Re: Streams for one-time single source replication

  • From: Chris Taylor <christopherdtaylor1994@xxxxxxxxx>
  • To: rjoralist3@xxxxxxxxxxxxxxxxxxxxx
  • Date: Thu, 6 Feb 2014 14:36:36 -0600

Well, that definitely complicates things...

The standby idea would be an all or nothing.  I was thinking about SQL
Apply but again you wouldn't have the ability to narrow it down to specific
schemas IIRC.

So, uhm, good luck?  ;)

How big are these schemas?  I imagine we're talking about a lot of data for
each?  If not, and I say this hesitantly, you could create a schema of
views on the new AIX 7.1 database pointed back to your original database on
AIX 5.x and run MERGE commands on the new database to merge the changes
until cutover.  I'd only consider that for fairly small schemas though.

Other than that, it sounds like a streams approach might be your only
option.

So, every schema in the existing AIX 5.x 11.x database is not moving over
to your new AIX 7 box?  I'm curious why only some schemas would be moved I
guess...

Chris



On Thu, Feb 6, 2014 at 2:22 PM, Rich Jesse <rjoralist3@xxxxxxxxxxxxxxxxxxxxx
> wrote:

> Chris writes:
>
> > You could do a standby 11.2.0.3.x on the AIX 7.1 box - keep the
> archivelogs
> > shipping over and the standby up to date.  Then on the cutover date, you
> > shutdown the primary, resync the last transactions and then patch to
> > 11.2.0.3.8.  Although, you might be able to do a standby from 11.2.0.3.x
> to
> > 11.2.0.3.8 since only the patch release version is changing (haven't
> tested
> > that scenario).
>
> Would this work for a logical copy?  I can only have a subset of schemas
> cloned.
>
> Thanks!
> Rich
>
> --
> //www.freelists.org/webpage/oracle-l
>
>
>

Other related posts: