If the original master is recoverable, shouldn't it have the list of
what has not yet been propagated? Why not simply apply these logs? Are
you worried about the logs causing mail to be deleted from the replica?
This is technically easier than trying to reconstruct from the logs
what was not propagated, especially since I've found the log files to
be not very informative even in verbose mode.
Fixing the logs to easily show what's been propagated seems trackable.
:wes
I am unable to apply the left behind transaction logs of the
sync_client. Perhaps I'm just not sure how to do it.
I found another way to do this. Within postfix is a function called
"always_bcc" which will copy all incoming email into one mailbox. What
I am planning on doing is that if we have a failover and the master is
recoverable, I would look at the /var/log/maillog to determine when the
last sync entry was. Then go into my mailbox with all of the emails and
then bounce the ones that didn't make it through the sync process.
Of course there are logistics to this method such as we have to purge
the bcc mailbox often to keep its size down, etc. But I'm working out
those details now.
-David
--
--------------------------------------------------------
David Korpiewski Phone: 413-545-4319
Software Specialist I Fax: 413-577-2285
Department of Computer Science ICQ: 7565766
University of Massachusetts Amherst
--------------------------------------------------------
----
Cyrus Home Page: http://asg.web.cmu.edu/cyrus
Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu
List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html