[pkg-fetchmail-maint] Bug#314509: fetchmail: IMAP timeouts & wrong message counting

Matthias Andree matthias.andree at gmx.de
Sat Oct 29 10:26:06 UTC 2005


On Sat, 13 Aug 2005, Stepan Golosunov wrote:

> Package: fetchmail
> Version: 6.2.5-12sarge1
> Followup-For: Bug #314509
> 
> I have seen this bug several times. To reproduce it I sent 20 messages with
> $ for ((i=1;i<=20;++i)); do mail -s "Test#$i" stepan at golosunov.pp.ru < /dev/null ; done
> Then while fetching these messages I started
> $ for ((i=1;i<=10;++i)); do mail -s "Test#X$i" stepan at golosunov.pp.ru < /dev/null ; done
> I have got wrong message message count (17 messages per minute is too
> high for me) and timeout after fetching 10 Test#X messages (f21.log).
> On the next run I have got only 2 messages (f22.log).
> 
> None of test messages were lost.
> 
> I have also reproduced this bug using dovecot-imapd as a server (within
> qemu). In this case timeout comes before fetching additional messages
> and they are fetched later; some messages have X-IMAPbase: header and
> once I have seen ")" (#312415?) in the message body (generated from /dev/null).

Please use Bourne or POSIX shell syntax for your demonstration scripts.
This one is simple enough to redo in Bourne, so nevermind this time, but
please remember next time.

Chances are this has long since been fixed, please ask all bug
submitters to re-test with 6.2.9-rc6.

Besides that, I have - for now - been unable to reproduce this, I have
however only Dovecot and Cyrus here to test.

If one of you can reproduce this problem, please let me know which IMAP
server you have been using. I don't have CommuniGate servers to test
with.

-- 
Matthias Andree



More information about the pkg-fetchmail-maint mailing list