[Tickets #711] Segfault in IMAP_Tree

[bugs at bugs.horde.org]

DO NOT REPLY TO THIS MESSAGE. THIS EMAIL ADDRESS IS NOT MONITORED.

Ticket URL: http://bugs.horde.org/ticket/?id=711
-----------------------------------------------------------------------
 Ticket     | 711
 Updated By | Michael Slusarz <slusarz at mail.curecanti.org>
 Summary    | Segfault in IMAP_Tree
 Queue      | Horde Base
 Version    | HEAD
 State      | Assigned
 Priority   | 3. High
 Type       | Bug
 Owners     | Michael Slusarz
-----------------------------------------------------------------------


Michael Slusarz <slusarz at mail.curecanti.org> (2004-12-14 23:21) wrote:

Unfortunately, I still can't produce a segfault so that I can get a
backtrace so it remains very frustrating to try and track this down.  ugh.

Since it is not reproducible, it makes me think that maybe it doesn't have
anything to do with the logic of the code.  Maybe some kind of concurrent
access to the IMAP_tree object that is somehow changing the object in
mid-stream of one of the occurrances, which is affecting the other
occurrance (any kind of change in the object/data structure during a sort()
call, for example, would very easily cause a segfault in there).  Maybe
accessing the IMP_Tree object in Horde_block_imp_summary could be done via a
copy of the object, for example, which would fix this problem?  I'm starting
to run short of ideas (actually, I've been running short of ideas for
months...)