Vcenter Alerts while doing Veeam Backups

[skumar]

When I do Veeam Backups on my Vcenter Server, I get alert messages regarding the health status. Below is the alert that I receive

Event details:
vCenter Service overall health changed from 'green' to 'red'

The alert message is generated as soon as the backup is completed. The backup gets completed successfully.
In looking at the events at my vcenter server, it looks like this happens when Veeam is trying to remove the snapshot. Any idea why this is happening?

[Gostev]

Actually, Veeam does not perform snapshot removal. All we do is ask vCenter to delete snapshot by issuing RemoveSnapshot() API call, the rest is beyond our control. We cannot affect or control how vCenter/ESX process our request. If your vCenter has some troubles processing snapshot removal, it is best to open support case with VMware and have them take a look at vCenter logs.

I certainly do not recall seeing this issue reported ever before, however to my knowledge most activity during snapshot removal happens on ESX server hosting corresponding VM, and not on vCenter - so I am not so sure why and how this could affect vCenter even theoretically (unlike, for example, monitoring/reporting products which may load vCenter with performance and configuration queries quite noticeably).

[Bunce]

When I do Veeam Backups on my Vcenter Server, I get alert messages regarding the health status. Below is the alert that I receive

Event details:
vCenter Service overall health changed from 'green' to 'red'

The alert message is generated as soon as the backup is completed. The backup gets completed successfully.
In looking at the events at my vcenter server, it looks like this happens when Veeam is trying to remove the snapshot. Any idea why this is happening?

You need to look at your VCenter configuration to see what alerts you have configued and what their tolerances are.

Its likely that you have an alert which is triggered when CPU usage (or similar) reaches something like 80% (either at VM or host level). These are configurable.

[nickf]

I think ours does this when the replication of the vCenter VM server is happening, it causes the vCenter server to briefly stop communicating with the ESX hosts and generates an alarm for each host, we just ignore so I don't have a workaround...

[Gostev]

I thought the question was about backup activities affecting physical vCenter VM. If above is related to processing vCenter VM itself, then yes this could be due to VM stun on snapshot removal... there is dedicated topic about issues with snapshot removal.

[Rohail2004]

[Merged into existing discussion]

Whenever vCenter backup runs, then vCenter sent out false alerts, such as below. This issue started happening few days ago, and I’m wondering if it is a known issue. False alarm does not generated, unless the vCenter backup is running.


Alarm Definition:
([Yellow state Is equal to disconnected; Red state Is equal to notResponding] AND [Red state Not equal to standBy])

Current values for metric/state:
State = Not responding AND State = Unknown

[Alexey D.]

Rohail, please refer to the posts above. Does your problem also seem to be related to snapshot removal phase?

[Rohail2004]

No, as soon as vCenter backup starts, then false alerts are being generated. how can I stop this? the host vCenter is running lower than 20 % in memory, and 14% CPU

[Gostev]

Hi Rohail, I understand that your vCenter server reports having an issue processing certain public API calls. With this issue, best would be to contact VMware support and ask them to find out what health issue specifically vCenter server is reporting. We do not possess vCenter source code to looks up the possible issue, nor sufficient knowledge to make sense of vCenter server logs. Thanks.