[ mailman-Patches-594831 ] Incomplete DSN bounce tracking

SourceForge.net noreply at sourceforge.net
Tue Mar 11 16:30:47 EST 2003


Patches item #594831, was opened at 2002-08-14 02:26
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594831&group_id=103

Category: bounce processing
Group: Mailman 2.0.x
>Status: Closed
>Resolution: Out of Date
Priority: 1
Submitted By: Jesús Cea Avión (jcea)
Assigned to: Nobody/Anonymous (nobody)
Summary: Incomplete DSN bounce tracking

Initial Comment:
Mailman 2.0.13 here. I didn´t check Mailman 2.1.*.

I'm very surprised nobody found & correct the problem
before.

Problem: Some valid DSN bounce notifications are not
correctly parsed.

A DSN message with several "packets", last packet is
ignored.

Reproduce: not neccesary. Code inspection should show
this patch as trivially correct.

Patch: In file Mailman/Bouncers/DSN.py, simply add two
lines:

--- DSN.py      Tue Aug 13 13:22:39 2002
+++ DSN2.py     Wed Aug 14 01:57:09 2002
@@ -105,6 +105,10 @@
         headers[string.lower(hdr)] = string.strip(val)
     # now go through all the blocks, finding the recip
address that is being
     # reported.
+
+    if len(headers) :
+         blocks.append(headers)
+
     addrs = []
     for headers in blocks:
         if string.lower(headers.get('action', '')) <>
'failed':


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

>Comment By: Thomas Wouters (twouters)
Date: 2003-03-12 01:30

Message:
Logged In: YES 
user_id=34209

I don't think this applies to Mailman 2.1, as it uses full
VERP (optionally, but enabled by default, and strongly
encouraged) and bounce detection by DSN parsing was
rewritten as well, and seems to do the right thing. But
thanks for the report, if you notice it is still current,
please file another patch or bugreport :-)


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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=300103&aid=594831&group_id=103



More information about the Mailman-coders mailing list