Re: DR technique (part 2)

From: Andrew Kerber <andrew.kerber_at_gmail.com>
Date: Wed, 14 Aug 2013 14:07:09 -0500
Message-ID: <CAJvnOJYaCQtr3dQOEOXKvB76pMgUKxwNsSAJxOrSbNT5e2v5NA_at_mail.gmail.com>

I cant help but think you are doing this the hard way. Why not just set up a rac one node image to restore, with ASM already there, and maybe the spfile file from your rac instance already converted to single instance, so that all you have to do is restore that, then use rman to restore the data? If the idea is to make this reasonably quick and easy, I would think you want as much done as possible before hand. However, if I didnt absolutely have to, I wouldnt bother with all that. I would just set up a rac to single instance dataguard standby and be done with it.

On Wed, Aug 14, 2013 at 11:32 AM, Hubler, Daniel <daniel.hubler_at_aurora.org>wrote:

> First: thanks to all the folks who responded.
>
>
> Next: After reading through the replies that I received, I think I did a
> poor job of explaining what our idea is.
>
> So I am going to try again . . . .
>
>
>
> This would be Oracle DB & Grid v11.2.0.3 on AIX v6.1
>
>
>
> Our idea is to go to the DR hot site,
>
> with just a single node to run from, instead of the 2 that exist in our
> RAC.
>
>
>
> That node would NOT be equipped with a switch, or a private network, . . .
> the entities that I labeled "RAC stuff".
>
>
>
> We would restore the node using a mksysb of one of our RAC nodes.
>
> (mksysb = root volume restore)
>
> We would then lay down all the other file system stuff from a tape backup
> of that same node.
>
> This should give us the operating system software and the Oracle software
> needed to run.
>
>
>
> After that, I think it would go something like:
>
> a) stop CRS (because it might have been started during a reboot)
>
> b) start CRS in exclusive mode (I believe this allows ASM to start
> without a voting disk)
>
> c) use sqlplus (for ASM) to create the disk group that will house the OCR
> (create diskgroup xxxx disk 'ssssss')
>
> d) restore the OCR from the latest OCR backup
>
> e) recreate the voting disk (crsctl replace votedisk +XXXXX)
>
> f) create ASM pfile/spfile parameters
>
> g) shutdown CRS
>
> h) start CRS
>
> i) rebuild other disk groups in ASM
>
> j) start DB instance
>
> k) restore/recover database using RMAN
>
>
>
> Again. . . . .any comments, thoughts, experiences would be welcome.
>
>
>
> Thanks.
>
>
>
>
> Daniel P. Hubler
> Database Administrator
> Aurora Health Care
> Daniel.hubler_at_aurora.org
>
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

-- 
Andrew W. Kerber

'If at first you dont succeed, dont take up skydiving.'


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

Received on Wed Aug 14 2013 - 21:07:09 CEST