Shrinking the VeeamBackup Database

[hongman]

Hi all

Our Veeam install has been extremely slow recently, and after trying every remedy under the sun I think this may be to do with the Database.

Our VeeamBackup DB is nearly 2GB. I have checked other veeam installs we have and they are couple hundred MB max...this is a huge difference?

I tried shrink etc in the Management Studio and it didnt take much off, couple MB.

This server has 18 jobs and about 6 backup copy jobs.

Running latest version of B&R, and ESXi 5.5 latest patch.

4 vCPU @ 2.13Ghz per core
6GB RAM
Windows Server 2008 R2 fully patched
E1000 NIC all offloading disabled (from VMXNET3)

[hongman]

Also, there are multiple instances of the Veeam Backup Service process in Task Manager (counted 6 last time) - is this normal??

[veremin]

In order to purge VB&R database you might want to set the database mode to simple. Additionally, you might want set the history retention period in accordance with your needs. Setting smaller value in Menu -> Options -> History -> Session history retention should result in decreased size of database.

As to the processes you see, how many running jobs you had at that time? Also, is the backup server running any other roles apart from management one? Proxy, repository, etc.?

Thanks.

[Vitaliy S.]

Can you please clarify what do you mean by saying "Veeam install is slow"? Do you refer to the GUI or the jobs themselves?

[hongman]

Hi

I ended up reinstalling Veeam and repatching up to date.

With a blank install, the GUI is fast (navigating between items etc) so I know the actual server itself is ok.

I restored the VeeamBackup DB and it has slowed down a lot, but still an improvement. Ultimately though navigating around the GUI is still quite slow (maybe 10s+ to go from item to item).

Job rates are a also a bit slower than usual and running into the day.

As I mentioned as well, there is 1 backup job running at the moment but Task Manager is showing 9 instances of Veeam.Backup.Manager.exe

[veremin]

Does the VB&R server host any other roles? Proxy, repository, etc.? The processes you see might be related to source/target services being executed on this VM in case of all-in deployment.

Thanks.

[Vitaliy S.]

Jobs performance does not depend on the database size, as to the GUI - can you please tell if you have a particular tab that is slow or all of them behave the same? Just trying to narrow down the scope of possible reasons for this.

[hongman]

The Veeam Server is the proxy but all repositories are remote (Windows and CIFS Shares)

The GUI problem is literally just clicking between the items on the left hand navigation menu (Jobs, Backups, Last 24 Hrs, etc) - this affects all tabs.

The green arrow depicting if a job is running or not also seems to not refresh itself, i.e I will log on and it will look like multiple jobs are running - but then if I click onto the job it becomes aapparent the job has finished, but the green arrow is still there. Also the same thing with the Running tab - it will show something like (3) jobs but when I click onto it there's only 1 job running. Eventually it will refresh itself but this takes quite a long time sometimes!

[veremin]

The Veeam Server is the proxy but all repositories are remote (Windows and CIFS Shares)

In case of CIFS repository, what server acts as proxying server? The backup server? If so, the target service is also executed on the backup server. So, another additional process.

Thanks.

[Vitaliy S.]

The GUI problem is literally just clicking between the items on the left hand navigation menu (Jobs, Backups, Last 24 Hrs, etc) - this affects all tabs.

Unfortunately, I'm out of ideas on what can be causing this, so I would appreciate if you could open a support case and demonstrate this behavior to our support team.

[doalwa]

Hello everybody,

we're running Veeam B&R 11 on a dedicated Dell T440 Server backing up a Microsoft HyperV 2012 R2 Failover cluster with two nodes.

Ever since upgrading to B&R 11 a few weeks ago, I'm being greeted by the following error message when logging on to the Veeam server:

Code: Select all

Configuration Database Issue
The configuration database size is approaching the limit. If this size limit is exceeded, the backup server will stop functioning.
Cick this notification for more information.

Clicking the notification does not do anything, however.

Looking at the database files located at C:\Program Files\Microsoft SQL Server\MSSQL13.VEEAMSQL2016\MSSQL\DATA
I can see that the VeeamBackup.mdf file is indeed 9,5 GB in size.

So we're indeed close on reaching the 10GB limit of the SQL Server Express edition.

The VeeamBackup_log.ldf file is a whopping 54GB in size?!? The database is running under the Simple Recovery model so I'm at a loss how this Log file got so big in the first place?

Anyway, what are my options in decreasing the size of the MDF file? Purchasing a SQL Server Standard license just to keep our backup software running isn't really in the cards.

Thanks in adcance,
Dominik

[Egor Yakovlev]

Hi Dominik,

I have merged your post into existing thread.
Please check answers above.

/Thanks!

[Gostev]

Anyway, what are my options in decreasing the size of the MDF file? Purchasing a SQL Server Standard license just to keep our backup software running isn't really in the cards.

Reduce the session history retention, as this is what consumes the bulk of the configuration database.

[DonZoomik]

LDF you can just shrink, if you're in simple recovery (IIRC you had to do full backup after switching from Full, don't remember exactly). Maybe at some point it was in full recovery mode or somebody changed it "just for kicks" and later changed back?
Make sure that your used space in DB is also high SSMS, MDF never shrinks by default and there can be a lot of free space in database depending on various conditions.
If your SQL Express is up to date (2016 SP1+), you can use row/page compression on database. In my experience, it compresses down about 2-3x but YMMV. It was discussed here: topic70033.html