Hi,

One warning about copying the current controlfile:

There was a bug (not very critical and thus a good candidate for the
reinvention in a version or two timeframe :-)). The bug was related to
the checkpoint number(s) in the controlfile and datafile headers. That
number may be different due to incremental checkpoints on the primary
which are absent on the "standby". So in some cases in was not possible
to open the database with such a copy of current controlfile from
production database.

The walkaround is trivial - recreate the controlfile.

My 2 kopejka

Andrey
--
Disclaimer: Opinions are of my own and not necessar[-il]y...

Paul Drake wrote:
> Sandra,
>
> "Graceful Switchover" refers to a term in a paper by Lawrence To back
> in the 7.3.4 days.
>
> Both the online redo log and current controlfile had to be copied to
> the new location of the primary database (as well as any archived redo
> logs that were not applied).
> Basically, the standby is no longer a standby database. It is not
> activated. It simply is recovered and opened at the primary database,
> albeit on the server that used to house the standby database.
>
> Paul

--
http://www.freelists.org/webpage/oracle-l