-
Bug
-
Resolution: Done
-
Major
-
None
-
CNV v4.18.0
-
Future Sustainability
-
3
-
False
-
-
False
-
None
-
Green
-
-
Yes
-
Important
-
None
Red Hat Documentation states the process of hot plugging cpu/memory but it does not clearly mention the max limit before a VM restart is required to take this into effect. This is important for the users to plan in advance about the initial VM sizing and what max they can hot plug before a restart is required.
Existing doc links:
- https://docs.redhat.com/en/documentation/openshift_container_platform/4.18/html-single/virtualization/index#virt-hot-plugging-memory_virt-edit-vms
- https://docs.redhat.com/en/documentation/openshift_container_platform/4.18/html-single/virtualization/index#virt-hot-plugging-cpu_virt-edit-vms
We may write something like this to aware the users in both CPU and Memory sections linked above:
The system applies these changes immediately. If the VM is migratable, a live migration is triggered. If not, or if the changes cannot be live-updated, a RestartRequired condition is added to the VM. Each time a VM is started, there is a maxSockets(cpu) and maxGuest (memory) value that is calculated and defined in the XML for the VM. You cannot hotplug above these values without restarting. By default that value is 4x whatever we start the VM with. A 1 CPU / 2 GB VM can hotplug up to 4 cpu / 8GB max before a restart is required for the VM. This default behavior can be modified cluster-wide by `maxHotplugRatio` which modifies this calculation, or can be set per VM basis.