Re: benefits of multiple switch logfile calls?

lsattle_at_yahoo.com (lsattle) wrote in message news:<95898986.0312110545.4f316b14_at_posting.google.com>...
> drak0nian_at_yahoo.com (Paul Drake) wrote in message news:<1ac7c7b3.0312081614.7d4515d9_at_posting.google.com>...
> > lsattle_at_yahoo.com (lsattle) wrote in message news:<95898986.0312040755.16c6f7ac_at_posting.google.com>...
> > > Brian Peasland <dba_at_remove_spam.peasland.com> wrote in message news:<3FCE5C98.1EDE2220_at_remove_spam.peasland.com>...
> > > > > 3 log groups? Make sure all switched?
> > > >
> > > > Sure...but why? After all this is said and done, one still has one
> > > > active log group. No different than if you just issued one switch.
> > > >
> > > > Cheers,
> > > > Brian
> > > >
> > > >
> > > > --
> > > > ===================================================================
> > > >
> > > > Brian Peasland
> > > > dba_at_remove_spam.peasland.com
> > > >
> > > > Remove the "remove_spam." from the email address to email me.
> > > >
> > > >
> > > > "I can give it to you cheap, quick, and good. Now pick two out of
> > > > the three"
> > >
> > >
> > > I'm not sure of where you are mentally on all this. The switch log
> > > files / archive log files, to my knowledge is intending to get any
> > > current redo out of the online redo logs and into the archive redo
> > > logs.
> > >
> > > I'm not completely clear on switching vs archiving. I would think
> > > that just archiving would be enough.
> > >
> > > What I am certain of is that while your tblspaces are in "begin
> > > backup" and you are gziping or copy data files, transactions and / or
> > > checkpoints are occuring in your system. After you do the "end
> > > backup" it is wise to get all redo out to the archive directory.
> > >
> > > Then when you restore and do the recovery command (our recovery
> > > command is "recover database using backup controlfile until cancel")
> > > the recovery will only need archived redo logs.
> > >
> > > In fact what we do is not restore the online redo logs at all, we let
> > > the database rebuild them during the recovery. I will explain why.
> > >
> > > We do backups similar to you, but use netbackup to copy the files to
> > > tape. This copy can take from 10 minutes to 6 hours depending on the
> > > size of the db. If we let the system backup the redo logs, who knows
> > > where in the time frame they will get backed up. If we back them up
> > > after the "end backup" they will contain advanced data. So we like to
> > > control the whole situation by doing the "end backup" followed by:
> > >
> > > alter system checkpoint;
> > > alter system switch logfile;
> > > host sleep 10
> > > alter database backup controlfile to '$gOUT_DIR/cntrlfile';
> > > alter database backup controlfile to trace;
> > >
> > >
> > > Note I now believe these last steps are out of order. I think we
> > > should get the controlfile backup first, and then get the log switch.
> > > Note we do restore the backup of the control file before the recovery
> > > because again who knows where in the backup process the controlfile
> > > was backed up.
> >
> > just snipping at one point:
> > you will be better off, if you backup controlfile to <dest>
> > and then archive the current logfile afterwards.
> >
> > if you use the binary version of the backup controlfile, it will be
> > asking for the logfile to be applied that has not yet been archived,
> > in other words, file#1 needs more recovery.
> >
> > if you decide not to create a binary backup of the controlfile, please
> > ignore this post.
> >
> > Pd
>
> Paul,
> Thanks for your response. With all the hoopla with HJR, I missed your
> response on this.
> Yes I understand what you are saying. I think I brought this same
> issue up in some of my dialogue with HJR. I will explain why we don't
> get burned (Howard answered it in other posts to this thread). When
> we do the recovery, with the parm "using backup controlfile until
> cancel" the recovery goes until it runs out of archive redo logs.
> Howard's argument to rebuild the controlfile during the recovery with
> the bu to trace file will produce the same results. If we happen to
> have 2 more hours of redo (or even a couple of days worth) we can keep
> the recovery going till oracle can't the next numbered one.
>
> Lynn Sattler

Lynn,

If you do not apply at least the log that was current while the backup controlfile was created, you will receive a message that "file 1 requires recovery". you must apply at least the current log during recovery, as that SCN is either in the backup controlfile somewhere (I should dump it and see) or in the datafile header (or both).

Pd Received on Tue Jan 06 2004 - 17:57:15 CST