Re: Patching in an environment used for dataguard
From: joel garry <joel-garry_at_home.com>
Date: Mon, 27 Sep 2010 14:13:49 -0700 (PDT)
Message-ID: <505b2eee-8193-4575-891e-0383452d2569_at_a4g2000prm.googlegroups.com>
On Sep 27, 12:24 pm, Steve Howard <stevedhow..._at_gmail.com> wrote:
> On Sep 27, 11:14 am, dba cjb <chris.br..._at_providentinsurance.co.uk>
> wrote:
>
>
>
> > 9777078 Patch Number 38
>
> > Oracle 10.2.0.4 Enterprise / windows 2003 server
>
> > I am looking to plan a patch to production and then a patch to
> > dataguard environment
>
> > My general ( simplified ! ) plan for dataguard / following production
> > patch will be as follows
>
> > * Stop database
> > * Apply software
> > * Re-start database ...with the hope that continued application of
> > archive logs will apply the patch database changes already applied to
> > production
>
> > cheers
> > Chris B
>
> Hi Chris,
>
> It's always safer to ask Oracle, but we have used the steps in your
> plan several times with no issues, so this should work. The database
> script changes run on the production database will be propagated to
> the standby via the redo stream. If you are *really* concerned, you
> can try to tie the application of the patch as close as possible to
> when you stop applying redo on the standby, so they "mostly" match.
>
> However, as noted, we have never had an issue when doing this.
>
> Regards,
>
> Steve
Date: Mon, 27 Sep 2010 14:13:49 -0700 (PDT)
Message-ID: <505b2eee-8193-4575-891e-0383452d2569_at_a4g2000prm.googlegroups.com>
On Sep 27, 12:24 pm, Steve Howard <stevedhow..._at_gmail.com> wrote:
> On Sep 27, 11:14 am, dba cjb <chris.br..._at_providentinsurance.co.uk>
> wrote:
>
>
>
> > 9777078 Patch Number 38
>
> > Oracle 10.2.0.4 Enterprise / windows 2003 server
>
> > I am looking to plan a patch to production and then a patch to
> > dataguard environment
>
> > My general ( simplified ! ) plan for dataguard / following production
> > patch will be as follows
>
> > * Stop database
> > * Apply software
> > * Re-start database ...with the hope that continued application of
> > archive logs will apply the patch database changes already applied to
> > production
>
> > cheers
> > Chris B
>
> Hi Chris,
>
> It's always safer to ask Oracle, but we have used the steps in your
> plan several times with no issues, so this should work. The database
> script changes run on the production database will be propagated to
> the standby via the redo stream. If you are *really* concerned, you
> can try to tie the application of the patch as close as possible to
> when you stop applying redo on the standby, so they "mostly" match.
>
> However, as noted, we have never had an issue when doing this.
>
> Regards,
>
> Steve
Interesting tidbit from Tanel about standby binaries I happened to notice this morning while looking for something else: http://www.freelists.org/post/oracle-l/FUZZY-column-in-VDATAFILE-HEADER,4
jg
-- _at_home.com is bogus. "ASM adds another layer of distraction in that it creates and manages the files internally so that from an OS point of view you may have very few files (assign whole disks to ASM and let if parcel out the space to the databases)." - Mark Powell I agree, and not in the abstract! :-)Received on Mon Sep 27 2010 - 16:13:49 CDT