Re: unrecoverable_time column in v$datafile not being refreshed

Hello Jonathan,

Thanks for your reply.

The context of the question is that of ensuring that a Standby database is kept viable.

As you know, one transfers archived redo logs from a "Master" database to the Standby server where they are applied to the Standby database.

Our shipping is scripted, but is checked by a monitor to make sure, among other things, that no direct logging has taken place. If it does, the Monitor stops the shipping immediatly.

The Monitor does this by checking the unrecoverable_time column in v$datafile.

Using your example, the problem would arise this way:

11:00 - A direct load takes place. Log Shipping to the Standby is stopped because a change to the unrecoverable_time column has been detected. Now the Standby has to be refreshed and restarted with a backup from the Master database. This is done.

11:55 Everything up and running again.

12:00 Another direct load takes place. This time, though, it is not detected because the Monitor has detected no change to the unrecoverable_time column (because there hasn't been one even though a direct load has taken place) and the redo logs are transferred and applied to the Standby.

This I am told will cause the Standby database to become unusable.

So thats where my problem lies

I'll check out the unrecoverable_change# tomorrow. That may prove to be my workaround.

Thanks again for your reply.

regards

Steve Roper
Database Administrator
EDS (UK) Ltd.

In article <972497451.3757.2.nnrp-09.9e984b29_at_news.demon.co.uk>,   "Jonathan Lewis" <jonathan_at_jlcomp.demon.co.uk> wrote:
>
> Does the unrecoverable_change# get updated, or
> does it behave the same way ?
>
> Personally I would expect the manual to be wrong.
> What good does it do you to know when the LAST
> unrecoverable event occurred ? Surely you need to
> know when the FIRST one occurred ? After all,
> if the file goes unrecoverable after 11:00 because of
> a direct load, it is still notionally unrecoverable after
> 11:00 even if you do another direct load at 12:00.
>
> Must be another case of the left hand tower not
> talking to the right hand tower at Redwood.
>
> --
>
> Jonathan Lewis
> Yet another Oracle-related web site: http://www.jlcomp.demon.co.uk
>
> Practical Oracle 8i: Building Efficient Databases - Publishers
 Addison
> Wesley Longman
> Book bound date now 1st Dec 2000
>

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Wed Oct 25 2000 - 17:04:21 CDT