On 03/06/2017 10:31 AM, Bruce Harrison wrote: > I'm running Mailman 2.1.7 > > In the past month or two. I'm seeing Mailman delaying sending out messages. > In the logs I see where it's posted, then in /var/logs/maillog I see sending > start immediately and then somewhere along the line a delay of an hour or > two, then it finishes sending. Or it delays initially then starts sending. > This hasn't been happening in the past. We in the process of moving a lot of > our mail users to O365 > Not sure what's going on or how best to troubleshoot. Any ideas?
I assume from the above that your concern is the delays following this > Post Log > Mar 03 13:34:57 2017 (1223) post to information from mun...@utm.edu, > size=37061, > message-id=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > success log entry and not with the delay between this > /var/logs/maillog > Mar 3 09:11:23 mailman sendmail[17197]: v23FBNsO017197: > from=<mun...@utm.edu>, size=35583, class=0, nrcpts=1, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, > relay=mail-dm3nam03lp0023.outbound.protection.outlook.com [207.46.163.23] and the rest. When the post log entry is written at 13:34:57, Mailman has finished the delivery of the message for all recipients to sendmail. Further delays are entirely within sendmail. > Mar 3 13:34:56 mailman sendmail[22801]: v23JYu9r022801: > from=<information-boun...@mailman.utm.edu>, size=37061, class=-30, > nrcpts=150, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1] (may be forged) > Mar 3 13:34:56 mailman sendmail[22801]: v23JYu9s022801: > from=<information-boun...@mailman.utm.edu>, size=37061, class=-30, > nrcpts=500, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1] (may be forged) > Mar 3 13:34:57 mailman sendmail[22801]: v23JYu9t022801: > from=<information-boun...@mailman.utm.edu>, size=37061, class=-30, > nrcpts=500, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1] (may be forged) > Mar 3 13:34:57 mailman sendmail[22801]: v23JYu9u022801: > from=<information-boun...@mailman.utm.edu>, size=37061, class=-30, > nrcpts=500, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1] (may be forged) > Mar 3 13:34:57 mailman sendmail[22801]: v23JYu9v022801: > from=<information-boun...@mailman.utm.edu>, size=37061, class=-30, nrcpts=8, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1] (may be forged) > Mar 3 13:34:57 mailman sendmail[22801]: v23JYu9w022801: > from=<information-boun...@mailman.utm.edu>, size=37061, class=-30, nrcpts=7, > msgid=<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu>, > proto=ESMTP, daemon=MTA, relay=localhost [127.0.0.1] (may be forged) The above are sendmail's receipt of the delivered "chunks" from mailman. Mailman is dividing the recipient list into chunks by top level domain. See the comments in the code near the beginning of the definition of the process function in Mailman/Handlers/SMTPDirect.py and the chunkify function for how this is done. In your case, SMTP_MAX_RCPTS is the default 500 and your recipients for this message are split into 6 chunks of 150, 500, 500, 500, 8 and 7 recipients respectively and those are delivered to sendmail in 6 smtp transactions, each with the respective number of RCPT TO recipients. Then sendmail begins delivery. > Mar 3 13:35:02 mailman sendmail[22811]: v23JYu9v022801: > to=<trinitypr...@frontiernet.net>,<mcoch...@wljt.org>,<khi...@utfcu.org>,<trinitypas...@frontiernet.net>,<mshum...@wljt.org>,<mnpe...@comcast.net>,<mre...@wljt.org>,<mfi...@charter.net>, > delay=00:00:05, xdelay=00:00:05, mailer=relay, pri=331061, > relay=xmail.utm.edu. [10.51.0.154], dsn=2.0.0, stat=Sent > (<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu> > [InternalId=7717462] Queued mail for delivery) This chunk with 8 recipients in various domains is delivered quickly > Mar 3 13:35:02 mailman sendmail[22805]: v23JYu9s022801: > to=<thibb...@utm.edu>,<clove...@utm.edu>,<dahen...@utm.edu>,<mbro...@utm.edu>,<afull...@utm.edu>,<kroud...@utm.edu>,<bjohn...@utm.edu>,<nda...@utm.edu>,<jimer...@utm.edu>,<sburl...@utm.edu>,<thutc...@utm.edu>,<hche...@utm.edu>,<nashb...@utm.edu>,<mpari...@utm.edu>,<jjohn...@utm.edu>,<tgall...@utm.edu>,<sgold...@utm.edu>,<agard...@utm.edu>,<ddietr...@utm.edu>,<jfish...@utm.edu>,<ahuff...@utm.edu>,<kwhee...@utm.edu>,<lmccur...@utm.edu>,<kelmw...@ut.utm.edu>,<khamm...@utm.edu>,<jscot...@utm.edu>,<arich...@utm.edu>,<rprui...@utm.edu>,<ewhee...@utm.edu>,<lgib...@utm.edu>,<shalk...@utm.edu>,<akel...@utm.edu>,<phew...@utm.edu>,<da...@utm.edu>,<drya...@utm.edu>,<jda...@utm.edu>, > [more], delay=00:00:06, xdelay=00:00:06, mailer=relay, pri=15091061, > relay=xmail.utm.edu. [10.51.0.154], dsn=2.0.0, stat=Sent > (<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu> > [InternalId=7717459] Queued mail for delivery) as is this chunk with a large number of utm recipients. > . > . > . > Mar 3 13:35:30 mailman sendmail[22805]: v23JYu9s022801: > to=<jjone...@utm.edu>,<pwatk...@utm.edu>,<dmccu...@utm.edu>,<dwilliams-b...@utm.edu>,<jbyr...@utm.edu>,<ajohn...@utm.edu>,<jgoy...@utm.edu>,<sh...@utm.edu>,<mgib...@utm.edu>,<jhend...@utm.edu>,<bgi...@utm.edu>,<lwar...@utm.edu>,<mtray...@utm.edu>,<ble...@utm.edu>,<jjohn...@utm.edu>,<rjo...@utm.edu>,<sd...@utm.edu>,<sperr...@utm.edu>,<djord...@utm.edu>,<cposa...@utm.edu>,<tharr...@utm.edu>,<gfe...@utm.edu>,<cnich...@utm.edu>,<jlava...@utm.edu>,<jscho...@utm.edu>,<vo...@utm.edu>, > delay=00:00:34, xdelay=00:00:06, mailer=relay, pri=15091061, > relay=xmail.utm.edu. [10.51.0.154], dsn=2.0.0, stat=Sent > (<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu> > [InternalId=7717507] Queued mail for delivery) and this one with more utm recips. > Mar 3 14:41:17 mailman sendmail[24288]: v23JYu9w022801: > to=<chaamerw2...@yahoo.com>,<jesse.kowalew...@sodexo.com>,<kevin_mclem...@hotmail.com>,<msud...@worldtrav.com>,<njanemille...@gmail.com>,<sully12...@aol.com>,<vlhi...@gmail.com>, > delay=01:06:20, xdelay=00:00:02, mailer=relay, pri=481061, > relay=xmail.utm.edu. [10.51.0.154], dsn=2.0.0, stat=Sent > (<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu> > [InternalId=7719913] Queued mail for delivery) this chunk with 7 recips in various domains is delayed over an hour. I don't know much about sendmail, but this could be because there was some issue contacting the MX for one of the seven and the whole message was deferred. > Mar 3 14:42:43 mailman sendmail[24288]: v23JYu9r022801: > to=<dsimm...@utm.edu>,<emo...@utm.edu>,<eprit...@utm.edu>,<eree...@utm.edu>,<esimm...@utm.edu>,<espi...@utm.edu>,<ewhit...@utm.edu>,<ffack...@utm.edu>,<gbr...@utm.edu>,<geck...@utm.edu>,<gib...@utm.edu>,<hk...@utm.edu>,<ja...@utm.edu>,<jcl...@utm.edu>,<jdona...@utm.edu>,<jesbe...@utm.edu>,<jfors...@utm.edu>,<jgreenw...@utm.edu>,<jh...@utm.edu>,<jjone...@utm.edu>,<jla...@utm.edu>,<jlong...@utm.edu>,<jm...@utm.edu>,<jmath...@utm.edu>,<jonrm...@ut.utm.edu>,<jstre...@utm.edu>,<jyo...@utm.edu>,<kcur...@utm.edu>,<kelli...@utm.edu>,<keric...@utm.edu>,<kevan...@utm.edu>,<khe...@utm.edu>,<kprzyb...@utm.edu>,<krimf...@ut.utm.edu>,<kwill...@utm.edu>,<lda...@utm.edu>,<lea...@utm.edu>, > [more], delay=01:07:47, xdelay=00:00:05, mailer=relay, pri=4771061, > relay=xmail.utm.edu. [10.51.0.154], dsn=2.0.0, stat=Sent > (<320e686fea4720449a309fc08fc6f49d01023d5...@exch2010mbox2.utm.edu> > [InternalId=7719987] Queued mail for delivery) And similarly for this one with utm recips. Maybe one recip had a transient mailbox issue trhat caused the entire message to be deferred. There are some older articles in the faq at <https://wiki.list.org/>. Search titles for performance. I recommend: Consider upgrading Mailman. 2.1.7 is very old. Upgrades won't address this issue specifically, but many new features and bug fixes have occurred since 2.1.7 was released over 11 years ago. Set VERP_DELIVERY_INTERVAL = 1 in mm_cfg.py. This will improve bounce recognition, but mainly it has the same effect as setting SMTP_MAX_RCPTS = 1 so each recipient is sent in a single SMTP transaction so sendmail delivery to a large chunk of recips is not impacted by one delay. If you don't want to go that far, or if this causes Mailman -> sendmail delivery to take too long. at least try something like SMTP_MAX_RCPTS = 10 The bottom line is this is all in sendmail and this list is not the best resource for sendmail issues. Most of us are more versed in Postfix or Exim. -- Mark Sapiro <m...@msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan ------------------------------------------------------ Mailman-Users mailing list Mailman-Users@python.org https://mail.python.org/mailman/listinfo/mailman-users Mailman FAQ: http://wiki.list.org/x/AgA3 Security Policy: http://wiki.list.org/x/QIA9 Searchable Archives: http://www.mail-archive.com/mailman-users%40python.org/ Unsubscribe: https://mail.python.org/mailman/options/mailman-users/archive%40jab.org
