[Mailman-Users] mailman gets stuck, stops sending messages

[Brad Knowles]

At 11:29 AM -0500 2004-11-17, Steven J. Owens wrote:

>       Even with -auxw I get nothing.

	You may need to toss in a second "w".  I know that I need to do 
this on some of the systems I admin.

>       Where is mailmanctl normally supposed to be started from?  I
>  don't see any reference to mailmanctl in /etc/cron.d/mailman.

	It is not started from cron.

>       I do see mailmanctl in /etc/init.d/mailman.  Does that mean that
>  the mailmanctl process should be started on system startup and should
>  remain running until system shutdown?

	Correct, it is started on boot and should remain running until shutdown.

>       I still want to solve the problem, but meanwhile, would there be
>  any negative effect to adding an hourly cron job to restart
>  mailmanctl?  Or perhaps to run a script to check that
>  /var/lib/mailman/data/master-qrunner.pid is a valid process ID, else
>  restart mailmanctl?

	The latter might be a good idea, but I think the former could be a bad one.


	If possible, I would want that script to gather as much 
information as possible about the mailmanctl process that died or 
does not currently exist (including all left-over children, etc...), 
before doing the actual restart.

	There may be something else going on that would help you solve 
the real problem, if you can figure out what that is.

>  17:07:30, puff at darksleep:~> ps -auxw| grep mailmanctl
>  bash: pipe error: Too many open files in system

	Ouch!

>       Could be the system is running out of filehandles and mailman's
>  falling over as a result?

	That could certainly cause problems for any process on the 
system, Mailman included.

>       Could mailman be consuming a whole lot of filehandles?

	It will consume filehandles while it is opening, closing, and 
otherwise processing messages, or while pipermail is accessing the 
system archive as a response to a user request.  But it should only 
be processing a small number of messages at any one given moment in 
time, and shouldn't be running out of filehandles.

	It sounds to me like there is a more serious problem going on 
here that needs to be located and resolved before Mailman will be 
able to function normally.

-- 
Brad Knowles, <brad at stop.mail-abuse.org>

"Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety."

     -- Benjamin Franklin (1706-1790), reply of the Pennsylvania
     Assembly to the Governor, November 11, 1755

   SAGE member since 1995.  See <http://www.sage.org/> for more info.