Hi Mark, thanks for your quick answer!
Am 09.05.19 um 14:31 schrieb Mark Sapiro: > I assume there's nothing in /var/lib/mailman/logs/smtp-failure since if > there were, these would also be recorded as bounces. /var/lib/mailman/logs/smtp-failure does not have any entries related to the issue. >> But overall I'm only able to find roughly half in either /var/log/mail or >> /var/lib/mailman/logs/*. [...] > Mailman did complete SMTP to Postfix for all 416 recipients. The > /var/lib/mailman/logs/smtp log entry verifies that. What happened after > that is a Postfix question. Ack. IIUC it seems in deed as if messages were lost between Mailman and Postfix, without any of them noticing/logging it. The only indication I could find that there /might/ be problem was this warning with a suitable timestamp: --- snip --- ... postfix/master...: warning: service ... has reached its process limit "10": new clients may experience noticeable delays --- snap --- >> Do I have any chance of finding out what happened to the mails to the >> addresses which >> don't appear in any of the log files? > > > I don't think this should happen without Postfix logging something, but > if I have to guess, I'd guess you are not VERPing delivery to Postfix > and the list is not personalized so Mailman is delivering to Postfix in > transactions with up to SMTP_MAX_RCPTS (default 500) per transaction and > Postfix's smtpd_recipient_limit is less. Postfix is configured with 1000: smtpd_recipient_limit = 1000 Neither SMTP_MAX_RCPTS nor VERP_DELIVERY_INTERVAL was set explicitly, so I guess the defaults from /usr/lib/mailman/Mailman/Defaults.py are being used: SMTP_MAX_RCPTS = 500 VERP_DELIVERY_INTERVAL = 0 > I suggest you set SMTP_MAX_RCPTS = 10 or 20 in mm_cfg.py or better set > VERP_DELIVERY_INTERVAL = 1 to enable VERP on all deliveries. I will try with VERP_DELIVERY_INTERVAL set to 1, to see whether all messages are accounted for then. IIRC, setting VERP_DELIVERY_INTERVAL = 1 effectively means to ignore SMTP_MAX_RCPTS and only submit 1 mail at a time to the MTA. Curiously for the message that was not delivered to the specific user I did /not/ find a message by Postfix like the following in /var/log/mail: --- snip --- ... to=<LISTADDR>, relay=local, delay=0.58, delays=0.01/0/0/0.57, dsn=2.0.0, status=sent (delivered to command: /usr/lib/mailman/mail/mailman post <LIST>) --- snap --- To me that looks as if Postfix did in deed not complete sending all messages. I will monitor the situation with these reduced rate settings applied. Thanks again -- Till -- Dipl.-Inform. Till Dörges doer...@pre-sense.de Tel. +49 - 40 - 244 2407 - 0 Fax +49 - 40 - 244 2407 - 24 PRESENSE Technologies GmbH Sachsenstr. 5, D-20097 HH Geschäftsführer/Managing Directors AG Hamburg, HRB 107844 Till Dörges, Jürgen Sander USt-IdNr.: DE263765024 Wir sind wieder auf dem BSI IT-Sicherheitskongress 21.-23. Mai 2019 – Bonn ------------------------------------------------------ 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