![](https://secure.gravatar.com/avatar/65a96637782467256a7b7bfe5591cdf3.jpg?s=120&d=mm&r=g)
I thought everything was going great(after I removed all the queue files, after having fixed mm_cfg.py and permissions), until things just ground to a halt again. 400+ files in out, mail's extremely slow- its taking well over 10 minutes from the time postfix calls bin/mailman to when something's logged in logs/post. The qrunner processes are running, and taking a few percent of CPU time, but apparently doing nothing, or next to it.
The only error we're getting is every 15 minutes: Nov 20 22:02:55 2003 (18412) No such list "[list": ...with one space after the :
I've been unsuccessful in tracking down where that could possibly be coming from....
Brett
"They that give up essential liberty to obtain temporary safety deserve neither liberty nor safety." - Ben Franklin http://www.users.cloud9.net/~brett/
![](https://secure.gravatar.com/avatar/8391289e5979e9b962979e9d20b9a344.jpg?s=120&d=mm&r=g)
On Thu, 20 Nov 2003, Brett Dikeman wrote:
The only error we're getting is every 15 minutes: Nov 20 22:02:55 2003 (18412) No such list "[list": ...with one space after the :
I've been unsuccessful in tracking down where that could possibly be coming from....
That message might come from any of these cgi accesses:
./Mailman/Cgi/admindb.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/confirm.py: syslog('error', 'No such list "%s": %s', listname, e) ./Mailman/Cgi/edithtml.py: syslog('error', 'No such list "%s": %s', listname, e) ./Mailman/Cgi/listinfo.py: syslog('error', 'No such list "%s": %s', listname, e) ./Mailman/Cgi/options.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/private.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/rmlist.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/subscribe.py: syslog('error', 'No such list "%s": %s\n', listname, e)
Is it precisely every 15 minutes? If so it sounds more like a cron job...
...which suggests yet another item that has changed dramatically over the versions--have you adjusted the crontab entries appropriately? Another random thought--have you created the mailman mailing list? (There's a fairly new requirement that you have that single specifically named list from which certain messages get sent.)
Dale Newfield <Dale@Newfield.org>
"They that can give up essential liberty to obtain a little safety deserve neither liberty nor safety." - Benjamin Franklin, on the Statue of Liberty
![](https://secure.gravatar.com/avatar/01aa7d6d4db83982a2f6dd363d0ee0f3.jpg?s=120&d=mm&r=g)
On Fri, 2003-11-21 at 08:46, Dale Newfield wrote:
On Thu, 20 Nov 2003, Brett Dikeman wrote:
The only error we're getting is every 15 minutes: Nov 20 22:02:55 2003 (18412) No such list "[list": ...with one space after the :
I've been unsuccessful in tracking down where that could possibly be coming from....
That message might come from any of these cgi accesses:
./Mailman/Cgi/admindb.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/confirm.py: syslog('error', 'No such list "%s": %s', listname, e) ./Mailman/Cgi/edithtml.py: syslog('error', 'No such list "%s": %s', listname, e) ./Mailman/Cgi/listinfo.py: syslog('error', 'No such list "%s": %s', listname, e) ./Mailman/Cgi/options.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/private.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/rmlist.py: syslog('error', 'No such list "%s": %s\n', listname, e) ./Mailman/Cgi/subscribe.py: syslog('error', 'No such list "%s": %s\n', listname, e)
I suspect the error message he's seeing is unrelated to his performance problem (not positive about that though).
Is it precisely every 15 minutes? If so it sounds more like a cron job...
As I mentioned to Brett in pvt email, it could also be the BounceRunner that's hosing things. By default, every 15 minutes it processes any queued bounces, and if these were bounces to the site list, it will lock every list on the system to register the bounce. If he's getting tons of bounces, or there's some bug that causes the BounceRunner to crash (or the Python process to core dump???), he could be leaving stale locks around that would wedge his lists.
There should be some evidence of this in the logs/bounce file, or in stale locks in the lock directory. He could also fiddle with the REGISTER_BOUNCES_EVERY variable in BounceRunner.py to see if that corresponds to the length of time until he's hosed.
If that turns out to be the case, then we'll need to investigate further to find out what's giving his bounce runner grief.
-Barry
participants (3)
-
Barry Warsaw
-
Brett Dikeman
-
Dale Newfield