gha-runner-scale-set helm chart broken with custom template.spec in 0.12.0

[rajesh-dhakad]

Checks

• I've already read https://docs.github.com/en/actions/hosting-your-own-runners/managing-self-hosted-runners-with-actions-runner-controller/troubleshooting-actions-runner-controller-errors and I'm sure my issue is not covered in the troubleshooting guide.
• I am using charts that are officially provided

Controller Version

0.12.0

Deployment Method

Helm

Checks

• This isn't a question or user support case (For Q&A and community support, go to Discussions).
• I've read the Changelog before submitting this issue and I'm sure it's not due to any recently-introduced backward-incompatible changes

To Reproduce

Create a runner scale set with a custom template spec, and you will get this error:

template:
  spec:
    initContainers:
    - name: init-dind-externals
      image: ghcr.io/actions/actions-runner:latest
      command: ["cp", "-r", "/home/runner/externals/.", "/home/runner/tmpDir/"]
      volumeMounts:
        - name: dind-externals
          mountPath: /home/runner/tmpDir
    - name: dind
      image: docker:dind
      args:
        - dockerd
        - --host=unix:///var/run/docker.sock
        - --group=$(DOCKER_GROUP_GID)
        - --registry-mirror=https://mirror.gcr.io
      env:
        - name: DOCKER_GROUP_GID
          value: "123"
      securityContext:
        privileged: true
      restartPolicy: Always
      startupProbe:
        exec:
          command:
            - docker
            - info
        initialDelaySeconds: 0
        failureThreshold: 24
        periodSeconds: 5
      volumeMounts:
        - name: work
          mountPath: /home/runner/_work
        - name: dind-sock
          mountPath: /var/run
        - name: dind-externals
          mountPath: /home/runner/externals
    containers:
    - name: runner
      image: ghcr.io/actions/actions-runner:latest
      command: ["/home/runner/run.sh"]
      env:
        - name: DOCKER_HOST
          value: unix:///var/run/docker.sock
        - name: RUNNER_WAIT_FOR_DOCKER_IN_SECONDS
          value: "120"
      volumeMounts:
        - name: work
          mountPath: /home/runner/_work
        - name: dind-sock
          mountPath: /var/run
    volumes:
    - name: work
      emptyDir: {}
    - name: dind-sock
      emptyDir: {}
    - name: dind-externals
      emptyDir: {}

Describe the bug

AutoscalingRunnerSet.actions.github.com "general-low" is invalid: [spec.template.spec.initContainers[2]: Duplicate value: map[string]interface {}{"name":"init-dind-externals"}, spec.template.spec.initContainers[3]: Duplicate value: map[string]interface {}{"name":"dind"}]

Describe the expected behavior

It should create the dind container with the custom template value --registry-mirror=https://mirror.gcr.io

Additional Context

The helm values are already provided above.

Controller Logs

helm chart issue

Runner Pod Logs

helm chart issue

[github-actions]

Hello! Thank you for filing an issue.

The maintainers will triage your issue shortly.

In the meantime, please take a look at the troubleshooting guide for bug reports.

If this is a feature request, please review our contribution guidelines.

[rajesh-dhakad]

This PR ( #4128 )claims to solve the syntax issue, but I am seeing the same issue when trying to override the args using a custom spec for dind?

[nikola-jokic]

Hey everyone, as noted in the values.yaml file, if you want to modify the spec, leave containerMode commented out.
Ref:

actions-runner-controller/charts/gha-runner-scale-set/values.yaml

Line 115 in aa14f50

## If any customization is required for dind or kubernetes mode, containerMode should remain

Can you please confirm that you have containerMode.type set to dind?

[rajesh-dhakad]

Hey everyone, as noted in the values.yaml file, if you want to modify the spec, leave containerMode commented out. Ref:

actions-runner-controller/charts/gha-runner-scale-set/values.yaml

Line 115 in aa14f50

If any customization is required for dind or kubernetes mode, containerMode should remain

Can you please confirm that you have containerMode.type set to dind?

Yes it was set to dind

containerMode:
  type: "dind"

[nikola-jokic]

Then this is the problem. We would expand the spec, duplicating container names.
Please leave the container mode empty if you use the full spec under template related to dind.

[rajesh-dhakad]

How about overriding only args values? My goal is just to add another argument.

[rojomisin]

our runners have been down all day

in response I've tried to update helm chart 0.11 -> 0.12.1 and now the arc-controller is showing

k logs -f arc-controller-gha-rs-controller-7756b77fd8-mv78b 
2025-07-01T20:14:42Z	INFO	Using options	{"runner-max-concurrent-reconciles": 2}
2025-07-01T20:14:42Z	INFO	Update strategy set to:	{"updateStrategy": "immediate"}
2025-07-01T20:14:42Z	INFO	AutoscalingListener logging parameters changed	{"LogLevel": "debug", "LogFormat": "text"}
2025-07-01T20:14:42Z	INFO	Registering scale set metrics
2025-07-01T20:14:42Z	INFO	Resource builder initializing
2025-07-01T20:14:42Z	INFO	starting manager	{"version": "0.12.1"}
W0701 20:14:42.821941       1 reflector.go:569] k8s.io/client-go@v0.32.3/tools/cache/reflector.go:251: failed to list *v1alpha1.EphemeralRunner: json: cannot unmarshal bool into Go struct field EphemeralRunnerStatus.items.status.failures of type string
E0701 20:14:42.821976       1 reflector.go:166] "Unhandled Error" err="k8s.io/client-go@v0.32.3/tools/cache/reflector.go:251: Failed to watch *v1alpha1.EphemeralRunner: failed to list *v1alpha1.EphemeralRunner: json: cannot unmarshal bool into Go struct field EphemeralRunnerStatus.items.status.failures of type string" logger="UnhandledError"

looks like 0.12.x has breaking changes, and says to delete all the CRDs but doesn't point to how to do this please help asap

these only?

 k get customresourcedefinitions.apiextensions.k8s.io | grep github
autoscalinglisteners.actions.github.com                 2025-06-05T20:07:49Z
autoscalingrunnersets.actions.github.com                2025-06-05T20:07:51Z
ephemeralrunners.actions.github.com                     2025-06-05T20:07:52Z
ephemeralrunnersets.actions.github.com                  2025-06-05T20:07:53Z

For posterity we had to delete the orphaned objects too...

# patch to remove finalizer
for r in $(kubectl get ephemeralrunner -o jsonpath='{range .items[*]}{.metadata.name}{"\n"}{end}'); do      
  kubectl patch ephemeralrunner "$r" --type=merge -p '{"metadata":{"finalizers":[]}}';sleep 2
done

# delete them
k delete ephemeralrunners.actions.github.com --all

[nikola-jokic]

Hey @rojomisin,

Just for the future reference, patch releases should not contain breaking changes, so you can upgrade without upgrading CRDs. Minor releases can contain breaking changes, so you should read the release notes and see if there are any breaking changes in the CRDs. If there are, you should clean up the old ones.

We documented the upgrade procedure here.

[nikola-jokic]

Hey @rajesh-dhakad,

That would be a customisation of the default spec, which means you just leave the container mode empty, uncomment out the part we documented under the template for the dind, and apply modifications you want.

Closing this one now since it works as expected.