On Mon, Jan 07, 2013 at 08:46:19PM +0000, Bernhard Reiter wrote:
> When running with postgresql, the problem is that the history is not set
> to the right value. Yes, I would expect a locking mechanism in the database
> to prevent this, a transaction would be cool. However it does not prevent it.
> Maybe that is the result of line 2996 of rdbms_common.py:
>    oldvalues = copy.deepcopy(self.db.getnode(self.classname, itemid))

Yes, if this happens outside the transaction this could be the reason
that all the transactions get the same history. If it happens inside the
transaction postgres should note that the issue is *read* and see a
conflict.

I think I'll have to look into this more closely...

Ralf