[ZODB-Dev] Mysterious traceback

[Chris McDonough]

Does this happen when you're not using a mounted storage (or a session data
manager)?

I've noticed this when using mounted storages, and there was a bugfix to
Mount.py which supposedly made it all better maybe two months ago or so.
But I noticed it lately when playing around with mounts, so the problem is
still there.

(Just FYI:  The session data manager "internal data container" is backed by
a mounted storage, but it doesn't use Mount.py code.  I however, did notice
this same problem in a customer site with core session tracking without any
mounted storages which used Mount.py)

----- Original Message -----
From: "Randall F. Kern" <randy@spoke.net>
To: <jim@digicool.com>; <akuchlin@mems-exchange.org>
Cc: <zodb-dev@zope.org>
Sent: Thursday, March 01, 2001 6:29 PM
Subject: RE: [ZODB-Dev] Mysterious traceback


I have some code running under Zope that causes the "AttributeError:
'None' object has no attribute load" error pretty frequently.  Sometimes
I also get the tpc_begin error.  I've patch Connection.py to dump error
messages to a log if it ever sets _storage to None, but the only time
that happens is on close().

If you have any suggestions for debugging, that would sure help.
-Randy

_______________________________________________
Bug reports, feature requests, etc. go in the ZEO Tracker:
http://www.zope.org/Products/ZEO/Tracker

Conversations etc. can take place in the Wiki:
http://www.zope.org/Products/ZEO/Wiki

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