Re: Two questions

Keith,

I've come in at the tail of the thread so please forgive me if my reply is not to the point.

I think, put simply, you are relying on no (or very little) activity on the database *during* the backup. If you are backing up ana ctive database and it doesn't take too long - and the redo logs are the last thing you back up - then, yes, I guess a recovery is like doing a recovery from a sort of hot backup: the database may have moved on from the point at which you backed up the first file, but you can redo those changes.

However, you have no guarantee of consistency. If enough activity occurs on the database, the redo log activity by the time you back the logs up my have gone right round the logs and over-written those that were active when you backed up the first files. If this is the case, you won't be able to roll forward on recovery and your stuck.

I presume you are confident that everyone (including batch activity) is off the database and you won't hit this scenario. Okay! But your job is to *guarantee* data consistency, if you're responsible for backups. Can you really do this hand on heart with "inconsistent backups". If you must keep the database active while backing up, for little additional effort you can do hot backups. But, okay, then you need to have archiving on.

If what you're doing works, no-one can argue with that. But you need to be sure it will always work - or, at least, always enough.

Len

>I started this as a new thread to avoid giving the impression that I was
>just continuing that argument.
>
>Tom Kyte very kindly put a reference to the Oracle Docs re inconsistent
>backups which I've just read, but they don't seem to make sense to me:
>
>"If you run your database in NOARCHIVELOG mode, only back it up when you
>have closed it cleanly using the IMMEDIATE or NORMAL options. Inconsistent
>whole database backups of databases running in NOARCHIVELOG mode are usable
>only if the redo logs containing the changes made prior to the backup are
>available when you
>restore it--an unlikely occurrence. "
>
>Why is this unlikely? The set of non-archived redo logs and the database
>data files are sufficient to redo any outstanding changes - that's why we
>can't reuse a redo log until the database checkpoint is complete. There is
>no need to use archived redo logs. Surely, if you were doing a cold backup,
>you would be backing up both the database data files and the redo logs?
>
>"The reason that NOARCHIVELOG inconsistent backups are not recommended is
>that the datafile headers of the backed up files contain different SCNs (a
>normal shutdown guarantees the consistency of these SCNs), and because the
>database is in NOARCHIVELOG mode, no archived redo logs are available to
>apply the lost changes. For this reason, RMAN does not allow you to back up
>a database that has been running in NOARCHIVELOG mode and shut down
>abnormally because the backup is not usable for recovery. "
>
>There is *never* a need to use archived redo logs to perform instance
>recovery after a shutdown abort, so why this statement?
>
>
>
>
>
Received on Sun Aug 19 2001 - 15:18:36 CDT