Re: DataGuard and closed/cold backups
- From: Carel-Jan Engel <cjpengel.dbalert@xxxxxxxxx>
- To: David Sharples <davidsharples@xxxxxxxxx>
- Date: Tue, 31 May 2005 21:42:34 +0200
Hi David,
Unskyld for being unclear about that.
We were consolidating data centres. The Singapore database was moved to
The Hague. So, the database in Singapore needed to be shutdown anyway
;-)
For that, we needed to shutdown/startup exclusive, perform the last
logswitches, and then activate the standby after the last redo was
applied. By then, the apps were run in The Hague, using Citrix clients
in Singapore.
Further, I didn't mention it was Data Guard on 8i. And that version
didn't support dataloss free switchover at all: the only means of log
transport was by ARCH. But, given the available bandwidth (or the cost
of the otherwise needed bandwidth), on 9i/10g I would have choosen ARCH
as log transport mechanism also. And, in that case, 9i/10g would have
required the same approach on switchover.
Best regards,
Carel-Jan Engel
===
If you think education is expensive, try ignorance. (Derek Bok)
===
On Tue, 2005-05-31 at 21:01, David Sharples wrote:
> Hi, one question about that - why was the database taken down at all.
>
> If you took a hot backup and starting sending the archives right away
> - why the shutdown?
>
> Chers - Dave
>
> On 5/31/05, Carel-Jan Engel <cjpengel.dbalert@xxxxxxxxx> wrote:
> > Jackie,
> > A couple of years ago I was able to move a database of 90 GB from
> > Singapore to The Hague, using Data Guard, through only a 128 kilobit
> > dual ISDN dial up line. The downtime of the database was 45 minutes. Of
> > course this was in ASYNC mode, logtransfer by ARCH.
> >
> > The trick was calculating the average amount of redo produced weekly,
> > and have enough bandwidth for just that. The outage was planned
> > carefully, including stopping / starting the apps it took 1.5 hour.
--
//www.freelists.org/webpage/oracle-l
Other related posts:
