Re: 11.2.0.3: Re-install grid and in-place upgrade of DB
- From: Martin Berger <martin.a.berger@xxxxxxxxx>
- To: yong321@xxxxxxxxx
- Date: Thu, 29 Sep 2011 21:29:19 +0200
Yong,
see
Note:337737.1<https://support.oracle.com/CSP/main/article?cmd=show&type=NOT&doctype=BULLETIN&id=337737.1>
:
The Oracle Grid Infrastructure/Clusterware (CRS) version must be the highest
version down to the 4th digit in the possible combinations at all times.
This means you have to upgrade Grid Infrastructure/Clusterware first.
That's the best I found - at least you follow this rule.
If you deinstall/reinstall the full GI stack and reinstall it afterwards,
you will be able to use all the ASM disks - nothing to change there. Just
the creation of cluster resources for DBs/Instances is needed.
I'd install in a new OH and do a rolling upgrade. Even not done to 11.2.0.3
I've done it several times up to 11.2.0.2.
hth
Martin
On Thu, Sep 29, 2011 at 20:27, Yong Huang <yong321@xxxxxxxxx> wrote:
> I have a 2-node RAC 11.2.0.2 running on 64-bit Red Hat Linux.
> Documentation says we can't do in-place upgrade of the grid infrastructure,
> and we don't want to do out-of-place upgrade. So the only option is to
> deinstall 11.2.0.2 and reinstall 11.2.0.3. Can the new grid infrastructure
> work with the existing 11.2.0.2 database (which can be upgraded later)? If
> you know the page in the documentation, please share the URL. Thank you.
>
> Downtime is not an issue. We can also extract ASM metadata with md_backup
> command inside asmcmd. But we don't know the data it creates can be restored
> by md_restore of the newly installed grid.
>
> Yong Huang
> --
> //www.freelists.org/webpage/oracle-l
>
>
>
--
//www.freelists.org/webpage/oracle-l
Other related posts:
