Let's try to implement alert that will fullfil the request. Then it needs to be decided if it has to be backported.
If you decide to add an alert, please clone this issue or create a new issue for the documentation team to add the alert to the product documentation. Thank you
We have two distinct personas here: - The cluster admin that installs our operator. We should cover it with the health metric - A regular user that tries to start a VM. Maybe make the error more evident also here
We are moving this to 4.14 due to capacity, planning and priority.
Tracked and closed in https://issues.redhat.com/browse/CNV-20923
With CNV-22382 a new alert `KubeVirtNoAvailableNodesToRunVMs` got introduced and will fire if no nodes suitable for VMs are available. Runbook: https://github.com/kubevirt/monitoring/blob/main/docs/runbooks/KubeVirtNoAvailableNodesToRunVMs.md Please note that the alert is checking for bare metal nodes OR emulation mode being enabled.