[pkg-fetchmail-maint] fetchmail: bad-headers accept breaks correct lmtp behavior

Matthias Andree matthias.andree at gmx.de
Fri Jan 16 19:21:54 UTC 2015


Am 30.06.2013 um 12:40 schrieb Pfannenschmid, P. (Binarus):
> Matthias,
> 
> I have to apologize for my stupidity.
> 
> Looking again at the excerpt which was in my last Email (sse below), I
> now think that fetchmail's lmtp _partner_ (in my case cyrus-imapd)
> rejects the message with the bad headers (at least that's the way I am
> understanding "fetchmail: LMTP< 554 5.6.0 Message contains invalid
> header" - that is an INPUT to fetchmail, right?). Thus, to solve my
> problem, I guess I'd have to configure cyrus-imapd in a way that makes
> it accept messages with bad headers via lmtp.

That's one option.

Has there ever been a bug for this? I can't seem to find one.

> But some questions remain:
> 
> Why does fetchmail try to deliver via SMTP after having got the reject
> from the lmtp partner?

It tries to create a bounce message which needs to go back to the sender
(rather than forward into the mailbox), and therefore fetchmail tries SMTP.

> And what does "Rewritten version is from ..." mean (by the way, in the
> first place, that line in the output led me to the assumption that
> fetchmail would rewrite (heal) the bad header somehow before forwarding
> the message, and therefore I didn't come to the idea that cyrus-imapd
> could actually be the cause of the problem).

That is debug tracing for how message headers got translated, for the
"is ... here" part.




More information about the pkg-fetchmail-maint mailing list