The cluster supports configuring QoS policies for virtual machines to limit or guarantee the vCPU scheduling time slices available to virtual machines at runtime. When the total CPU resources reserved by QoS policies for all virtual machines exceed the maximum capacity of the physical CPUs on the host, the alert is triggered.
The CPU QoS reservation for virtual machine { .labels.vm_name } failed due to insufficient node resources.
The affected virtual machines cannot obtain the minimum CPU resources as configured, which may result in performance degradation.
After a virtual machine triggers HA, the host where it resides cannot meet its configured CPU QoS resource requirements.
In AOC, select the affected virtual machine. On the virtual machine details page, go to Edit basic information > vCPU and memory > Advanced attribute, select QoS policy, and verify whether the reserved CPU resources match the expected values. If the reserved CPU resources are as expected, try using hot migration to move the virtual machine to another host with sufficient CPU resources.
If CPU resources are tight across the entire cluster and hot migration cannot free up sufficient resources for the host, consider shutting down some virtual machines of lower priority first.
If neither of the above approaches is feasible in your scenario, consider adding nodes to expand the cluster.
