i agree with your approach, if i was using NFS or something similar. i plan on using a clustered filesystem for the shared location of the archive logs. we are using GPFS as our clustered filesystem. am i missing something by using a clustered filesystem for the archive logs? any performance issues? thanks and happy holidays to everyone! may all your instances stay up ;) haroon ----- Original Message ---- From: "Bobak, Mark" <Mark.Bobak@xxxxxxxxxxxxxxx> To: kevin.lidh@xxxxxxxxx; rjamya@xxxxxxxxx Cc: haroon_a_qureshi@xxxxxxxxx; oracle-l@xxxxxxxxxxxxx Sent: Friday, December 22, 2006 7:59:57 AM Subject: RE: RAC Archivelog mode and RMAN That's what we do as well. Each node has it's own, but each node can see all destinations. Seems like a reasonable compromise. -----Original Message----- From: oracle-l-bounce@xxxxxxxxxxxxx on behalf of Kevin Lidh Sent: Fri 12/22/2006 8:07 AM To: rjamya@xxxxxxxxx Cc: haroon_a_qureshi@xxxxxxxxx; oracle-l@xxxxxxxxxxxxx Subject: Re: RAC Archivelog mode and RMAN In my last job, we had a 3-node 9.2.0.4 RAC on Tru64 5.1b and we had a separate archive log destination for performance reasons but each node could see each other's destination. On Fri, 2006-12-22 at 07:58 -0500, rjamya wrote: > grrrr .... about assigning shared location for archive logs. Perhaps > it is just me. If that shared dest goes away for some reason, (most > likely) all nodes using that will go down. We intentionally assign > separate archive log dest for each node. If required we do > cross-instance-archiving, else run a cron job to move archived logs to > a common dest manually. > > Then again, it is just me. > Raj -- //www.freelists.org/webpage/oracle-l -- //www.freelists.org/webpage/oracle-l