Anne Wainwright wrote: > >Sorry if this is a bit vague.
And my guesses at what's going on may be a bit vague too. >I give an excerpt from a log file today. One of many with slightly >different wordings depending on which server was responding. Some >reporting me as 'exceeding the limit for bounced messages'. I don't know about 'exceeding the limit for bounced messages', but see below. >Oct 30 16:02:39 jason postfix/smtp[11547]: B6883E81A7: >to=<[email protected]>, >relay=smtp.isdsl.net[196.26.208.193]:25, conn_use=5, delay=45, >delays=0.02/0/0.24/45, dsn=5.0.0, status=bounced (host >smtp.isdsl.net[196.26.208.193] said: 550 Sorry, I don't accept bounce >messages with a invalid recipient. (in reply to RCPT TO command)) The recipient here is [email protected] which smtp.isdsl.net thinks is an invalid address. Apparently smtp.isdsl.net is one of the names of one of the IPs of default-mx.imaginet.co.za which is the MX for the fables.co.za domain. >(some of the bounce addresses _were_ valid, I checked one, got an >answer back from the addressee, but that could be a different issue) The address that smtp.isdsl.net thinks is invalid is [email protected], not [email protected]. The most likely reason for this is that Postfix's main.cf on smtp.isdsl.net (if the MTA on smtp.isdsl.net is Postfix) is lacking recipient_delimiter = + This is required for VERP to work with the default delimiter (+). It tell Postfix that the recipient fables-list-bounces+dmorris=inext.co.za is actually fables-list-bounces. Without this, Postfix attempts to deliver to the full VERPed recipient fables-list-bounces+dmorris=inext.co.za which is invalid. >Verp is on as you can see. But on a run last week without verp we were >left in the same situation as described here. Then there's more to it than what I'm saying. Does smtp.isdsl.net (and the other relays) know how to deliver or relay to all 10 of the fables-list addresses including fables-list-bounces, fables-list-confirm, fables-list-request, fables-list-owner, fables-list-join, fables-list-leave, fables-list-subscribe and fables-list-unsubscribe? >There were a lot similar to this. I note that get_bounce_info shows no >bounce information when I ran it after the run. Because none of the bounce DSNs ever got delivered to Mailman. [...] >Further, since I am running this through a relay, the relevant email >account fables-list-bounces@ shows no activity whatsoever today. No >bounce mail passed through there. I'm not sure what you're saying, but I would expect something in the MTA log of smtp.isdsl.net at Oct 30 16:02:39 because there was at least a connect, a MAIL FROM and a RCPT TO from jason. OTOH, if all you are saying is that no mail was relayed to fables-list-bounces@, that's because it was all rejected at the relay. >Not a doubt that I should have had a lot of bounces on this list which >has been dormant for months, and the log file did show plenty. > >Any idea where I might find the holdup? I suspect either the postfix >configuration or something to do with the smtp relay isp. I don't think >the bounces get back to the isp email account, are then not downloaded, >are then not processed. ??? Right. They get rejected as undeliverable by smtp.isdsl.net, either because it doesn't know how to relay for [email protected] or it doesn't recognize that [email protected] should be relayed the same as [email protected]. If it's the latter, and you don't control the MTA there and can't teach it that '+' is a valid separator between a local part and a suffix, it might recognize '-' in this role in which case see the FAQ at <http://wiki.list.org/x/D4CE>. -- Mark Sapiro <[email protected]> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan ------------------------------------------------------ Mailman-Users mailing list [email protected] http://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: http://mail.python.org/mailman/options/mailman-users/archive%40jab.org
