You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Reject conflicting healthcheck ports in the DRA kubelet-plugin daemonset #2979
When ComputeDomains are enabled in the GPUCluster CR (draDriver.computeDomains.enabled=true), the kubelet-plugin daemonset will have two main containers (one for GPUs and one for ComputeDomains). Both containers run a gRPC healthcheck service that is checked by a liveness probe (see here and here). The GPUCluster controller should validate that different ports are configured for the two healthcheck services, and return an error if there is a collision.
Hello all. This is my first contribution to this project. If my submitted PR for this issue fix is incorrectly formatted please advise. I have been able to replicate this issue with a local test bed and in the same test environment, validate both good and bad configs in the fixed version proposed. Thanks
Activity
jramkissoon commentedon Oct 8, 2026
Hello all. This is my first contribution to this project. If my submitted PR for this issue fix is incorrectly formatted please advise. I have been able to replicate this issue with a local test bed and in the same test environment, validate both good and bad configs in the fixed version proposed. Thanks
Reject reserved listener env vars in DRA kubelet-plugin config
Reject reserved listener env vars in DRA kubelet-plugin config
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Cover computeDomains healthcheck collisions with metrics ports in tests
Add METRICS_PATH scenarios to the issue NVIDIA#2979 test bed
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Drop TestDRADriverKubeletPluginReservedEnv duplicated by the merge
Reproduce the HTTP_ENDPOINT override regression in the test bed
Validate the resolved HTTP_ENDPOINT instead of reserving it
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Validate the HTTP_ENDPOINT fix on the test bed
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Validate the distinct-hosts fix on the test bed
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Validate the $(VAR) expansion fix on the test bed
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Validate the service-name fix on the test bed
Add an e2e runner for the issue NVIDIA#2979 test bed
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Merge branch 'issue/NVIDIA#2979' into gpu-operator-issue-NVIDIA#2979
Validate the Devin review fixes on the test bed
Reject conflicting DRA kubelet-plugin healthcheck ports
Reject conflicting DRA kubelet-plugin healthcheck ports
Add the issue NVIDIA#2979 test bed
Reject HEALTHCHECK_PORT in DRA kubelet-plugin env