[Mailman-Users] Mailman log error
msapiro at value.net
Fri Feb 3 20:12:14 CET 2006
Clovis Tristao wrote:
>I am having problem with one of my Mailman lists, it stopped to function
>to one lists.
>This appearing errors and several bounce-events-????.pck, when send
>e-mail to list.
Errors in the error log? What are they?
>Fedora Core 4
>Jan 29 05:09:49 2006 (1858) Low level smtp error: Server not connected,
>msgid: <43DA0C85.6070909 at mydomain.com>
>Jan 29 05:09:49 2006 (1858) delivery to users at mydomain.com failed with
>code -1: Server not connected
This was presumably a transient error in communicating with your
outgoing MTA. You could check your MTA logs for the same time to see
if there's more info there.
>-rw-rw-r-- 1 mailman mailman 4335 Fev 3 14:25
>-rw-rw-rw- 1 mailman mailman 0 Fev 3 13:43 bounce-events-21966.pck
>-rw-rw-r-- 1 mailman mailman 716831 Fev 3 12:48 heldmsg-docentes-l-79.pck
>-rw-rw-r-- 1 mailman mailman 45113 Fev 3 12:45
>-rw-rw-r-- 1 mailman mailman 76828 Fev 3 08:58 heldmsg-posgrad-l-41.pck
>-rw-rw-r-- 1 mailman mailman 34012 Fev 2 13:52 heldmsg-posgrad-l-40.pck
>-rw-rw-rw- 1 mailman mailman 5645 Fev 1 08:24 bounce-events-02358.pck
>-rw-rw-rw- 1 mailman mailman 0 Jan 30 10:37 bounce-events-01858.pck
>-rw-rw-rw- 1 mailman mailman 1169 Jan 20 14:27 bounce-events-02362.pck
>What they are these archives(held*,bounce*)? I have some archives of
>bounce? It's normal?
The BounceRunner queues bounces as they are received and only processes
them at intervals determined by the Defaults.py/mm_cfg.py setting
REGISTER_BOUNCES_EVERY which defaults to 15 minutes. The bounces are
queued in the bounce-events-*.pck files. If the bounce runner is
stopped or restarted when there is an outstanding file, the file can
be orphaned and the bounces lost. All the older bounce-events-*.pck
files are such and can be deleted.
The heldmsg-<listname>-*.pck files are messages for <listname> waiting
moderator action. When the moderator discards, rejects or approves the
messages, the files will be removed.
None of these things would be responsible for one list not functioning.
In Mailman 2.1.5, it is possible that a malformed message of some kind
in the lists/<listname>/digest.mbox file can stop a list from
processing, but there should be 'shunted' messages and tracebacks in
Mailman's 'error' log if this is the case. What's in the 'error' log?
If there is a lists/<listname>/digest.mbox file for the problem list,
you could try moving it out of the lists/<listname>/ directory and see
if that allows the list to work. The real solution, if this is the
problem, is to find the bad message in lists/<listname>/digest.mbox
and fix or delete it.
Mark Sapiro <msapiro at value.net> The highway is for gamblers,
San Francisco Bay Area, California better use your sense - B. Dylan
More information about the Mailman-Users