997 06:28:10 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B8B3A.88534DC0.139@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: Problem with one domain in MLF list > From: MX%"MX-List@MadGoat.com" 13-AUG-1997 06:22:53.84 > To: MX%"MX-List@MadGoat.com" > CC: > Subj: Problem with one domain in MLF list > On Wed, 13 Aug 1997 09:16:41 EDT, Irv Eisen said: Irv Eisen writes: > We use MLF to serve a list of over 1100 addresses. All of the addressees > at one particular institution, domains OSU.EDU and MEDCTR.OSU.EDU, claim > that they have never received any of our broadcasts even though we can > send email directly to these people. > > MLF has never returned any errors for these domains, even though it is > appropriately returning errors for many other bad addresses. I turned on > debugging for ROUTER, SMTP and MLF and can see no indication of problems > with the broadcasts. > > It's as if MLF broadcasts, but not regular mail, are being handled at OSU > but not redistributed. > > Is there something else we can do on this end to debug this? Does this > sound like something MLF could be doing to the addresses? Any other > suggestions would be appreciated. > > Thanks for your help, > ================================================================ > Irv Eisen, Systems Manager > Cancer Center Information Systems Email: IRV@CCSTAT.MC.DUKE.EDU > Duke University Medical Center > DUMC Box 3958 Voice: (919) 681-5405 > Durham, NC 27710 Fax : (919) 681-8028 > ================================================================ > Irv, If the SMTP debug log shows the mail getting into their mail system, then I would give them that log and tell them to work on it. If possible, can you get an actual dump of the TCP SMTP transaction between your host and their host? That would prove that you are handing the mail off to them. -HWM ================================================================================ Archive-Date: Wed, 13 Aug 1997 12:07:48 -0500 Sender: Date: Wed, 13 Aug 1997 10:05:52 PST8PDT7,M4.1.0,M10.5.0 From: James Wilkinson Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: wilkinson@hsq.com Message-ID: <009B8B58.F2136AEC.6@hsq.com> Subject: Selective delivery based on From address I need to implement selective delivery based on a message's From address. So we have the following: Local users all have accounts on a local Alpha running OVMS 7.1 MX 4.2 UCX 4.1e5. Most users access mail via UCX pop server, some use VMSmail directly on the server. If any local user sends to any other local user, normal default mail delivery is what we of course want. If external mail is delivered to a local user, for some users just do normal default delivery. For other users, message is redirected to another predefined account. Or even better, normal delivery plus a copy to the other predefined account. But redirecting is sufficient. local domain = mycom.com bob@mycom.com sends to will@mycom.com -> goes to local mbx will roger@outside.com sends to will@mycom.com -> goes to local mbx support (& will) roger@outside.com sends to bob@mycom.com -> goes to local mbx bob roger@outside.com sends to hank@mycom.com -> goes to local mbx sales (& hank) It would also be beneficial to handle outbound mail in a likewise manner. bob@mycom.com sends to roger@outside.com -> smtp delivery to roger@outside.com will@mycom.com sends to roger@outside.com -> smtp delivery to roger@outside.com and local CC deliver to support will@mycom.com sends to bob@mycom.com -> local delivery to mbx bob hank@mycom.com sends to roger@outside.com -> smtp delivery to roger@outside.com and local CC deliver to sales I need all the intelligence in the SMTP server, not in the mail clients. I would like to keep everything on one server to simplify management. So how to go about this? James James Wilkinson HSQ Technology 1435 Huntington Avenue South San Francisco, CA 94080-5999 ph: +1-650-952-4310 fx: +1-650-952-7206 em: wilkinson@hsq.com ================================================================================ Archive-Date: Thu, 14 Aug 1997 03:58:37 -0500 Sender: Date: Thu, 14 Aug 1997 10:59:30 -0200 From: "Mario Meyer, Phys.-Techn. Bundesanstalt" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8C29.9AA7FD00.43@ChbRB.Berlin.PTB.De> Subject: Wish for MLF request handler Hi Hunter, as MX becomes more and more perfect, I'd like to add another wish to the ever growing list. Could You , please , modify the Request handler so that mail signatures (line starting with "--" and following text) are ignored ? Thanks Mario --,------------------------------------------------------.-- . , | Mario Meyer Physikalisch-Technische Bundesanstalt | _QQ__ | ............. Institut Berlin Proj. IB.T02 | __( U, )__ | : wide area : Abbestr. 2-12, D - 10587 Berlin | /// `---' \\\ | : networker : tel. (+49 30) 3481 472, fax. ... 490 | /||\ /||\ --| :...........: SMTP: MMeyer@Berlin.PTB.De |------------------ `------------------------------------------------------' ================================================================================ Archive-Date: Thu, 14 Aug 1997 04:13:12 -0500 Sender: Date: Thu, 14 Aug 1997 02:13:03 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B8BE0.0F5DBB66.53@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: Wish for MLF request handler > From: MX%"MX-List@MadGoat.com" 14-AUG-1997 02:02:17.77 > To: MX%"MX-List@MadGoat.com" > CC: > Subj: Wish for MLF request handler > On Thu, 14 Aug 1997 10:59:30 -0200, "Mario Meyer, Phys.-Techn. Bundesanstalt" said: "Mario Meyer, Phys.-Techn. Bundesanstalt" writes: Mario, > Hi Hunter, > > as MX becomes more and more perfect, I'd like to add another wish to the ever > growing list. Could You , please , modify the Request handler so that mail > signatures (line starting with "--" and following text) are ignored ? > > Thanks > Mario > If the sender of the request puts the word: QUIT as the only word in a line to the request address, any text after that point is ignored by the MLF processor. > --,------------------------------------------------------.-- . , > | Mario Meyer Physikalisch-Technische Bundesanstalt | _QQ__ > | ............. Institut Berlin Proj. IB.T02 | __( U, )__ > | : wide area : Abbestr. 2-12, D - 10587 Berlin | /// `---' \\\ > | : networker : tel. (+49 30) 3481 472, fax. ... 490 | /||\ /||\ > --| :...........: SMTP: MMeyer@Berlin.PTB.De |------------------ > `------------------------------------------------------' -HWM ================================================================================ Archive-Date: Thu, 14 Aug 1997 04:44:46 -0500 Sender: Date: Thu, 14 Aug 1997 11:45:37 -0200 From: "Mario Meyer, Phys.-Techn. Bundesanstalt" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8C30.0B9FE9E0.5@ChbRB.Berlin.PTB.De> Subject: RE: Wish for MLF request handler Henry W. Miller answered advised to insert the command QUIT. Ok., this works as well as a line /NOSIGNATURE, but nearly any normal user steps in that trap and asks me about these "strange" error messages from MLF. They say MLF has a bug and think it is a lousy program. You know this situation. I think that implementation of automatic signature stripping wouldn't be such a complicated task, and this is no urgent problem, rather some kind of cosmetics. But when I have a look at some actual software developements on the market - aren't most of them cosmetics ? Regards Mario --,------------------------------------------------------.-- . , | Mario Meyer Physikalisch-Technische Bundesanstalt | _QQ__ | ............. Institut Berlin Proj. IB.T02 | __( U, )__ | : wide area : Abbestr. 2-12, D - 10587 Berlin | /// `---' \\\ | : networker : tel. (+49 30) 3481 472, fax. ... 490 | /||\ /||\ --| :...........: SMTP: MMeyer@Berlin.PTB.De |------------------ `------------------------------------------------------' ================================================================================ Archive-Date: Thu, 14 Aug 1997 05:51:48 -0500 Sender: Date: Thu, 14 Aug 1997 03:51:39 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B8BED.D586AC82.3@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: Wish for MLF request handler > From: MX%"MX-List@MadGoat.com" 14-AUG-1997 02:47:45.67 > To: MX%"MX-List@MadGoat.com" > CC: > Subj: RE: Wish for MLF request handler > Mario, On Thu, 14 Aug 1997 11:45:37 -0200, "Mario Meyer, Phys.-Techn. Bundesanstalt" said: "Mario Meyer, Phys.-Techn. Bundesanstalt" writes: > Henry W. Miller answered advised to insert the command QUIT. > Ok., this works as well as a line /NOSIGNATURE, but nearly any normal user steps > in that trap and asks me about these "strange" error messages from MLF. > They say MLF has a bug and think it is a lousy program. You know this situation. > I think that implementation of automatic signature stripping wouldn't be such a > complicated task, and this is no urgent problem, rather some kind of cosmetics. > But when I have a look at some actual software developements on the market - > aren't most of them cosmetics ? > The problem is, I don't think that there is really a standard signature seperator. Sure, most people DO use at least two dashes, but it you ascede to that demand, someone else is sure to come along with something else they they think should be supported. I have had arguements with subscribers on some lista about their formatting of text - basically each paragraph being one long line of hundreds or thousands or characters without a line break of any sort. This one person, for instance, pulled a standard out of the air (actually, I think that he pulled it out from someplace lower and to the rear, if you take my meaning...) stating that the user agent should be able to reformat his crap. (Note that NO encoding was used at all...) And this guy RUNS a company that produces email software. Incredible. Anyway, I do see your point. But at some point, you have to draw the line on what you support in regards to something that does not have a standard. (Note: I am assuming here that there is no RFC defining a "standard" signature block - someone please correct me if I am wrong.) I've actually had one unixhead state that he prefers the handling of mail and mailing lists via VMS and MX than anything available on unix systems! I suppost that there could be a "/NOGACK" setting on the MLF, meaning to not return obtuse error messages when the user sends something ridiculous to the list server, but on the other hand, that would make it difficult to notify the bonehead users when they make a serious fubar. I warn the users to either not send a signature with the message to the "-request" addresses, or if they have to, to add QUIT after the last command. And, when people bitch about the software, which still runs better than any unix based system I've seen, I tell them that when they pay the bills for the machine, it's support, the software, and my time, then they can dictate what is used to run the mailing lists. But, then again, that's just me... > Regards > Mario > > --,------------------------------------------------------.-- . , > | Mario Meyer Physikalisch-Technische Bundesanstalt | _QQ__ > | ............. Institut Berlin Proj. IB.T02 | __( U, )__ > | : wide area : Abbestr. 2-12, D - 10587 Berlin | /// `---' \\\ > | : networker : tel. (+49 30) 3481 472, fax. ... 490 | /||\ /||\ > --| :...........: SMTP: MMeyer@Berlin.PTB.De |------------------ > `------------------------------------------------------' -HWM ================================================================================ Archive-Date: Thu, 14 Aug 1997 09:16:22 -0500 Sender: Message-ID: <199708141413.PAA28684@zaphod.CFMfel01.icl.co.uk> From: "Jamie Jones" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Date: Thu, 14 Aug 1997 15:16:08 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: Re: Wish for MLF request handler On 14 Aug 97 at 11:45:37 -0200, Mario Meyer wrote: > Henry W. Miller answered advised to insert the command QUIT. > Ok., this works as well as a line /NOSIGNATURE, but nearly any normal user steps > in that trap and asks me about these "strange" error messages from MLF. > They say MLF has a bug and think it is a lousy program. You know this situation. > I think that implementation of automatic signature stripping wouldn't be such a > complicated task, and this is no urgent problem, rather some kind of cosmetics. .... and as the code already exists to accept "QUIT", it should be relatively easy to also accept "-- " (startline-hyphen-hyphen-space-endline) as an alias for it. Regards, Jamie ================================================================================ Archive-Date: Thu, 14 Aug 1997 10:05:31 -0500 Sender: Date: Thu, 14 Aug 1997 11:04:22 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8C2A.48B52DA0.16@swdev.si.com> Subject: RE: Wish for MLF request handler Mario, what makes you think all signatures start with "--"? -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Fri, 15 Aug 1997 04:54:09 -0500 Sender: Date: Fri, 15 Aug 1997 11:54:54 -0200 From: "Mario Meyer, Phys.-Techn. Bundesanstalt" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8CFA.82419380.10@ChbRB.Berlin.PTB.De> Subject: RE: Wish for MLF request handler Brian Tillman asked what made me thinking that all signatures start with "--". This is a good question and I feared that someone would ask. I have no definitive answer. It's just one of these old wisdoms that are taken from generation to generation and nobody knows who they originate :-). I surfed a bit through the Web without much success... In http://www.ecnet.net/users/xwww/ecn/tutorials/signature.html I found: Some programs recognize a signature by signature dashes, which form the first line of a signature. They consist of two dashes and a trailing space ("-- ") and act as a separator, allowing easy recognition of signatures by both people and computers. It is a courtesy to preface your signature with signature dashes. At another place was stated that a line containing only "--" is known as the "kill line" that would mark the end of any kind of mail command scripts (see http://megadodo.com/articles/2R205.html ). Anyway - it seems to be a commonly applied method to have at least two dashes at the start of mail signature blocks. Or why is your signature also separated this way, Brian ? Mario --,------------------------------------------------------.-- . , | Mario Meyer Physikalisch-Technische Bundesanstalt | _QQ__ | ............. Institut Berlin Proj. IB.T02 | __( U, )__ | : wide area : Abbestr. 2-12, D - 10587 Berlin | /// `---' \\\ | : networker : tel. (+49 30) 3481 472, fax. ... 490 | /||\ /||\ --| :...........: SMTP: MMeyer@Berlin.PTB.De |------------------ `------------------------------------------------------' ================================================================================ Archive-Date: Fri, 15 Aug 1997 07:22:03 -0500 Sender: Date: Fri, 15 Aug 1997 08:21:53 EST From: Bruce Bowler Reply-To: MX-List@MadGoat.com To: MX-List@madgoat.com CC: BBOWLER@BIGELOW.ORG Message-ID: <009B8CDC.C0469D5B.5@phyto.bigelow.org> Subject: RE: Wish for MLF request handler "Mario Meyer, Phys.-Techn. Bundesanstalt" writes: >Some programs recognize a signature by signature dashes, which form the first >line of a signature. They consist of two dashes and a trailing space ("-- ") And what does Mario's signature start with? 2 dashes and a trailing comma... (yes, I realize he goes on to talk about just 2 dashes without specifying any trailing character). >--,------------------------------------------------------.-- . , > | Mario Meyer Physikalisch-Technische Bundesanstalt | _QQ__ > | ............. Institut Berlin Proj. IB.T02 | __( U, )__ > | : wide area : Abbestr. 2-12, D - 10587 Berlin | /// `---' \\\ > | : networker : tel. (+49 30) 3481 472, fax. ... 490 | /||\ /||\ >--| :...........: SMTP: MMeyer@Berlin.PTB.De |------------------ > `------------------------------------------------------' +---------------------+---------------------------------------+-------------+ | Bruce Bowler | I like long walks, especially when | Usual | | 1.207.633.9600 | they're taken by people who annoy me. | Disclaimers | | BBowler@Bigelow.org | Fred Allen | Apply | +---------------------+---------------------------------------+-------------+ ================================================================================ Archive-Date: Fri, 15 Aug 1997 09:03:36 -0500 Sender: Date: Fri, 15 Aug 1997 10:02:43 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8CEA.D6679460.1@swdev.si.com> Subject: RE: Wish for MLF request handler >Anyway - it seems to be a commonly applied method to have at least two dashes >at the start of mail signature blocks. Or why is your signature also separated >this way, Brian ? Um, because I used to have my signature surrounded by a drawn box like this: +------------------------------+ | This is where I had my sig | +------------------------------+ But I got tired of the noise. You'll notice that there is no blank following the two dashes I use now. Also, I notice that, though you advocate MX recognizing two dashes followed by a space, your own sig does NOT follow that convention so that, even if MX were to recognize it as you ask, it would fail to recognize your own sig. -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Fri, 15 Aug 1997 09:22:14 -0500 Sender: Date: Fri, 15 Aug 1997 10:19:57 -1300 From: "Charles T. Smith, Jr." Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8CED.3EAFD440.4@dragon.com> Subject: Listserv Interface I have a cluster where several machines run MX processes. I was recently doing some testing with listserv, and noted that nodes running the router process other than the node running listserv were not finding the listserv lists, even though I'd done a cluster-wide reset. Are there any restrictions on what nodes can run the router with listserv? How does listserv process communicate the lists it has with the other MX modules? ================================================================================ Archive-Date: Fri, 15 Aug 1997 09:50:27 -0500 Sender: Date: Fri, 15 Aug 1997 09:50:20 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B8CE9.1B3E9938.42@ALPHA.WKU.EDU> Subject: RE: Listserv Interface "Charles T. Smith, Jr." writes: > >I have a cluster where several machines run MX processes. I was recently doing >some testing with listserv, and noted that nodes running the router process >other than the node running listserv were not finding the listserv lists, >even though I'd done a cluster-wide reset. > >Are there any restrictions on what nodes can run the router with listserv? >How does listserv process communicate the lists it has with the other MX >modules? > LISTSERV mailing list names are located by MX Router by searching for: LISTSERV_LISTS_DIR:*.LIST In order for all of your MX Router processes to find the LISTSERV lists, you must make sure LISTSERV_LISTS_DIR is defined on all the nodes (and that the device is available to all nodes). Sounds kind of cheesy, I know, but that's how LISTSERV itself determines the list names (or, at least, that's how we were told to do the check). Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Fri, 15 Aug 1997 09:52:24 -0500 Sender: Date: Fri, 15 Aug 1997 10:52:13 -0400 (EDT) From: "Richard S. Shuford" Reply-To: MX-List@MadGoat.com To: MX-List@madgoat.com Subject: RE: Wish for MLF request handler Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII About signatures: historically, the Unix news-posting program that people began using in "B" News days (circa 1981) appended a signature following a separator line of the form hyphen-hyphen-space: -- Notice how you can't see that the space is there. Think how many teenage developers of shareware Windows email and news-posting programs also didn't notice that the space is there. Imagine how much Microsoft cares about any standards that anybody else in the world invented. -- ...Richard S. Shuford (at my current workplace) ================================================================================ Archive-Date: Fri, 15 Aug 1997 15:02:30 -0500 Sender: Date: Fri, 15 Aug 1997 22:01:57 +0200 From: "GWDVMS::MOELLER" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Subject: RE: Wish for MLF request handler Come on, folks ...I thought the topic was "commands" that would terminate the scan of the mailing list processor? How many _legal_ commands will it ever have? Why not make the scanner stop on _any_ non-alphabetic, if at least one legal command has been recognized earlier? This would still take care of E-Mail systems which force some "header" in front of their user's input, and change nothing for sources of all garbage input. I can't think of someone putting 'funny' lines in between legal commands, and being surprised when the subsequent commands were not acted upon at all ... Wolfgang J. Moeller, Tel. +49 551 2011516 or -510, moeller@gwdvms.dnet.gwdg.de GWDG, D-37077 Goettingen, F.R.Germany | Disclaimer: No claim intended! ----- ----- ================================================================================ Archive-Date: Mon, 18 Aug 1997 16:30:50 -0500 Sender: Date: Mon, 18 Aug 1997 17:30:39 -1300 From: william@fractl.tam.cornell.edu Reply-To: MX-List@MadGoat.com To: MX-LIST@MADGOAT.COM Message-ID: <009B8F84.E89B7970.7@fractl.tam.cornell.edu> Subject: Using a Site Transport Interface Hi I am trying to use a site transport interface. I created the following command file and called it MX_EXE:SITE_DELIVER.COM $ OPEN/WRITE ZZZ FRACTL_1:[HOLMES]FOO.DAT $ WRITE ZZZ P1 $ WRITE ZZZ P2 $ WRITE ZZZ P3 $ WRITE ZZZ P4 $ CLOSE ZZZ $ STATUS = 1 $ EXIT 'STATUS I then set the path command to SITE for my local host and sent a message. The message is received and ends up in the queue with a path of site. However its stays in the inprog state and the file foo.dat never gets created. Am I missing somthing ? Thanks ======================================================================= ___Bill william@fractl.tam.cornell.edu wth1@Cornell.edu ======================================================================= ================================================================================ Archive-Date: Mon, 18 Aug 1997 16:34:54 -0500 Sender: Date: Mon, 18 Aug 1997 16:34:46 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: WILLIAM@FRACTL.TAM.CORNELL.EDU Message-ID: <009B8F7D.1A5FFC2D.41@ALPHA.WKU.EDU> Subject: RE: Using a Site Transport Interface william@fractl.tam.cornell.edu writes: > >I then set the path command to SITE for my local host and sent a message. >The message is received and ends up in the queue with a path of site. >However its stays in the inprog state and the file foo.dat never gets >created. Am I missing somthing ? > Are you running the MX Site interface? Did you install it when you installed MX? Is the agent there (MCP STATUS)? Did you do MCP RESET after you defined the path? Have you tried enabling MX_SITE_DEBUG to see what's happening? Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Tue, 19 Aug 1997 07:08:21 -0500 Sender: From: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) Subject: Re: Problem with MX queue and leftover files. Date: 19 Aug 1997 11:50:17 GMT Message-ID: <5tc19p$asl@paperboy.ccc.nottingham.ac.uk> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU Following my message yesterday, it seems that MX on our VAX cluster is almost never deleting the files created by passing mail. I stopped MX when there were no entries in the queue and created a new queue file which has the desired effect of removing whatever was making mcp queue stat give screwy output. It has also restarted the queue entry numbers and the command $ dir [queue.%]% gives at least one file in each subdirectory. I've put LOCAL debugging on and there are no error messages logged. The only thing that I can think of is that we have a rather odd MX_ROOT logical, to wit MX_DEVICE:[CCCCOMMON.MX.] with MX_DEVICE defined as $1$DUA400:, which is the system disk (an RZ29 connected to an HSD30 controller, if that's relevant). The oldest files that I could find dated back to June and I've no idea what could have changed at that time. Surely this isn't the expected behaviour for MX? Baffled, Andy ------------------------------------------------------------------------ Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB Nottingham NG7 2RD, UK Phone: +44 115 951 3328 ================================================================================ Archive-Date: Tue, 19 Aug 1997 10:47:21 -0500 Sender: Date: Tue, 19 Aug 1997 16:19:27 GMT From: Chris.Sharman@ccagroup.co.uk Reply-To: MX-List@MadGoat.com To: MX-List@wkuvx1.wku.edu CC: Chris.Sharman@ccagroup.co.uk Message-ID: <009B9044.20F1B57C.2@ccagroup.co.uk> Subject: Name aliasing of local usernames I've been trying to persuade mx to put full names (eg Chris.Sharman) rather than usernames on all local addresses on outgoing mail. [VMS mail forwarding handles incoming mail for forename.surname]. I've tried address_rewriter and name_conversion, with rather mixed results. I've now got name_conversion more or less working, but my convert routine is getting called with non-local usernames (such as mx-list). This isn't what I was expecting - is it right or is there a configuration problem of some kind ? If so, what mechanism should I be using ? I don't want chris@somewhere.else translated, and the convert routine doesn't get to see the @somewhere.else Thanks ______________________________________________________________________ Chris Sharman Chris.Sharman@CCAgroup.co.uk Tel: +(44) 1772 662883 Fax: +(44) 1772 662893 CCA Stationery Ltd, Eastway, Fulwood, Preston, Lancs, PR2 9WS, ENGLAND ================================================================================ Archive-Date: Tue, 19 Aug 1997 11:24:06 -0500 Sender: Date: Tue, 19 Aug 1997 12:23:51 EDT From: "Jonathan E. Hardis" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: william@fractl.tam.cornell.edu, hardis@garnet.nist.gov Message-ID: <009B9023.379DDF60.9@garnet.nist.gov> Subject: RE: Using a Site Transport Interface > I am trying to use a site transport interface. I created the following > command file and called it MX_EXE:SITE_DELIVER.COM > > $ OPEN/WRITE ZZZ FRACTL_1:[HOLMES]FOO.DAT > $ WRITE ZZZ P1 > $ WRITE ZZZ P2 > $ WRITE ZZZ P3 > $ WRITE ZZZ P4 > $ CLOSE ZZZ > $ STATUS = 1 > $ EXIT 'STATUS > > I then set the path command to SITE for my local host and sent a message. > The message is received and ends up in the queue with a path of site. > However its stays in the inprog state and the file foo.dat never gets > created. Am I missing somthing ? Is the SITE agent installed and running? (Enter: $ MCP STATUS) - Jonathan ================================================================================ Archive-Date: Wed, 20 Aug 1997 12:46:51 -0500 Sender: From: welchb@woods.uml.edu (Brendan Welch, W1LPG) Reply-To: MX-List@MadGoat.com Subject: Re: Problem with MX queue and leftover files. Date: 19 Aug 97 08:52:48 -0500 Message-ID: <1997Aug19.085248.1@aspen> To: MX-List@WKUVX1.WKU.EDU In article <5tc19p$asl@paperboy.ccc.nottingham.ac.uk>, cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > Following my message yesterday, it seems that MX on our VAX cluster is > almost never deleting the files created by passing mail. I stopped MX > when there were no entries in the queue and created a new queue file > which has the desired effect of removing whatever was making mcp queue > stat give screwy output. It has also restarted the queue entry numbers > and the command > $ dir [queue.%]% > gives at least one file in each subdirectory. > > I've put LOCAL debugging on and there are no error messages logged. > > The only thing that I can think of is that we have a rather odd MX_ROOT > logical, to wit MX_DEVICE:[CCCCOMMON.MX.] with MX_DEVICE defined as > $1$DUA400:, which is the system disk (an RZ29 connected to an HSD30 > controller, if that's relevant). > > The oldest files that I could find dated back to June and I've no idea > what could have changed at that time. > > Surely this isn't the expected behaviour for MX? > > Baffled, Try looking at some "unshown" messages to see if they are on OPERHOLD. While looping over values of "i" in DCL, $mcp que sho/br 'i Just yesterday I recovered over 1000 messages $set ver $i = $loop: $i = i + 1 $if i .gt. then exit (and set nover) $mcp que sho/br 'i $mcp que ready 'i $wait 00:00:02 $goto loop but I do not know if I am being clever or stupid. -- Brendan Welch, system analyst, UMass-Lowell, W1LPG, welchb@woods.uml.edu ================================================================================ Archive-Date: Wed, 20 Aug 1997 12:46:59 -0500 Sender: From: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) Subject: Re: Problem with MX queue and leftover files. Date: 20 Aug 1997 15:28:27 GMT Message-ID: <5tf2er$n0c@paperboy.ccc.nottingham.ac.uk> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU In article <1997Aug19.085248.1@aspen>, welchb@woods.uml.edu (Brendan Welch, W1LPG) writes: [...] >Try looking at some "unshown" messages to see if they are on OPERHOLD. >While looping over values of "i" in DCL, > $mcp que sho/br 'i > >Just yesterday I recovered over 1000 messages > $set ver > $i = > $loop: > $i = i + 1 > $if i .gt. then exit (and set nover) > $mcp que sho/br 'i > $mcp que ready 'i > $wait 00:00:02 > $goto loop >but I do not know if I am being clever or stupid. Well, it's the most helpful reply that I've had so far. It does partially seem to solve the problem in that running the above DCL and then forcing a purge of the queue reduces the number of files left lying around. However, it seems to have the side-effect of re-sending messages that have already been delivered. My alternative seems to be - $mcp shutdown/clu $mcp queue compress $backup mx_root:[queue.old_%]*.*;*/del nla0:no.where/save $delete mx_root:[queue]old_%.dir; $mc sysman set e/c do @sys$startup:mx_startup which would be OK in the middle of the night. But why are these files left there in the first place? Andy ------------------------------------------------------------------------ Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB Nottingham NG7 2RD, UK Phone: +44 115 951 3328 ================================================================================ Archive-Date: Wed, 20 Aug 1997 14:31:28 -0500 Sender: Date: Wed, 20 Aug 1997 14:31:22 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B90FE.31EEF826.3@ALPHA.WKU.EDU> Subject: Re: Problem with MX queue and leftover files. cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > >In article <1997Aug19.085248.1@aspen>, welchb@woods.uml.edu (Brendan Welch, W1LPG) writes: >[...] >>Try looking at some "unshown" messages to see if they are on OPERHOLD. >>While looping over values of "i" in DCL, >> $mcp que sho/br 'i >> >>Just yesterday I recovered over 1000 messages >> $set ver >> $i = >> $loop: >> $i = i + 1 >> $if i .gt. then exit (and set nover) >> $mcp que sho/br 'i >> $mcp que ready 'i >> $wait 00:00:02 >> $goto loop >>but I do not know if I am being clever or stupid. > I don't either. You shouldn't be having these problems!! I'm at a loss to know why you see them. >Well, it's the most helpful reply that I've had so far. It does >partially seem to solve the problem in that running the above DCL and >then forcing a purge of the queue reduces the number of files left lying >around. However, it seems to have the side-effect of re-sending messages >that have already been delivered. > Yep. >My alternative seems to be - > > $mcp shutdown/clu > $mcp queue compress > $backup mx_root:[queue.old_%]*.*;*/del nla0:no.where/save > $delete mx_root:[queue]old_%.dir; > $mc sysman > set e/c > do @sys$startup:mx_startup > >which would be OK in the middle of the night. But why are these files >left there in the first place? > There is a bug in the FLQ (the queue management code) that can cause this problem after the agents have been up for a while. My guess is that that's what you're seeing. This problem has been fixed in V5.0, which we hope to have ready for release RSN. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Wed, 20 Aug 1997 17:07:16 -0500 Sender: From: welchb@woods.uml.edu (Brendan Welch, W1LPG) Reply-To: MX-List@MadGoat.com Subject: Re: Problem with MX queue and leftover files. Date: 20 Aug 97 16:06:40 -0500 Message-ID: <1997Aug20.160640.1@aspen> To: MX-List@WKUVX1.WKU.EDU > Well, it's the most helpful reply that I've had so far. It does > partially seem to solve the problem in that running the above DCL and > then forcing a purge of the queue reduces the number of files left lying > around. You could change the script so that it would only READY those numbers whose status is OPERHOLD. I just did not get tricky enough with the DCL. Maybe you could force the que to be purged ( ? correct nomenclature ?) after every few messages made ready, or if this procedure was running at night, wait a huge number of seconds inbetween each (so that the que would get cleaned up eventually anyway). In practice, we seem to have huge runs of contiguous message numbers on hold, when we are in trouble. >However, it seems to have the side-effect of re-sending messages > that have already been delivered. not for us > > My alternative seems to be - > > $mcp shutdown/clu > $mcp queue compress > $backup mx_root:[queue.old_%]*.*;*/del nla0:no.where/save > $delete mx_root:[queue]old_%.dir; > $mc sysman > set e/c > do @sys$startup:mx_startup > > which would be OK in the middle of the night. But why are these files > left there in the first place? > "Most of the time" the queue compress will not be allowed, because file is locked. (Maybe that is the crux of our whole problem.) Yet, at that instant, $mcp sta does indeed show that nothing is running. In the past I used to be able to run that compression in the middle of the night, and it often would work. I presumed that if it did not work, the reason was that mail was being sent or delivered at that instant, and I would just let it try again some other night. -- Brendan Welch, system analyst, UMass-Lowell, W1LPG, welchb@woods.uml.edu ================================================================================ Archive-Date: Wed, 20 Aug 1997 17:13:59 -0500 Sender: Date: Wed, 20 Aug 1997 17:13:52 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B9114.E5AC4AA1.51@ALPHA.WKU.EDU> Subject: Re: Problem with MX queue and leftover files. welchb@woods.uml.edu (Brendan Welch, W1LPG) writes: > >You could change the script so that it would only READY those numbers >whose status is OPERHOLD. OPERHOLD should only occur when there's a problem with an entry, specifically, if one of the files is missing for it, or if it points back to a non-matching entry (say, an SMTP entry points back to the wrong ROUTER entry). >I just did not get tricky enough with the DCL. >Maybe you could force the que to be purged ( ? correct nomenclature ?) >after every few messages made ready, or if this procedure >was running at night, wait a huge number of seconds inbetween each (so >that the que would get cleaned up eventually anyway). In practice, we >seem to have huge runs of contiguous message numbers on hold, when we >are in trouble. > I can't remember; do you have MX_FLQ_AUTOPURGE_FIN defined so messages are purged as they're marked FIN? I'm telling you, you should never see messages in OPERHOLD. And if you do, there were corresponding OPCOM messages, if you have MX_EVENT_OPER_CLASS defined. What do these messages show? >"Most of the time" the queue compress will not be allowed, because file >is locked. (Maybe that is the crux of our whole problem.) Yet, at that >instant, $mcp sta does indeed show that nothing is running. > Then you probably have a user sitting in VMS Mail in the process of sending a message. SHOW DEVICE/FILE MX_FLQ_DIR: will show which process has the queue file open. >In the past I used to be able to run that compression in the middle of the >night, and it often would work. I presumed that if it did not work, the >reason was that mail was being sent or delivered at that instant, and I >would just let it try again some other night. Doing MCP QUEUE COMPRESS *SHOULD* be something you do only in emergencies. I certainly wouldn't recommend doing it every night or anything like that (it should work OK, but it's extra work that should never be necessary). Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Wed, 20 Aug 1997 18:12:01 -0500 Sender: From: welchb@woods.uml.edu (Brendan Welch, W1LPG) Reply-To: MX-List@MadGoat.com Subject: Re: Problem with MX queue and leftover files. Date: 20 Aug 97 16:10:06 -0500 Message-ID: <1997Aug20.161006.1@aspen> To: MX-List@WKUVX1.WKU.EDU > My alternative seems to be - > > $mcp shutdown/clu > $mcp queue compress > $backup mx_root:[queue.old_%]*.*;*/del nla0:no.where/save > $delete mx_root:[queue]old_%.dir; Beware of problem which can arise if the old directories have multiple version numbers; you have to do a series of renamings and deletions. > $mc sysman > set e/c > do @sys$startup:mx_startup -- Brendan Welch, system analyst, UMass-Lowell, W1LPG, welchb@woods.uml.edu ================================================================================ Archive-Date: Thu, 21 Aug 1997 01:21:52 -0500 Sender: Date: Wed, 20 Aug 1997 23:21:42 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B9148.484709B5.1@CVOBKU.CVO.MP.USBR.GOV> Subject: Re: Problem with MX queue and leftover files. > From: MX%"MX-List@MadGoat.com" 20-AUG-1997 10:52:02.17 > To: MX%"MX-List@WKUVX1.WKU.EDU" > CC: > Subj: Re: Problem with MX queue and leftover files. > On 20 Aug 1997 15:28:27 GMT, cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) said: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > In article <1997Aug19.085248.1@aspen>, welchb@woods.uml.edu (Brendan Welch, W1LPG) writes: > [...] > >Try looking at some "unshown" messages to see if they are on OPERHOLD. > >While looping over values of "i" in DCL, > > $mcp que sho/br 'i > > > >Just yesterday I recovered over 1000 messages > > $set ver > > $i = > > $loop: > > $i = i + 1 > > $if i .gt. then exit (and set nover) > > $mcp que sho/br 'i > > $mcp que ready 'i > > $wait 00:00:02 > > $goto loop > >but I do not know if I am being clever or stupid. > > Well, it's the most helpful reply that I've had so far. It does > partially seem to solve the problem in that running the above DCL and > then forcing a purge of the queue reduces the number of files left lying > around. However, it seems to have the side-effect of re-sending messages > that have already been delivered. > > My alternative seems to be - > > $mcp shutdown/clu > $mcp queue compress > $backup mx_root:[queue.old_%]*.*;*/del nla0:no.where/save > $delete mx_root:[queue]old_%.dir; > $mc sysman > set e/c > do @sys$startup:mx_startup > > which would be OK in the middle of the night. But why are these files > left there in the first place? > > Andy > > ------------------------------------------------------------------------ > Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk > University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB > Nottingham NG7 2RD, UK Phone: +44 115 951 3328 Andy, I would ask if you are running: MCP> QUEUE PURGE Manually, in conjunction with letting the FLQ or ROUTER processes doing that in the background? I have seen instances of queue corruption and orphaned messages like that if the manual QUEUE PURGE happens to occur at the same time the automatic process is happening in the background. Also, I don't know what effect a disk defragmenter might have if attempted to defragment your MX disk at the same time a purge was occuring - a badly behaving defragmenter might cause some problems. As a general treatment, if I start seeing problems like this, or if I have had a system crash, or even an abrupt system shutdown without MX having been shutdown properly, I do this: $ MCP MCP> SHUTDOWN/CLUSTER MCP> QUEUE SYNCH/LOG MCP> QUEUE PURGE/LOG MCP> EXIT Then, from DCL: $ ANALYZE DISK/REPAIR Then, after making certain that the disk structure is consistant, then I restart MX on all nodes. Hope this helps. -HWM ================================================================================ Archive-Date: Thu, 21 Aug 1997 08:24:02 -0500 Sender: From: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) Subject: Leftover files again. Date: 21 Aug 1997 11:33:17 GMT Message-ID: <5th91t$ar2@paperboy.ccc.nottingham.ac.uk> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU Curiouser and curiouser ... First, answers to a few of the questions - Refining Brendan's DCL a little bit, I observe that $mcp queue show /all lists between zero and a dozen entries in the queue (usually all finished) A succession of mcp queue show/br where n goes from 0 to 1000, indicates that the queue has getting on for 800 entries which are marked as FIN. There are no entries marked OPERHOLD. In answer to Henry Miller's questions, Yes, I have done a manual purge of the queue while MX is running but I'm fairly sure that is not the source of the problem - in that I have only been tinkering with MX since the start of the week and these files appear to have been accumulating since June at least. Maybe I did a manual purge in June? I honestly cannot remember. In answer to the question about defraggers, I find (to my chagrin, once again) that I haven't actually defragmented the disk for some considerable time. I used to use the DFO system with an RDB database but abandoned it when we changed from RA disks to these RZ's. I meant to put a sequence of jobs in to run overnight defragging each disk in turn but it looks as if I forgot to do that as well. So... the disk with the MX queues on is pretty badly fragmented. I'll try a QUEUE SYNCH and QUEUE PURGE over lunch today with MX shut down. ... and finally How soon is RSN for MX 5.0, Hunter? Thank you all for your interest and help. Andy ------------------------------------------------------------------------ Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB Nottingham NG7 2RD, UK Phone: +44 115 951 3328 ================================================================================ Archive-Date: Thu, 21 Aug 1997 09:19:46 -0500 Sender: From: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) Subject: Leftover files yet again. Date: 21 Aug 1997 13:54:33 GMT Message-ID: <5thhap$c3r@paperboy.ccc.nottingham.ac.uk> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU I stopped MX, cleared out all the [.old*] files and directories, did a queue synch, queue purge and then a queue compress. I then again cleared the [.old*] directories so that I had ten empty directories under [queue]. I restarted MX and went for lunch. On my return, I discover that a small number of files have been created and apparently left after the entries have gone from the queue. These are files like 1.SMTP_INFO and there is something very, very odd going on. Doing mcp queu sho /all gives Entry# Status Size Source Agent Entry# Status Size ------ ------ ------- ------ ------- ------ ------ ------- 12 FINISH 540 MAIL 16 FINISH 515 MAIL 19 FINISH 64 MAIL 2 FINISH 857 SMTP 4 FINISH 857 LOCAL 6 FINISH 915 SMTP 8 FINISH 1681 SMTP <52121889@pbi.net> 10 FINISH 2886 LOCAL %MCP-I-QENTRIES, total matching entries: 8 Note that there's no entry number 1. Doing mcp queue sho 1 gives $ mcp queue sho 1 Entry# Status Size Source Agent Entry# Status Size ------ ------ ------- ------ ------- ------ ------ ------- 19 FINISH 64 MAIL So something that should point to 1 is pointing to 19 (or vice versa) SHOW/FULL on entries 1 and 19 give identical output - $ mcp queue sho 1/full Entry: 19, Origin: [Local] Status: FINISHED, size: 64 bytes Created: 21-AUG-1997 14:35:40.83, expires 20-SEP-1997 14:35:40.83 Last modified 21-AUG-1997 14:36:14.81 Recipient #1: %MCP-I-QENTRIES, total matching entries: 1 $ mcp queue sho 19/full Entry: 19, Origin: [Local] Status: FINISHED, size: 64 bytes Created: 21-AUG-1997 14:35:40.83, expires 20-SEP-1997 14:35:40.83 Last modified 21-AUG-1997 14:36:14.81 Recipient #1: %MCP-I-QENTRIES, total matching entries: 1 I think I'm going mad. Andy ------------------------------------------------------------------------ Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB Nottingham NG7 2RD, UK Phone: +44 115 951 3328 ================================================================================ Archive-Date: Thu, 21 Aug 1997 09:22:58 -0500 Sender: Date: Thu, 21 Aug 1997 07:22:47 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B918B.7D09F6CE.236@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: Leftover files again. > From: MX%"MX-List@MadGoat.com" 21-AUG-1997 06:33:38.33 > To: MX%"MX-List@WKUVX1.WKU.EDU" > CC: > Subj: Leftover files again. > On 21 Aug 1997 11:33:17 GMT, cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) said: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > Curiouser and curiouser ... > > First, answers to a few of the questions - > > Refining Brendan's DCL a little bit, I observe that > > $mcp queue show /all lists between zero and a dozen entries in > the queue (usually all finished) > > A succession of mcp queue show/br where n goes from 0 > to 1000, indicates that the queue has getting on for 800 entries > which are marked as FIN. There are no entries marked OPERHOLD. > > In answer to Henry Miller's questions, > > Yes, I have done a manual purge of the queue while MX is running > but I'm fairly sure that is not the source of the problem - in > that I have only been tinkering with MX since the start of the > week and these files appear to have been accumulating since June > at least. Maybe I did a manual purge in June? I honestly cannot > remember. > I know the feeling. Sometimes it's difficult to remember what I did yesterday... > In answer to the question about defraggers, I find (to my > chagrin, once again) that I haven't actually defragmented the > disk for some considerable time. I used to use the DFO system > with an RDB database but abandoned it when we changed from RA > disks to these RZ's. I meant to put a sequence of jobs in to run > overnight defragging each disk in turn but it looks as if I > forgot to do that as well. So... the disk with the MX queues on > is pretty badly fragmented. > Hmm, I wonder if this could be the problem - I have noticed anomalies with some programs dealing with badly fragmented disks. Maybe you should try to defrag it while you have MX down for a while? > I'll try a QUEUE SYNCH and QUEUE PURGE over lunch today with MX > shut down. > > ... and finally > > How soon is RSN for MX 5.0, Hunter? > > Thank you all for your interest and help. > > Andy > > ------------------------------------------------------------------------ > Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk > University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB > Nottingham NG7 2RD, UK Phone: +44 115 951 3328 -HWM ================================================================================ Archive-Date: Thu, 21 Aug 1997 09:37:09 -0500 Sender: Date: Thu, 21 Aug 1997 07:36:55 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B918D.76A76DC8.307@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: Leftover files yet again. > From: MX%"MX-List@MadGoat.com" 21-AUG-1997 07:26:41.31 > To: MX%"MX-List@WKUVX1.WKU.EDU" > CC: > Subj: Leftover files yet again. > On 21 Aug 1997 13:54:33 GMT, cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) said: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > I stopped MX, cleared out all the [.old*] files and directories, did a > queue synch, queue purge and then a queue compress. I then again cleared > the [.old*] directories so that I had ten empty directories under > [queue]. I restarted MX and went for lunch. > > On my return, I discover that a small number of files have been created > and apparently left after the entries have gone from the queue. These > are files like 1.SMTP_INFO and there is something very, very odd going > on. > > Doing mcp queu sho /all gives > > Entry# Status Size Source Agent Entry# Status Size > ------ ------ ------- ------ ------- ------ ------ ------- > 12 FINISH 540 MAIL > 16 FINISH 515 MAIL > 19 FINISH 64 MAIL > 2 FINISH 857 SMTP > 4 FINISH 857 LOCAL > 6 FINISH 915 SMTP > 8 FINISH 1681 SMTP <52121889@pbi.net> > 10 FINISH 2886 LOCAL > > %MCP-I-QENTRIES, total matching entries: 8 > > Note that there's no entry number 1. > > Doing mcp queue sho 1 gives > > $ mcp queue sho 1 > > Entry# Status Size Source Agent Entry# Status Size > ------ ------ ------- ------ ------- ------ ------ ------- > 19 FINISH 64 MAIL > > So something that should point to 1 is pointing to 19 (or vice versa) > SHOW/FULL on entries 1 and 19 give identical output - > > $ mcp queue sho 1/full > > Entry: 19, Origin: [Local] > Status: FINISHED, size: 64 bytes > Created: 21-AUG-1997 14:35:40.83, expires 20-SEP-1997 14:35:40.83 > Last modified 21-AUG-1997 14:36:14.81 > Recipient #1: > > %MCP-I-QENTRIES, total matching entries: 1 > $ mcp queue sho 19/full > > Entry: 19, Origin: [Local] > Status: FINISHED, size: 64 bytes > Created: 21-AUG-1997 14:35:40.83, expires 20-SEP-1997 14:35:40.83 > Last modified 21-AUG-1997 14:36:14.81 > Recipient #1: > > %MCP-I-QENTRIES, total matching entries: 1 > > I think I'm going mad. > > Andy > > > ------------------------------------------------------------------------ > Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk > University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB > Nottingham NG7 2RD, UK Phone: +44 115 951 3328 As Hunter indicated yesterday, maybe QUEUE COMPRESS is not the best choice. If you can get the queue completely clear, I'd try a QUEUE CREATE to create a completely new master queue file. -HWM ================================================================================ Archive-Date: Thu, 21 Aug 1997 09:37:18 -0500 Sender: Date: Thu, 21 Aug 1997 07:36:57 PST From: syscorn@CCVAX.CCS.CSUS.EDU Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B918D.77F22EC0.1@CCVAX.CCS.CSUS.EDU> Subject: RE: Leftover files yet again. Andy, I have had a similiar problem with leftover files in the mx queues, although my problem started with a power failure and a corrupt queue. I resolved the problem by doing the "queue compress" AND defining the logical MX_FLQ_AUTOPURGE_FIN TRUE as documented in section 6.4 of the Message Exchange Management Guide for V4.1 (although I am running V4.2 and hope to upgrade to V4.3 soon). Hope that helps! Phyllis Cornett Operating Systems Analyst CSU, Sacramento ================================================================================ Archive-Date: Thu, 21 Aug 1997 09:46:23 -0500 Sender: Date: Thu, 21 Aug 1997 09:46:14 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: ANDY.JACK@NOTTINGHAM.AC.UK Message-ID: <009B919F.874FBCE6.52@ALPHA.WKU.EDU> Subject: RE: Leftover files yet again. cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > >On my return, I discover that a small number of files have been created >and apparently left after the entries have gone from the queue. These >are files like 1.SMTP_INFO and there is something very, very odd going >on. > Not at all! >Doing mcp queu sho /all gives > [...] >Note that there's no entry number 1. > >Doing mcp queue sho 1 gives > >$ mcp queue sho 1 > >Entry# Status Size Source Agent Entry# Status Size >------ ------ ------- ------ ------- ------ ------ ------- > 19 FINISH 64 MAIL > >So something that should point to 1 is pointing to 19 (or vice versa) >SHOW/FULL on entries 1 and 19 give identical output - > Yes, they're designed to do that. Entry 19 was processed by MX Router, which determined that it needed to be processed by MX SMTP. It created the .SMTP_INFO using the next available queue entry (1), thus creating 1.SMTP_INFO. Entry 1 was delivered and marked as FINISHED. SHOW QUEUE and SHOW QUEUE/FULL only show the *original* router entry when they find an entry marked FIN, so doing SHOW QUEUE 1 and SHOW QUEUE 19 will both show 19. You can use MCP SHOW QUEUE/ALL/BRIEF to see *all* the entries in the queue. >I think I'm going mad. > When MX FLQ Manager or MX Router does it next PURGE pass, both the files for entry 1 and entry 19 should be deleted. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Thu, 21 Aug 1997 09:55:51 -0500 Sender: Date: Thu, 21 Aug 1997 10:54:36 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B91A9.145D21A0.24@swdev.si.com> Subject: RE: Leftover files yet again. Andy.Jack@nottingham.ac.uk writes: >Note that there's no entry number 1. > >Doing mcp queue sho 1 gives > >$ mcp queue sho 1 > >Entry# Status Size Source Agent Entry# Status Size >------ ------ ------- ------ ------- ------ ------ ------- > 19 FINISH 64 MAIL > >So something that should point to 1 is pointing to 19 (or vice versa) I had this very problem and discovered it was due to a corrupt MX_SYSTEM_QUEUE.FLQ_CTL file. To fix it, I completely stopped MX and used a QUEUE CREATE command to create a new file. That completely fixed the problem. Also, make _sure_ all your cluster nodes are using the exact same files, just to be on the safe side. -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Thu, 21 Aug 1997 10:00:04 -0500 Sender: Date: Thu, 21 Aug 1997 09:59:57 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B91A1.719FAAE4.75@ALPHA.WKU.EDU> Subject: RE: Leftover files yet again. "Brian Tillman, x8425" writes: > >Andy.Jack@nottingham.ac.uk writes: > >>Note that there's no entry number 1. >> >>Doing mcp queue sho 1 gives >> >>$ mcp queue sho 1 >> >>Entry# Status Size Source Agent Entry# Status Size >>------ ------ ------- ------ ------- ------ ------ ------- >> 19 FINISH 64 MAIL >> >>So something that should point to 1 is pointing to 19 (or vice versa) > >I had this very problem and discovered it was due to a corrupt >MX_SYSTEM_QUEUE.FLQ_CTL file. To fix it, I completely stopped MX and used a >QUEUE CREATE command to create a new file. That completely fixed the problem. > Actually, in the case Andy described, there is no queue corruption. >Also, make _sure_ all your cluster nodes are using the exact same files, just to >be on the safe side. But this is a good point I should mention: it is especially important that all cluster members accessing MX files have the same value for MX_FLQ_NODE. That logical value is used to create lock names, and if MX_FLQ_NODE is different on your systems, you won't be locking properly.... Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Thu, 21 Aug 1997 10:06:36 -0500 Sender: Date: Thu, 21 Aug 1997 08:06:24 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B9191.94EBF7D7.13@CVOBKU.CVO.MP.USBR.GOV> Subject: RE: Leftover files yet again. > From: MX%"MX-List@MadGoat.com" 21-AUG-1997 08:03:17.90 > To: MX%"MX-List@MadGoat.com" > CC: > Subj: RE: Leftover files yet again. > On Thu, 21 Aug 1997 09:59:57 -0500, Hunter Goatley said: Hunter Goatley writes: > "Brian Tillman, x8425" writes: > > > >Andy.Jack@nottingham.ac.uk writes: > > > >>Note that there's no entry number 1. > >> > >>Doing mcp queue sho 1 gives > >> > >>$ mcp queue sho 1 > >> > >>Entry# Status Size Source Agent Entry# Status Size > >>------ ------ ------- ------ ------- ------ ------ ------- > >> 19 FINISH 64 MAIL > >> > >>So something that should point to 1 is pointing to 19 (or vice versa) > > > >I had this very problem and discovered it was due to a corrupt > >MX_SYSTEM_QUEUE.FLQ_CTL file. To fix it, I completely stopped MX and used a > >QUEUE CREATE command to create a new file. That completely fixed the problem. > > > Actually, in the case Andy described, there is no queue corruption. > > >Also, make _sure_ all your cluster nodes are using the exact same files, just to > >be on the safe side. > > But this is a good point I should mention: it is especially important > that all cluster members accessing MX files have the same value for > MX_FLQ_NODE. That logical value is used to create lock names, and if > MX_FLQ_NODE is different on your systems, you won't be locking > properly.... > Which COULD lead to queue file corruption... > Hunter > ------ > Hunter Goatley, Process Software, http://www.process.com > TCPware & MultiNet: The Best TCP/IP for OpenVMS > http://www.madgoat.com/hunter.html -HWM ================================================================================ Archive-Date: Thu, 21 Aug 1997 11:07:26 -0500 Sender: From: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) Subject: RE: Leftover files yet again. Date: 21 Aug 1997 15:04:10 GMT Message-ID: <5thlda$ckr@paperboy.ccc.nottingham.ac.uk> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU In article <009B919F.874FBCE6.52@ALPHA.WKU.EDU>, Hunter Goatley writes: >cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: >> >>On my return, I discover that a small number of files have been created >>and apparently left after the entries have gone from the queue. These >>are files like 1.SMTP_INFO and there is something very, very odd going >>on. >> >Not at all! [...] >When MX FLQ Manager or MX Router does it next PURGE pass, both the >files for entry 1 and entry 19 should be deleted. And so they were! I hesitate to say it, but it's beginning to look as if things are functioning normally again. It's now nearly three hours since I last restarted MX and all temporary files seem to be being deleted properly. >> $ dir/before="-0-0:20:00" mx_root:[queue.%] >> No files found Those sweet words! "No files found"! Thanks again, everybody. Andy ------------------------------------------------------------------------ Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB Nottingham NG7 2RD, UK Phone: +44 115 951 3328 ================================================================================ Archive-Date: Fri, 22 Aug 1997 07:08:57 -0500 Sender: From: cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) Subject: RE: Leftover files yet again. Date: 22 Aug 1997 11:53:46 GMT Message-ID: <5tjuka$1nk@paperboy.ccc.nottingham.ac.uk> Reply-To: MX-List@MadGoat.com To: MX-List@WKUVX1.WKU.EDU In article <009B91A1.719FAAE4.75@ALPHA.WKU.EDU>, Hunter Goatley writes: [...] > >But this is a good point I should mention: it is especially important >that all cluster members accessing MX files have the same value for >MX_FLQ_NODE. That logical value is used to create lock names, and if >MX_FLQ_NODE is different on your systems, you won't be locking >properly.... Is this something else that I've missed? NONE of the members of our cluster have that logical defined. MX_FLQ_NODE_NAME is defined, with startling originality, as "VAX". Andy ------------------------------------------------------------------------ Cripps Computing Centre, RFC-822: Andy.Jack@nottingham.ac.uk University of Nottingham, X.400: G=Andy;S=Jack;O=Nottingham;P=UK.AC;C=GB Nottingham NG7 2RD, UK Phone: +44 115 951 3328 ================================================================================ Archive-Date: Fri, 22 Aug 1997 07:33:20 -0500 Sender: Date: Fri, 22 Aug 1997 07:33:14 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B9256.1CF643AB.6@ALPHA.WKU.EDU> Subject: RE: Leftover files yet again. cczanj@axp0.ccc.nottingham.ac.uk (Andy Jack) writes: > >Is this something else that I've missed? NONE of the members of our >cluster have that logical defined. MX_FLQ_NODE_NAME is defined, with >startling originality, as "VAX". > My mistake. The logical is MX_FLQ_NODE_NAME. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Mon, 25 Aug 1997 09:47:08 -0500 Sender: Message-ID: From: "Worlton, Thomas" Reply-To: MX-List@MadGoat.com To: "'MX-List@MadGoat.com'" Subject: MX Wish List--Mailing Lists Date: Mon, 25 Aug 1997 09:47:39 -0500 MIME-Version: 1.0 Content-Type: text/plain Hunter: I run an international mailing list and just returned from an international conference where it was suggested we hold regional meetings two years from now. It would be very helpful if I could send mail to a sub-set of the list--say those users from particular countries or institutions (i.e *.jp, *.edu, *.anl.gov, etc.). Maybe the only way to do that would be to create a new list with the subset of addresses, but if you have any other ideas, let me know. I have written a routine to sort the list and maybe I could use that to create a file with "add" commands to create a new sub-list. The problem with that is that it is not dynamic. Tom Worlton ================================================================================ Archive-Date: Mon, 25 Aug 1997 09:58:54 -0500 Sender: Date: Mon, 25 Aug 1997 09:58:48 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B94C5.F229599B.14@ALPHA.WKU.EDU> Subject: RE: MX Wish List--Mailing Lists "Worlton, Thomas" writes: > >Hunter: I run an international mailing list and just returned from an >international conference where it was suggested we hold regional >meetings two years from now. It would be very helpful if I could send >mail to a sub-set of the list--say those users from particular countries >or institutions (i.e *.jp, *.edu, *.anl.gov, etc.). Maybe the only way >to do that would be to create a new list with the subset of addresses, >but if you have any other ideas, let me know. I have written a routine >to sort the list and maybe I could use that to create a file with "add" >commands to create a new sub-list. The problem with that is that it is >not dynamic. > Yes, but that's the only way to do it now. I'll add this to the wish list, but I don't see any way right off the top of my head to make this practical. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Mon, 25 Aug 1997 11:06:06 -0500 Sender: Date: Mon, 25 Aug 1997 12:05:54 -1300 From: william@fractl.tam.cornell.edu Reply-To: MX-List@MadGoat.com To: mx-list@madgoat.com Message-ID: <009B94D7.B3AAB2D2.7@fractl.tam.cornell.edu> Subject: Forwarding mail using the site interface. Hello I am working on a site interface that will forward mail during certain time of the day based on a configuration file that a user can set up. One of the "features" is to both forward the mail and deliver it locally. What I am doing right now is taking the parameters from site_deliver.com procedure and then using a program that uses the VMS Callable Mail Routines to deliver the mail to the local recipient. I was then using mx_site_in.exe to forward the message. mx_site_in takes at least two parameters and up to three. The third paramter is described as follows: [origin-address] should be the RFC822 route address of the sender including the surrounding angle brackets. If the third parameter is ommitted the address of the user runing the program will be used as the origin of the message. Now what I am after is to have the origional sender's address to be in the from field. So that it may be directly replied to. So I used the the mx_site_in as follows: $mx_enter :== $mx_exe:mx_site_in $mx_enter message.tmp destination.tmp "" message.tmp and destination.tmp were both created by MX and copied by my very simple site_deliver.com file. however when the mail is sent the following headers result: the joe address ends up in the return path but not in the from address. the results are the same on my SUN-OS system. I want the message to appear form the JOE address not from my own. Otherwise all the forwarded mail will appear from my MX machine rather than from its true source. I don't want to have to look at the headers to figure out where the message really came from. Can someone explain how I can accomplish what I am after using MX ? From: MX%"william@FRACTL.TAM.CORNELL.EDU" 19-AUG-1997 14:00:49.88 To: MX%"WTH1@cornell.edu" CC: Subj: SITE Test Return-Path: Received: from cornell.edu (132.236.56.6) by fractl.tam.cornell.edu (MX F5.0) with ESMTP; Tue, 19 Aug 1997 14:00:43 -1300 Received: (from daemon@localhost) by cornell.edu (8.8.5/8.8.5) id OAA08712 for william@fractl.tam.cornell.edu; Tue, 19 Aug 1997 14:00:38 -0400 (EDT) Received: from fractl.tam.cornell.edu (FRACTL.TAM.CORNELL.EDU [128.84.253.91]) by cornell.edu (8.8.5/8.8.5) with ESMTP id OAA08703 for ; Tue, 19 Aug 1997 14:00:37 -0400 (EDT) Received: by fractl.tam.cornell.edu (MX F5.0) with SITE; Tue, 19 Aug 1997 14:00:32 -1300 Received: from cornell.edu (132.236.56.6) by fractl.tam.cornell.edu (MX F5.0) with ESMTP; Tue, 19 Aug 1997 13:47:06 -1300 Received: (from daemon@localhost) by cornell.edu (8.8.5/8.8.5) id NAA28859 for william@fractl.tam.cornell.edu; Tue, 19 Aug 1997 13:47:05 -0400 (EDT) X-PH: V4.1@cornell.edu (Cornell Modified) From: william@FRACTL.TAM.CORNELL.EDU Received: from fractl.tam.cornell.edu (FRACTL.TAM.CORNELL.EDU [128.84.253.91]) by cornell.edu (8.8.5/8.8.5) with ESMTP id NAA28839 for ; Tue, 19 Aug 1997 13:47:03 -0400 (EDT) X-PH: V4.1@cornell.edu (Cornell Modified) Received: by fractl.tam.cornell.edu (MX F5.0) id 1; Tue, 19 Aug 1997 13:47:00 -1300 Date: Tue, 19 Aug 1997 13:46:59 -1300 To: WTH1@cornell.edu (William T Holmes) Message-ID: <009B902E.D482D7F8.1@fractl.tam.cornell.edu> Subject: SITE Test ================================================================================ Archive-Date: Mon, 25 Aug 1997 11:55:56 -0500 Sender: Date: Mon, 25 Aug 1997 9:55:48 -0700 From: Matt Madison Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: william@fractl.tam.cornell.edu Message-ID: <970825095548.2029bf2e@Cisco.COM> Subject: RE: Forwarding mail using the site interface. >I am working on a site interface that will forward mail during certain >time of the day based on a configuration file that a user can set up. > >One of the "features" is to both forward the mail and deliver it locally. OK, with you so far. >What I am doing right now is taking the parameters from site_deliver.com >procedure and then using a program that uses the VMS Callable Mail Routines >to deliver the mail to the local recipient. OK, although you could have MX do that for you, too. >I was then using mx_site_in.exe to forward the message. mx_site_in takes >at least two parameters and up to three. The third paramter is described >as follows: > >[origin-address] should be the RFC822 route address of the sender including >the surrounding angle brackets. If the third parameter is ommitted the address >of the user runing the program will be used as the origin of the message. > >Now what I am after is to have the origional sender's address to be in >the from field. So that it may be directly replied to. And that should be what you're getting, if you're leaving the message file that was given to you by MX_SITE intact. >So I used the the mx_site_in as follows: > >$mx_enter :== $mx_exe:mx_site_in >$mx_enter message.tmp destination.tmp "" > >message.tmp and destination.tmp were both created by MX and copied by >my very simple site_deliver.com file. > >however when the mail is sent the following headers result: > >the joe address ends up in the return path but not in the from address. >the results are the same on my SUN-OS system. I want the message to appear >form the JOE address not from my own. Otherwise all the forwarded mail will >appear from my MX machine rather than from its true source. I don't want to >have to look at the headers to figure out where the message really came >from. If you want a From: header in the message, you'll have to massage the message file yourself; the third parameter to MX_SITE_IN affects only the Return-Path (aka the envelope return address). But now I'm confused, because it sounds as though you want the forwarding to happen transparently -- i.e., without any modification to the message at all. If you want to forward the message on transparently, all you do is create a file containing the new destination address, and pass to MX_SITE_IN the original message, the name of the file where you put the new destination address, and the original origin address as handed to you by MX_SITE. -Matt -- Matthew Madison | | madison@cisco.com Cisco Systems | 101 Cooper St. | Santa Cruz, CA 95060 USA | +1 408 457 5390 ================================================================================ Archive-Date: Mon, 25 Aug 1997 14:14:29 -0500 Sender: Date: Mon, 25 Aug 1997 15:14:15 -1300 From: william@fractl.tam.cornell.edu Reply-To: MX-List@MadGoat.com To: MX-LIST@MADGOAT.COM Message-ID: <009B94F2.03A28142.1@fractl.tam.cornell.edu> Subject: Site interface Question > From: MX%"MX-List@MadGoat.com" 25-AUG-1997 13:07:12.75 > Subj: RE: Forwarding mail using the site interface. > > Return-Path: > Received: from alpha.wku.edu (161.6.18.4) by fractl.tam.cornell.edu (MX F5.0) > with ESMTP; Mon, 25 Aug 1997 13:07:08 -1300 > X-ListName: Message Exchange Discussion List > Warnings-To: <> > Errors-To: > Sender: > Date: Mon, 25 Aug 1997 9:55:48 -0700 > From: Matt Madison > Reply-To: MX-List@MadGoat.com > To: MX-List@MadGoat.com > CC: william@fractl.tam.cornell.edu > Message-ID: <970825095548.2029bf2e@Cisco.COM> > Subject: RE: Forwarding mail using the site interface. > X-List-Subscribe: > X-List-Unsubscribe: > X-List-Help: > > >I am working on a site interface that will forward mail during certain > >time of the day based on a configuration file that a user can set up. > > > >One of the "features" is to both forward the mail and deliver it locally. > > OK, with you so far. > > >What I am doing right now is taking the parameters from site_deliver.com > >procedure and then using a program that uses the VMS Callable Mail Routines > >to deliver the mail to the local recipient. > > OK, although you could have MX do that for you, too. > What would I have to do for this ? I assume I would have to create another host entry in the the MX database and have its path set to local rather than site. Is this correct ? > >I was then using mx_site_in.exe to forward the message. mx_site_in takes > >at least two parameters and up to three. The third paramter is described > >as follows: > > > >[origin-address] should be the RFC822 route address of the sender including > >the surrounding angle brackets. If the third parameter is ommitted the address > >of the user runing the program will be used as the origin of the message. > > > >Now what I am after is to have the origional sender's address to be in > >the from field. So that it may be directly replied to. > > And that should be what you're getting, if you're leaving the message file > that was given to you by MX_SITE intact. I assume by this statement that in order to change the from address that I must change the From line that is contained in the message file in order to have a transparent forward. Is this correct ? Then I would change the destination file to be the address that I wish to have the mail forwarded to. Is this also correct ? > > >So I used the the mx_site_in as follows: > > > >$mx_enter :== $mx_exe:mx_site_in > >$mx_enter message.tmp destination.tmp "" > > > >message.tmp and destination.tmp were both created by MX and copied by > >my very simple site_deliver.com file. > > > >however when the mail is sent the following headers result: > > > >the joe address ends up in the return path but not in the from address. > >the results are the same on my SUN-OS system. I want the message to appear > >form the JOE address not from my own. Otherwise all the forwarded mail will > >appear from my MX machine rather than from its true source. I don't want to > >have to look at the headers to figure out where the message really came > >from. > > If you want a From: header in the message, you'll have to massage the > message file yourself; the third parameter to MX_SITE_IN affects only > the Return-Path (aka the envelope return address). But now I'm confused, > because it sounds as though you want the forwarding to happen transparently -- > i.e., without any modification to the message at all. That is what I am after. > > If you want to forward the message on transparently, all you do is create a > file containing the new destination address, and pass to MX_SITE_IN the > original message, the name of the file where you put the new destination > address, and the original origin address as handed to you by MX_SITE. > But that's eactly what I did. Here is my site_deliver.com: $ OPEN/WRITE ZZZ FRACTL_1:[HOLMES.MXTEST]FOO.DAT $ WRITE ZZZ P1 $ WRITE ZZZ P2 $ WRITE ZZZ P3 $ WRITE ZZZ P4 $ CLOSE ZZZ $ COPY 'P2' FRACTL_1:[HOLMES.MXTEST]MESSAGE.TMP $ COPY 'P3' FRACTL_1:[HOLMES.MXTEST]ADDRESS.TMP $ STATUS = 1 $ EXIT 'STATUS Here is foo.dat fractl.tam.cornell.edu MX_ROOT:[SITE]SITE_MSG_DA7B9F28_009B902E_202000D2.TMP;1 MX_ROOT:[SITE]SITE_ADR_DA7B9F28_009B902E_202000D2.TMP;1 Here is Message.tmp: Received: from cornell.edu (132.236.56.6) by fractl.tam.cornell.edu (MX F5.0) with ESMTP; Tue, 19 Aug 1997 13:47:06 -1300 Received: (from daemon@localhost) by cornell.edu (8.8.5/8.8.5) id NAA28859 for william@fractl.tam.cornell.edu; Tue, 19 Aug 1997 13:47:05 -0400 (EDT) From: william@fractl.tam.cornell.edu Received: from fractl.tam.cornell.edu (FRACTL.TAM.CORNELL.EDU [128.84.253.91]) by cornell.edu (8.8.5/8.8.5) with ESMTP id NAA28839 for ; Tue, 19 Aug 1997 13:47:03 -0400 (EDT) X-PH: V4.1@cornell.edu (Cornell Modified) Received: by fractl.tam.cornell.edu (MX F5.0) id 1; Tue, 19 Aug 1997 13:47:00 -1300 Date: Tue, 19 Aug 1997 13:46:59 -1300 To: WTH1@CORNELL.EDU (William T Holmes) Message-ID: <009B902E.D482D7F8.1@fractl.tam.cornell.edu> Subject: SITE Test Hi This is a site test Hope it works ======================================================================= ___Bill william@fractl.tam.cornell.edu wth1@Cornell.edu ======================================================================= And Here is Address.tmp (which I modified from to > -Matt > > -- > Matthew Madison | | madison@cisco.com > Cisco Systems | 101 Cooper St. | Santa Cruz, CA 95060 USA | +1 408 457 5390 ======================================================================= ___Bill william@fractl.tam.cornell.edu wth1@Cornell.edu ======================================================================= ================================================================================ Archive-Date: Wed, 27 Aug 1997 10:35:38 -0500 Sender: Message-ID: <3.0.32.19970827103552.00ac4210@moses.acu.edu> Date: Wed, 27 Aug 1997 10:35:53 -0500 To: mx-list@madgoat.com From: Tom Dolan Reply-To: MX-List@MadGoat.com Subject: Searchable Archives MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" MX People, I would like to make my mailing list archives searchable like the ones at: http://www2.wku.edu:81/scripts/mxarchive/as_init.com?MX-List Is the software for this already available or will it be included in MX V5? Tom Dolan Dolan@Bible.acu.edu 202 Bible Building Systems Manager ACU Box 29454 College of Biblical Studies Abilene, TX 79699 Abilene Christian University 915.674.3706 ================================================================================ Archive-Date: Wed, 27 Aug 1997 10:47:36 -0500 Sender: Date: Wed, 27 Aug 1997 10:47:29 -0500 From: Hunter Goatley Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B965F.144923CC.4@ALPHA.WKU.EDU> Subject: RE: Searchable Archives Tom Dolan writes: > >I would like to make my mailing list archives searchable like the >ones at: > >http://www2.wku.edu:81/scripts/mxarchive/as_init.com?MX-List > >Is the software for this already available or will it be included >in MX V5? > Yes, sort of. I created the indexes using the WWWVMSINDEX stuff that came with the OSU HTTP server. However, I didn't like their .COM files, so I made my own. If I have the time, I'll include my files in MX V5.0, and possibly the relevant WWWVMSINDEX files too. Hunter ------ Hunter Goatley, Process Software, http://www.process.com TCPware & MultiNet: The Best TCP/IP for OpenVMS http://www.madgoat.com/hunter.html ================================================================================ Archive-Date: Thu, 28 Aug 1997 23:37:54 -0500 Sender: From: geoffrob@stmarks.pp.catholic.edu.au (Geoff Roberts) Subject: POP3 problems - Password hashing? Date: Fri, 29 Aug 1997 02:04:49 GMT Message-ID: <34062b81.2859976@news.adl.aone.net.au> Reply-To: MX-List@MadGoat.com MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: MX-List@WKUVX1.WKU.EDU Hi all, Having a problem with our mail system, running MX 4.2 and IUPOP3 (CMUIP Variant) It appears that the version of IUPOP I am using cannot handle VMS 6.0 password hashing correctly. It worked fine initially, but now after a password change, it won't allow the user to POP in to check/send mail from their PC's running Pegasus. Mine is still working, but I haven't changed my password since I put the POP server on last week either. ;^) The user can telnet or login from a terminal ok. Does anyone have/know of a VMS 6.0 capable version of IUPOP3 (or any CMUIP compatible POP3 server for that case) I tried to cheat by deleting the users account and recreating it, but get the same result. Maybe using a different UIC would help? To make matters worse, we don't have a (usable) C compiler (unless you count GNU's port, which I don't) ie we don't have VAX or DEC C. All we have is Macro, Bliss (which I am starting to play with) and Cobol (erk) and I'm a tech not a programmer anyway. Any help appreciated. I see a bit about MX 5.0 being mentioned, is it perhaps possible it will have inbuilt POP3 support? Please? Pretty please??? Thanks all ================================================================================ Archive-Date: Fri, 29 Aug 1997 04:37:07 -0500 Sender: Date: Fri, 29 Aug 1997 10:35:57 BST From: Andy Harper Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: A.HARPER@kcl.ac.uk Message-ID: <009B97EF.CC89D8BC.1444@alder.cc.kcl.ac.uk> Subject: RE: POP3 problems - Password hashing? >Having a problem with our mail system, running MX 4.2 and IUPOP3 >(CMUIP Variant) > >It appears that the version of IUPOP I am using cannot handle VMS 6.0 >password hashing correctly. It worked fine initially, but now after a >password change, it won't allow the user to POP in to check/send mail from their PC's running Pegasus. >Mine is still working, but I haven't changed my password since I put >the POP server on last week either. ;^) >The user can telnet or login from a terminal ok. Does anyone >have/know of a VMS 6.0 capable version of IUPOP3 (or any CMUIP >compatible POP3 server for that case) >I tried to cheat by deleting the users account and recreating it, but >get the same result. Maybe using a different UIC would help? You dont say what version if IUPOP3 you are using but, if its the old CMU hacked version, then it doeas not support the later password hashing algorithm and you will need to upgrade. A new version V1.9 Beta14, with many extras and proper CMU support (VIA NETLIB) is available from my ftp server, as well as from the main support site. (although a beta, its very stable): ftp://ftp2.kcl.ac.uk/zip/iupop3_v1_9_beta14.zip >To make matters worse, we don't have a (usable) C compiler (unless you >count GNU's port, which I don't) ie we don't have VAX or DEC C. >All we have is Macro, Bliss (which I am starting to play with) and >Cobol (erk) and I'm a tech not a programmer anyway. >Any help appreciated. Well you need to specify what type of system and what VMS version you've got so that we know what binaries to try and find. You should also install NETLIB from the madgoat ftp site (this comes with binaries) I can supply you with binaries for OpenVMS 6.2 on both VAX and ALPHA built with DEC C and liniked against NETLIB 2.1 >I see a bit about MX 5.0 being mentioned, is it perhaps possible it >will have inbuilt POP3 support? Please? Pretty please??? Unlikely, MX concerns itself with SMTP based mail transfer and local delivery. POP is a means to access a mailbox from a remote system and is COMPLETELY independent of MX. A combination of MX and IUPOP3 V1.9 works very well though. Regards, Andy Harper Kings College London ================================================================================ Archive-Date: Fri, 29 Aug 1997 08:45:40 -0500 Sender: Date: Fri, 29 Aug 1997 09:44:25 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B97E8.99A0A2C0.4@swdev.si.com> Subject: RE: POP3 problems - Password hashing? Geoff Roberts (geoffrob@stmarks.pp.catholic.edu.au) writes: >Does anyone have/know of a VMS 6.0 capable version of IUPOP3 (or any CMUIP >compatible POP3 server for that case)? Since CMUIP now includes a UCX compatibility mode, I'd try using the POP server that comes with UCX. Also, the version of IUPOP3 on the latest freeware CD may work (although I have no personal experience). -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company ================================================================================ Archive-Date: Fri, 29 Aug 1997 09:59:51 -0500 Sender: Date: Fri, 29 Aug 1997 16:54:38 +0200 Message-ID: <97082916543830@vw.tci.uni-hannover.de> From: stenns@vw.tci.uni-hannover.de (Michael Stenns) Reply-To: MX-List@MadGoat.com To: MX-LIST@wkuvx1.wku.edu Subject: Re: POP3 problems - Password hashing? Geoff Roberts (geoffrob@stmarks.pp.catholic.edu.au) wrote: >Hi all, >Having a problem with our mail system, running MX 4.2 and IUPOP3 >(CMUIP Variant) > <...> >To make matters worse, we don't have a (usable) C compiler (unless you >count GNU's port, which I don't) ie we don't have VAX or DEC C. >All we have is Macro, Bliss (which I am starting to play with) and >Cobol (erk) and I'm a tech not a programmer anyway. >Any help appreciated. >I see a bit about MX 5.0 being mentioned, is it perhaps possible it >will have inbuilt POP3 support? Please? Pretty please??? >Thanks all There is a new IUPOP3 version (still in beta state) that can be used via NETLIB 2.x on OpenCMU. The IUPOP3 beta is available from http://www.tci.uni-hannover.de/pub/iupop3/index.htmlx or ftp://www.tci.uni-hannover.de/pub/iupop3/ or ftp://ftp2.kcl.ac.uk/zip/iupop3_v1_9_beta14.zip It can be build with the GNU port of gcc 2.7.x (or VAXC/DECC of course). For questions regarding the IUPOP3 server a mailing list (iupop3-users@indiana.edu , subscription address is majordomo@indiana.edu) does exist. Brian Tillman, x8425 (tillman@swdev.si.com) wrote: >Since CMUIP now includes a UCX compatibility mode, I'd try using the POP server >that comes with UCX. Also, the version of IUPOP3 on the latest freeware CD may >work (although I have no personal experience). AFAIK the current version of OpenCMU does not have a UCX compatibility mode. ====================================================================================== I had send this letter first to the newsgroup vmsnet.mail.mx, but I got an access denied message from the list processor. The problem might be that I am subscribed as stenns@vw.tci.uni-hannover.de, while the news client generates the return address STENNS@vw.tci.uni-hannover.de. Is there any chance to avoid the rejection? Michael -- Michael Stenns Email: stenns@vw.tci.uni-hannover.de ================================================================================ Archive-Date: Fri, 29 Aug 1997 11:04:49 -0500 Sender: Date: Fri, 29 Aug 1997 12:03:38 -0400 From: "Brian Tillman, x8425" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com Message-ID: <009B97FC.0C34EEA0.1@swdev.si.com> Subject: Re: POP3 problems - Password hashing? >AFAIK the current version of OpenCMU does not have a UCX compatibility mode. You're not well-informed, then. Here's the announcement (from 28-Feb-1996!), from Andy Harper. -- Brian Tillman Internet: tillman_brian at si.com Smiths Industries, Inc. tillman at swdev.si.com 4141 Eastern Ave., MS239 Addresses modified to prevent Grand Rapids, MI 49518-8727 SPAM. Replace "at" with "@" This opinion doesn't represent that of my company I've added Clive Nicolson's LIBCMUII socket library to my ftp server. It provides a socket library that has UCX compatibility, to ease the porting problems that have beset CMU users over the years. I've included clive's summary of the package below. This description is also available as LIBCMUII.DOC in the server directory. Thanks to clive for his work on this. URL: ftp://ftp2.kcl.ac.uk/cmu-tcpip/libcmuii_*.bck Note, with the current abysmal state of the transatlantic links that access our site, I'ld strongly recommened that you access us when the US and UK are asleep - that's roughly between 3AM and 8AM GMT. Otherwise, you'll spend days trying! (There's hope on the horizon though; the UK links are being upgraded in the near future) Regards, Andy Harper Kings College London LIBCMU-II/UCX$IPC_SHR, "UCX Compatibility" for CMU-OpenVMS-IP Author: Clive Nicolson clive@bedroom.gen.nz 25-FEB-1996 This package is built on LIBCMU and SOCKETSHR, in the sense that I've just put a new roof on the house. For best performance of this package one should also use the 6.6-5K version of CMUIP's IPACP.EXE. LIBCMU-II is package that contains a sharable image that provides the socket entry points as does LIBCMU, and also entry points that implement DEC's UCX$IPC_SHR interface. LIBCMU is a socket interface for CMU-OpenVMS-IP. The UCX$IPC_SHR interface is used by VAXCRTL.EXE and DECC$SHR.EXE (and their corresponding object libraries) to provide a IP stack independent socket interface at the user level. ie BSD select is offered by VAXCRTL.EXE as vaxc$select (or by DECC$SHR.EXE as decc$select) which dynamically links to ucx$select via the logical UCX$IPC_SHR. The UCX$IPC_SHR interface provided should match the forth coming DECC RTL 7.0 which has (latent) support for a complete BSD socket interface. "UCX Compatibility" is also provided by (I think): i) DEC's UCX (of course) ii) PathWay Runtime (twg/Wollongong) iii) ? LIBCMU-II also has sharable images (VAXC_RTL_SHR and DECC_RTL_SHR) to extend, with VAXCRTL and DECC$SHR, the user interface to near complete BSD socket i/o, this is done via calls to UCX$IPC_SHR and extra code. This could be extended to be better, if I knew what DEC have announced for this. Sources for these bits and pieces will help providers of other socket stacks/interfaces build a UCX$IPC_SHR interface. This stuff should enable the user interface to all socket/IP packages under VAX/VMS to be raised to the C RTL user level. Some 3 party stacks may already support UCX$IPC_SHR. This is similar to the aims of SOCKETSHR and NETLIB. I've tested (using LIBCMU_IPC_SHR) the NCSA Mosaic versions 2.4 through 2.7b2a that DEC distribute with recent releases of DW-Motif, it hardly knows that DEC's UCX in not really being used. Latest version found at: http://www.openvms.digital.com/openvms/freeware/mosaic/index.html I've also built HTTPD1-9B to use the VAXCRTL socket interface, ie so that it is independant of IP stack. ------------------------------------------------------------------------------ - LIBCMUII is distributed in 3 (disjoint) pieces: LIBCMUII_exe.bck, Contains the shareable images, the netdb data files and minimum installation files, currently linked against VMS 6.1. This is all you need to use LIBCMUII. for VMS 6.1 or later, well maybe you should also find IPACP.EXE 6.6-5K or better (in CMUTEK665K.bck). LIBCMUII_src.bck Contains the source files to build the executables found in LIBCMUII_exe.bck . LIBCMUII_bld.bck Contains the object files, and libraries built with files found in LIBCMUII_src.bck, these are not required but short circuit initial build work. ================================================================================ Archive-Date: Sat, 30 Aug 1997 01:02:38 -0500 Sender: From: "Geoffrey D. Roberts" To: mx-list@madgoat.com Date: Sat, 30 Aug 1997 15:35:15 +0930 MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Subject: RE: POP3 problems - Password hashing? Reply-To: MX-List@MadGoat.com Thanks for the response guys, > You dont say what version if IUPOP3 you are using but, if its the old CMU > hacked version, then it doeas not support the later password hashing algorithm > and you will need to upgrade. Sorry, my oversight, V1.7 for CMU by Andrew Greer. > A new version V1.9 Beta14, with many extras and proper CMU support > ftp://ftp2.kcl.ac.uk/zip/iupop3_v1_9_beta14.zip Will give this version a try. Thanks for that. > Well you need to specify what type of system and what VMS version you've got > so that we know what binaries to try and find. You should also install > NETLIB from the madgoat ftp site (this comes with binaries) Vax 6000-310, VMS 6.0, CMUIP 6.6-5A NETLIB 2.1 MX 4.2 No DECNET. (well it's on there somewhere, but have nothing for it to talk to, so it isn't loaded.) I haven't put the 6.6-5K upgrade in as it is linked against VMS 6.1 and is binaries only (no obj or source) I have netlib 2.1 installed. > I can supply you with binaries for OpenVMS 6.2 on both VAX and ALPHA built > with DEC C and liniked against NETLIB 2.1 I gather 6.2 binaries won't work on a 6.0 system? Or will they? > MX concerns itself with SMTP based mail transfer and local > delivery. POP is a means to access a mailbox from a remote system and is > COMPLETELY independent of MX. Yes, although some PC based systems include a POP3 server. (Mercury for Netware is a good example) Just thought MX might be going to do the same. No harm in asking. ;^) > A combination of MX and IUPOP3 V1.9 works very > well though. Initially, V1.7 was too, I think it's just this password thing that is causing me problems, although I have noted some of the known bugs with this version (although they have not occurred here yet) Geoffrey D. Roberts Computer Systems Manager Saint Marks College Port Pirie, South Australia geoffrob@stmarks.pp.catholic.edu.au Phone Work:61-8-8632-2800 Home:61-8-8633-0619 Fax :61-8-8633-0104 Mobile:61-17-160-166 ================================================================================ Archive-Date: Sat, 30 Aug 1997 04:58:10 -0500 Sender: Date: Sat, 30 Aug 1997 02:58:01 -1300 From: "Henry W. Miller" Reply-To: MX-List@MadGoat.com To: MX-List@MadGoat.com CC: henrym@SACTO.MP.USBR.GOV Message-ID: <009B9878.FE4BE95B.90@CVOBKU.CVO.MP.USBR.GOV> Subject: Re: POP3 problems - Password hashing? > From: MX%"MX-List@MadGoat.com" 29-AUG-1997 08:04:42.75 > To: MX%"MX-LIST@wkuvx1.wku.edu" > CC: > ====================================================================================== > I had send this letter first to the newsgroup vmsnet.mail.mx, but I got an > access denied message from the list processor. > The problem might be that I am subscribed as stenns@vw.tci.uni-hannover.de, > while the news client generates the return address STENNS@vw.tci.uni-hannover.de. > > Is there any chance to avoid the rejection? > > Michael > -- > Michael Stenns > Email: stenns@vw.tci.uni-hannover.de Michael, You may have to subscribe again from the new address, and then send a: SET NOMAIL to the server from your new, alternate address to avoid getting duplicate copies of the messages. (I was forced to close the CMU list several months ago because of SPAM, so anyone who posts via USENET and wants the messages to propagagte to the list must be registered on the list, even if they just turn around and set it to NOMAIL). -HWM