[Mailman-Users] Problems after changing server network setup

[Mark Sapiro]

Aaron Crosman wrote:

>Earlier today, for various reasons, we had to move our Mailman server to
>a different segment of our network.  This changed the internal IP, and
>several other network details (although I don't think any others that
>impact mailman).  Starting just after I made the move, I am not getting
>a large number of error messages in the error log that look like this:
>
> 
>
>Mar 03 12:40:53 2008 (2881) Uncaught runner exception: No terminating
>boundary and no trailing empty line

<snip>

>Mar 03 12:40:53 2008 (2881) Ignoring unparseable message:
>1178048805.0749691+0654fbb7bc97eff421d692fe6978466fca6d68c1
>
> 
>
>Mail appears to be going through, but since these messages are new, they
>make me nervous that I missed something in the move.  I reconfigured
>exim and Mailman and replaced all references to the old IP with the new.
>All the other functions appear to be working.  Any suggestions about
>what's going on, and if a I need to be worried?


These are caused by malformed MIME messages that can't be parsed by the
Python email library. They are almost certainly spam. Perhaps your
prior configuration did a better job of spam filtering ahead of
Mailman.

In your case, whatever message was queued as
1178048805.0749691+0654fbb7bc97eff421d692fe6978466fca6d68c1.pck is
irrecoverably gone, but as I said, it was most likely spam anyway.

Beginning in Mailman 2.1.10, the queue entry is moved aside so you can
actually see the message if you want.

If you are concerned, you could find the MTA log entries that show
delivery to Mailman, and at least see the sender of the message.

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