Re: Data Guard RMAN deletion policy
- From: Justin Mungal <justin@xxxxxxx>
- To: Marcin Przepiorowski <pioro1@xxxxxxxxx>
- Date: Fri, 30 Aug 2013 05:02:11 -0500
Hi Marcin, I don't recall him saying he is stopping the primary. Where are
you seeing that?
Tiran, I'm wondering if RMAN is incorrectly marking the logs as obsolete
when the FRA is under space pressure. Or, they could be getting marked
obsolete once they've been backed up even if they haven't yet been shipped.
You could test that theory by deferring the log destination, switching the
redo log, running an archive log backup to tape, and then finally checking
to see if the backed up archive log has been marked as obsolete.
-How is the space pressure managed in the Flash Recovery Area - An Example.
(Doc ID 315098.1)
Also, there have been numerous bugs relating to similar issues.
-RMAN Archived Redo Logs Are Deleted Before Being Applied at Standby
Database (Doc ID 740322.1)
-Bug 7433541 : ARCHIVED LOGS ARE DELETED BY RMAN (AT PRIMARY) EVEN THOUGH
THEY ARE NOT APPLIED
-Bug 6216036 : RMAN+DG ARCHIVELOG DELETION POLICY APPLIED ON STANDBY NOT
RESPECTED
It would be very interesting if you opened an SR with Oracle to see what
they say; there could be a simple fix for this.
-Justin
On Fri, Aug 30, 2013 at 3:07 AM, Marcin Przepiorowski <pioro1@xxxxxxxxx>wrote:
> Hi,
> Easiest solution move archive logs from FRA to other directory and write
> your own RMAN scripts to control when you will delete archive logs.
>
> Anyway idea of stopping primary when standby is down due to archive log
> issue seems strange for me. Even if you will lost archive logs from primary
> you can always use RMAN incremental backup to synchronize your standby.
>
> regards,
> --
> Marcin Przepiorowski
> http://oracleprof.blogspot.com
>
>
>
--
//www.freelists.org/webpage/oracle-l
Other related posts:
