Storage consumption on copypools increasing a lot since some weeks

[Spex]

Hello everyone,

I am faced with a problem that makes me nervous: We use 2 data domains in one SOBR for copy pools and the space requirement on our data domains has been increasing steadily since the beginning of August. In 2 or 3 weeks we will reach our limits. We have had a very stable growth rate for years (approx. 1200 VMs in the backup - individual VMs should not change the overall picture so much).

I have no idea why this is happening - what I checked:
- No need to increase the primary backup storage
- no changes to the retention period etc.
- no OS problems with DDs
- no known changes in vms to use encryption etc.
- Deduplication level doesn't change much with DDs
- no shifting of backup chains from one dd to another

What we did in July was an update to V10 P20210606 (from V10)
One of my theories is that with an MS patch, all of our VMs need additional storage space, like a paging file, a hibernating file that changes daily.

Do you see an increase in your copy pools since the beginning of August or do you have ideas where you could look elsewhere ...

Thank you for your input

[Mildur]

What is the daily change rate of your VM Jobs?

like a paging file, a hibernating file

Veeam should not backup page or hybernating files, if you have configured the exclusion. The default is to exclude this files.
https://helpcenter.veeam.com/docs/backu ... ml?ver=110

[Spex]

What is the daily change rate of your VM Jobs?

I'm not sure how to measure this, but since the space needed by our primary backups didn't change over the last weeks I would stay it is nearly constant.
Regarding hibernation or pagefile I'm aware of this it was only an example for that type of files that could make the difference if comming new from all our vms.

[Mildur]

You can see the change rate in the Job Detail view:
https://helpcenter.veeam.com/docs/backu ... ml?ver=110
- Processed
- Read
- Transferred



Is it possible, that a job has created a new backup chain on the second datadomain?
Does the Data Domain garbage collection run without any issues?

[Spex]

Is it possible, that a job has created a new backup chain on the second datadomain?

I checked this for the largest jobs - no occurances.

Does the Data Domain garbage collection run without any issues?

Yes on both dd's ok

Daily change rate for some jobs: (seems reasonable)
9.5TB 1.8TB 0.56TB - 55 vms
10.4TB 1.4TB 0.5TB - 60 vms
46 .3TB 2.7TB 0.61TB - 89 vms
22.6TB 4.7TB 1.2TB - 59 vms

[Spex]

No 100% confirmed by support, but there seems to be a difference in storage consumption for copyjobs depending on whether you are using "read from source" or "sythnetic" processing. The second setting does not fill space in the former backups that is becoming free. It appends new data to the existing backups.
Good to know.....
Will post final confirmation when available.