missing Cluster / Host

[felyjos]

Hello,
I am missing:
- One Cluster
- Several hosts from this cluster 4 out of 10 ONLY!!!
... Hosts were moved from one existing cluster to the one missing about a month ago...

Any idea?
Thanks,
Dom

[sergey.g]

Hello Dominique,

Do you use split clusters functionality in our extensions service? This could explain the issue if one of your collectors with one part of the cluster (cluster job and 6 missing hosts) are on the other collector and this collector is overloaded or cannot process the topology because of other issue. For instance if account which is being used for monitoring is not authorized to get data from other hosts or if this other collector with missing hosts is behind the firewall of some sort.

We would highly appreciate additional information to help you solve the issue. Do you see any alerts from collectors or extension service?

Thanks.

[felyjos]

Hello Dominique,

Do you use split clusters functionality in our extensions service? This could explain the issue if one of your collectors with one part of the cluster (cluster job and 6 missing hosts) are on the other collector and this collector is overloaded or cannot process the topology because of other issue. For instance if account which is being used for monitoring is not authorized to get data from other hosts or if this other collector with missing hosts is behind the firewall of some sort.

We would highly appreciate additional information to help you solve the issue. Do you see any alerts from collectors or extension service?

Thanks.

Let me review this

Thanks,
Dom

[felyjos]

Do you use split clusters functionality in our extensions service? YES

YES the collectors are heavily loaded (Health Service (Monitoring Host) Handle Count has exceeded the threshold) not sure overloaded...

from 0% (VOPVEEAMMGR1 - VRPVEEAMCOL1 - VRPVEEAMCOL2) to 2.25% (VOPVEEAMCOL1) to 3.56% (VOPVEEAMCOL2) Monitoring Load on instance _Total
from 0% (VOPVEEAMMGR1 - VRPVEEAMCOL1 - VRPVEEAMCOL2) to 0.3% (VOPVEEAMCOL2) to 0.8 % (VOPVEEAMCOL1 with peak at 5%) processor Time on instance HealthService
from 0% (VOPVEEAMMGR1 - VRPVEEAMCOL1 - VRPVEEAMCOL2) to 0.25 (VOPVEEAMCOL1) to 0.5 % (VOPVEEAMCOL2) Processor Time for the MonitoringHost instance on one collector..

continuing to check the performances... on collectors
Thanks,
Dom

[felyjos]

thanks for the clues as apparently there is one collector which is not communicating well with the rest of the Veeam environment. Trying to find out why. I moved several hosts to another collector and they popped up in SCOM...
DCOM was unable to communicate with the computer VRPVEAOMGR1.ad using any of the configured protocols.
in progress
After sometime the ESX disappeared again!!!

Thanks,
Dom

[sergey.g]

Hi,

Do you have any updates? Were you able to bring the other collector to life?

Thanks.