[Date Prev][Date Next]
Re: (ITS#5451) syncprov_checkpoint deadlock bugfix?
> On Wed, 9 Apr 2008, email@example.com wrote:
>> Does this help? From my fiddling with ITS#5340 (REP_ENTRY_MODIFIABLE).
>> I do not understand syncprov's handling of REP_ENTRY_MUSTRELEASE though.
>> (For one thing it seems to assume that REP_ENTRY_MUSTRELEASE is set if
>> and only if rs.sr_entry->e_private != NULL. Which is possibly true with
>> back-bdb but seems a shaky assumption in general.)
> Almost, except that it triggers an abort.. Using be_entry_release_r() was
> not correct, as it ended up calling entry_free(). It works if I changes
> it to use overlay_entry_release_ov() instead, as in the alternative patch
> at the end.
> I have put a test script that shows this deadlock on:
> It deadlocks without this patch, it and the rest of the test suite succeed
> with it :-) I haven't tested it in production though..
This last patch is now committed to HEAD. I am ignoring the first patch; it is
unsafe because it unlocks the si_csn_rwlock too soon. (si_ctxcsn is being used
directly in the modify request and must be held consistent until that completes.)
-- Howard Chu
CTO, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc/
Chief Architect, OpenLDAP http://www.openldap.org/project/