onbar performance and restore

Ihave problem with onbar performance. It takes more than 2 hours to backup all database using onbar -b -w. Therefore I use onbar -b <all dbspaces list> to back up dbspaces parallely. It is fast. However there is a problem restoring database from above backup up to the poit in time or up to logical log # using onbar -r -n <log number>. It simply doesn' work! Is there any solution to have fast backup and possibility to restore with rollforward using onbar? Thanks. Tom

We had a similar problem for point-in-time recovery using OnBAR with Omniback ( on HP-11 box ). We had to get around it using point-in-log recovery which DID work fine. You need to watch the parallel archiving aspect though because sometimes the engine will freeze with a message ( when you do onstat - ) like CKPT REQUIRTED: ARCHIVE. This is due to OnBAR issuing multiple threads, each one requesting a checkpoint. If two of the threads issue the chkpoint request at the same time, only one will be recognised, freezing the other one, and the engine as well. Attempting to kill the process using onmode -z is futile ! Point-in-time is very flakey in 7.24.UC5 - we've been through technical support who gave us a work-around but wasn't very clever ( messing with the ixbar.0 file ) What problems are you getting when attempting to do point-in-log restore ? Sean chacalat@istar.ca wrote in message <36A60B32.83698AA1@istar.ca>... >Ihave problem with onbar performance. It takes more than 2 hours to >backup all database using onbar -b -w. Therefore I use onbar -b <all >dbspaces list> to back up dbspaces parallely. It is fast. However >there is a problem restoring database from above backup up to the poit >in time or up to logical log # using onbar -r -n <log number>. It >simply doesn' work! Is there any solution to have fast backup and >possibility to restore with rollforward using onbar? > >Thanks. > >Tom > > > >