Tend to Pending Moderator Requests breaks on a few lists

[Lenhatfield]

After upgrading from MM 2.0.3 to 2.1.5, I find that on
just a couple of my lists when I access the web-admin
interface and try to follow the "Tend to Pending
Moderator Requests" link, I get the following bug
announcement:

Bug in Mailman version 2.1.5

We're sorry, we hit a bug!

If you would like to help us identify the problem,
please email a copy of this page to the webmaster for
this site with a description of what happened. Thanks!
Traceback:

Traceback (most recent call last):
  File "/c/mm2/scripts/driver", line 87, in run_main
    main()
  File "/c/mm2/Mailman/Cgi/admindb.py", line 231, in main
    num = show_pending_subs(mlist, form)
  File "/c/mm2/Mailman/Cgi/admindb.py", line 274, in
show_pending_subs
    pendingsubs = mlist.GetSubscriptionIds()
  File "/c/mm2/Mailman/ListAdmin.py", line 138, in
GetSubscriptionIds
    return self.__getmsgids(SUBSCRIPTION)
  File "/c/mm2/Mailman/ListAdmin.py", line 130, in
__getmsgids
    ids = [k for k, (op, data) in self.__db.items() if
op == rtype]
ValueError: unpack tuple of wrong size

Python information:

Variable Value
sys.version 2.3.4 (#1, Jul 21 2004, 23:31:00) [GCC
2.95.3 20010315 (release)]
sys.executable /usr/bin/python
sys.prefix /usr/local
sys.exec_prefix /usr/local
sys.path /usr/local
sys.platform linux2

Environment variables:

Variable Value
HTTP_REFERER http://wiz.cath.vt.edu/mailman/admin/hel-l
SERVER_SOFTWARE Apache
SCRIPT_NAME /mailman/admindb
SERVER_SIGNATURE
REQUEST_METHOD GET
HTTP_KEEP_ALIVE 300
SERVER_PROTOCOL HTTP/1.1
QUERY_STRING
HTTP_ACCEPT_CHARSET ISO-8859-1,utf-8;q=0.7,*;q=0.7
HTTP_USER_AGENT Mozilla/5.0 (Macintosh; U; PPC Mac OS
X Mach-O; en-US; rv:1.7) Gecko/20040628 Firefox/0.9.1
HTTP_CONNECTION keep-alive
HTTP_COOKIE
hel-l+admin=280200000069299d05417328000000346333373433316630373764643863663
53661356532393638663262343937613764313365663234
SERVER_NAME wiz.cath.vt.edu
REMOTE_ADDR 68.185.125.105
PATH_TRANSLATED /www/hel-l
SERVER_PORT 80
SERVER_ADDR 128.173.51.243
DOCUMENT_ROOT /www
PYTHONPATH /c/mm2
SCRIPT_FILENAME /c/mm2/cgi-bin/admindb
SERVER_ADMIN <email address hidden>
HTTP_HOST wiz.cath.vt.edu
REQUEST_URI /mailman/admindb/hel-l
HTTP_ACCEPT
text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q
=0.8,image/png,*/*;q=0.5
GATEWAY_INTERFACE CGI/1.1
REMOTE_PORT 61615
HTTP_ACCEPT_LANGUAGE en-us,en;q=0.5
HTTP_ACCEPT_ENCODING gzip,deflate
UNIQUE_ID QQWdLICtM-MAAEji16U
PATH_INFO /hel-l

[http://sourceforge.net/tracker/index.php?func=detail&aid=998384&group_id=103&atid=100103]

[Zot (zot)]

I am seeing the same exact thing.

Traceback (most recent call last):
  File "/var/mailman/scripts/driver", line 87, in run_main
    main()
  File "/var/mailman/Mailman/Cgi/admindb.py", line 231, in main
    num = show_pending_subs(mlist, form)
  File "/var/mailman/Mailman/Cgi/admindb.py", line 274, in
show_pending_subs
    pendingsubs = mlist.GetSubscriptionIds()
  File "/var/mailman/Mailman/ListAdmin.py", line 138, in
GetSubscriptionIds
    return self.__getmsgids(SUBSCRIPTION)
  File "/var/mailman/Mailman/ListAdmin.py", line 130, in
__getmsgids
    ids = [k for k, (op, data) in self.__db.items() if op ==
rtype]
ValueError: unpack tuple of wrong size

sys.version 2.3.3 (#1, May 7 2004, 10:31:40) [GCC 3.3.3
20040412 (Red Hat Linux 3.3.3-7)]
sys.executable /usr/bin/python
sys.prefix /usr
sys.exec_prefix /usr
sys.path /usr
sys.platform linux2

The environment vars were a bit off for

PATH_TRANSLATED

and

DOCUMENT_ROOT

But evertyhing else seems to be working.

Is anyone looking at this bug?

[Followme (followme)]

I've stumbled upon this too. I've sent a patch to mailman-devel without
checking for the same bug reported, so I'll try ot attach it here too.
As soon as I find out how to.

[Followme (followme)]

Third time's a charm! I can't find a way to attach file to the report,
so
here's the patch:
http://list.iskon.hr/~jelly/mailman-2.1.5-_UpdateRequests.diff

Here's my mail to -devel:
http://mail.python.org/pipermail/mailman-developers/2005-March/017956.
html

After applying, running _UpdateRecords() on the affected lists using
withlist
fixes the problem here. Running bin/update would probably work too.

Bug 986186 might be the same thing.

[Knut Auvor Grythe (knut-a-grythe)]

This bug still exists in mailman 2.1.11rc2. It is caused by the upgrade script storing broken data whenever the structure of an admin request needs to be changed. The attached patch both fixes the error and cleans up the data so mailman no longer crashes. Thus either running bin/update or running _UpdateRecords() using withlist should solve the problem, as mentioned in the previous post.

PS: For all I know, this might be the exact same patch mentioned earlier. However, all links above seem dead.

[Mark Sapiro (msapiro)]

The above link to the Mailman-Developers post is valid. It is just split across lines. The unsplit link is
  http://mail.python.org/pipermail/mailman-developers/2005-March/017956.html

I have committed the fix part of the patch on both the 2.1 and 2.2 branches. There is an ongoing discussion about the recovery part of the patch on Mailman-Users starting at
  http://mail.python.org/pipermail/mailman-users/2008-September/063339.html