Cri-timeout is not propagate to health-checker

[lenglet-k]

Hello,

We use this following custom plugin config:

      health-checker-containerd.json: |
        {
          "plugin": "custom",
          "pluginConfig": {
            "invoke_interval": "10s",
            "timeout": "3m",
            "max_output_length": 80,
            "concurrency": 1
          },
          "source": "health-checker",
          "metricsReporting": true,
          "conditions": [
            {
              "type": "ContainerRuntimeUnhealthy",
              "reason": "ContainerRuntimeIsHealthy",
              "message": "Container runtime on the node is functioning properly"
            }
          ],
          "rules": [
            {
              "type": "permanent",
              "condition": "ContainerRuntimeUnhealthy",
              "reason": "ContainerdUnhealthy",
              "path": "/home/kubernetes/bin/health-checker",
              "args": [
                "--component=cri",
                "--enable-repair=true",
                "--cooldown-time=2m",
                "--cri-timeout=10s",
                "--health-check-timeout=60s"
              ],
              "timeout": "3m"
            }
          ]
        }

and the --cri-timeout is not propagate to health_checker.go:194:

I0313 15:49:00.940744 3782720 health_checker.go:194] command /usr/bin/crictl --timeout=2s --runtime-endpoint=unix:///var/run/containerd/containerd.sock pods --latest

how to pass this configuration properly ?

[santoshkal]

I tried to reproduce this locally. I see the --cri-timeout propogated correctly:

$ ./bin/health-checker \
--component=cri \
--enable-repair=false \
--cri-timeout=10s \
--health-check-timeout=60s \
-v=4 2>&1 | grep "timeout"
I0317 20:33:35.343653 1932571 health_checker.go:194] command /usr/bin/crictl --timeout=10s --runtime-endpoint=unix:///var/run/containerd/containerd.sock pods --latest failed: exit status 1, time="2026-03-17T20:33:35+05:30" level=debug msg="Get runtime connection"
time="2026-03-17T20:33:35+05:30" level=debug msg="Using runtime connection timeout: 10s"

[lenglet-k]

I tried to reproduce this locally. I see the --cri-timeout propogated correctly:

$ ./bin/health-checker
--component=cri
--enable-repair=false
--cri-timeout=10s
--health-check-timeout=60s
-v=4 2>&1 | grep "timeout"
I0317 20:33:35.343653 1932571 health_checker.go:194] command /usr/bin/crictl --timeout=10s --runtime-endpoint=unix:///var/run/containerd/containerd.sock pods --latest failed: exit status 1, time="2026-03-17T20:33:35+05:30" level=debug msg="Get runtime connection"
time="2026-03-17T20:33:35+05:30" level=debug msg="Using runtime connection timeout: 10s"

Yes when i run manually the binary the configuration it's fine. But not when we use a custom rule like that:

          "rules": [
            {
              "type": "permanent",
              "condition": "ContainerRuntimeUnhealthy",
              "reason": "ContainerdUnhealthy",
              "path": "/home/kubernetes/bin/health-checker",
              "args": [
                "--component=cri",
                "--enable-repair=true",
                "--cooldown-time=2m",
                "--cri-timeout=10s",
                "--health-check-timeout=60s"
              ],
              "timeout": "3m"
            }

--> "--cri-timeout=10s", it's not propagate.