Database Software

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

Checkpoint time per log very long

Orna WeismanApr 5 2011 — edited Apr 15 2011

Hi

We are in the process of creating a Dataguard to our very active 10.2.0.4 2 node RAC cluster. ( Sun Solaris SPARC 64 bit )
this database generates around a Terabyte of archive data a day.
It has 589 data files.

The dataGuard is not able to catch up with the redo generation.
the issue is due to a very long checkpoint time , that is done at the end of each archive log applied.

the actual apply is pretty fast, but then 2-3 minutes are spent waiting for the checkpoint to complete, before moving to the next log

select item,units,sofar from v$recovery_progress where start_time = ( select max(start_time) from v$recovery_progress) and item = 'Checkpoint Time per Log'

Checkpoint Time per Log Seconds 102 '

My questiion is
Is there a way to tune these checkpoints ? or control them in any way ?

I assume that reducing the number of datafile will probably help , but this is not a solution we can currently persue

thanks
Orna

Locked Post

New comments cannot be posted to this locked post.

Locked on May 13 2011

Added on Apr 5 2011

#data-guard, #performance-availability

6 comments

1,510 views