Re: Resize redo logs
From: ddf <oratune_at_msn.com>
Date: Mon, 21 Sep 2009 12:05:05 -0700 (PDT)
Message-ID: <5b50843d-a94e-4e32-a98d-36b7fcb7c047_at_p9g2000vbl.googlegroups.com>
On Sep 21, 1:52 pm, ddf <orat..._at_msn.com> wrote:
> On Sep 21, 12:53 pm, Tom76 <br..._at_hotmail.com> wrote:
>
> > On Sep 21, 1:50 pm, Tom76 <br..._at_hotmail.com> wrote:
>
> > > Need to resize online and standby redo logs for a DataGuard(primary
> > > and physical dataguard). What is the procedure?
>
> > primary and physical standby database.
>
> To keep things working properly through Data Guard you'll need to
> create the new logs/groups of the proper size on the primary then
> create new logs/groups on the standby. Once you have all of your new,
> larger redo logs/groups in place on both servers you can switch
> logfiles until the the first of the newer, larger logs is in use.
> Usually you could simply switch logfiles until the first group is no
> longer active, drop it, and recreate it with a larger size, but the
> standby database relies upon the fact that the standby redo logs
> exactly mirror the primary redo logs. This is why you must create
> the larger logs first, on both systems, so that when the log switch on
> the primary gets to the larger log a matching log on the standby will
> be found. When the smaller redo logs are shown as INACTIVE you should
> be able to drop them from both the primary and the standby leaving
> only the 'resized' logs in place. Yes, this 'upsets' your log group
> numbering but that won't matter to Oracle as long as the logs/groups
> match on both primary and standby databases.
>
> David Fitzjarrell
Date: Mon, 21 Sep 2009 12:05:05 -0700 (PDT)
Message-ID: <5b50843d-a94e-4e32-a98d-36b7fcb7c047_at_p9g2000vbl.googlegroups.com>
On Sep 21, 1:52 pm, ddf <orat..._at_msn.com> wrote:
> On Sep 21, 12:53 pm, Tom76 <br..._at_hotmail.com> wrote:
>
> > On Sep 21, 1:50 pm, Tom76 <br..._at_hotmail.com> wrote:
>
> > > Need to resize online and standby redo logs for a DataGuard(primary
> > > and physical dataguard). What is the procedure?
>
> > primary and physical standby database.
>
> To keep things working properly through Data Guard you'll need to
> create the new logs/groups of the proper size on the primary then
> create new logs/groups on the standby. Once you have all of your new,
> larger redo logs/groups in place on both servers you can switch
> logfiles until the the first of the newer, larger logs is in use.
> Usually you could simply switch logfiles until the first group is no
> longer active, drop it, and recreate it with a larger size, but the
> standby database relies upon the fact that the standby redo logs
> exactly mirror the primary redo logs. This is why you must create
> the larger logs first, on both systems, so that when the log switch on
> the primary gets to the larger log a matching log on the standby will
> be found. When the smaller redo logs are shown as INACTIVE you should
> be able to drop them from both the primary and the standby leaving
> only the 'resized' logs in place. Yes, this 'upsets' your log group
> numbering but that won't matter to Oracle as long as the logs/groups
> match on both primary and standby databases.
>
> David Fitzjarrell
Make sure the standby redo logs you want to drop are not marked CLEARING or CLEARING_CURRENT as they cannot be dropped in that state. You can expedite the clearing process by issuing an 'alter database clear logfile group <n>;' where <n> is the group number you want to drop.
David Fitzjarrell Received on Mon Sep 21 2009 - 14:05:05 CDT