Skip to content

CNV-62298#4.18: CPU hotplug with networkInterfaceMultiqueue enabled #96182

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 17, 2025
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions modules/virt-hot-plugging-cpu.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -17,9 +17,12 @@ You can increase or decrease the number of CPU sockets allocated to a virtual ma
. Select the *vCPU* radio button.
. Enter the desired number of vCPU sockets and click *Save*.
+
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.

[NOTE]
====
* You can apply memory hot-plug changes immediately by using an automatic migration. If the live migration is not triggered, you must restart the VM. You can adjust this behavior cluster-wide by using the `maxHotplugRatio` parameter or override it in your VM.
* By default, you can hot-plug a VM with up to 4 times its initial CPU or memory configuration. For example, a VM with 1 vCPU and 2 GiB memory can scale up to 4 vCPUs and 8 GiB before a restart is needed. You can adjust this behavior cluster-wide by using the `maxHotplugRatio` parameter or override it in your VM.
If a VM has the `spec.template.spec.domain.devices.networkInterfaceMultiQueue` field enabled and CPUs are hot plugged, the following behavior occurs:
* Existing network interfaces that you attach before the CPU hot plug retain their original queue count, even after you add more virtual CPUs (vCPUs). The underlying virtualization technology causes this expected behavior.
* To update the queue count of existing interfaces to match the new vCPU configuration, you can restart the VM. A restart is only necessary if the update improves performance.
* New VirtIO network interfaces that you hot plugged after the CPU hotplug automatically receive a queue count that matches the updated vCPU configuration.
====