Prepare 0.12.1 release (#4153)

.github/actions/setup-arc-e2e/action.yaml

@@ -36,8 +36,8 @@ runs:
         driver-opts: image=moby/buildkit:v0.10.6
 
     - name: Build controller image
-      # https://github.com/docker/build-push-action/releases/tag/v6.15.0
-      uses: docker/build-push-action@471d1dc4e07e5cdedd4c2171150001c434f0b7a4
+      # https://github.com/docker/build-push-action/releases/tag/v6.18.0
+      uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83
       with:
         file: Dockerfile
         platforms: linux/amd64

.github/workflows/gha-e2e-tests.yaml

@@ -16,7 +16,7 @@ env:
   TARGET_ORG: actions-runner-controller
   TARGET_REPO: arc_e2e_test_dummy
   IMAGE_NAME: "arc-test-image"
-  IMAGE_VERSION: "0.12.0"
+  IMAGE_VERSION: "0.12.1"
 
 concurrency:
   # This will make sure we only apply the concurrency limits on pull requests

.github/workflows/gha-publish-chart.yaml

@@ -93,8 +93,8 @@ jobs:
           password: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Build & push controller image
-        # https://github.com/docker/build-push-action/releases/tag/v6.15.0
-        uses: docker/build-push-action@471d1dc4e07e5cdedd4c2171150001c434f0b7a4
+        # https://github.com/docker/build-push-action/releases/tag/v6.18.0
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83
         with:
           file: Dockerfile
           platforms: linux/amd64,linux/arm64
@@ -103,8 +103,6 @@ jobs:
           tags: |
             ghcr.io/${{ steps.resolve_parameters.outputs.repository_owner }}/gha-runner-scale-set-controller:${{ inputs.release_tag_name }}
             ghcr.io/${{ steps.resolve_parameters.outputs.repository_owner }}/gha-runner-scale-set-controller:${{ inputs.release_tag_name }}-${{ steps.resolve_parameters.outputs.short_sha }}
-          cache-from: type=gha
-          cache-to: type=gha,mode=max
 
       - name: Job summary
         run: |

.github/workflows/gha-validate-chart.yaml

@@ -75,8 +75,8 @@ jobs:
           version: latest
 
       - name: Build controller image
-        # https://github.com/docker/build-push-action/releases/tag/v6.15.0
-        uses: docker/build-push-action@471d1dc4e07e5cdedd4c2171150001c434f0b7a4
+        # https://github.com/docker/build-push-action/releases/tag/v6.18.0
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83
         if: steps.list-changed.outputs.changed == 'true'
         with:
           file: Dockerfile

.github/workflows/global-publish-canary.yaml

@@ -123,8 +123,8 @@ jobs:
 
       # Unstable builds - run at your own risk
       - name: Build and Push
-        # https://github.com/docker/build-push-action/releases/tag/v6.15.0
-        uses: docker/build-push-action@471d1dc4e07e5cdedd4c2171150001c434f0b7a4
+        # https://github.com/docker/build-push-action/releases/tag/v6.18.0
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83
         with:
           context: .
           file: ./Dockerfile

charts/gha-runner-scale-set-controller/Chart.yaml

@@ -15,13 +15,13 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.12.0
+version: 0.12.1
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "0.12.0"
+appVersion: "0.12.1"
 
 home: https://github.com/actions/actions-runner-controller
 

charts/gha-runner-scale-set/Chart.yaml

@@ -15,13 +15,13 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 0.12.0
+version: 0.12.1
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "0.12.0"
+appVersion: "0.12.1"
 
 home: https://github.com/actions/actions-runner-controller
 

charts/gha-runner-scale-set/templates/_helpers.tpl

@@ -113,9 +113,9 @@ startupProbe:
     command:
       - docker
       - info
-    initialDelaySeconds: 0
-    failureThreshold: 24
-    periodSeconds: 5
+  initialDelaySeconds: 0
+  failureThreshold: 24
+  periodSeconds: 5
 {{- end }}
 volumeMounts:
   - name: work

charts/gha-runner-scale-set/templates/autoscalingrunnerset.yaml

@@ -166,11 +166,11 @@ spec:
       initContainers:
         {{- if eq $containerMode.type "dind" }}
       - name: init-dind-externals
-        {{- include "gha-runner-scale-set.dind-init-container" . | nindent 8 }}
-        {{- end }}
-        {{- if (ge (.Capabilities.KubeVersion.Minor | int) 29) }}
+          {{- include "gha-runner-scale-set.dind-init-container" . | nindent 8 }}
+          {{- if (ge (.Capabilities.KubeVersion.Minor | int) 29) }}
       - name: dind
-        {{- include "gha-runner-scale-set.dind-container" . | nindent 8 }}
+          {{- include "gha-runner-scale-set.dind-container" . | nindent 8 }}
+          {{- end }}
         {{- end }}
         {{- with .Values.template.spec.initContainers }}
       {{- toYaml . | nindent 6 }}

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

@@ -326,9 +326,16 @@ template:
   ##           command:
   ##             - docker
   ##             - info
-  ##           initialDelaySeconds: 0
-  ##           failureThreshold: 24
-  ##           periodSeconds: 5
+  ##         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

cmd/ghalistener/metrics/metrics.go

@@ -468,7 +468,7 @@ func (e *exporter) PublishStatistics(stats *actions.RunnerScaleSetStatistic) {
 	e.setGauge(MetricAssignedJobs, e.scaleSetLabels, float64(stats.TotalAssignedJobs))
 	e.setGauge(MetricRunningJobs, e.scaleSetLabels, float64(stats.TotalRunningJobs))
 	e.setGauge(MetricRegisteredRunners, e.scaleSetLabels, float64(stats.TotalRegisteredRunners))
-	e.setGauge(MetricBusyRunners, e.scaleSetLabels, float64(float64(stats.TotalBusyRunners)))
+	e.setGauge(MetricBusyRunners, e.scaleSetLabels, float64(stats.TotalBusyRunners))
 	e.setGauge(MetricIdleRunners, e.scaleSetLabels, float64(stats.TotalIdleRunners))
 }
 

controllers/actions.github.com/ephemeralrunner_controller.go

@@ -201,12 +201,16 @@ func (r *EphemeralRunnerReconciler) Reconcile(ctx context.Context, req ctrl.Requ
 	backoffDuration := failedRunnerBackoff[len(ephemeralRunner.Status.Failures)]
 	nextReconciliation := lastFailure.Add(backoffDuration)
 	if !lastFailure.IsZero() && now.Before(&metav1.Time{Time: nextReconciliation}) {
+		requeueAfter := nextReconciliation.Sub(now.Time)
 		log.Info("Backing off the next reconciliation due to failure",
 			"lastFailure", lastFailure,
 			"nextReconciliation", nextReconciliation,
-			"requeueAfter", nextReconciliation.Sub(now.Time),
+			"requeueAfter", requeueAfter,
 		)
-		return ctrl.Result{RequeueAfter: now.Sub(nextReconciliation)}, nil
+		return ctrl.Result{
+			Requeue:      true,
+			RequeueAfter: requeueAfter,
+		}, nil
 	}
 
 	secret := new(corev1.Secret)
@@ -293,28 +297,10 @@ func (r *EphemeralRunnerReconciler) Reconcile(ctx context.Context, req ctrl.Requ
 		}
 		return ctrl.Result{}, nil
 
-	default:
-		// pod succeeded. We double-check with the service if the runner exists.
-		// The reason is that image can potentially finish with status 0, but not pick up the job.
-		existsInService, err := r.runnerRegisteredWithService(ctx, ephemeralRunner.DeepCopy(), log)
-		if err != nil {
-			log.Error(err, "Failed to check if runner is registered with the service")
-			return ctrl.Result{}, err
-		}
-		if !existsInService {
-			// the runner does not exist in the service, so it must be done
-			log.Info("Ephemeral runner has finished since it does not exist in the service anymore")
-			if err := r.markAsFinished(ctx, ephemeralRunner, log); err != nil {
-				log.Error(err, "Failed to mark ephemeral runner as finished")
-				return ctrl.Result{}, err
-			}
-			return ctrl.Result{}, nil
-		}
-
-		// The runner still exists. This can happen if the pod exited with 0 but fails to start
-		log.Info("Ephemeral runner pod has finished, but the runner still exists in the service. Deleting the pod to restart it.")
-		if err := r.deletePodAsFailed(ctx, ephemeralRunner, pod, log); err != nil {
-			log.Error(err, "failed to delete a pod that still exists in the service")
+	default: // succeeded
+		log.Info("Ephemeral runner has finished successfully")
+		if err := r.markAsFinished(ctx, ephemeralRunner, log); err != nil {
+			log.Error(err, "Failed to mark ephemeral runner as finished")
 			return ctrl.Result{}, err
 		}
 		return ctrl.Result{}, nil
@@ -752,35 +738,6 @@ func (r *EphemeralRunnerReconciler) updateRunStatusFromPod(ctx context.Context,
 	return nil
 }
 
-// runnerRegisteredWithService checks if the runner is still registered with the service
-// Returns found=false and err=nil if ephemeral runner does not exist in GitHub service and should be deleted
-func (r EphemeralRunnerReconciler) runnerRegisteredWithService(ctx context.Context, runner *v1alpha1.EphemeralRunner, log logr.Logger) (found bool, err error) {
-	actionsClient, err := r.GetActionsService(ctx, runner)
-	if err != nil {
-		return false, fmt.Errorf("failed to get Actions client for ScaleSet: %w", err)
-	}
-
-	log.Info("Checking if runner exists in GitHub service", "runnerId", runner.Status.RunnerId)
-	_, err = actionsClient.GetRunner(ctx, int64(runner.Status.RunnerId))
-	if err != nil {
-		actionsError := &actions.ActionsError{}
-		if !errors.As(err, &actionsError) {
-			return false, err
-		}
-
-		if actionsError.StatusCode != http.StatusNotFound ||
-			!actionsError.IsException("AgentNotFoundException") {
-			return false, fmt.Errorf("failed to check if runner exists in GitHub service: %w", err)
-		}
-
-		log.Info("Runner does not exist in GitHub service", "runnerId", runner.Status.RunnerId)
-		return false, nil
-	}
-
-	log.Info("Runner exists in GitHub service", "runnerId", runner.Status.RunnerId)
-	return true, nil
-}
-
 func (r *EphemeralRunnerReconciler) deleteRunnerFromService(ctx context.Context, ephemeralRunner *v1alpha1.EphemeralRunner, log logr.Logger) error {
 	client, err := r.GetActionsService(ctx, ephemeralRunner)
 	if err != nil {

controllers/actions.github.com/ephemeralrunner_controller_test.go

@@ -675,53 +675,6 @@ var _ = Describe("EphemeralRunner", func() {
 			).Should(BeEquivalentTo(true))
 		})
 
-		It("It should re-create pod on exit status 0, but runner exists within the service", func() {
-			pod := new(corev1.Pod)
-			Eventually(
-				func() (bool, error) {
-					if err := k8sClient.Get(ctx, client.ObjectKey{Name: ephemeralRunner.Name, Namespace: ephemeralRunner.Namespace}, pod); err != nil {
-						return false, err
-					}
-					return true, nil
-				},
-				ephemeralRunnerTimeout,
-				ephemeralRunnerInterval,
-			).Should(BeEquivalentTo(true))
-
-			pod.Status.ContainerStatuses = append(pod.Status.ContainerStatuses, corev1.ContainerStatus{
-				Name: v1alpha1.EphemeralRunnerContainerName,
-				State: corev1.ContainerState{
-					Terminated: &corev1.ContainerStateTerminated{
-						ExitCode: 0,
-					},
-				},
-			})
-			err := k8sClient.Status().Update(ctx, pod)
-			Expect(err).To(BeNil(), "failed to update pod status")
-
-			updated := new(v1alpha1.EphemeralRunner)
-			Eventually(func() (bool, error) {
-				err := k8sClient.Get(ctx, client.ObjectKey{Name: ephemeralRunner.Name, Namespace: ephemeralRunner.Namespace}, updated)
-				if err != nil {
-					return false, err
-				}
-				return len(updated.Status.Failures) == 1, nil
-			}, ephemeralRunnerTimeout, ephemeralRunnerInterval).Should(BeEquivalentTo(true))
-
-			// should re-create after failure
-			Eventually(
-				func() (bool, error) {
-					pod := new(corev1.Pod)
-					if err := k8sClient.Get(ctx, client.ObjectKey{Name: ephemeralRunner.Name, Namespace: ephemeralRunner.Namespace}, pod); err != nil {
-						return false, err
-					}
-					return true, nil
-				},
-				ephemeralRunnerTimeout,
-				ephemeralRunnerInterval,
-			).Should(BeEquivalentTo(true))
-		})
-
 		It("It should not set the phase to succeeded without pod termination status", func() {
 			pod := new(corev1.Pod)
 			Eventually(

docs/gha-runner-scale-set-controller/README.md

@@ -43,6 +43,14 @@ You can follow [this troubleshooting guide](https://docs.github.com/en/actions/h
 
 ## Changelog
 
+### 0.12.1
+
+1. Fix indentation of startupProbe attributes in dind sidecar [#4126](https://github.com/actions/actions-runner-controller/pull/4126)
+1. Remove duplicate float64 call [#4139](https://github.com/actions/actions-runner-controller/pull/4139)
+1. Fix dind sidecar template [#4128](https://github.com/actions/actions-runner-controller/pull/4128)
+1. Remove check if runner exists after exit code 0 [#4142](https://github.com/actions/actions-runner-controller/pull/4142)
+1. Explicitly requeue during backoff ephemeral runner [#4152](https://github.com/actions/actions-runner-controller/pull/4152)
+
 ### 0.12.0
 
 1. Allow use of client id as an app id [#4057](https://github.com/actions/actions-runner-controller/pull/4057)