[ mailman-Bugs-1598087 ] Error mail may be accumulated in qfiles

SourceForge.net noreply at sourceforge.net
Fri Nov 17 06:36:15 CET 2006


Bugs item #1598087, was opened at 2006-11-16 15:55
Message generated for change (Comment added) made by msapiro
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=1598087&group_id=103

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: None
Group: None
Status: Open
Resolution: None
Priority: 5
Private: No
Submitted By: Tokio Kikuchi (tkikuchi)
Assigned to: Nobody/Anonymous (nobody)
Summary: Error mail may be accumulated in qfiles

Initial Comment:
Mailman/Queue/Runner.py is called by other runners and resposible for dequeueing from one queue and enqueue into another after processing.  In _oneloop() function it shunts error messages during processing but leaves them in the original queue directory by noting 'Ignoring ...' message in the error log.  These left errorneous messages are never treated again and may be accumulated and slow down the queue processing.  These messages should be shunted also.


----------------------------------------------------------------------

>Comment By: Mark Sapiro (msapiro)
Date: 2006-11-16 21:36

Message:
Logged In: YES 
user_id=1123998
Originator: NO

If I understand the issue correctly, the problem is that the .bak file is
left in the original queue after the parse error. This is not a
recoverable error as explained in the code comment, thus I don't think the
proper action is to shunt the message. Rather, I think the proper action is
to continue to log the event and just remove the .bak file. Not removing
the .bak file was an oversight in the original backup implementation.

My suggested patch is attached.

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=100103&aid=1598087&group_id=103


More information about the Mailman-coders mailing list