Excluded datastores

[rayruest]

I think I have discovered a minor issue with excluded datastores. For example: If you have a VM in a monitored datastore, but mount an ISO image from an excluded datastore, the entire VM will be excluded from monitoring. Obviously this can be avoided by remembering to dismount any ISO images, but I imaging there could be other scenarios where this could come into play.

Ray

[RomanK]

Hello Ray,

Thanks for the report.

Could you please let us know a bit more details about the settings applied? There are two tabs in Veeam ONE server settings: Monitored Datastores and Monitored VMs. Am I understanding correctly, that no rules are configured for the affected VM on the Monitored VMs tab?

Thanks.

[rayruest]

I can replicate it using this config:

Monitored datastores: A vSAN datastore and an excluded NAS datastore with ISOs.
Monitored VMs : Virtual Infrastructure

At this point, my test VM (on the vSan datastore) is not-excluded. It has an alarm for low disk space. I then mount an ISO from the excluded datastore. Within a few minutes, the alert disappears. I then dismount the ISO, give it a few minutes, and the alarm reappears.

[RomanK]

Hello Ray,

The monitored datastores tab should not affect VM monitoring in the way you've described.

The Guest disk space alarm uses the "Guest disk free space" rule, which in turn could detect relatively in % and absolute in GB.
It seems to me that the rule estimates all VM disks. When you mount ISO, you increase the total capacity of all disks. I already asked our QA team to reproduce that behavior.

As a workaround, could you please try to change the detection type to absolute values or set only system disks in the "Disk to monitor" field?

Thanks

[rayruest]

I did more experimenting, but I will play with the alerts as you suggest as well. This is more of a curiosity in behavior than a huge problem for me.

I added a 1TB empty disk to that VM. VM shows 2 space alerts on existing drives. I then migrated that new single disk to an excluded datastore. The alert disappeared. I know this isn't common anymore now that storage is cheap, but I used to do this back in the day to put vdiscs that needed capacity but not performance on cheaper storage volumes.

I migrated that disc back to the primary datastore and the alert returned.

[RomanK]

Hello Ray,

Just in case, do you have all defaults in the alarm rule? What is the version of Veeam ONE?

Thanks

[rayruest]

Yes, alert is default. I verified by clicking the default button and there was no change. We are on Version 12.0.1.2591, but I believe behavior was present in prior versions as well. Some of the VMs that were not throwing alerts had snapshots dating back prior to my upgrade, which is how I discovered the issue.

[RomanK]

Hello Ray,

Unfortunately, after our internal research, we can't say anything specific. Please open a support case and provide a case ID in this thread so we could collect all the additional details.

Thanks.