As a quick trial, I just replaced the mysql_backend.py in .10.4 with the one from r4342, and the issue went away. So it looks like the issue is in mysql_backend.py rather than MySQLdb or mysql. That's good news.
Like you said, all the changes are there for a reason, so the next step is for me to determine exactly what difference in mysql_backend.py is connected with the issue. I'll work on that today and post here when we have a candidate patch. Christian, thank you. -Aaron On Jun 1, 1:04 am, Christian Boos <[EMAIL PROTECTED]> wrote: > Aaron Maxwell - SnapLogic wrote: > > > Hi, > > > There is a highest/blocker bug about sessions expiring in .10.4, which > > we are attempting to fix. Having dug pretty deep into the internals, > > it appears that a specific sql query is misbehaving. I've posted a > > writeup here: > > >http://trac-hacks.org/ticket/1493#comment:7 > > > Please look at it and let me know if you have a suggestion. It seems > > like we're pretty close to the solution for this issue; hopefully we > > can come up with a patch that fixes it. > > Here are some ideas to try out... > > Looking at the changes from 0.10.3.1 to 0.10.4, I don't see anything > that could have introduced this "interesting" new behavior, except > perhaps at the level of the mysql_backend support itself: > > http://trac.edgewall.org/changeset?new=branches%2F0.10-stable%2Ftrac%... > > Try to undo those changes (which, OTOH, were done for a reason, so you > might have other issues), to check if this solves the "auto-logout" issue. > > If this is not the case, then please setup a test env using something > else than MySQL (e.g. sqlite), and see if you experience the issue as well. > > If the problem is related to MySQL as I'd expect, then perhaps you would > have to debug the mysql bindings themselves, in order to understand why > that SELECT fails when it shouldn't (don't forget to activate the print > statements in trac/db/util.py, to monitor the SQL statements and their > arguments, maybe it's simply the auth_cookie value changing for some > other reason). > > -- Christian --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Trac Development" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en -~----------~----~----~----~------~----~------~--~---
