Does Veeam 9 stop and start ORACLE instances?

[Gostev]

Well, it depends on the data size and nothing stops you from backing up the VM but not the Oracle data. We did this a lot when I was a customer because the Oracle data was backed up by RMAN (or some other method) and many of are larger customers continue to use this approach. Veeam backs up the OS and ORACLE_HOME. Recovery times are amazing because you instant restore the server with OS/Oracle installed exactly as it was and hand the DBAs an empty system. They restore the database using whatever means they have. For a batch system that is reloading it's data from scratch daily anyway, this is actually a very awesome approach.

Yes, the OS part is easy. I was thinking total database rebuild time for a server with multiple large databases (such as big ERP system) from those dumps.

[ostandal]

Hi
Until now v8 we have used RMAN and Veeam in combination, thus Veeam giving us crash-consitant backups and RMAN dumps to NAS.
The schedules are not overalapping

In v9, can we enable application aware processing for our Oracle dbs and still continue to use RMAN?
Regards,
Standal

[foggy]

You would need to disable AAIP for those VMs in the job settings, as mentioned above.

[cxwsrei]

I changed intentionally the archivelog mode on 2 instances due to some initialization runs that cause a heavy amount of archive log files that I wanted to avoid.
But veeam is shutting down the db, switching to archivelog, performing the backup, and at the end switching back to noarchivelog (2nd shutdown) !?!
AND THIS ALL WITHOUT WARNING IN THE BACKUP LOG FILE !!!!!
REALLY A POOOOOR CONCEPT/TOOL DESIGN FROM MY POINT OF VIEW !!
Bytheway: running processes on the database are cancelled !!
WITHOUT NOTIFICATION OF AN ADMINISTRATOR THAT veeam IS CHANGING THE OPERATION MODE IS NOT ACCEPTABLE!
I´m disappointed - so many "surprises" all the time on the oracle database backup feature of veeam - not really a proven feature of veeam´s backup&recovery suite.

[PTide]

Hi,

Although I totally understand your frustration, I have to point out that if the DB in noarchivelog mode had not been shut down, that would have been against what Oracle recommends to do, resulting in a crash-consistent backup. Please see Gostev's post for explanation and links.

There are two options in this situation - you should either switch AAIP off for those VMs, or switch back to archivelog mode.

Thank you

[cxwsrei]

I understand your point.
Nevertheless it´s poor logging of the backup execution from my point of view when on the one hand veeam is shutting down the database to be able to perform the backup but on the other hand is not throwing a warning.
This is a must from my point of view.

[Gostev]

Our customers are always extremely frustrated to receive warnings during "normal" product operations. Every time we (often accidentally) were adding a warning before, this resulted in a multi-page topic with complaints within weeks of the update release, forcing us to roll the warning back. This particular case is no different - it would be impossible to explain our customers why their job does not complete with Success when it does everything according to Oracle requirements, and there are no failures or problems observed. And I understand their frustration too, because this means they will NEVER have this job complete with Success ever again, accordingly forcing them to check its reports for "real" warnings.

[cxwsrei]

Hi Gostev,
I understand. Then I suggest to add it as an option for the next release.
Each admin can decide if those warnings should be supressed or not.
Please make your product MORE flexible - give the admins MORE options.

Regards

[Gostev]

Please help me understand - given that this warning will have to be disabled by default as per my explanation ago, how having this option will help the admins?

[Regnor]

Such warnings could be shown only in the first initial backup, like the warning regarding RDMs or independent disks. On the other hand probably no one would care the warning if it disappeared in the next run.
I'm glad that I've found this topic as I'm in the process of configuring backups for some Oracle VMs.

[leodias]

Any changes to this discussion on Veeam 10?

AAP was shutting down (momentarily) our Oracle databases and causing various scheduled procedures to fail.
So we've removed these VMs from the Veeam job entirely, and are backing them up through other means.

Curious if we should:
a) Enable the Veeam job for these VMs, and backup the VM but not the Oracle data (as indicated here: post180418.html#p180418 ); or
b) Enable the Veeam job and disable AAP for these Oracle VMs, therefore keeping crash-consistent (but not application-consistent) backups

What is Veeam's latest recommendation on this front?

[leodias]

Would anything here help set up a backup approach for Oracle databases that avoids the shutdown?
Lots of ideas, just not clear if any of these actually address the issue.

https://www.veeam.com/oracle-backup-rec ... lorer.html
https://www.veeam.com/blog/how-to-backu ... tices.html

[Gostev]

Normally, Veeam performs hot backup and does not shut down Oracle databases during AAP, unless the database is in the NOARCHIVELOG mode - in which case it is the requirement by Oracle themselves to shut down the database (otherwise, the backup will be inconsistent).

If you don't use the NOARCHIVELOG mode, but see your database stopping during AAP for a brief moment, then you will want to investigate this with support to understand the root cause. Once that is known, we can go from there and give some recommendations.