Simple DR Question

[jluckhoff]

Hi,

Planning to implement a standby database on 9i for DR support.

One particular application does some hefy db writing and reading and because if one of the processes goes wrong in it it can corrupt the entire schema's data - Yes... very bad coding indeed

Anyway to secure this data we normaly take a schema export before running one of these jobs so if it goes wrong we just drop the schema and restore the old one. I have found that this is normally the fastest way to recover rather than trying to do a PITR and figure exactly when the user kicked off the job.

My question is. If I'm running a standby db and I have to do a recovery on my prod database like described above will there be any obvious problems.
Imp.exe should simply write the data as DDL and DML which is recorded int the redo logs and can therefore be applied to the standby DB.

Right?

Thanks

j

[DaPi]

Sounds right to me - but that doesn't mean you don't have to test it Beware of the volumes of archivelogs that can be produced and that will (I presume) have to traverse your network - I hung my db by filling the archlog disk with a bunch of imp's

[rad_jen]

You are on right track. I did exp/imp recently in my DR and I din't see any problem.

[ljayanth]

Hey put ur archivelog off before you import or else if you run out of disk space ur databse will get skewd up and u need to restart the database.

[rad_jen]

ljayanth:- you aren't right. 'J' has Dataguard.