[PLUG] Weird Mailman Problems

Keith Lofstrom keithl at kl-ic.com
Wed May 18 00:56:43 UTC 2005


On Tue, May 17, 2005 at 12:12:29PM -0700, R Haack wrote:
...
> 1) First off I have a user that sent a message to four of our lists.  
> When I reviewed the logs her message only went to 3 of the 4 lists.  The 
> first thing I done and redone is check this users permissions and they 
> are correct.  I'm using the patch to allow us to create a list of all 
> members that can post to a list and she is a member of this list.

Make sure she is actually sending from the same address she is signed
up as.  This has caught me before.  To do this, you have to look at 
the incoming mail spool, then trace the message into mailman.

> 2) The second odd thing that is happening is that I seem to have to 
> start and stop mailman multiple times a day to get it to send messages.  
> What makes this one odd is sometimes a message will come in and go out 
> without a problem.  Other times it just gets stuck and the only way to 
> make it go is to stop mailman, clean up the locks, and start mailman.  
> I've looked at the contents of the lock files and find nothing unusual.  
> I've check that the qrunners and they seem to be running as expected.

I recently had problems with a mailman list that got too big for the
MIRV settings I had configured for postfix.  That caused some things
to snag.   I set "SMTP_MAX_RCPT=8" in /var/mailman/Mailman/mm_cfg.py
to fix that.

Your problems may be in the interaction with the SMTP program
(sendmail/postfix/exim) rather than mailman itself.   Again, look
at all the logs, all the way through the system.

I figure those answers have about a 20% chance of solving your
problems, and an additional 30% chance of inducing a D'oh/headslap
moment.  Beyond that, I will leave it to experts like Wil.

Keith

-- 
Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs



More information about the PLUG mailing list