We just set up and tested our standby server and it's really quite
straight forward :

1. Shutdown the Live database and take a "cold backup" - however you
want. OS or RMAN.
2. Set up the Standby Database by copying all datafiles etc (via SCP or
whatever) to the standby machine.
3. Create a standby control file on the live database (any time while
running ) :
SQL> alter database create standby controlfile as
'/u01/app/oracle/standby.ctl';
4. Replace all the multiplexed control files on the standby machine with
the above "standby control file"
5. On the Standby machine :
5.a. SQL> startup nomount
5.b. SQL> alter database mount standby database;
5.c. SQL> recover standby database until cancel
Incorporate the above into a cron or equivalent to run periodicaly
in a script answering "AUTO" - this will aplly any archived redo logs
not yet applied.
5.d. Ship the Archived Redo Logs across from the live machine - we use
"rsync" with a script just before the step 5.c. above

6. If you want to access the standby you can open it for read only as
follows (closing it after otherwise the "recover" won't run) :
SQL> alter database open read only
SQL> select blah blah from blih blih......
SQL> alter database close

7. When disaster strikes you activate the standby database - NO GOING
BACK ! IT's NOW LIVE

7.a MAKE SURE ALL ARCHIVE LOGS have been shipped across
- if you can on live "LIVE : SQL> alter system switch logfile - x times
where x = number of Redo Log Groups - to close the Online Redo Logs and
write Archived Redo Logs !
7.b Standby : SQL> recover standby database until cancel
7.c Standby : SQL> alter database activate standby database
7.d Standby : SQL> shutdown immediate
7.c Standby : SQL> startup

I hope this helps !


If you have knowledge, let others light their candles at it.

Margaret Fuller (1810 - 1850)

ROLFE HARE
Senior Database Administrator
Qpass - Amdocs Digital Commerce Division
Neutorgasse 12/9, 1010 Vienna, Austria
o: +43 1 5326464 0 | m: +43 676 5688 515| [Email address protected]


-----Original Message-----
From: karthik ganesan
[mailto:oracledba-ezmlmshield-x85276318.[Email address protected]
Sent: Donnerstag, 05. Oktober 2006 10:27
To: LazyDBA Discussion
Subject: Regarding RMAN restoration - urgent - please!

Hi all,

I'm using oracle RMAN for hot backup and recovery, its working fine.

My current practice-
I take the RMAN hot backup every night and then restore the contents on
another server. During this phase the ATM machine won't be accessible
due to resource crunching that takes place because of hot backup.

My proposed view -
Once the data is restored, take the archive logs every 2 hours and just
keep running, Recover database; The advantage, I need not take hot
backup every night in the Live, instead transfer only the archive logs
and apply it on the backup server database. When situation demands use
ALTER DATABASE OPEN RESETLOGS;

Will there be any flaw or data inconsistency will it occur?

Regards,
KARTHIK GANESAN
IT Consultant
Uganda




---------------------------------------------------------------------
TO REPLY TO EVERYBODY , PLEASE CLICK REPLY-ALL, NOT JUST REPLY To post a
dba job: http://jobs.lazydba.com To Subscribe : http://www.LazyDBA.com
To unsubscribe: http://www.lazydba.com/unsubscribe.html




Oracle LazyDBA home page