On Dec 21, 2006, at 4:11 AM, Christian Theune wrote:

Hi,

Chris Withers wrote:
Christian Theune wrote:
I've tried to analyse the situation for this a bit. Some annotations are in http://www.zope.org/Collectors/Zope/2151 and maybe this can trigger
more input.
Thanks for looking into this, any help is much appreciated!
Note: I was a bad boy and used ZODB trunk for the analysis and didn't
remember that Chris was talking about 2.9.2 until I was done. I hope
this area didn't change too much, otherwise I'd have to repeat that
analysis. :/
zodb doesn't seem to have gotten that much attention since 2.9.x, unless I've missed something.

You have. I spend a fair bit if time on it for the 2.10/3.3 releases. This was mainly to chase a problem on the Mac but I ended up cleaning up some internal messiness
quite a bit.  Of course, there's also the blob work.

I also spent some time on this issue. Chris and I went back on this in August:

  http://mail.zope.org/pipermail/zodb-dev/2006-August/010355.html

Unfortunately, when I made time to look at this, Chris was away on vacation.

I see you raised some possibilities in terms of what might be causing the problem. How should we go about eliminating these?

That's the plan. This bug takes me quite some time to analyse, so I don't promise any quick success here. If anybody (Jim?) can comment on my findings to prevent me running in the wrong direction that would be very helpful.

See the note from Chris lined above. It responds to and fortunately quotes a note from me
that I can't find. :/  I found a race condition that was fairly benign.


Next step for me will be to verify the assertion that when calling settid() it must be passed a strictly larger tid than the one before.

There is no test currently assuring this or the opposite. Once I find out, I'll update the tests.

At this point, I don't really remember many details.

Jim

--
Jim Fulton                      mailto:[EMAIL PROTECTED]                Python 
Powered!
CTO                             (540) 361-1714                  
http://www.python.org
Zope Corporation        http://www.zope.com             http://www.zope.org



_______________________________________________
For more information about ZODB, see the ZODB Wiki:
http://www.zope.org/Wikis/ZODB/

ZODB-Dev mailing list  -  ZODB-Dev@zope.org
http://mail.zope.org/mailman/listinfo/zodb-dev

Reply via email to