[Mailman-Users] trouble running mailman in a virtual environment

[Mark Sapiro]

wiren wrote:
>
>Version 2.1.7, the latest available.


The current release is 2.1.8 <http://sourceforge.net/projects/mailman>.
For 2.1.7 also see
<http://sourceforge.net/forum/forum.php?forum_id=545674>.


>Then a problem arose, mailman got stuck. My consultant went to see the 
>log and noticed several *interrupted system calls*.


Which log?


>As a workaround we put a script to restart mailman every 5 minuts.


Not necessarily a good idea. See
<http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq04.068.htp>


>All went smooth for a few days until last friday, when mailman got stuck 
>again and restarting it did not solve the problem.
>
>My consultants watched the logs more carefully and they reported several 
>"interrupted system calls".


Which logs?


>After a couple of days, mailman started to run again (we guess it was 
>due to the cron script) and released all the previous mail.
>
>Incoming mail is always received regularly, no user receive errors and 
>it is always immediately put into the on line archive of the lists, it 
>just do not get delivered untile mailman starts delivering again...


This would appear to be a problem with OutgoingRunner or SMTP delivery
to the outgoing MTA.


>Do you know of any problem relating to running mailman in virtual 
>environment? Any other hint? Unfortunately, log files are too concise to 
>get a good idea of what the problem is related to...


What log files are you looking at? What is in Mailman's 'error' and
'smtp-failure' logs?

When Mailman 'stops', has OutgoingRunner died? What's in Mailman's
'qrunner' log?

See
<http://www.python.org/cgi-bin/faqw-mm.py?req=show&file=faq03.014.htp>

-- 
Mark Sapiro <msapiro at value.net>       The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan