Right you are. It's Postfix's 'sendmail' binary that appears to be generating the error, although I'm still perplexed that it claims to be in response to a non-LMTP response from lmtpd.Some investigation confirmed my initial suspicions: that Cyrus lmtpd is returning to Postfix's lmtp delivery daemon the message: "sendmail: fatal: sender@xxxxxxxxxxx(76): Message file too big" Since that doesn't start with [2-5]\d\d, it's not a valid LMTP response, and Postfix is doing the right thing by complaining. Why would Cyrus return that? Obviously something is wrong with the delivery, but under what circumstances would a) lmtpd invoke the sendmail binary?
That would happen if the user in question has a sieve redirect rule.
b) sendmail fail thusly?
I don't use Postfix, but maybe it has a separate config file for direct submission that (in your case) doesn't allow mails of that size. Just guessing ...
c) lmtpd pass the failure message, unadorned, back to the sender? The latter in particular seems like broken behavior; shouldn't lmtpd return something like '5xx sendmail: fatal: ...'?
Sure, but that's a known bug :-) At least I seem to remember that it came up some time ago.
-- Sebastian Hagedorn - RZKR-R1 (Flachbau), Zi. 18, Robert-Koch-Str. 10 Zentrum für angewandte Informatik - Universitätsweiter Service RRZK Universität zu Köln / Cologne University - Tel. +49-221-478-5587
Attachment:
pgp57SwCMoKdQ.pgp
Description: PGP signature
---- Cyrus Home Page: http://cyrusimap.web.cmu.edu/ Cyrus Wiki/FAQ: http://cyrusimap.web.cmu.edu/twiki List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
