Skip to content

[state-driver] add new toleration to handle driver upgrades#2408

Open
tariq1890 wants to merge 1 commit intomainfrom
k8s-dm-delete-pod-privs
Open

[state-driver] add new toleration to handle driver upgrades#2408
tariq1890 wants to merge 1 commit intomainfrom
k8s-dm-delete-pod-privs

Conversation

@tariq1890
Copy link
Copy Markdown
Contributor

@tariq1890 tariq1890 commented Apr 28, 2026

This change is added in anticipation of a new enhancement
in k8s-driver-manager where the driver-manager will add
a taint to a node where a gpu driver upgrade is taking
place. This taint helps the driver-manager to evict
third party gpu client pods to ensure driver module
unloads are successful during an upgrade cycle.

Related PR: NVIDIA/k8s-driver-manager#177

@coveralls
Copy link
Copy Markdown

coveralls commented Apr 28, 2026

Coverage Status

coverage: 28.31% (-0.02%) from 28.327% — k8s-dm-delete-pod-privs into main

This change is added in anticipation of a new enhancement
in k8s-driver-manager where the driver-manager will add
a taint to a node where a gpu driver upgrade is taking
place. This taint helps the driver-manager to evict
third party gpu client pods to ensure driver module
unloads are successful during an upgrade cycle.

Signed-off-by: Tariq Ibrahim <tibrahim@nvidia.com>
@tariq1890 tariq1890 force-pushed the k8s-dm-delete-pod-privs branch from f8b27d4 to a9ccf2a Compare April 30, 2026 00:47
@rajathagasthya rajathagasthya changed the title [state-driver] add delete pod privileges in cluster-scope [state-driver] add new toleration to handle driver upgrades Apr 30, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants