Re: recover standby database failure
Date: Wed, 14 May 2008 18:16:10 -0400
Managed Standby in 8.1.7 is/was an Enterprise Edition feature. In 10g, that became affectionately known as DataGuard.
I would think that in standard edition, the use of:
SQL> recover managed standby database disconnect from session;
should return a message of
"feature not enabled"
If the attempted managed recovery statement does succeed, I would check the view dba_features_usage for the resulting hit. RDA will likely show this usage as well.
SQL> recover standby database;
on standard edition and you should do just fine. You'll get used to seeing that error message every time manual recovery is executed.
On Wed, May 14, 2008 at 5:38 PM, <krish.hariharan_at_quasardb.com> wrote:
> I would have to be
> To have it run the background: recover managed standby database disconnect
> from session;
> To cancel: recover managed standby database cancel;
> To execute in current session: recover managed standby database; --
> would block).
> Or at the least that is how I have done it. I haven't looked at the newer
> commands and capabilities. If you monitor the alert log it will indicate
> wait for the next archive log and an apply when you switch logs in the
> primary. Recover automatic would/should give the same error.
> The error as you said is what would be expected since it is attempting and
> unbounded recovery.
> -----Original Message-----
> From: oracle-l-bounce_at_freelists.org [mailto:oracle-l-bounce_at_freelists.org]
> On Behalf Of Finn Jorgensen
> Sent: Wednesday, May 14, 2008 3:26 PM
> To: mike.mitchell_at_etouchpoint.com
> Cc: oracle-l_at_freelists.org
> Subject: Re: recover standby database failure
> If I'm understanding you correctly then this is normal behaviour. It
> means the database has been recovered up until the most recent
> available archive log. I haven't set up a manual standby db since
> Oracle 7.1 so I'm not sure about this, but I would think at this point
> you just type "cancel", wait for more archive logs to arrive and then
> "recover automatic standby database;" again. Repeat.
> On 5/14/08, Michael Mitchell <mike.mitchell_at_etouchpoint.com> wrote:
> > I'm trying to build a physical standby server and have been following an
> > article I found at
> > 'http://www.dbasupport.com/oracle/ora10g/manual_standby.shtml'.
> > Everything is great until I execute the recover command.
> > Here is the back ground info:
> > - Two Linux (Centos 5) servers
> > - Oracle Standard Edition version 10.2.0g on both servers (we're young
> > poor :)
> > - Both database servers were installed at the same time with the same
> > (SID etc...)
> > - The primary database is configured for ARCHIVELOG
> > - I've done a hot backup of all tablespaces in the database where the
> > 'included_in_database_backup' value is set to 'YES' (select name from
> > v$tablespace where included_in_database_backup = 'YES';)
> > - I've built a standby controlfile
> > - Set the conversion parameters in the standby servers pfile to point to
> > correct location for database (.dbf) archives (.arc) files
> > The problem occurs on the 'recover standby database' command where I get
> > following errors 'Linux Error: 2: No such file of directory' on an
> > file that isn't part of the archive set. (See the full output below)
> > - There are 7 archive files in the standby's archivelog directory (arch2
> > arch8) but the recover command keeps looking for the 8th arch file which
> > doesn't exist on the primary.
> > Here's the full transcript from command. Note that the directory path is
> > expected and the archive file naming convention is correct, it's just
> > looking for a file that I don't seem to have yet.
> > Any suggestions will be greatly appreciated and I thank you all in
> > Mike
-- http://www.completestreets.org/faq.html http://safety.fhwa.dot.gov/ped_bike/docs/pamanual.pdf -- http://www.freelists.org/webpage/oracle-lReceived on Wed May 14 2008 - 17:16:10 CDT