[Mailman-Users] high memory usage and long running time for mailman Python process

[Mark Sapiro]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Terrence Brannon wrote:
|
| Hello, I am running Debian/etch and the user "list" has a Python process
| which takes up 83.2% of my memory and has been running for more than 40
| minutes.
|
| My suspicion is that some sort of queue of moderator messages is very
| large.
|
| In fact, /var/lib/mailman/ is very large... I cannot access the
| administrator interface via the web:
|
| 6733    ./qfiles
| 21551   ./messages
| 60071   ./lists/asciidoc-discuss
| 60093   ./lists
| 288500  ./data
| 380881  .
|
|
| In looking through archives, I discovered I have more than 60,000 held
| messages for one mailing list:
|
| schemelab at li2-168:/var/lib/mailman/data$ ls -l | wc -l
| 64688
|
| but for some reason attempting to discard even one is not working... it
| just sits there and hangs:
|
| schemelab at li2-168:/var/lib/mailman/data$ sudo ../bin/discard
| heldmsg-asciidoc-discuss-59527.pck
| Password:
|
| (no return to shell after 5 minutes)
|

and

| I posted this
| <http://www.mail-archive.com/mailman-users%40python.org/msg49373.html>
| and basically the whole list is having other discussions like I never
| even posted...


The second item has been addressed. I just want to add that sometimes we
even go on vacation and don't take our computers.

As far as your first issue is concerned, bin/discard is the best tool to
delete these held messages. If you can't wait for it to complete, or for
some reason it won't complete, you can remove
lists/asciidoc-discuss/request.pck which is the 'moderator queue'. This
will allow you to access your list. As Brad suggested
(<http://mail.python.org/pipermail/mailman-users/2008-May/061765.html>),
it would be good to stop Mailman, remove the request.pck, remove all the
data/heldmsg-asciidoc-discuss-*.pck files, move aside and remake the
mailman/data directory to compact it, and start Mailman.

Then, consider setting the lists General Options->max_days_to_hold to
some small, non-zero value so this doesn't happen again

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

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (MingW32)

iD8DBQFIRLyAVVuXXpU7hpMRAjatAKCwzrUW+FABCW97aImX5k1YnmU6qwCdERg9
sHysnF/4e2CzxKJjYoHCjbw=
=pNeK
-----END PGP SIGNATURE-----