We have a ZODB which is about 12 - 14 gigs when packed. We use FileStorage, and when anyone clicks Undo eventually times out. This is better now that the undoLog does not lock the database.
However if there are not enough transactions to fill the batch Undo is useless. So, everytime we upgrade to a new version of Zope I patch FileStorage.py to add a timeout. Used to be 3 seconds when the ZODB locked, now I've bumped it up to 10 seconds.
Has anyone else had this problem? Would adding the timeout, possibly set by an environment variable, be considered by the maintainers?
Thanks, Brian Brinegar Engineering Computer Network (ECN) Purdue University
Zope-Dev maillist - [EMAIL PROTECTED]
** No cross posts or HTML encoding! **
(Related lists - http://mail.zope.org/mailman/listinfo/zope-announce