RE: ASM or not to ASM
Date: Mon, 11 Jul 2011 13:53:32 -0500
Message-ID: <F077F09A0E11504D9E720358BEE994D108262F11_at_APSW0553EVS.ms.ds.uhc.com>
We currently use standbys for our recovery strategy along with netapps for storage.
If the only reason you are contemplating RAC is for failover to a secondary instance, then I would reconsider that thought.
Depending on your downtime requirements, standbys might work for you.
Our situation:
2 colocations.
2 Netapps, one in each colocation.
1 Production Database in first Colocation.
1 Standby database in second Colocation.
Automated log transport and application makes sure we are never more than 15 minutes behind on the logs.
Log storage with redundant drives.
The netapps themselves handle their own drive redundancy.
For a failure of production, we push any remaining logs and bring up
standby .
Failure of production netapp, push any remaining logs and bring up
standby.
Failure of Production Server, bring up standby (with the understanding
that we might have up to 15 minutes of logs not applied. Worst scenario
but ok in our case.)
Failure of Production Colocation, bring up standy with same note as
above.
Our backup methodology is to use the netapps snapshot abilities to backup files during daily hot backups plus a weekly cold backup. Downtime approximately 5 minutes total. We also take the standby down once a week for the same 5 minutes and make snapshots of it. Plus, as a final contingency, we take tape backups of the snapshot offsite after the cold backups.
Since the system has been setup the biggest non-automated actions that have to be done is if we have to bring up the standby. Its been a very stable configuration.
-----Original Message-----
From: oracle-l-bounce_at_freelists.org
[mailto:oracle-l-bounce_at_freelists.org] On Behalf Of Rich Jesse
Sent: Monday, July 11, 2011 11:38 AM
To: Oracle L
Subject: Re: ASM or not to ASM
Hi Robert,
> My question is why would I, on my single server with a single
> instance, using a SAN for my storage, bother with ASM? Why do all of
> this overhead just for a single instance?
I wouldn't use ASM, not only for the overhead, but because of the unnecessary risk of introducing bugs, some of which are (were?) severe.
I'm interested to see if anyone has any advantages of ASM in your scenario, which is the same boat I'm in. Seems that it would duplicate the functions, especially if the SAN storage has a front-end like IBM's SVC (which we have).
Rich
-- http://www.freelists.org/webpage/oracle-l This e-mail, including attachments, may include confidential and/or proprietary information, and may be used only by the person or entity to which it is addressed. If the reader of this e-mail is not the intended recipient or his or her authorized agent, the reader is hereby notified that any dissemination, distribution or copying of this e-mail is prohibited. If you have received this e-mail in error, please notify the sender by replying to this message and delete this e-mail immediately. -- http://www.freelists.org/webpage/oracle-lReceived on Mon Jul 11 2011 - 13:53:32 CDT