Re: Oracle Fail Safe Question

More to the point, your instances should be identical on each side of the cluster, this means your problem will probably still occur again after a fail-over.

But if you wish to persist, and the error appears in the alert log, schedule a batch script job to poll the alert log and on fidning the error (or any error you wish) have the script 'kill' the Oracle thread and force OFS/MCS to initiate and instance recovery on the other node...If you dont like killing the thread, use svrmgrl to do a shutdown abort...

Not too sure if this will cause a restart of the service on the same node, but iirc you can can configure the service in the way it is either restarted or failed-over on failure detection.

"Terry Banicki" <banicki_at_home.com> wrote in message news:3992D34B.532639E4_at_home.com...
> With my limited knowledge...
> I do not believe you can specify a particular message to fail on.
> If this is a message right before the oracle instance fails, OFS will
> handle the fail over.
> Actually, I just browsed metalink. Looks like there are several bugs
> related to Ora-04030. You probably should resolve the root cause of the
> ORA-4030.
> OFS works great for us.
> Terry
>
> shayko_at_my-deja.com wrote:
>
> > Does any one know if there is a way to tell Oracle Fail Safe
> > to perform a fail over when a specific Oracle error occures
> > for example "ORA-04030: out of process memory" ?
> >
> > Thanks a lot.
> >
> > Sent via Deja.com http://www.deja.com/
> > Before you buy.
>
Received on Fri Aug 11 2000 - 00:00:00 CDT