I was just wondering:

While I could not generate any more deadlocks with the WebDAV level
locking there might be people who do not want it and only want to rely
on the database locks. In this case it is just a fact that deadlocks
can occur.

Slide's policy has been to report this to the client/user, but why not
just retry internally? You could throw a TimeoutException or Deadlock
exception (RuntimeException?), catch it in the WebDAV layer, rollback
the old transaction, start a new one and repeat. Should work with
reasonable effort...

Of this switching this on should be configurable. 

Using externally controlled transactions should remain unchanged.

Oliver

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to