Re: RMAN performance issue after patching OS

I just want to thank everyone who has responded. We are reviewing the drivers this morning.

Leng - to answer your questions:
We did run a backup to a local disk, and it ran fine. I have been reviewing the v$ tables - our backups are using async IO. This is the command we are using:
backup device type disk filesperset = 1 tag 'BACKUP_TEST_082521091605' database;

Thank you,
Jay

On Tue, Aug 24, 2021 at 7:21 PM Leng Burgess <lkaing_at_gmail.com> wrote:

> Hi Jay,
>
> Do you get the same problem when backing to local disk? Have you compared
> Unix cp vs rman backup as copy? Have you looked at V$BACKUP_SYNC_IO and
> V$BACKUP_ASYNC_IO
>
> What command are you using?
>
> Cheers,
>
> Leng.
>
> On 25 Aug 2021, at 3:17 am, Jay Hostetter <hostetter.jay_at_gmail.com> wrote:
>
> Good afternoon,
>
> Many of our database servers are running Oracle Linux Kernel Release 5
> Update 4. We have been patching these servers and discovered our RMAN
> performance (to NFS shares) has suddenly dropped ( from 100+ mb/s to 10
> mb/s). It took us quite some time to figure out that it is actually the OS
> patch that is causing the issue. We even applied the latest patch (Kernel
> Release 6 Update 2), but we are still running into the same issues.
> Has anybody else encountered this issue? We have not made any changes
> to the mount options for our shares. We haven't made any changes to the
> NFS server itself. We ran a backup to local disk, moved underlying storage
> for NFS, etc. We finally determined that it is caused by our OS patch.
> Any help would be appreciated.
>
> Thank you,
> Jay Hostetter
>
>
>

--
http://www.freelists.org/webpage/oracle-l