Skip to content

Update GKE CA-certificate mitigation DaemonSet for preloaded GPU installer and NPD - #80

Merged
SergeyKanzhelev merged 1 commit into
GoogleCloudPlatform:masterfrom
chenglongtang1:cos-ca-cert-mitigation-gpu-npd
Oct 23, 2026
Merged

SergeyKanzhelev merged 1 commit into
GoogleCloudPlatform:masterfrom
chenglongtang1:cos-ca-cert-mitigation-gpu-npd

Conversation

@chenglongtang1

Copy link
Copy Markdown
Contributor

Updates cos-ca-cert-mitigation/gke/cos-ca-cert-mitigation.yaml and cos-ca-cert-mitigation/README.md with three improvements for GKE nodes:

  • Retag preloaded GPU installer image (cos-nvidia-installer:fixed): On GKE GPU nodes where docker.io/library/cos-nvidia-installer:fixed is preloaded in containerd (k8s.io namespace), pulls gcr.io/cos-cloud/cos-gpu-installer:v2.7.11 and retags docker.io/library/cos-nvidia-installer:fixed and gcr.io/cos-cloud/cos-gpu-installer:fixed to the patched v2.7.11 image.
  • Restart gke-node-problem-detector: Adds gke-node-problem-detector alongside node-problem-detector in the active systemd daemon restart loop so GKE's Node Problem Detector service reloads the updated system CA trust store.
  • Add priorityClassName: system-node-critical: Ensures the cos-ca-cert-mitigator DaemonSet pod is scheduled on fully allocated nodes and protected from preemption.

…aller and NPD

- Retag preloaded `docker.io/library/cos-nvidia-installer:fixed` and `gcr.io/cos-cloud/cos-gpu-installer:fixed` in containerd (`k8s.io` namespace) to `gcr.io/cos-cloud/cos-gpu-installer:v2.7.11` when present on GKE GPU nodes.
- Add `gke-node-problem-detector` to the active systemd daemon restart list.
- Add `priorityClassName: system-node-critical` to the DaemonSet pod spec.
@chenglongtang1
chenglongtang1 force-pushed the cos-ca-cert-mitigation-gpu-npd branch from dd0f58f to c7f213c Compare October 23, 2026 07:01
@SergeyKanzhelev
SergeyKanzhelev merged commit d623151 into GoogleCloudPlatform:master Oct 23, 2026
7 checks passed
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.

2 participants