Have either of you checked other logs, like error?
Can either of you run qrunner under something like truss or strace? (on Solaris, truss python ~mailman/cron/qrunner will run qrunner once under system-call tracing, so you can see what it's doing)
Running python -v ~mailman/cron/qrunner might be worth a try too
Harold Paulson wrote:
Rodrigo,
We seem to be having similar problems, and no one is really making any good guesses. One weird thing that I noticed is that if I kill all current qrunners, and remove all locks:
killall qrunner; rm ~mailman/locks/*then a new qrunner will pop up in a minute (from cron). If I watch it in top:
top -qUmailmanIt uses a little CPU for a minute or two, and then slowly dies off to 0% CPU usage. It never does anything that I can see, and runs for a very long time.
Could I have a corrupted msg that is hanging qrunner? The biggest file in qfiles is about 6k, so it's not a size issue...
- HWARNING: 'killall' may work very differently on various unix flavors. Don't do that on a Solaris box!
On Oct 26, 2000 at 12:53, Rodrigo Moya wrote:
I've installed mailman on a RH 6.2 machine. The POP3 server in that machine works perfectly, the messages I send to the mailing lists are logged in the logs/post file. But, no mail goes to the list members.
POP3? SMTP would be what you want to check.
Some guesses...
Check mailman's crontab. The queue runner should be a cron job, AIUI.
it seems ok. The qrunner binary is set to be executed every minute (at least this is what "crontab -u mailman -l" says.
And no /etc/cron.{allow,deny} exists, so mailman should be allowed to run cron jobs.
And "tail -f logs/qrunner" shows 2 log messages every minute. So, everything related to cron seems to work.
But still, list members don't get any message from the lists.
What else could be?
cheers
Harold Paulson Sierra Web Design haroldp@sierraweb.com http://www.sierraweb.com VOICE: 775.833.9500 FAX: 810.314.1517
Mailman-Users maillist - Mailman-Users@python.org http://www.python.org/mailman/listinfo/mailman-users