[Mailman-Users] Uncaught Runner Exception

Mark Sapiro mark at msapiro.net
Sun Apr 18 00:33:30 CEST 2010

James Kemp wrote:
>Over the last couple of weeks I've noticed that a number of legitimate 
>e-mails have failed to be delivered. All of these have been shunted and 
>an error like the one repeated below is written to the error log.
>Apr 17 21:22:28 2010 (2558) Uncaught runner exception: [Errno 104] 
>Connection reset by peer
>   File "/var/lib/mailman/Mailman/Queue/IncomingRunner.py", line 153, in 
>     sys.modules[modname].process(mlist, msg, msgdata)
>   File "/var/lib/mailman/Mailman/Handlers/SpamAssassin.py", line 76, in 
>     score, symbols = check_message(mlist, str(msg))
>   File "/var/lib/mailman/Mailman/Handlers/SpamAssassin.py", line 62, in 
>     res = connection.check(spamd.SYMBOLS, message)
>   File "/var/lib/mailman/Mailman/Handlers/spamd.py", line 122, in check
>     response = fp.readline()
>   File "/usr/lib/python2.6/socket.py", line 406, in readline
>     data = self._sock.recv(self._rbufsize)
>error: [Errno 104] Connection reset by peer

It appears that you (or someone) has installed the
Spamassassin.py/spamd.py custom handler from
<https://bugs.launchpad.net/mailman/+bug/557991> and included it in
the GLOBAL_PIPELINE or a lists' pipeline.

The "Connection reset by peer" error is occurring while spamd.py is
waiting for a response from the spamd daemon.

I suspect this is occurring because of a time out. Prior to SA 3.3,
there was no limit on how long spamd could take to process a message.
If there were DNS issues with RBL lookups for example, there could be
very long delays. SA 3.3 has introduced a 300 second time limit, after
which it just returns what it found plus a hit on the

If you have SA 3.3, you could configure a shorter time_limit and see if
that helps.

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

More information about the Mailman-Users mailing list