Add releasenote for 0.26.0 (#1796)

Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

.github/ISSUE_TEMPLATE/bug_report.md

@@ -1,36 +0,0 @@
----
-name: Bug report
-about: Create a report to help us improve
-title: ''
-assignees: ''
-
----
-
-**Describe the bug**
-A clear and concise description of what the bug is.
-
-**Checks**
-
-- [ ] My actions-runner-controller version (v0.x.y) does support the feature
-- [ ] I'm using an unreleased version of the controller I built from HEAD of the default branch
-
-**To Reproduce**
-Steps to reproduce the behavior:
-1. Go to '...'
-2. Click on '....'
-3. Scroll down to '....'
-4. See error
-
-**Expected behavior**
-A clear and concise description of what you expected to happen.
-
-**Screenshots**
-If applicable, add screenshots to help explain your problem.
-
-**Environment (please complete the following information):**
- - Controller Version [e.g. 0.18.2]
- - Deployment Method [e.g. Helm and Kustomize ]
- - Helm Chart Version [e.g. 0.11.0, if applicable]
-
-**Additional context**
-Add any other context about the problem here.

.github/ISSUE_TEMPLATE/bug_report.yml

@@ -0,0 +1,177 @@
+name: Bug Report
+description: File a bug report
+title: "Bug"
+labels: ["bug"]
+body:
+- type: input
+  id: controller-version
+  attributes:
+    label: Controller Version
+    description: Refer to semver-like release tags for controller versions. Any release tags prefixed with `actions-runner-controller-` are for chart releases
+    placeholder: ex. 0.18.2 or git commit ID
+  validations:
+    required: true
+- type: input
+  id: chart-version
+  attributes:
+    label: Helm Chart Version
+    description: Run `helm list` and see what's shown under CHART VERSION. Any release tags prefixed with `actions-runner-controller-` are for chart releases
+    placeholder: ex. 0.11.0
+- type: input
+  id: cert-manager-version
+  attributes:
+    label: CertManager Version
+    description: Run `kubectl get po -o yaml $CERT_MANAGER_POD` and see the image tag, or run `helm list` and see what's shown under APP VERSION for your cert-manager Helm release.
+    placeholder: ex. 1.8
+- type: dropdown
+  id: deployment-method
+  attributes:
+    label: Deployment Method
+    description: Which deployment method did you use to install ARC?
+    options:
+      - Helm
+      - Kustomize
+      - ArgoCD
+      - Other
+  validations:
+    required: true
+- type: textarea
+  id: cert-manager
+  attributes:
+    label: cert-manager installation
+    description: Confirm that you've installed cert-manager correctly by answering a few questions
+    placeholder: |
+      - Did you follow https://github.com/actions-runner-controller/actions-runner-controller#installation? If not, describe the installation process so that we can reproduce your environment.
+      - Are you sure you've installed cert-manager from an official source?
+      (Note that we won't provide user support for cert-manager itself. Make sure cert-manager is fully working before testing ARC or reporting a bug
+  validations:
+    required: true
+- type: checkboxes
+  id: checks
+  attributes:
+    label: Checks
+    description: Please check the boxes below before submitting
+    options:
+    - label: This isn't a question or user support case (For Q&A and community support, go to [Discussions](https://github.com/actions-runner-controller/actions-runner-controller/discussions). It might also be a good idea to contract with any of contributors and maintainers if your business is so critical and therefore you need priority support
+      required: true
+    - label: I've read [releasenotes](https://github.com/actions-runner-controller/actions-runner-controller/tree/master/docs/releasenotes) before submitting this issue and I'm sure it's not due to any recently-introduced backward-incompatible changes
+      required: true
+    - label: My actions-runner-controller version (v0.x.y) does support the feature
+      required: true
+    - label: I've already upgraded ARC (including the CRDs, see charts/actions-runner-controller/docs/UPGRADING.md for details) to the latest and it didn't fix the issue
+      required: true
+- type: textarea
+  id: resource-definitions
+  attributes:
+    label: Resource Definitions
+    description: "Add copy(s) of your resource definition(s) (RunnerDeployment or RunnerSet, and HorizontalRunnerAutoscaler. If RunnerSet, also include the StorageClass being used)"
+    render: yaml
+    placeholder: |
+      apiVersion: actions.summerwind.dev/v1alpha1
+      kind: RunnerDeployment
+      metadata:
+        name: example
+      spec:
+        #snip
+      ---
+      apiVersion: actions.summerwind.dev/v1alpha1
+      kind: RunnerSet
+      metadata:
+        name: example
+      spec:
+        #snip
+      ---
+      apiVersion: storage.k8s.io/v1
+      kind: StorageClass
+      metadata:
+        name: example
+      provisioner: ...
+      reclaimPolicy: ...
+      volumeBindingMode: ...
+      ---
+      apiVersion: actions.summerwind.dev/v1alpha1
+      kind: HorizontalRunnerAutoscaler
+      metadata:
+        name:
+      spec:
+        #snip
+  validations:
+    required: true
+- type: textarea
+  id: reproduction-steps
+  attributes:
+    label: To Reproduce
+    description: "Steps to reproduce the behavior"
+    render: markdown
+    placeholder: |
+      1. Go to '...'
+      2. Click on '....'
+      3. Scroll down to '....'
+      4. See error
+  validations:
+    required: true
+- type: textarea
+  id: actual-behavior
+  attributes:
+    label: Describe the bug
+    description: Also tell us, what did happen?
+    placeholder: A clear and concise description of what happened.
+  validations:
+    required: true
+- type: textarea
+  id: expected-behavior
+  attributes:
+    label: Describe the expected behavior
+    description: Also tell us, what did you expect to happen?
+    placeholder: A clear and concise description of what the expected behavior is.
+  validations:
+    required: true
+- type: textarea
+  id: controller-logs
+  attributes:
+    label: Controller Logs
+    description: "NEVER EVER OMIT THIS! Include logs from `actions-runner-controller`'s controller-manager pod"
+    render: shell
+    placeholder: |
+      PROVIDE THE LOGS VIA A GIST LINK (https://gist.github.com/), NOT DIRECTLY IN THIS TEXT AREA
+      
+      To grab controller logs:
+
+      # Set NS according to your setup
+      NS=actions-runner-system
+
+      # Grab the pod name and set it to $POD_NAME
+      kubectl -n $NS get po
+
+      kubectl -n $NS logs $POD_NAME > arc.log
+  validations:
+    required: true
+- type: textarea
+  id: runner-pod-logs
+  attributes:
+    label: Runner Pod Logs
+    description: "Include logs from runner pod(s)"
+    render: shell
+    placeholder: |
+      PROVIDE THE LOGS VIA A GIST LINK (https://gist.github.com/), NOT DIRECTLY IN THIS TEXT AREA
+      
+      To grab the runner pod logs:
+
+      # Set NS according to your setup. It should match your RunnerDeployment's metadata.namespace.
+      NS=default
+
+      # Grab the name of the problematic runner pod and set it to $POD_NAME
+      kubectl -n $NS get po
+
+      kubectl -n $NS logs $POD_NAME -c runner > runnerpod_runner.log
+      kubectl -n $NS logs $POD_NAME -c docker > runnerpod_docker.log
+  validations:
+    required: true
+- type: textarea
+  id: additional-context
+  attributes:
+    label: Additional Context
+    description: |
+      Add any other context about the problem here.
+
+      Tip: You can attach images or log files by clicking this area to highlight it and then dragging files in.

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,15 @@
+# Blank issues are mainly for maintainers who are known to write complete issue descriptions without need to following a form
+blank_issues_enabled: true
+contact_links:
+- name: Sponsor ARC Maintainers
+  about: If your business relies on the continued maintainance of actions-runner-controller, please consider sponsoring the project and the maintainers.
+  url: https://github.com/actions-runner-controller/actions-runner-controller/tree/master/CODEOWNERS
+- name: Ideas and Feature Requests
+  about: Wanna request a feature? Create a discussion and collect :+1:s first.
+  url: https://github.com/actions-runner-controller/actions-runner-controller/discussions/new?category=ideas
+- name: Questions and User Support
+  about: Need support using ARC? We use Discussions as the place to provide community support.
+  url: https://github.com/actions-runner-controller/actions-runner-controller/discussions/new?category=questions
+- name: Need Paid Support?
+  about: Consider contracting with any of the actions-runner-controller maintainers and contributors.
+  url: https://github.com/actions-runner-controller/actions-runner-controller/tree/master/CODEOWNERS

.github/actions/setup-docker-environment/action.yaml

@@ -29,23 +29,23 @@ runs:
       shell: bash
 
     - name: Set up QEMU
-      uses: docker/setup-qemu-action@v1
+      uses: docker/setup-qemu-action@v2
 
     - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v1
+      uses: docker/setup-buildx-action@v2
       with:
         version: latest
 
     - name: Login to DockerHub
-      if: ${{ github.ref == 'master' && github.event.pull_request.merged == true }}
-      uses: docker/login-action@v1
+      if: ${{ github.event_name == 'release' || github.event_name == 'push' && github.ref == 'refs/heads/master' }}
+      uses: docker/login-action@v2
       with:
         username: ${{ inputs.username }}
         password: ${{ inputs.password }}
 
     - name: Login to GitHub Container Registry
-      uses: docker/login-action@v1
-      if: ${{ github.ref == 'master' && github.event.pull_request.merged == true }}
+      if: ${{ github.event_name == 'release' || github.event_name == 'push' && github.ref == 'refs/heads/master' }}
+      uses: docker/login-action@v2
       with:
         registry: ghcr.io
         username: ${{ inputs.ghcr_username }}

.github/lock.yml

@@ -1,25 +0,0 @@
-# Configuration for Lock Threads
-# Repo: https://github.com/dessant/lock-threads-app
-# App: https://github.com/apps/lock
-
-# Number of days of inactivity before a closed issue or pull request is locked
-daysUntilLock: 7
-
-# Skip issues and pull requests created before a given timestamp. Timestamp must
-# follow ISO 8601 (`YYYY-MM-DD`). Set to `false` to disable
-skipCreatedBefore: false
-
-# Issues and pull requests with these labels will be ignored. Set to `[]` to disable
-exemptLabels: []
-
-# Label to add before locking, such as `outdated`. Set to `false` to disable
-lockLabel: false
-
-# Comment to post before locking. Set to `false` to disable
-lockComment: >
-  This thread has been automatically locked since there has not been
-  any recent activity after it was closed. Please open a new issue for
-  related bugs.
-
-# Assign `resolved` as the reason for locking. Set to `false` to disable
-setLockReason: true

.github/renovate.json5

@@ -13,11 +13,30 @@
     {
       // use https://github.com/actions/runner/releases
       "fileMatch": [
-        ".github/workflows/runners.yml"
-        ],
+        ".github/workflows/runners.yaml"
+      ],
       "matchStrings": ["RUNNER_VERSION: +(?<currentValue>.*?)\\n"],
       "depNameTemplate": "actions/runner",
       "datasourceTemplate": "github-releases"
+    },
+    {
+      "fileMatch": [
+        "runner/Makefile",
+        "Makefile"
+      ],
+      "matchStrings": ["RUNNER_VERSION \\?= +(?<currentValue>.*?)\\n"],
+      "depNameTemplate": "actions/runner",
+      "datasourceTemplate": "github-releases"
+    },
+    {
+      "fileMatch": [
+        "runner/actions-runner.dockerfile",
+        "runner/actions-runner-dind.dockerfile",
+        "runner/actions-runner-dind-rootless.dockerfile"
+      ],
+      "matchStrings": ["RUNNER_VERSION=+(?<currentValue>.*?)\\n"],
+      "depNameTemplate": "actions/runner",
+      "datasourceTemplate": "github-releases"
     }
   ]
-}
+}

.github/stale.yml

@@ -1,67 +0,0 @@
-# Configuration for probot-stale - https://github.com/probot/stale
-
-# Number of days of inactivity before an Issue or Pull Request becomes stale
-daysUntilStale: 30
-
-# Number of days of inactivity before an Issue or Pull Request with the stale label is closed.
-# Set to false to disable. If disabled, issues still need to be closed manually, but will remain marked as stale.
-daysUntilClose: 14
-
-# Only issues or pull requests with all of these labels are check if stale. Defaults to `[]` (disabled)
-onlyLabels: []
-
-# Issues or Pull Requests with these labels will never be considered stale. Set to `[]` to disable
-exemptLabels:
-  - pinned
-  - security
-  - enhancement
-  - refactor
-  - documentation
-  - chore
-  - bug
-  - dependencies
-  - needs-investigation
-
-# Set to true to ignore issues in a project (defaults to false)
-exemptProjects: false
-
-# Set to true to ignore issues in a milestone (defaults to false)
-exemptMilestones: false
-
-# Set to true to ignore issues with an assignee (defaults to false)
-exemptAssignees: false
-
-# Label to use when marking as stale
-staleLabel: stale
-
-# Comment to post when marking as stale. Set to `false` to disable
-markComment: >
-  This issue has been automatically marked as stale because it has not had
-  recent activity. It will be closed if no further activity occurs. Thank you
-  for your contributions.
-
-# Comment to post when removing the stale label.
-# unmarkComment: >
-#   Your comment here.
-
-# Comment to post when closing a stale Issue or Pull Request.
-# closeComment: >
-#   Your comment here.
-
-# Limit the number of actions per hour, from 1-30. Default is 30
-limitPerRun: 30
-
-# Limit to only `issues` or `pulls`
-# only: issues
-
-# Optionally, specify configuration settings that are specific to just 'issues' or 'pulls':
-# pulls:
-#   daysUntilStale: 30
-#   markComment: >
-#     This pull request has been automatically marked as stale because it has not had
-#     recent activity. It will be closed if no further activity occurs. Thank you
-#     for your contributions.
-
-# issues:
-#   exemptLabels:
-#     - confirmed

.github/workflows/publish-arc.yaml

@@ -1,26 +1,28 @@
-name: Publish Controller Image
+name: Publish ARC
 
 on:
   release:
-    types: [published]
+    types:
+      - published
+
+# https://docs.github.com/en/rest/overview/permissions-required-for-github-apps
+permissions:
+ contents: write 
+ packages: write
 
 jobs:
-  build:
-    runs-on: ubuntu-latest
+  release-controller:
     name: Release
+    runs-on: ubuntu-latest
     env:
       DOCKERHUB_USERNAME: ${{ secrets.DOCKER_USER }}
     steps:
-      - name: Set outputs
-        id: vars
-        run: echo ::set-output name=sha_short::${GITHUB_SHA::7}
-
       - name: Checkout
         uses: actions/checkout@v3
 
       - uses: actions/setup-go@v3
         with:
-          go-version: '^1.17.7'
+          go-version: '1.18.2'
 
       - name: Install tools
         run: |
@@ -39,31 +41,31 @@ jobs:
       - name: Upload artifacts
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        run: make github-release
+        run: |
+          make github-release
 
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v1
-
-      - name: Set up Docker Buildx
-        id: buildx
-        uses: docker/setup-buildx-action@v1
+      - name: Setup Docker Environment
+        id: vars
+        uses: ./.github/actions/setup-docker-environment
         with:
-          version: latest
-
-      - name: Login to DockerHub
-        uses: docker/login-action@v1
-        with:
-          username: ${{ secrets.DOCKER_USER }}
+          username: ${{ env.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKER_ACCESS_TOKEN }}
+          ghcr_username: ${{ github.actor }}
+          ghcr_password: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Build and Push
-        uses: docker/build-push-action@v2
+        uses: docker/build-push-action@v3
         with:
           file: Dockerfile
           platforms: linux/amd64,linux/arm64
+          build-args: VERSION=${{ env.VERSION }}
           push: true
           tags: |
             ${{ env.DOCKERHUB_USERNAME }}/actions-runner-controller:latest
             ${{ env.DOCKERHUB_USERNAME }}/actions-runner-controller:${{ env.VERSION }}
             ${{ env.DOCKERHUB_USERNAME }}/actions-runner-controller:${{ env.VERSION }}-${{ steps.vars.outputs.sha_short }}
-
+            ghcr.io/actions-runner-controller/actions-runner-controller:latest
+            ghcr.io/actions-runner-controller/actions-runner-controller:${{ env.VERSION }}
+            ghcr.io/actions-runner-controller/actions-runner-controller:${{ env.VERSION }}-${{ steps.vars.outputs.sha_short }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

.github/workflows/publish-canary.yaml

@@ -0,0 +1,59 @@
+name: Publish Canary Image
+
+on:
+  push:
+    branches:
+      - master
+    paths-ignore:
+      - '**.md'
+      - '.github/ISSUE_TEMPLATE/**'
+      - '.github/workflows/validate-chart.yaml'
+      - '.github/workflows/publish-chart.yaml'
+      - '.github/workflows/publish-arc.yaml'
+      - '.github/workflows/runners.yaml'
+      - '.github/workflows/validate-entrypoint.yaml'
+      - '.github/renovate.*'
+      - 'runner/**'
+      - '.gitignore'
+      - 'PROJECT'
+      - 'LICENSE'
+      - 'Makefile'
+
+# https://docs.github.com/en/rest/overview/permissions-required-for-github-apps
+permissions:
+  contents: read
+  packages: write
+
+jobs:
+  canary-build:
+    name: Build and Publish Canary Image
+    runs-on: ubuntu-latest
+    env:
+      DOCKERHUB_USERNAME: ${{ secrets.DOCKER_USER }}
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v3
+
+      - name: Setup Docker Environment
+        id: vars
+        uses: ./.github/actions/setup-docker-environment
+        with:
+          username: ${{ env.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKER_ACCESS_TOKEN }}
+          ghcr_username: ${{ github.actor }}
+          ghcr_password: ${{ secrets.GITHUB_TOKEN }}
+
+      # Considered unstable builds
+      # See Issue #285, PR #286, and PR #323 for more information
+      - name: Build and Push
+        uses: docker/build-push-action@v3
+        with:
+          file: Dockerfile
+          platforms: linux/amd64,linux/arm64
+          build-args: VERSION=canary-${{ github.sha }}
+          push: true
+          tags: |
+            ${{ env.DOCKERHUB_USERNAME }}/actions-runner-controller:canary
+            ghcr.io/${{ github.repository }}:canary
+          cache-from: type=gha,scope=arc-canary
+          cache-to: type=gha,mode=max,scope=arc-canary

.github/workflows/publish-chart.yaml

@@ -1,4 +1,4 @@
-name: Publish helm chart
+name: Publish Helm Chart
 
 on:
   push:
@@ -6,7 +6,7 @@ on:
       - master
     paths:
       - 'charts/**'
-      - '.github/workflows/on-push-master-publish-chart.yml'
+      - '.github/workflows/publish-chart.yaml'
       - '!charts/actions-runner-controller/docs/**'
       - '!**.md'
   workflow_dispatch:
@@ -15,10 +15,13 @@ env:
   KUBE_SCORE_VERSION: 1.10.0
   HELM_VERSION: v3.8.0
 
+permissions:
+  contents: read
+
 jobs:
   lint-chart:
-    runs-on: ubuntu-latest
     name: Lint Chart
+    runs-on: ubuntu-latest
     outputs:
       publish-chart: ${{ steps.publish-chart-step.outputs.publish }}
     steps:
@@ -28,7 +31,7 @@ jobs:
           fetch-depth: 0
 
       - name: Set up Helm
-        uses: azure/setup-helm@v2.0
+        uses: azure/setup-helm@v3.3
         with:
           version: ${{ env.HELM_VERSION }}
 
@@ -49,12 +52,12 @@ jobs:
               --enable-optional-test container-security-context-readonlyrootfilesystem
 
       # python is a requirement for the chart-testing action below (supports yamllint among other tests)
-      - uses: actions/setup-python@v3
+      - uses: actions/setup-python@v4
         with:
-          python-version: 3.7
+          python-version: '3.7'
 
       - name: Set up chart-testing
-        uses: helm/chart-testing-action@v2.2.1
+        uses: helm/chart-testing-action@v2.3.0
 
       - name: Run chart-testing (list-changed)
         id: list-changed
@@ -65,22 +68,23 @@ jobs:
           fi
 
       - name: Run chart-testing (lint)
-        run: ct lint --config charts/.ci/ct-config.yaml
+        run: |
+          ct lint --config charts/.ci/ct-config.yaml
 
       - name: Create kind cluster
-        uses: helm/kind-action@v1.2.0
         if: steps.list-changed.outputs.changed == 'true'
+        uses: helm/kind-action@v1.3.0
 
       # We need cert-manager already installed in the cluster because we assume the CRDs exist
       - name: Install cert-manager
+        if: steps.list-changed.outputs.changed == 'true'      
         run: |
           helm repo add jetstack https://charts.jetstack.io --force-update
           helm install cert-manager jetstack/cert-manager --set installCRDs=true --wait
-        if: steps.list-changed.outputs.changed == 'true'
 
       - name: Run chart-testing (install)
-        run: ct install --config charts/.ci/ct-config.yaml
         if: steps.list-changed.outputs.changed == 'true'
+        run: ct install --config charts/.ci/ct-config.yaml
 
       # WARNING: This relies on the latest release being inat the top of the JSON from GitHub and a clean chart.yaml
       - name: Check if Chart Publish is Needed
@@ -99,8 +103,11 @@ jobs:
   publish-chart:
     if: needs.lint-chart.outputs.publish-chart == 'true'
     needs: lint-chart
-    runs-on: ubuntu-latest
     name: Publish Chart
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write  # for helm/chart-releaser-action to push chart release and create a release
+    
 
     steps:
       - name: Checkout

.github/workflows/run-codeql.yaml

@@ -0,0 +1,32 @@
+name: Run CodeQL
+
+on:
+  push:
+    branches: 
+      - master
+  pull_request:
+    branches:
+      - master
+  schedule:
+    - cron: '30 1 * * 0'
+
+jobs:
+  analyze:
+    name: Analyze
+    runs-on: ubuntu-latest
+    permissions:
+      security-events: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v3
+
+      - name: Initialize CodeQL
+        uses: github/codeql-action/init@v2
+        with:
+          languages: go
+
+      - name: Autobuild
+        uses: github/codeql-action/autobuild@v2
+
+      - name: Perform CodeQL Analysis
+        uses: github/codeql-action/analyze@v2

.github/workflows/run-stale.yaml

@@ -0,0 +1,25 @@
+name: Run Stale Bot
+on:
+  schedule:
+    - cron: '30 1 * * *'
+
+permissions:
+  contents: read
+
+jobs:
+  stale:
+    name: Run Stale
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write         # for actions/stale to close stale issues
+      pull-requests: write  # for actions/stale to close stale PRs
+    steps:
+      - uses: actions/stale@v5
+        with:
+          stale-issue-message: 'This issue is stale because it has been open 30 days with no activity. Remove stale label or comment or this will be closed in 5 days.'
+          # turn off stale for both issues and PRs
+          days-before-stale: -1
+          # turn stale back on for issues only
+          days-before-issue-stale: 30
+          days-before-issue-close: 14
+          exempt-issue-labels: 'pinned,security,enhancement,refactor,documentation,chore,bug,dependencies,needs-investigation'

.github/workflows/runners.yaml

@@ -2,30 +2,41 @@ name: Runners
 
 on:
   pull_request:
-    types: 
+    types:
       - opened
       - synchronize
       - reopened
-      - closed
     branches:
       - 'master'
     paths:
       - 'runner/**'
-      - .github/workflows/runners.yml
+      - '!runner/Makefile'
+      - '.github/workflows/runners.yaml'
+      - '!**.md'
+  # We must do a trigger on a push: instead of a types: closed so GitHub Secrets 
+  # are available to the workflow run
+  push:
+    branches:
+      - 'master'
+    paths:
+      - 'runner/**'
+      - '!runner/Makefile'
+      - '.github/workflows/runners.yaml'
       - '!**.md'
 
 env:
-  RUNNER_VERSION: 2.289.2
+  RUNNER_VERSION: 2.296.2
   DOCKER_VERSION: 20.10.12
+  RUNNER_CONTAINER_HOOKS_VERSION: 0.1.2
   DOCKERHUB_USERNAME: summerwind
 
 jobs:
-  build:
+  build-runners:
+    name: Build ${{ matrix.name }}-${{ matrix.os-name }}-${{ matrix.os-version }}
     runs-on: ubuntu-latest
     permissions:
       packages: write
       contents: read
-    name: Build ${{ matrix.name }}-${{ matrix.os-name }}-${{ matrix.os-version }}
     strategy:
       fail-fast: false
       matrix:
@@ -33,11 +44,12 @@ jobs:
           - name: actions-runner
             os-name: ubuntu
             os-version: 20.04
-            dockerfile: Dockerfile
           - name: actions-runner-dind
             os-name: ubuntu
             os-version: 20.04
-            dockerfile: Dockerfile.dindrunner
+          - name: actions-runner-dind-rootless
+            os-name: ubuntu
+            os-version: 20.04
 
     steps:
       - name: Checkout
@@ -46,22 +58,23 @@ jobs:
       - name: Setup Docker Environment
         id: vars
         uses: ./.github/actions/setup-docker-environment
-        with: 
+        with:
           username: ${{ env.DOCKERHUB_USERNAME }}
           password: ${{ secrets.DOCKER_ACCESS_TOKEN }}
           ghcr_username: ${{ github.actor }}
           ghcr_password: ${{ secrets.GITHUB_TOKEN }}
 
       - name: Build and Push Versioned Tags
-        uses: docker/build-push-action@v2
+        uses: docker/build-push-action@v3
         with:
           context: ./runner
-          file: ./runner/${{ matrix.dockerfile }}
+          file: ./runner/${{ matrix.name }}.dockerfile
           platforms: linux/amd64,linux/arm64
-          push: ${{ github.ref == 'master' && github.event.pull_request.merged == true }}
+          push: ${{ github.event_name == 'push' && github.ref == 'refs/heads/master' }}
           build-args: |
             RUNNER_VERSION=${{ env.RUNNER_VERSION }}
             DOCKER_VERSION=${{ env.DOCKER_VERSION }}
+            RUNNER_CONTAINER_HOOKS_VERSION=${{ env.RUNNER_CONTAINER_HOOKS_VERSION }}
           tags: |
             ${{ env.DOCKERHUB_USERNAME }}/${{ matrix.name }}:v${{ env.RUNNER_VERSION }}-${{ matrix.os-name }}-${{ matrix.os-version }}
             ${{ env.DOCKERHUB_USERNAME }}/${{ matrix.name }}:v${{ env.RUNNER_VERSION }}-${{ matrix.os-name }}-${{ matrix.os-version }}-${{ steps.vars.outputs.sha_short }}
@@ -69,5 +82,5 @@ jobs:
             ghcr.io/${{ github.repository }}/${{ matrix.name }}:latest
             ghcr.io/${{ github.repository }}/${{ matrix.name }}:v${{ env.RUNNER_VERSION }}-${{ matrix.os-name }}-${{ matrix.os-version }}
             ghcr.io/${{ github.repository }}/${{ matrix.name }}:v${{ env.RUNNER_VERSION }}-${{ matrix.os-name }}-${{ matrix.os-version }}-${{ steps.vars.outputs.sha_short }}
-          cache-from: type=gha
-          cache-to: type=gha,mode=max
+          cache-from: type=gha,scope=build-${{ matrix.name }}
+          cache-to: type=gha,mode=max,scope=build-${{ matrix.name }}

.github/workflows/validate-arc.yaml

@@ -1,44 +1,59 @@
-name: CI
+name: Validate ARC
 
 on:
   pull_request:
     branches:
       - master
     paths-ignore:
-      - .github/workflows/runners.yml
-      - .github/workflows/on-push-lint-charts.yml
-      - .github/workflows/on-push-master-publish-chart.yml
-      - .github/workflows/release.yml
-      - .github/workflows/test-entrypoint.yml
-      - .github/workflows/wip.yml
-      - 'runner/**'
       - '**.md'
+      - '.github/ISSUE_TEMPLATE/**'
+      - '.github/workflows/publish-canary.yaml'
+      - '.github/workflows/validate-chart.yaml'
+      - '.github/workflows/publish-chart.yaml'
+      - '.github/workflows/runners.yaml'
+      - '.github/workflows/publish-arc.yaml'
+      - '.github/workflows/validate-entrypoint.yaml'
+      - '.github/renovate.*'
+      - 'runner/**'
       - '.gitignore'
+      - 'PROJECT'
+      - 'LICENSE'
+      - 'Makefile'
+
+permissions:
+  contents: read
 
 jobs:
-  test:
+  test-controller:
+    name: Test ARC
     runs-on: ubuntu-latest
-    name: Test
     steps:
     - name: Checkout
       uses: actions/checkout@v3
-    - uses: actions/setup-go@v3
+
+    - name: Set-up Go
+      uses: actions/setup-go@v3
       with:
-        go-version: '^1.17.7'
-    - run: go version
+        go-version: '1.18.2'
+        check-latest: false
+    
     - uses: actions/cache@v3
       with:
         path: ~/go/pkg/mod
         key: ${{ runner.os }}-go-${{ hashFiles('**/go.sum') }}
         restore-keys: |
           ${{ runner.os }}-go-
+
     - name: Install kubebuilder
       run: |
         curl -L -O https://github.com/kubernetes-sigs/kubebuilder/releases/download/v2.3.2/kubebuilder_2.3.2_linux_amd64.tar.gz
         tar zxvf kubebuilder_2.3.2_linux_amd64.tar.gz
         sudo mv kubebuilder_2.3.2_linux_amd64 /usr/local/kubebuilder
+
     - name: Run tests
-      run: make test
+      run: |
+        make test
+
     - name: Verify manifests are up-to-date
       run: |
         make manifests

.github/workflows/validate-chart.yaml

@@ -1,10 +1,10 @@
-name: Lint and Test Charts
+name: Validate Helm Chart
 
 on:
   push:
     paths:
       - 'charts/**'
-      - '.github/workflows/on-push-lint-charts.yml'
+      - '.github/workflows/validate-chart.yaml'
       - '!charts/actions-runner-controller/docs/**'
       - '!**.md'
   workflow_dispatch:
@@ -12,10 +12,13 @@ env:
   KUBE_SCORE_VERSION: 1.10.0
   HELM_VERSION: v3.8.0
 
+permissions:
+  contents: read
+
 jobs:
-  lint-test:
-    runs-on: ubuntu-latest
+  validate-chart:
     name: Lint Chart
+    runs-on: ubuntu-latest
     steps:
       - name: Checkout
         uses: actions/checkout@v3
@@ -23,7 +26,7 @@ jobs:
           fetch-depth: 0
 
       - name: Set up Helm
-        uses: azure/setup-helm@v2.0
+        uses: azure/setup-helm@v3.3
         with:
           version: ${{ env.HELM_VERSION }}
 
@@ -44,12 +47,12 @@ jobs:
               --enable-optional-test container-security-context-readonlyrootfilesystem
 
       # python is a requirement for the chart-testing action below (supports yamllint among other tests)
-      - uses: actions/setup-python@v3
+      - uses: actions/setup-python@v4
         with:
-          python-version: 3.7
+          python-version: '3.7'
 
       - name: Set up chart-testing
-        uses: helm/chart-testing-action@v2.2.1
+        uses: helm/chart-testing-action@v2.3.0
 
       - name: Run chart-testing (list-changed)
         id: list-changed
@@ -60,18 +63,20 @@ jobs:
           fi
 
       - name: Run chart-testing (lint)
-        run: ct lint --config charts/.ci/ct-config.yaml
+        run: |
+          ct lint --config charts/.ci/ct-config.yaml
 
       - name: Create kind cluster
-        uses: helm/kind-action@v1.2.0
+        uses: helm/kind-action@v1.3.0
         if: steps.list-changed.outputs.changed == 'true'
 
       # We need cert-manager already installed in the cluster because we assume the CRDs exist
       - name: Install cert-manager
+        if: steps.list-changed.outputs.changed == 'true'
         run: |
           helm repo add jetstack https://charts.jetstack.io --force-update
           helm install cert-manager jetstack/cert-manager --set installCRDs=true --wait
-        if: steps.list-changed.outputs.changed == 'true'
 
       - name: Run chart-testing (install)
-        run: ct install --config charts/.ci/ct-config.yaml
+        run: |
+          ct install --config charts/.ci/ct-config.yaml

.github/workflows/validate-runners.yaml

@@ -1,4 +1,4 @@
-name: Unit tests for entrypoint
+name: Validate Runners
 
 on:
   pull_request:
@@ -9,13 +9,17 @@ on:
       - 'test/entrypoint/**'
       - '!**.md'
 
+permissions:
+  contents: read
+
 jobs:
-  test:
-    runs-on: ubuntu-latest
+  test-runner-entrypoint:
     name: Test entrypoint
+    runs-on: ubuntu-latest
     steps:
     - name: Checkout
       uses: actions/checkout@v3
-    - name: Run unit tests for entrypoint.sh
+
+    - name: Run tests
       run: |
         make acceptance/runner/entrypoint

.github/workflows/wip.yml

@@ -1,51 +0,0 @@
-name: Publish Canary Image
-
-on:
-  push:
-    branches:
-      - master
-    paths-ignore:
-      - .github/workflows/runners.yml
-      - .github/workflows/on-push-lint-charts.yml
-      - .github/workflows/on-push-master-publish-chart.yml
-      - .github/workflows/release.yml
-      - .github/workflows/test-entrypoint.yml
-      - "runner/**"
-      - "**.md"
-      - ".gitignore"
-
-jobs:
-  build:
-    runs-on: ubuntu-latest
-    name: Build and Publish Canary Image
-    env:
-      DOCKERHUB_USERNAME: ${{ secrets.DOCKER_USER }}
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v3
-
-      - name: Set up QEMU
-        uses: docker/setup-qemu-action@v1
-
-      - name: Set up Docker Buildx
-        id: buildx
-        uses: docker/setup-buildx-action@v1
-        with:
-          version: latest
-
-      - name: Login to DockerHub
-        uses: docker/login-action@v1
-        with:
-          username: ${{ secrets.DOCKER_USER }}
-          password: ${{ secrets.DOCKER_ACCESS_TOKEN }}
-
-      # Considered unstable builds
-      # See Issue #285, PR #286, and PR #323 for more information
-      - name: Build and Push
-        uses: docker/build-push-action@v2
-        with:
-          file: Dockerfile
-          platforms: linux/amd64,linux/arm64
-          push: true
-          tags: |
-            ${{ env.DOCKERHUB_USERNAME }}/actions-runner-controller:canary

Actions-Runner-Controller-Overview.md

@@ -0,0 +1,132 @@
+## Introduction
+This document provides a high level overview of Actions Runner Controller (ARC). ARC enables running Github Actions Runners on Kubernetes (K8s) clusters.
+
+This document provides a background of Github Actions, self-hosted runners and ARC overview. By the end of the doc, the reader should have a foundation with basic scenarios and be capable of reviewing other advanced topics.
+
+## GitHub Actions
+[GitHub Actions](https://github.com/features/actions) is a continuous integration and continuous delivery (CI/CD) platform to automate your build, test, and deployment pipeline. 
+
+You can create workflows that build and test every pull request to your repository, or deploy merged pull requests to production. Your workflow contains one or more jobs which can run in sequential order or in parallel. Each job will run inside its own runner and has one or more steps that either run a script that you define or run an action, which is a reusable extension that can simplify your workflow. To learn more about about Actions - see "[Learn Github Actions](https://docs.github.com/en/actions/learn-github-actions)".
+
+## Runners
+Runners execute the job that is assigned to them by Github Actions workflow. There are two types of Runners:
+
+- [Github-hosted runners](https://docs.github.com/en/actions/using-github-hosted-runners/about-github-hosted-runners) - GitHub provides Linux, Windows, and macOS virtual machines to run your workflows. These virtual machines are hosted in the cloud by Github.
+- [Self-hosted runners](https://docs.github.com/en/actions/hosting-your-own-runners/about-self-hosted-runners) - you can host your own self-hosted runners in your own data center or cloud infrastructure. ARC deploys self-hosted runners.
+
+## Self hosted runners
+Self-hosted runners offer more control of hardware, operating system, and software tools than GitHub-hosted runners. With self-hosted runners, you can create custom hardware configurations that meet your needs with processing power or memory to run larger jobs, install software available on your local network, and choose an operating system not offered by GitHub-hosted runners.
+
+### Types of Self hosted runners
+Self-hosted runners can be physical, virtual, in a container, on-premises, or in a cloud.
+- Traditional Deployment is having a physical machine, with OS and apps on it. The runner runs on this machine and executes any jobs. It comes with the cost of owning and operating the hardware 24/7 even if it isn't in use that entire time. 
+- Virtualized deployments are simpler to manage. Each runner runs on a virtual machine (VM) that runs on a host. There could be multiple such VMs running on the same host. VMs are complete OS’s and might take time to bring up everytime a clean environment is needed to run workflows.
+- Containerized deployments are similar to VMs, but instead of bringing up entire VM’s, a container gets deployed.Kubernetes (K8s) provides a scalable and reproducible environment for containerized workloads. They are lightweight, loosely coupled, highly efficient and can be managed centrally. There are advantages to using Kubernetes (outlined "[here](https://kubernetes.io/docs/concepts/overview/what-is-kubernetes/)."), but it is more complicated and less widely-understood than the other options. A managed provider makes this much simpler to run at scale.
+
+*Actions Runner Controller(ARC) makes it simpler to run self hosted runners on K8s managed containers.*
+
+## Actions Runner Controller (ARC)
+ARC  is a K8s controller to create self-hosted runners on your K8s cluster. With few commands, you can set up self hosted runners that can scale up and down based on demand. And since these could be ephemeral and based on containers, new instances of the runner can be brought up rapidly and cleanly.
+
+### Deploying ARC
+We have a quick start guide that demonstrates how to easily deploy ARC into your K8s environment. For more details, see "[QuickStart Guide](https://github.com/actions-runner-controller/actions-runner-controller/blob/master/QuickStartGuide.md)."
+
+## ARC components
+ARC basically consists of a set of custom resources. An ARC deployment is applying these custom resources onto a K8s cluster. Once applied, it creates a set of Pods, with the Github Actions runner running within them. Github is now able to treat these Pods as self hosted runners and allocate jobs to them.
+
+### Custom resources 
+ARC consists of several custom resource definitions (Runner, Runner Set, Runner Deployment, Runner Replica Set and Horizontal Runner AutoScaler). For more information on CRDs, refer "[Kubernetes Custom Resources](https://kubernetes.io/docs/concepts/extend-kubernetes/api-extension/custom-resources/)."
+
+The helm command (in the QuickStart guide) installs the custom resources into the actions-runner-system namespace.
+```console
+helm install -f custom-values.yaml --wait --namespace actions-runner-system \
+  --create-namespace actions-runner-controller \
+  actions-runner-controller/actions-runner-controller
+ ```
+
+### Runner deployment
+Once the custom resources are installed, another command deploys ARC into your K8s cluster.
+
+![actions-runner-controller architecture](https://user-images.githubusercontent.com/53718047/183928236-ddf72c15-1d11-4304-ad6f-0a0ff251ca55.jpg)
+
+
+
+The `Deployment and Configure ARC` section in the `Quick Start guide` lists the steps to deploy ARC using a `runnerdeployment.yaml` file. Here, we will explain the details
+For more details, see "[QuickStart Guide](https://github.com/actions-runner-controller/actions-runner-controller/blob/master/QuickStartGuide.md)."
+
+```yaml
+apiVersion: actions.summerwind.dev/v1alpha1
+kind: RunnerDeployment
+metadata:
+  name: example-runnerdeploy
+spec:
+  replicas: 1
+  template:
+    spec:
+      repository: mumoshu/actions-runner-controller-ci
+```
+
+- `kind: RunnerDeployment`: indicates its a kind of custom resource RunnerDeployment.
+- `replicas: 1` : will deploy one replica. Multiple replicas can also be deployed ( more on that later).
+- `repository: mumoshu/actions-runner-controller-ci` : is the repository to link to when the pod comes up with the Actions runner (Note, this can be configured to link at the Enterprise or Organization level also).
+
+When this configuration is applied with `kubectl apply -f runnerdeployment.yaml` , ARC creates one pod `example-runnerdeploy-[**]` with 2 containers `runner` and `docker`.
+`runner` container has the github runner component installed, `docker` container has docker installed.
+
+
+### The Runner container image
+The GitHub hosted runners include a large amount of pre-installed software packages. For complete list, see "[Runner images](https://github.com/actions/virtual-environments/tree/main/images/linux)."
+
+ARC maintains a few runner images with `latest` aligning with GitHub's Ubuntu version. These images do not contain all of the software installed on the GitHub runners. They contain subset of packages from the GitHub runners: Basic CLI packages, git, docker and build-essentials. To install additional software, it is recommended to use the corresponding setup actions. For instance, `actions/setup-java` for Java or `actions/setup-node` for Node.
+
+## Executing workflows
+Now, all the setup and configuration is done. A workflow can be created in the same repository that could target the self hosted runner created from ARC. The workflow needs to have `runs-on: self-hosted` so it can target the self host pool. For more information on targeting workflows to run on self hosted runners, see "[Using Self-hosted runners](https://docs.github.com/en/actions/hosting-your-own-runners/using-self-hosted-runners-in-a-workflow)."
+
+## Scaling runners - statically with replicas count
+With a small tweak to the replicas count (for eg - `replicas: 2`) in the `runnerdeployment.yaml` file, more runners can be created. Depending on the count of replicas, those many sets of pods would be created. As before, Each pod contains the two containers.
+
+
+## Scaling runners - dynamically with Pull Driven Scaling
+ARC also allows for scaling the runners dynamically. There are two mechanisms for dynamically scaling - (1) Webhook driven scaling and (2) Pull Driven scaling, This document describes the Pull Driven scaling model.
+
+![actions-runner-controller architecture_2](https://user-images.githubusercontent.com/53718047/183928429-7000329d-38eb-4054-9879-41ae44e1ff85.jpg)
+
+
+
+You can enable scaling with 3 steps
+1) Enable `HorizontalRunnerAutoscaler` - Create a `deployment.yaml` file of type `HorizontalRunnerAutoscaler`. The schema for this file is defined below.
+2) Scaling parameters - `minReplicas` and `maxReplicas` indicates the min and max number of replicas to scale to.
+3) Scaling metrics - ARC currently supports `PercentageRunnersBusy` as a metric type. The `PercentageRunnersBusy` will poll GitHub for the number of runners in the `busy` state in the RunnerDeployment's namespace, it will then scale depending on how you have configured the scale factors.
+
+### Pull Driven Scaling Schema
+```yaml
+apiVersion: actions.summerwind.dev/v1alpha1
+kind: HorizontalRunnerAutoscaler
+metadata:
+  name: example-runner-deployment-autoscaler
+spec:
+  scaleTargetRef:
+    # Your RunnerDeployment Here
+    name: example-runnerdeploy
+    kind: RunnerDeployment
+  minReplicas: 1
+  maxReplicas: 5
+  metrics:
+  - type: PercentageRunnersBusy
+    scaleUpThreshold: '0.75'
+    scaleDownThreshold: '0.25'
+    scaleUpFactor: '2'
+    scaleDownFactor: '0.5'
+  ```
+
+For more details - please see "[Pull Driven Scaling](https://github.com/actions-runner-controller/actions-runner-controller#pull-driven-scaling)."
+
+*The period between polls is defined by the controller's `--sync-period` flag. If this flag isn't provided then the controller defaults to a sync period of `1m`, this can be configured in seconds or minutes.*
+
+## Other Configurations
+ARC supports several different advanced configuration. 
+- support for alternate runners : Setting up runner pods with Docker-In-Docker configuration.
+- managing runner groups : Managing a set of running with runner groups thus making it easy to manage different groups within enterprise
+- Webhook driven scaling.
+
+Please refer to the documentation in this repo for further details.

CODEOWNERS

@@ -0,0 +1,2 @@
+# actions-runner-controller maintainers
+* @mumoshu @toast-gear

Dockerfile

@@ -1,11 +1,10 @@
 # Build the manager binary
-FROM --platform=$BUILDPLATFORM golang:1.17 as builder
+FROM --platform=$BUILDPLATFORM golang:1.19.1 as builder
 
 WORKDIR /workspace
 
 # Make it runnable on a distroless image/without libc
 ENV CGO_ENABLED=0
-
 # Copy the Go Modules manifests
 COPY go.mod go.sum ./
 
@@ -25,7 +24,7 @@ RUN go mod download
 # With the above commmand,
 # TARGETOS can be "linux", TARGETARCH can be "amd64", "arm64", and "arm", TARGETVARIANT can be "v7".
 
-ARG TARGETPLATFORM TARGETOS TARGETARCH TARGETVARIANT
+ARG TARGETPLATFORM TARGETOS TARGETARCH TARGETVARIANT VERSION=dev
 
 # We intentionally avoid `--mount=type=cache,mode=0777,target=/go/pkg/mod` in the `go mod download` and the `go build` runs
 # to avoid https://github.com/moby/buildkit/issues/2334
@@ -37,7 +36,7 @@ env GOCACHE /build/${TARGETPLATFORM}/root/.cache/go-build
 RUN --mount=target=. \
   --mount=type=cache,mode=0777,target=${GOCACHE} \
   export GOOS=${TARGETOS} GOARCH=${TARGETARCH} GOARM=${TARGETVARIANT#v} && \
-  go build -o /out/manager main.go && \
+  go build -ldflags="-X 'github.com/actions-runner-controller/actions-runner-controller/build.Version=${VERSION}'" -o /out/manager main.go && \
   go build -o /out/github-webhook-server ./cmd/githubwebhookserver
 
 # Use distroless as minimal base image to package the manager binary

Makefile

@@ -4,7 +4,8 @@ else
 	NAME ?= summerwind/actions-runner-controller
 endif
 DOCKER_USER ?= $(shell echo ${NAME} | cut -d / -f1)
-VERSION ?= latest
+VERSION ?= dev
+RUNNER_VERSION ?= 2.296.2
 TARGETPLATFORM ?= $(shell arch)
 RUNNER_NAME ?= ${DOCKER_USER}/actions-runner
 RUNNER_TAG  ?= ${VERSION}
@@ -12,9 +13,8 @@ TEST_REPO ?= ${DOCKER_USER}/actions-runner-controller
 TEST_ORG ?=
 TEST_ORG_REPO ?=
 TEST_EPHEMERAL ?= false
-SYNC_PERIOD ?= 5m
+SYNC_PERIOD ?= 1m
 USE_RUNNERSET ?=
-RUNNER_FEATURE_FLAG_EPHEMERAL ?=
 KUBECONTEXT ?= kind-acceptance
 CLUSTER ?= acceptance
 CERT_MANAGER_VERSION ?= v1.1.1
@@ -56,6 +56,7 @@ GO_TEST_ARGS ?= -short
 # Run tests
 test: generate fmt vet manifests
 	go test $(GO_TEST_ARGS) ./... -coverprofile cover.out
+	go test -fuzz=Fuzz -fuzztime=10s -run=Fuzz* ./controllers
 
 test-with-deps: kube-apiserver etcd kubectl
 	# See https://pkg.go.dev/sigs.k8s.io/controller-runtime/pkg/envtest#pkg-constants
@@ -91,7 +92,7 @@ manifests: manifests-gen-crds chart-crds
 manifests-gen-crds: controller-gen yq
 	$(CONTROLLER_GEN) $(CRD_OPTIONS) rbac:roleName=manager-role webhook paths="./..." output:crd:artifacts:config=config/crd/bases
 	for YAMLFILE in config/crd/bases/actions*.yaml; do \
-		$(YQ) write --inplace "$$YAMLFILE" spec.preserveUnknownFields false; \
+		$(YQ) '.spec.preserveUnknownFields = false' --inplace "$$YAMLFILE" ; \
 	done
 
 chart-crds:
@@ -109,19 +110,16 @@ vet:
 generate: controller-gen
 	$(CONTROLLER_GEN) object:headerFile=./hack/boilerplate.go.txt paths="./..."
 
-# Build the docker image
-docker-build:
-	docker build -t ${NAME}:${VERSION} .
-	docker build -t ${RUNNER_NAME}:${RUNNER_TAG} --build-arg TARGETPLATFORM=${TARGETPLATFORM} runner
-
 docker-buildx:
-	export DOCKER_CLI_EXPERIMENTAL=enabled
+	export DOCKER_CLI_EXPERIMENTAL=enabled ;\
+	export DOCKER_BUILDKIT=1
 	@if ! docker buildx ls | grep -q container-builder; then\
 		docker buildx create --platform ${PLATFORMS} --name container-builder --use;\
 	fi
 	docker buildx build --platform ${PLATFORMS} \
 		--build-arg RUNNER_VERSION=${RUNNER_VERSION} \
 		--build-arg DOCKER_VERSION=${DOCKER_VERSION} \
+		--build-arg VERSION=${VERSION} \
 		-t "${NAME}:${VERSION}" \
 		-f Dockerfile \
 		. ${PUSH_ARG}
@@ -191,7 +189,6 @@ acceptance/deploy:
 	TEST_ORG=${TEST_ORG} TEST_ORG_REPO=${TEST_ORG_REPO} SYNC_PERIOD=${SYNC_PERIOD} \
 	USE_RUNNERSET=${USE_RUNNERSET} \
 	TEST_EPHEMERAL=${TEST_EPHEMERAL} \
-	RUNNER_FEATURE_FLAG_EPHEMERAL=${RUNNER_FEATURE_FLAG_EPHEMERAL} \
 	acceptance/deploy.sh
 
 acceptance/tests:
@@ -226,7 +223,7 @@ ifeq (, $(wildcard $(GOBIN)/controller-gen))
 	CONTROLLER_GEN_TMP_DIR=$$(mktemp -d) ;\
 	cd $$CONTROLLER_GEN_TMP_DIR ;\
 	go mod init tmp ;\
-	go get sigs.k8s.io/controller-tools/cmd/controller-gen@v0.7.0 ;\
+	go install sigs.k8s.io/controller-tools/cmd/controller-gen@v0.7.0 ;\
 	rm -rf $$CONTROLLER_GEN_TMP_DIR ;\
 	}
 endif
@@ -246,7 +243,7 @@ ifeq (, $(wildcard $(GOBIN)/yq))
 	YQ_TMP_DIR=$$(mktemp -d) ;\
 	cd $$YQ_TMP_DIR ;\
 	go mod init tmp ;\
-	go get github.com/mikefarah/yq/v3@3.4.0 ;\
+	go install github.com/mikefarah/yq/v4@v4.25.3 ;\
 	rm -rf $$YQ_TMP_DIR ;\
 	}
 endif

QuickStartGuide.md

@@ -0,0 +1,136 @@
+## Introduction
+
+GitHub Actions can be run in GitHub-hosted cloud or self hosted environments. Self-hosted runners offer more control of hardware, operating system, and software tools than GitHub-hosted runners provide.
+
+With just a few steps, you can set up your kubernetes (K8s) cluster to be a self-hosted environment.
+In this guide, we will setup prerequistes, deploy Actions Runner controller (ARC) and then target that cluster to run GitHub Action workflows.
+
+<p align="center">
+  <img src="https://user-images.githubusercontent.com/53718047/181159115-dbf41416-89a7-408c-b575-bb0d059a1a36.png" />
+</p>
+
+
+
+## Setup your K8s cluster
+
+<details><summary><sub>Create a K8s cluster, if not available.</sub></summary>
+   <sub>
+If you don't have a K8s cluster, you can install a local environment using minikube. For more information, see "[Installing minikube](https://minikube.sigs.k8s.io/docs/start/)."
+     
+     "[Using workflows](/actions/using-workflows)."
+   </sub>
+</details>
+
+:one: Install cert-manager in your cluster. For more information, see "[cert-manager](https://cert-manager.io/docs/installation/)."
+
+```shell
+kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/v1.8.2/cert-manager.yaml
+```
+<sub> *note:- This command uses v1.8.2. Please replace with a later version, if available.</sub>
+
+
+>You may also install cert-manager using Helm. For instructions, see "[Installing with Helm](https://cert-manager.io/docs/installation/helm/#installing-with-helm)."
+
+
+:two: Next, Generate a Personal Access Token (PAT) for ARC to authenticate with GitHub.
+   - Login to GitHub account and Navigate to https://github.com/settings/tokens/new.
+   - Select  **repo**.
+   - Click **Generate Token** and then copy the token locally ( we’ll need it later).
+
+
+
+
+## Deploy and Configure ARC
+1️⃣ Deploy  and configure ARC on your K8s cluster. You may use Helm or Kubectl.
+
+
+<details><summary>Helm deployment</summary>
+
+##### Add repository
+```shell
+helm repo add actions-runner-controller https://actions-runner-controller.github.io/actions-runner-controller
+```
+
+##### Install Helm chart
+```shell
+helm upgrade --install --namespace actions-runner-system --create-namespace\
+  --set=authSecret.create=true\
+  --set=authSecret.github_token="REPLACE_YOUR_TOKEN_HERE"\
+  --wait actions-runner-controller actions-runner-controller/actions-runner-controller
+```
+<sub> *note:- Replace REPLACE_YOUR_TOKEN_HERE with your PAT that was generated in Step 1 </sub>
+</details>
+
+<details><summary>Kubectl deployment</summary>
+
+##### Deploy ARC
+```shell
+kubectl apply -f \
+https://github.com/actions-runner-controller/actions-runner-controller/\
+releases/download/v0.22.0/actions-runner-controller.yaml
+```
+<sub> *note:- Replace "v0.22.0" with the version you wish to deploy </sub>
+ 
+
+##### Configure Personal Access Token
+```shell
+kubectl create secret generic controller-manager \
+    -n actions-runner-system \
+    --from-literal=github_token=REPLACE_YOUR_TOKEN_HERE
+````
+<sub> *note:- Replace REPLACE_YOUR_TOKEN_HERE with your PAT that was generated in Step 1. </sub>
+  
+  </details>
+
+2️⃣ Create the GitHub self hosted runners and configure to run against your repository.
+
+Create a `runnerdeployment.yaml` file containing..
+
+```yaml
+apiVersion: actions.summerwind.dev/v1alpha1
+kind: RunnerDeployment
+metadata:
+  name: example-runnerdeploy
+spec:
+  replicas: 1
+  template:
+    spec:
+      repository: mumoshu/actions-runner-controller-ci
+````
+<sub> *note:- Replace mumoshu/actions-runner-controller-ci with the full path to your github repository. </sub>
+
+Apply this file to your K8s cluster.
+```shell
+kubectl apply -f runnerdeployment.yaml
+````
+ 
+
+>
+>🎉 We are done - now we should have self hosted runners running in K8s configured to your repository.  🎉
+> 
+> Up Next - lets verify and execute some workflows.
+ 
+## Verify and execute workflows
+:one: Verify your setup is successful with.. 
+```shell
+$ kubectl get runners
+NAME                             REPOSITORY                             STATUS
+example-runnerdeploy2475h595fr   mumoshu/actions-runner-controller-ci   Running
+
+$ kubectl get pods
+NAME                           READY   STATUS    RESTARTS   AGE
+example-runnerdeploy2475ht2qbr 2/2     Running   0          1m
+````
+Also, this runner has been registered directly to the specified repository, you can see it in repository settings. For more information, see "[settings](https://docs.github.com/en/actions/hosting-your-own-runners/monitoring-and-troubleshooting-self-hosted-runners#checking-the-status-of-a-self-hosted-runner)."
+
+:two: You are ready to execute workflows against this self hosted runner. 
+GitHub documentation lists the steps to target Actions against self hosted runners. For more information, see "[Using self-hosted runners in a workflow - GitHub Docs](https://docs.github.com/en/actions/hosting-your-own-runners/using-self-hosted-runners-in-a-workflow#using-self-hosted-runners-in-a-workflow)."
+
+There's also has a quick start guide to get started on Actions, For more information, see "[Quick start Guide to GitHub Actions](https://docs.github.com/en/actions/quickstart)."
+
+## Next steps
+ARC provides several interesting features and capabilities. For more information, see "[readme](https://github.com/actions-runner-controller/actions-runner-controller/blob/master/README.md)."
+
+
+
+ 

SECURITY.md

@@ -0,0 +1,22 @@
+# Security Policy
+
+##  Sponsoring the project
+
+This project is maintained by a small team of two and therefore lacks the resource to provide security fixes in a timely manner.
+
+If you have important business(es) that relies on this project, please consider sponsoring the project so that the maintainer(s) can commit to providing such service.
+
+Please refer to https://github.com/sponsors/actions-runner-controller for available tiers.
+
+## Supported Versions
+
+| Version | Supported          |
+| ------- | ------------------ |
+| 0.23.0  | :white_check_mark: |
+| < 0.23.0| :x:                |
+
+## Reporting a Vulnerability
+
+To report a security issue, please email ykuoka+arcsecurity(at)gmail.com with a description of the issue, the steps you took to create the issue, affected versions, and, if known, mitigations for the issue.
+
+A maintainer will try to respond within 5 working days. If the issue is confirmed as a vulnerability, a Security Advisory will be opened. This project tries to follow a 90 day disclosure timeline.

TROUBLESHOOTING.md

@@ -1,10 +1,93 @@
 # Troubleshooting
 
-* [Invalid header field value](#invalid-header-field-value)
-* [Runner coming up before network available](#runner-coming-up-before-network-available)
-* [Deployment fails on GKE due to webhooks](#deployment-fails-on-gke-due-to-webhooks)
+* [Tools](#tools)
+* [Installation](#installation)
+  * [InternalError when calling webhook: context deadline exceeded](#internalerror-when-calling-webhook-context-deadline-exceeded)
+  * [Invalid header field value](#invalid-header-field-value)
+  * [Helm chart install failure: certificate signed by unknown authority](#helm-chart-install-failure-certificate-signed-by-unknown-authority)
+* [Operations](#operations)
+  * [Stuck runner kind or backing pod](#stuck-runner-kind-or-backing-pod)
+  * [Delay in jobs being allocated to runners](#delay-in-jobs-being-allocated-to-runners)
+  * [Runner coming up before network available](#runner-coming-up-before-network-available)
+  * [Outgoing network action hangs indefinitely](#outgoing-network-action-hangs-indefinitely)
+  * [Unable to scale to zero with TotalNumberOfQueuedAndInProgressWorkflowRuns](#unable-to-scale-to-zero-with-totalnumberofqueuedandinprogressworkflowruns)
 
-## Invalid header field value
+## Tools
+
+A list of tools which are helpful for troubleshooting
+
+* https://github.com/rewanthtammana/kubectl-fields Kubernetes resources hierarchy parsing tool
+* https://github.com/stern/stern Multi pod and container log tailing for Kubernetes
+
+## Installation
+
+Troubeshooting runbooks that relate to ARC installation problems
+
+### InternalError when calling webhook: context deadline exceeded
+
+**Problem**
+
+This issue can come up for various reasons like leftovers from previous installations or not being able to access the K8s service's clusterIP associated with the admission webhook server (of ARC).
+
+```
+Internal error occurred: failed calling webhook "mutate.runnerdeployment.actions.summerwind.dev":
+Post "https://actions-runner-controller-webhook.actions-runner-system.svc:443/mutate-actions-summerwind-dev-v1alpha1-runnerdeployment?timeout=10s": context deadline exceeded
+```
+
+**Solution**
+
+First we will try the common solution of checking webhook leftovers from previous installations:
+
+1.  ```bash
+    kubectl get validatingwebhookconfiguration -A
+    kubectl get mutatingwebhookconfiguration -A
+    ```
+2.  If you see any webhooks related to actions-runner-controller, delete them:
+    ```bash
+    kubectl delete mutatingwebhookconfiguration actions-runner-controller-mutating-webhook-configuration
+    kubectl delete validatingwebhookconfiguration actions-runner-controller-validating-webhook-configuration
+    ```
+
+If that didn't work then probably your K8s control-plane is somehow unable to access the K8s service's clusterIP associated with the admission webhook server:
+1. You're running apiserver as a binary and you didn't make service cluster IPs available to the host network. 
+2. You're running the apiserver in the pod but your pod network (i.e. CNI plugin installation and config) is not good so your pods(like kube-apiserver) in the K8s control-plane nodes can't access ARC's admission webhook server pod(s) in probably data-plane nodes.
+
+
+Another reason could be due to GKEs firewall settings you may run into the following errors when trying to deploy runners on a private GKE cluster: 
+
+To fix this, you may either:
+
+1. Configure the webhook to use another port, such as 443 or 10250, [each of
+   which allow traffic by default](https://cloud.google.com/kubernetes-engine/docs/how-to/private-clusters#add_firewall_rules).
+
+   ```sh
+   # With helm, you'd set `webhookPort` to the port number of your choice
+   # See https://github.com/actions-runner-controller/actions-runner-controller/pull/1410/files for more information
+   helm upgrade --install --namespace actions-runner-system --create-namespace \
+                --wait actions-runner-controller actions-runner-controller/actions-runner-controller \
+                --set webhookPort=10250
+   ```
+
+2. Set up a firewall rule to allow the master node to connect to the default
+   webhook port. The exact way to do this may vary, but the following script
+   should point you in the right direction:
+
+   ```sh
+   # 1) Retrieve the network tag automatically given to the worker nodes
+   # NOTE: this only works if you have only one cluster in your GCP project. You will have to manually inspect the result of this command to find the tag for the cluster you want to target
+   WORKER_NODES_TAG=$(gcloud compute instances list --format='text(tags.items[0])' --filter='metadata.kubelet-config:*' | grep tags | awk '{print $2}' | sort | uniq)
+
+   # 2) Take note of the VPC network in which you deployed your cluster
+   # NOTE this only works if you have only one network in which you deploy your clusters
+   NETWORK=$(gcloud compute instances list --format='text(networkInterfaces[0].network)' --filter='metadata.kubelet-config:*' | grep networks | awk -F'/' '{print $NF}' | sort | uniq)
+
+   # 3) Get the master source ip block
+   SOURCE=$(gcloud container clusters describe <cluster-name> --region <region> | grep masterIpv4CidrBlock| cut -d ':' -f 2 | tr -d ' ')
+
+   gcloud compute firewall-rules create k8s-cert-manager --source-ranges $SOURCE --target-tags $WORKER_NODES_TAG  --allow TCP:9443 --network $NETWORK
+   ```
+
+### Invalid header field value
 
 **Problem**
 
@@ -23,7 +106,88 @@ Your base64'ed PAT token has a new line at the end, it needs to be created witho
 * `echo -n $TOKEN | base64`
 * Create the secret as described in the docs using the shell and documented flags
 
-## Runner coming up before network available
+### Helm chart install failure: certificate signed by unknown authority
+
+**Problem**
+
+```
+Error: UPGRADE FAILED: failed to create resource: Internal error occurred: failed calling webhook "webhook.cert-manager.io": failed to call webhook: Post "https://cert-manager-webhook.cert-manager.svc:443/mutate?timeout=10s": x509: certificate signed by unknown authority
+```
+
+Apparently, it's failing while `helm` is creating one of resources defined in the ARC chart and the cause was that cert-manager's webhook is not working correctly, due to the missing or the invalid CA certficate.
+
+You'd try to tail logs from the `cert-manager-cainjector` and see it's failing with an error like:
+
+```
+$ kubectl -n cert-manager logs cert-manager-cainjector-7cdbb9c945-g6bt4
+I0703 03:31:55.159339       1 start.go:91] "starting" version="v1.1.1" revision="3ac7418070e22c87fae4b22603a6b952f797ae96"
+I0703 03:31:55.615061       1 leaderelection.go:243] attempting to acquire leader lease  kube-system/cert-manager-cainjector-leader-election...
+I0703 03:32:10.738039       1 leaderelection.go:253] successfully acquired lease kube-system/cert-manager-cainjector-leader-election
+I0703 03:32:10.739941       1 recorder.go:52] cert-manager/controller-runtime/manager/events "msg"="Normal"  "message"="cert-manager-cainjector-7cdbb9c945-g6bt4_88e4bc70-eded-4343-a6fb-0ddd6434eb55 became leader" "object"={"kind":"ConfigMap","namespace":"kube-system","name":"cert-manager-cainjector-leader-election","uid":"942a021e-364c-461a-978c-f54a95723cdc","apiVersion":"v1","resourceVersion":"1576"} "reason"="LeaderElection"
+E0703 03:32:11.192128       1 start.go:119] cert-manager/ca-injector "msg"="manager goroutine exited" "error"=null
+I0703 03:32:12.339197       1 request.go:645] Throttling request took 1.047437675s, request: GET:https://10.96.0.1:443/apis/storage.k8s.io/v1beta1?timeout=32s
+E0703 03:32:13.143790       1 start.go:151] cert-manager/ca-injector "msg"="Error registering certificate based controllers. Retrying after 5 seconds." "error"="no matches for kind \"MutatingWebhookConfiguration\" in version \"admissionregistration.k8s.io/v1beta1\""
+Error: error registering secret controller: no matches for kind "MutatingWebhookConfiguration" in version "admissionregistration.k8s.io/v1beta1"
+```
+
+**Solution**
+
+Your cluster is based on a new enough Kubernetes of version 1.22 or greater which does not support the legacy `admissionregistration.k8s.io/v1beta1` API anymore, and your `cert-manager` is not up-to-date hence it's still trying to use the leagcy Kubernetes API.
+
+In many cases, it's not an option to downgrade Kubernetes. So, just upgrade `cert-manager` to a more recent version that does have have the support for the specific Kubernetes version you're using.
+
+See https://cert-manager.io/docs/installation/supported-releases/ for the list of available cert-manager versions.
+
+## Operations
+
+Troubeshooting runbooks that relate to ARC operational problems
+
+### Stuck runner kind or backing pod
+
+**Problem**
+
+Sometimes either the runner kind (`kubectl get runners`) or it's underlying pod can get stuck in a terminating state for various reasons. You can get the kind unstuck by removing its finaliser using something like this:
+
+**Solution**
+
+Remove the finaliser from the relevent runner kind or pod
+
+```
+# Get all kind runners and remove the finalizer
+$ kubectl get runners --no-headers | awk {'print $1'} | xargs kubectl patch runner --type merge -p '{"metadata":{"finalizers":null}}'
+
+# Get all pods that are stuck terminating and remove the finalizer
+$ kubectl -n get pods | grep Terminating | awk {'print $1'} | xargs kubectl patch pod -p '{"metadata":{"finalizers":null}}'
+```
+
+_Note the code assumes you have already selected the namespace your runners are in and that they 
+are in a namespace not shared with anything else_
+
+### Delay in jobs being allocated to runners
+
+**Problem**
+
+ARC isn't involved in jobs actually getting allocated to a runner. ARC is responsible for orchestrating runners and the runner lifecycle. Why some people see large delays in job allocation is not clear however it has been https://github.com/actions-runner-controller/actions-runner-controller/issues/1387#issuecomment-1122593984 that this is caused from the self-update process somehow.
+
+**Solution**
+
+Disable the self-update process in your runner manifests
+
+```yaml
+apiVersion: actions.summerwind.dev/v1alpha1
+kind: RunnerDeployment
+metadata:
+  name: example-runnerdeployment-with-sleep
+spec:
+  template:
+    spec:
+      ...
+      env:
+        - name: DISABLE_RUNNER_UPDATE
+          value: "true"
+```
+
+### Runner coming up before network available
 
 **Problem**
 
@@ -61,40 +225,48 @@ metadata:
 spec:
   template:
     spec:
+      ...
       env:
-        # This runner's entrypoint script will have a 5 seconds delay 
-        # as a first action within the entrypoint script
         - name: STARTUP_DELAY_IN_SECONDS
           value: "5"
 ```
 
-## Deployment fails on GKE due to webhooks
+## Outgoing network action hangs indefinitely
 
 **Problem**
 
-Due to GKEs firewall settings you may run into the following errors when trying to deploy runners on a private GKE cluster:
+Some random outgoing network actions hangs indefinitely. This could be because your cluster does not give Docker the standard MTU of 1500, you can check this out by running `ip link` in a pod that encounters the problem and reading the outgoing interface's MTU value. If it is smaller than 1500, then try the following.
 
-```
-Internal error occurred: failed calling webhook "mutate.runner.actions.summerwind.dev": 
-Post https://webhook-service.actions-runner-system.svc:443/mutate-actions-summerwind-dev-v1alpha1-runner?timeout=10s: 
-context deadline exceeded
+**Solution**
+
+Add a `dockerMTU` key in your runner's spec with the value you read on the outgoing interface. For instance:
+
+```yaml
+apiVersion: actions.summerwind.dev/v1alpha1
+kind: RunnerDeployment
+metadata:
+  name: github-runner
+  namespace: github-system
+spec:
+  replicas: 6
+  template:
+    spec:
+      dockerMTU: 1400
+      repository: $username/$repo
+      env: []
 ```
 
-**Solution**<br />
+There may be more places you need to tweak for MTU.
+Please consult issues like #651 for more information.
 
-To fix this, you need to set up a firewall rule to allow the master node to connect to the webhook port.
-The exact way to do this may wary, but the following script should point you in the right direction:
+## Unable to scale to zero with TotalNumberOfQueuedAndInProgressWorkflowRuns
 
-```
-# 1) Retrieve the network tag automatically given to the worker nodes
-# NOTE: this only works if you have only one cluster in your GCP project. You will have to manually inspect the result of this command to find the tag for the cluster you want to target
-WORKER_NODES_TAG=$(gcloud compute instances list --format='text(tags.items[0])' --filter='metadata.kubelet-config:*' | grep tags | awk '{print $2}' | sort | uniq)
+**Problem**
 
-# 2) Take note of the VPC network in which you deployed your cluster
-# NOTE this only works if you have only one network in which you deploy your clusters
-NETWORK=$(gcloud compute instances list --format='text(networkInterfaces[0].network)' --filter='metadata.kubelet-config:*' | grep networks | awk -F'/' '{print $NF}' | sort | uniq)
+HRA doesn't scale the RunnerDeployment to zero, even though you did configure HRA correctly, to have a pull-based scaling metric `TotalNumberOfQueuedAndInProgressWorkflowRuns`, and set `minReplicas: 0`.
 
-# 3) Get the master source ip block
-SOURCE=$(gcloud container clusters describe <cluster-name> --region <region> | grep masterIpv4CidrBlock| cut -d ':' -f 2 | tr -d ' ')
-gcloud compute firewall-rules create k8s-cert-manager --source-ranges $SOURCE --target-tags $WORKER_NODES_TAG  --allow TCP:9443 --network $NETWORK
-```
+**Solution**
+
+You very likely have some dangling workflow jobs stuck in `queued` or `in_progress` as seen in [#1057](https://github.com/actions-runner-controller/actions-runner-controller/issues/1057#issuecomment-1133439061).
+
+Manually call [the "list workflow runs" API](https://docs.github.com/en/rest/actions/workflow-runs#list-workflow-runs-for-a-repository), and [remove the dangling workflow job(s)](https://docs.github.com/en/rest/actions/workflow-runs#delete-a-workflow-run).

acceptance/argotunnel.sh

@@ -0,0 +1,97 @@
+#!/usr/bin/env bash
+
+# See https://developers.cloudflare.com/cloudflare-one/tutorials/many-cfd-one-tunnel/
+
+kubectl create ns tunnel || :
+
+kubectl -n tunnel delete secret tunnel-credentials || :
+
+kubectl -n tunnel create secret generic tunnel-credentials \
+  --from-file=credentials.json=$HOME/.cloudflared/${TUNNEL_ID}.json || :
+
+cat <<MANIFEST | kubectl -n tunnel ${OP} -f -
+---
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: cloudflared
+spec:
+  selector:
+    matchLabels:
+      app: cloudflared
+  replicas: 2 # You could also consider elastic scaling for this deployment
+  template:
+    metadata:
+      labels:
+        app: cloudflared
+    spec:
+      containers:
+      - name: cloudflared
+        image: cloudflare/cloudflared:latest
+        args:
+        - tunnel
+        # Points cloudflared to the config file, which configures what
+        # cloudflared will actually do. This file is created by a ConfigMap
+        # below.
+        - --config
+        - /etc/cloudflared/config/config.yaml
+        - run
+        livenessProbe:
+          httpGet:
+            # Cloudflared has a /ready endpoint which returns 200 if and only if
+            # it has an active connection to the edge.
+            path: /ready
+            port: 2000
+          failureThreshold: 1
+          initialDelaySeconds: 10
+          periodSeconds: 10
+        volumeMounts:
+        - name: config
+          mountPath: /etc/cloudflared/config
+          readOnly: true
+        # Each tunnel has an associated "credentials file" which authorizes machines
+        # to run the tunnel. cloudflared will read this file from its local filesystem,
+        # and it'll be stored in a k8s secret.
+        - name: creds
+          mountPath: /etc/cloudflared/creds
+          readOnly: true
+      volumes:
+      - name: creds
+        secret:
+          secretName: tunnel-credentials
+      # Create a config.yaml file from the ConfigMap below.
+      - name: config
+        configMap:
+          name: cloudflared
+          items:
+          - key: config.yaml
+            path: config.yaml
+---
+# This ConfigMap is just a way to define the cloudflared config.yaml file in k8s.
+# It's useful to define it in k8s, rather than as a stand-alone .yaml file, because
+# this lets you use various k8s templating solutions (e.g. Helm charts) to
+# parameterize your config, instead of just using string literals.
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: cloudflared
+data:
+  config.yaml: |
+    # Name of the tunnel you want to run
+    tunnel: ${TUNNEL_NAME}
+    credentials-file: /etc/cloudflared/creds/credentials.json
+    # Serves the metrics server under /metrics and the readiness server under /ready
+    metrics: 0.0.0.0:2000
+    # Autoupdates applied in a k8s pod will be lost when the pod is removed or restarted, so
+    # autoupdate doesn't make sense in Kubernetes. However, outside of Kubernetes, we strongly
+    # recommend using autoupdate.
+    no-autoupdate: true
+    ingress:
+    # The first rule proxies traffic to the httpbin sample Service defined in app.yaml
+    - hostname: ${TUNNEL_HOSTNAME}
+      service: http://actions-runner-controller-github-webhook-server.actions-runner-system:80
+    # This rule matches any traffic which didn't match a previous rule, and responds with HTTP 404.
+    - service: http_status:404
+MANIFEST
+
+kubectl -n tunnel delete po -l app=cloudflared || :

acceptance/deploy.sh

@@ -41,8 +41,23 @@ TEST_ID=${TEST_ID:-default}
 
 if [ "${tool}" == "helm" ]; then
   set -v
+
+  CHART=${CHART:-charts/actions-runner-controller}
+
+  flags=()
+  if [ "${IMAGE_PULL_SECRET}" != "" ]; then
+    flags+=( --set imagePullSecrets[0].name=${IMAGE_PULL_SECRET})
+    flags+=( --set image.actionsRunnerImagePullSecrets[0].name=${IMAGE_PULL_SECRET})
+    flags+=( --set githubWebhookServer.imagePullSecrets[0].name=${IMAGE_PULL_SECRET})
+  fi
+  if [ "${CHART_VERSION}" != "" ]; then
+    flags+=( --version ${CHART_VERSION})
+  fi
+
+  set -vx
+
   helm upgrade --install actions-runner-controller \
-    charts/actions-runner-controller \
+    ${CHART} \
     -n actions-runner-system \
     --create-namespace \
     --set syncPeriod=${SYNC_PERIOD} \
@@ -51,6 +66,7 @@ if [ "${tool}" == "helm" ]; then
     --set image.tag=${VERSION} \
     --set podAnnotations.test-id=${TEST_ID} \
     --set githubWebhookServer.podAnnotations.test-id=${TEST_ID} \
+    ${flags[@]} --set image.imagePullPolicy=${IMAGE_PULL_POLICY} \
     -f ${VALUES_FILE}
   set +v
   # To prevent `CustomResourceDefinition.apiextensions.k8s.io "runners.actions.summerwind.dev" is invalid: metadata.annotations: Too long: must have at most 262144 bytes`
@@ -76,56 +92,3 @@ kubectl -n actions-runner-system wait deploy/actions-runner-controller --for con
 
 # Adhocly wait for some time until actions-runner-controller's admission webhook gets ready
 sleep 20
-
-RUNNER_LABEL=${RUNNER_LABEL:-self-hosted}
-
-if [ -n "${TEST_REPO}" ]; then
-  if [ "${USE_RUNNERSET}" != "false" ]; then
-    cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ENTERPRISE= TEST_ORG= RUNNER_MIN_REPLICAS=${REPO_RUNNER_MIN_REPLICAS} NAME=repo-runnerset envsubst | kubectl apply -f -
-  else
-    echo 'Deploying runnerdeployment and hra. Set USE_RUNNERSET if you want to deploy runnerset instead.'
-    cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ENTERPRISE= TEST_ORG= RUNNER_MIN_REPLICAS=${REPO_RUNNER_MIN_REPLICAS} NAME=repo-runnerdeploy envsubst | kubectl apply -f -
-  fi
-else
-  echo 'Skipped deploying runnerdeployment and hra. Set TEST_REPO to "yourorg/yourrepo" to deploy.'
-fi
-
-if [ -n "${TEST_ORG}" ]; then
-  if [ "${USE_RUNNERSET}" != "false" ]; then
-    cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} NAME=org-runnerset envsubst | kubectl apply -f -
-  else
-    cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} NAME=org-runnerdeploy envsubst | kubectl apply -f -
-  fi
-
-  if [ -n "${TEST_ORG_GROUP}" ]; then
-    if [ "${USE_RUNNERSET}" != "false" ]; then
-      cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ORG_GROUP} NAME=orgroupg-runnerset envsubst | kubectl apply -f -
-    else
-      cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ORG_GROUP} NAME=orggroup-runnerdeploy envsubst | kubectl apply -f -
-    fi
-  else
-    echo 'Skipped deploying enterprise runnerdeployment. Set TEST_ORG_GROUP to deploy.'
-  fi
-else
-  echo 'Skipped deploying organizational runnerdeployment. Set TEST_ORG to deploy.'
-fi
-
-if [ -n "${TEST_ENTERPRISE}" ]; then
-  if [ "${USE_RUNNERSET}" != "false" ]; then
-    cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} NAME=enterprise-runnerset envsubst | kubectl apply -f -
-  else
-    cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} NAME=enterprise-runnerdeploy envsubst | kubectl apply -f -
-  fi
-
-  if [ -n "${TEST_ENTERPRISE_GROUP}" ]; then
-    if [ "${USE_RUNNERSET}" != "false" ]; then
-      cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ENTERPRISE_GROUP} NAME=enterprisegroup-runnerset envsubst | kubectl apply -f -
-    else
-      cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ENTERPRISE_GROUP} NAME=enterprisegroup-runnerdeploy envsubst | kubectl apply -f -
-    fi
-  else
-    echo 'Skipped deploying enterprise runnerdeployment. Set TEST_ENTERPRISE_GROUP to deploy.'
-  fi
-else
-  echo 'Skipped deploying enterprise runnerdeployment. Set TEST_ENTERPRISE to deploy.'
-fi

acceptance/deploy_runners.sh

@@ -0,0 +1,60 @@
+#!/usr/bin/env bash
+
+set -e
+
+OP=${OP:-apply}
+
+RUNNER_LABEL=${RUNNER_LABEL:-self-hosted}
+
+cat acceptance/testdata/kubernetes_container_mode.envsubst.yaml  | NAMESPACE=${RUNNER_NAMESPACE} envsubst  | kubectl apply -f -
+
+if [ -n "${TEST_REPO}" ]; then
+  if [ "${USE_RUNNERSET}" != "false" ]; then
+    cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ENTERPRISE= TEST_ORG= RUNNER_MIN_REPLICAS=${REPO_RUNNER_MIN_REPLICAS} NAME=repo-runnerset envsubst | kubectl ${OP} -f -
+  else
+    echo "Running ${OP} runnerdeployment and hra. Set USE_RUNNERSET if you want to deploy runnerset instead."
+    cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ENTERPRISE= TEST_ORG= RUNNER_MIN_REPLICAS=${REPO_RUNNER_MIN_REPLICAS} NAME=repo-runnerdeploy envsubst | kubectl ${OP} -f -
+  fi
+else
+  echo "Skipped ${OP} for runnerdeployment and hra. Set TEST_REPO to "yourorg/yourrepo" to deploy."
+fi
+
+if [ -n "${TEST_ORG}" ]; then
+  if [ "${USE_RUNNERSET}" != "false" ]; then
+    cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} NAME=org-runnerset envsubst | kubectl ${OP} -f -
+  else
+    cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} NAME=org-runnerdeploy envsubst | kubectl ${OP} -f -
+  fi
+
+  if [ -n "${TEST_ORG_GROUP}" ]; then
+    if [ "${USE_RUNNERSET}" != "false" ]; then
+      cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ORG_GROUP} NAME=orggroup-runnerset envsubst | kubectl ${OP} -f -
+    else
+      cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ENTERPRISE= TEST_REPO= RUNNER_MIN_REPLICAS=${ORG_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ORG_GROUP} NAME=orggroup-runnerdeploy envsubst | kubectl ${OP} -f -
+    fi
+  else
+    echo "Skipped ${OP} on enterprise runnerdeployment. Set TEST_ORG_GROUP to ${OP}."
+  fi
+else
+  echo "Skipped ${OP} on organizational runnerdeployment. Set TEST_ORG to ${OP}."
+fi
+
+if [ -n "${TEST_ENTERPRISE}" ]; then
+  if [ "${USE_RUNNERSET}" != "false" ]; then
+    cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} NAME=enterprise-runnerset envsubst | kubectl ${OP} -f -
+  else
+    cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} NAME=enterprise-runnerdeploy envsubst | kubectl ${OP} -f -
+  fi
+
+  if [ -n "${TEST_ENTERPRISE_GROUP}" ]; then
+    if [ "${USE_RUNNERSET}" != "false" ]; then
+      cat acceptance/testdata/runnerset.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ENTERPRISE_GROUP} NAME=enterprisegroup-runnerset envsubst | kubectl ${OP} -f -
+    else
+      cat acceptance/testdata/runnerdeploy.envsubst.yaml | TEST_ORG= TEST_REPO= RUNNER_MIN_REPLICAS=${ENTERPRISE_RUNNER_MIN_REPLICAS} TEST_GROUP=${TEST_ENTERPRISE_GROUP} NAME=enterprisegroup-runnerdeploy envsubst | kubectl ${OP} -f -
+    fi
+  else
+    echo "Skipped ${OP} on enterprise runnerdeployment. Set TEST_ENTERPRISE_GROUP to ${OP}."
+  fi
+else
+  echo "Skipped ${OP} on enterprise runnerdeployment. Set TEST_ENTERPRISE to ${OP}."
+fi

acceptance/testdata/kubernetes_container_mode.envsubst.yaml

@@ -0,0 +1,86 @@
+# USAGE:
+#   cat acceptance/testdata/kubernetes_container_mode.envsubst.yaml  | NAMESPACE=default envsubst  | kubectl apply -f -
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: k8s-mode-runner
+rules:
+- apiGroups: [""]
+  resources: ["pods"]
+  verbs: ["get", "list", "create", "delete"]
+- apiGroups: [""]
+  resources: ["pods/exec"]
+  verbs: ["get", "create"]
+- apiGroups: [""]
+  resources: ["pods/log"]
+  verbs: ["get", "list", "watch",]
+- apiGroups: ["batch"]
+  resources: ["jobs"]
+  verbs: ["get", "list", "create", "delete"]
+- apiGroups: [""]
+  resources: ["secrets"]
+  verbs: ["get", "list", "create", "delete"]
+# Needed to report test success by crating a cm from within workflow job step
+- apiGroups: [""]
+  resources: ["configmaps"]
+  verbs: ["create", "delete"]
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: ClusterRole
+metadata:
+  name: runner-status-updater
+rules:
+- apiGroups: ["actions.summerwind.dev"]
+  resources: ["runners/status"]
+  verbs: ["get", "update", "patch"]
+---
+apiVersion: v1
+kind: ServiceAccount
+metadata:
+  name: ${RUNNER_SERVICE_ACCOUNT_NAME}
+  namespace: ${NAMESPACE}
+---
+# To verify it's working, try:
+#   kubectl auth can-i --as system:serviceaccount:default:runner get pod
+# If incomplete, workflows and jobs would fail with an error message like:
+#   Error: Error: The Service account needs the following permissions [{"group":"","verbs":["get","list","create","delete"],"resource":"pods","subresource":""},{"group":"","verbs":["get","create"],"resource":"pods","subresource":"exec"},{"group":"","verbs":["get","list","watch"],"resource":"pods","subresource":"log"},{"group":"batch","verbs":["get","list","create","delete"],"resource":"jobs","subresource":""},{"group":"","verbs":["create","delete","get","list"],"resource":"secrets","subresource":""}] on the pod resource in the 'default' namespace. Please contact your self hosted runner administrator.
+#   Error: Process completed with exit code 1.
+apiVersion: rbac.authorization.k8s.io/v1
+# This role binding allows "jane" to read pods in the "default" namespace.
+# You need to already have a Role named "pod-reader" in that namespace.
+kind: RoleBinding
+metadata:
+  name: runner-k8s-mode-runner
+  namespace: ${NAMESPACE}
+subjects:
+- kind: ServiceAccount
+  name: ${RUNNER_SERVICE_ACCOUNT_NAME}
+  namespace: ${NAMESPACE}
+roleRef:
+  kind: ClusterRole
+  name: k8s-mode-runner
+  apiGroup: rbac.authorization.k8s.io
+---
+apiVersion: rbac.authorization.k8s.io/v1
+kind: RoleBinding
+metadata:
+  name: runner-runner-stat-supdater
+  namespace: ${NAMESPACE}
+subjects:
+- kind: ServiceAccount
+  name: ${RUNNER_SERVICE_ACCOUNT_NAME}
+  namespace: ${NAMESPACE}
+roleRef:
+  kind: ClusterRole
+  name: runner-status-updater
+  apiGroup: rbac.authorization.k8s.io
+---
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: org-runnerdeploy-runner-work-dir
+  labels:
+    content: org-runnerdeploy-runner-work-dir
+provisioner: rancher.io/local-path
+reclaimPolicy: Delete
+volumeBindingMode: WaitForFirstConsumer

acceptance/testdata/runnerdeploy.envsubst.yaml

@@ -1,3 +1,13 @@
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: ${NAME}-runner-work-dir
+  labels:
+    content: ${NAME}-runner-work-dir
+provisioner: rancher.io/local-path
+reclaimPolicy: Delete
+volumeBindingMode: WaitForFirstConsumer
+---
 apiVersion: actions.summerwind.dev/v1alpha1
 kind: RunnerDeployment
 metadata:
@@ -19,11 +29,6 @@ spec:
 
       ephemeral: ${TEST_EPHEMERAL}
 
-      # Whether to pass --ephemeral (true) or --once (false, deprecated)
-      env:
-      - name: RUNNER_FEATURE_FLAG_EPHEMERAL
-        value: "${RUNNER_FEATURE_FLAG_EPHEMERAL}"
-
       #
       # dockerd within runner container
       #
@@ -44,10 +49,26 @@ spec:
       labels:
       - "${RUNNER_LABEL}"
 
+      env:
+      - name: ROLLING_UPDATE_PHASE
+        value: "${ROLLING_UPDATE_PHASE}"
+
       #
       # Non-standard working directory
       #
       # workDir: "/"
+
+      # # Uncomment the below to enable the kubernetes container mode
+      # # See https://github.com/actions-runner-controller/actions-runner-controller#runner-with-k8s-jobs
+      containerMode: ${RUNNER_CONTAINER_MODE}
+      workVolumeClaimTemplate:
+        accessModes:
+        - ReadWriteOnce
+        storageClassName: "${NAME}-runner-work-dir"
+        resources:
+          requests:
+            storage: 10Gi
+      serviceAccountName: ${RUNNER_SERVICE_ACCOUNT_NAME}
 ---
 apiVersion: actions.summerwind.dev/v1alpha1
 kind: HorizontalRunnerAutoscaler
@@ -57,7 +78,8 @@ spec:
   scaleTargetRef:
     name: ${NAME}
   scaleUpTriggers:
-  - githubEvent: {}
+  - githubEvent:
+      workflowJob: {}
     amount: 1
     duration: "10m"
   minReplicas: ${RUNNER_MIN_REPLICAS}

acceptance/testdata/runnerset.envsubst.yaml

@@ -1,3 +1,59 @@
+---
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: ${NAME}-runner-work-dir
+  labels:
+    content: ${NAME}-runner-work-dir
+provisioner: rancher.io/local-path
+reclaimPolicy: Delete
+volumeBindingMode: WaitForFirstConsumer
+---
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: ${NAME}
+# In kind environments, the provider writes:
+#   /var/lib/docker/volumes/KIND_NODE_CONTAINER_VOL_ID/_data/local-path-provisioner/PV_NAME
+# It can be hundreds of gigabytes depending on what you cache in the test workflow. Beware to not encounter `no space left on device` errors!
+# If you did encounter no space errorrs try:
+#   docker system prune
+#   docker buildx prune #=> frees up /var/lib/docker/volumes/buildx_buildkit_container-builder0_state
+#   sudo rm -rf /var/lib/docker/volumes/KIND_NODE_CONTAINER_VOL_ID/_data/local-path-provisioner #=> frees up local-path-provisioner's data
+provisioner: rancher.io/local-path
+reclaimPolicy: Retain
+volumeBindingMode: WaitForFirstConsumer
+---
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: ${NAME}-var-lib-docker
+  labels:
+    content: ${NAME}-var-lib-docker
+provisioner: rancher.io/local-path
+reclaimPolicy: Retain
+volumeBindingMode: WaitForFirstConsumer
+---
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: ${NAME}-cache
+  labels:
+    content: ${NAME}-cache
+provisioner: rancher.io/local-path
+reclaimPolicy: Retain
+volumeBindingMode: WaitForFirstConsumer
+---
+apiVersion: storage.k8s.io/v1
+kind: StorageClass
+metadata:
+  name: ${NAME}-runner-tool-cache
+  labels:
+    content: ${NAME}-runner-tool-cache
+provisioner: rancher.io/local-path
+reclaimPolicy: Retain
+volumeBindingMode: WaitForFirstConsumer
+---
 apiVersion: actions.summerwind.dev/v1alpha1
 kind: RunnerSet
 metadata:
@@ -56,14 +112,134 @@ spec:
       labels:
         app: ${NAME}
     spec:
+      serviceAccountName: ${RUNNER_SERVICE_ACCOUNT_NAME}
       containers:
       - name: runner
         imagePullPolicy: IfNotPresent
         env:
         - name: RUNNER_FEATURE_FLAG_EPHEMERAL
           value: "${RUNNER_FEATURE_FLAG_EPHEMERAL}"
-      #- name: docker
-      #  #image: mumoshu/actions-runner-dind:dev
+        - name: GOMODCACHE
+          value: "/home/runner/.cache/go-mod"
+        - name: ROLLING_UPDATE_PHASE
+          value: "${ROLLING_UPDATE_PHASE}"
+        # PV-backed runner work dir
+        volumeMounts:
+        # Comment out the ephemeral work volume if you're going to test the kubernetes container mode
+        # The volume and mount with the same names will be created by workVolumeClaimTemplate and the kubernetes container mode support.
+        # - name: work
+        #   mountPath: /runner/_work
+        # Cache docker image layers, in case dockerdWithinRunnerContainer=true
+        - name: var-lib-docker
+          mountPath: /var/lib/docker
+        # Cache go modules and builds
+        # - name: gocache
+        #   # Run `goenv | grep GOCACHE` to verify the path is correct for your env
+        #   mountPath: /home/runner/.cache/go-build
+        # - name: gomodcache
+        #   # Run `goenv | grep GOMODCACHE` to verify the path is correct for your env
+        #   # mountPath: /home/runner/go/pkg/mod
+        - name: cache
+          # go: could not create module cache: stat /home/runner/.cache/go-mod: permission denied
+          mountPath: "/home/runner/.cache"
+        - name: runner-tool-cache
+          # This corresponds to our runner image's default setting of RUNNER_TOOL_CACHE=/opt/hostedtoolcache.
+          #
+          # In case you customize the envvar in both runner and docker containers of the runner pod spec,
+          # You'd need to change this mountPath accordingly.
+          #
+          # The tool cache directory is defined in actions/toolkit's tool-cache module:
+          # https://github.com/actions/toolkit/blob/2f164000dcd42fb08287824a3bc3030dbed33687/packages/tool-cache/src/tool-cache.ts#L621-L638
+          #
+          # Many setup-* actions like setup-go utilizes the tool-cache module to download and cache installed binaries:
+          # https://github.com/actions/setup-go/blob/56a61c9834b4a4950dbbf4740af0b8a98c73b768/src/installer.ts#L144
+          mountPath: "/opt/hostedtoolcache"
+      # Valid only when dockerdWithinRunnerContainer=false
+      # - name: docker
+      #   # PV-backed runner work dir
+      #   volumeMounts:
+      #   - name: work
+      #     mountPath: /runner/_work
+      #   # Cache docker image layers, in case dockerdWithinRunnerContainer=false
+      #   - name: var-lib-docker
+      #     mountPath: /var/lib/docker
+      #   # image: mumoshu/actions-runner-dind:dev
+
+      #   # For buildx cache
+      #   - name: cache
+      #     mountPath: "/home/runner/.cache"
+        # Comment out the ephemeral work volume if you're going to test the kubernetes container mode
+      # volumes:
+      # - name: work
+      #   ephemeral:
+      #     volumeClaimTemplate:
+      #       spec:
+      #         accessModes:
+      #         - ReadWriteOnce
+      #         storageClassName: "${NAME}-runner-work-dir"
+      #         resources:
+      #           requests:
+      #             storage: 10Gi
+  volumeClaimTemplates:
+  - metadata:
+      name: vol1
+    spec:
+      accessModes:
+      - ReadWriteOnce
+      resources:
+        requests:
+          storage: 10Mi
+      storageClassName: ${NAME}
+      ## Dunno which provider supports auto-provisioning with selector.
+      ## At least the rancher local path provider stopped with:
+      ##  waiting for a volume to be created, either by external provisioner "rancher.io/local-path" or manually created by system administrator
+      # selector:
+      #    matchLabels:
+      #      runnerset-volume-id: ${NAME}-vol1
+  - metadata:
+      name: vol2
+    spec:
+      accessModes:
+      - ReadWriteOnce
+      resources:
+        requests:
+          storage: 10Mi
+      storageClassName: ${NAME}
+      # selector:
+      #    matchLabels:
+      #      runnerset-volume-id: ${NAME}-vol2
+  - metadata:
+      name: var-lib-docker
+    spec:
+      accessModes:
+      - ReadWriteOnce
+      resources:
+        requests:
+          storage: 10Mi
+      storageClassName: ${NAME}-var-lib-docker
+  - metadata:
+      name: cache
+    spec:
+      accessModes:
+      - ReadWriteOnce
+      resources:
+        requests:
+          storage: 10Mi
+      storageClassName: ${NAME}-cache
+  - metadata:
+      name: runner-tool-cache
+      # It turns out labels doesn't distinguish PVs across PVCs and the
+      # end result is PVs are reused by wrong PVCs.
+      # The correct way seems to be to differentiate storage class per pvc template.
+      # labels:
+      #   id: runner-tool-cache
+    spec:
+      accessModes:
+      - ReadWriteOnce
+      resources:
+        requests:
+          storage: 10Mi
+      storageClassName: ${NAME}-runner-tool-cache
 ---
 apiVersion: actions.summerwind.dev/v1alpha1
 kind: HorizontalRunnerAutoscaler
@@ -74,9 +250,17 @@ spec:
     kind: RunnerSet
     name: ${NAME}
   scaleUpTriggers:
-  - githubEvent: {}
+  - githubEvent:
+      workflowJob: {}
     amount: 1
     duration: "10m"
   minReplicas: ${RUNNER_MIN_REPLICAS}
   maxReplicas: 10
   scaleDownDelaySecondsAfterScaleOut: ${RUNNER_SCALE_DOWN_DELAY_SECONDS_AFTER_SCALE_OUT}
+  # Comment out the whole metrics if you'd like to solely test webhook-based scaling
+  metrics:
+  - type: PercentageRunnersBusy
+    scaleUpThreshold: '0.75'
+    scaleDownThreshold: '0.25'
+    scaleUpFactor: '2'
+    scaleDownFactor: '0.5'

acceptance/values.yaml

@@ -1,6 +1,18 @@
 # Set actions-runner-controller settings for testing
 logLevel: "-4"
+imagePullSecrets: []
+image:
+  # This needs to be an empty array rather than a single-item array with empty name.
+  # Otherwise you end up with the following error on helm-upgrade:
+  #   Error: UPGRADE FAILED: failed to create patch: map: map[] does not contain declared merge key: name && failed to create patch: map: map[] does not contain declared merge key: name
+  actionsRunnerImagePullSecrets: []
+runner:
+  statusUpdateHook:
+    enabled: true
+rbac:
+  allowGrantingKubernetesContainerModePermissions: true
 githubWebhookServer:
+  imagePullSecrets: []
   logLevel: "-4"
   enabled: true
   labels: {}

api/v1alpha1/horizontalrunnerautoscaler_types.go

@@ -60,6 +60,9 @@ type HorizontalRunnerAutoscalerSpec struct {
 	// The earlier a scheduled override is, the higher it is prioritized.
 	// +optional
 	ScheduledOverrides []ScheduledOverride `json:"scheduledOverrides,omitempty"`
+
+	// +optional
+	GitHubAPICredentialsFrom *GitHubAPICredentialsFrom `json:"githubAPICredentialsFrom,omitempty"`
 }
 
 type ScaleUpTrigger struct {
@@ -72,10 +75,12 @@ type GitHubEventScaleUpTriggerSpec struct {
 	CheckRun    *CheckRunSpec    `json:"checkRun,omitempty"`
 	PullRequest *PullRequestSpec `json:"pullRequest,omitempty"`
 	Push        *PushSpec        `json:"push,omitempty"`
+	WorkflowJob *WorkflowJobSpec `json:"workflowJob,omitempty"`
 }
 
 // https://docs.github.com/en/actions/reference/events-that-trigger-workflows#check_run
 type CheckRunSpec struct {
+	// One of: created, rerequested, or completed
 	Types  []string `json:"types,omitempty"`
 	Status string   `json:"status,omitempty"`
 
@@ -90,6 +95,10 @@ type CheckRunSpec struct {
 	Repositories []string `json:"repositories,omitempty"`
 }
 
+// https://docs.github.com/en/developers/webhooks-and-events/webhooks/webhook-events-and-payloads#workflow_job
+type WorkflowJobSpec struct {
+}
+
 // https://docs.github.com/en/actions/reference/events-that-trigger-workflows#pull_request
 type PullRequestSpec struct {
 	Types    []string `json:"types,omitempty"`
@@ -124,7 +133,7 @@ type ScaleTargetRef struct {
 
 type MetricSpec struct {
 	// Type is the type of metric to be used for autoscaling.
-	// The only supported Type is TotalNumberOfQueuedAndInProgressWorkflowRuns
+	// It can be TotalNumberOfQueuedAndInProgressWorkflowRuns or PercentageRunnersBusy.
 	Type string `json:"type,omitempty"`
 
 	// RepositoryNames is the list of repository names to be used for calculating the metric.
@@ -164,7 +173,7 @@ type MetricSpec struct {
 }
 
 // ScheduledOverride can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule.
-// A schedule can optionally be recurring, so that the correspoding override happens every day, week, month, or year.
+// A schedule can optionally be recurring, so that the corresponding override happens every day, week, month, or year.
 type ScheduledOverride struct {
 	// StartTime is the time at which the first override starts.
 	StartTime metav1.Time `json:"startTime"`

api/v1alpha1/runner_types.go

@@ -18,8 +18,10 @@ package v1alpha1
 
 import (
 	"errors"
+	"fmt"
 
 	"k8s.io/apimachinery/pkg/api/resource"
+	"k8s.io/apimachinery/pkg/util/validation/field"
 
 	corev1 "k8s.io/api/core/v1"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
@@ -71,6 +73,19 @@ type RunnerConfig struct {
 	VolumeSizeLimit *resource.Quantity `json:"volumeSizeLimit,omitempty"`
 	// +optional
 	VolumeStorageMedium *string `json:"volumeStorageMedium,omitempty"`
+
+	// +optional
+	ContainerMode string `json:"containerMode,omitempty"`
+
+	GitHubAPICredentialsFrom *GitHubAPICredentialsFrom `json:"githubAPICredentialsFrom,omitempty"`
+}
+
+type GitHubAPICredentialsFrom struct {
+	SecretRef SecretReference `json:"secretRef,omitempty"`
+}
+
+type SecretReference struct {
+	Name string `json:"name"`
 }
 
 // RunnerPodSpec defines the desired pod spec fields of the runner pod
@@ -135,6 +150,9 @@ type RunnerPodSpec struct {
 	// +optional
 	Tolerations []corev1.Toleration `json:"tolerations,omitempty"`
 
+	// +optional
+	PriorityClassName string `json:"priorityClassName,omitempty"`
+
 	// +optional
 	TerminationGracePeriodSeconds *int64 `json:"terminationGracePeriodSeconds,omitempty"`
 
@@ -145,7 +163,7 @@ type RunnerPodSpec struct {
 	HostAliases []corev1.HostAlias `json:"hostAliases,omitempty"`
 
 	// +optional
-	TopologySpreadConstraints []corev1.TopologySpreadConstraint `json:"topologySpreadConstraint,omitempty"`
+	TopologySpreadConstraints []corev1.TopologySpreadConstraint `json:"topologySpreadConstraints,omitempty"`
 
 	// RuntimeClassName is the container runtime configuration that containers should run under.
 	// More info: https://kubernetes.io/docs/concepts/containers/runtime-class
@@ -153,11 +171,33 @@ type RunnerPodSpec struct {
 	RuntimeClassName *string `json:"runtimeClassName,omitempty"`
 
 	// +optional
-	DnsConfig []corev1.PodDNSConfig `json:"dnsConfig,omitempty"`
+	DnsConfig *corev1.PodDNSConfig `json:"dnsConfig,omitempty"`
+
+	// +optional
+	WorkVolumeClaimTemplate *WorkVolumeClaimTemplate `json:"workVolumeClaimTemplate,omitempty"`
+}
+
+func (rs *RunnerSpec) Validate(rootPath *field.Path) field.ErrorList {
+	var (
+		errList field.ErrorList
+		err     error
+	)
+
+	err = rs.validateRepository()
+	if err != nil {
+		errList = append(errList, field.Invalid(rootPath.Child("repository"), rs.Repository, err.Error()))
+	}
+
+	err = rs.validateWorkVolumeClaimTemplate()
+	if err != nil {
+		errList = append(errList, field.Invalid(rootPath.Child("workVolumeClaimTemplate"), rs.WorkVolumeClaimTemplate, err.Error()))
+	}
+
+	return errList
 }
 
 // ValidateRepository validates repository field.
-func (rs *RunnerSpec) ValidateRepository() error {
+func (rs *RunnerSpec) validateRepository() error {
 	// Enterprise, Organization and repository are both exclusive.
 	foundCount := 0
 	if len(rs.Organization) > 0 {
@@ -179,6 +219,18 @@ func (rs *RunnerSpec) ValidateRepository() error {
 	return nil
 }
 
+func (rs *RunnerSpec) validateWorkVolumeClaimTemplate() error {
+	if rs.ContainerMode != "kubernetes" {
+		return nil
+	}
+
+	if rs.WorkVolumeClaimTemplate == nil {
+		return errors.New("Spec.ContainerMode: kubernetes must have workVolumeClaimTemplate field specified")
+	}
+
+	return rs.WorkVolumeClaimTemplate.validate()
+}
+
 // RunnerStatus defines the observed state of Runner
 type RunnerStatus struct {
 	// Turns true only if the runner pod is ready.
@@ -207,13 +259,60 @@ type RunnerStatusRegistration struct {
 	ExpiresAt    metav1.Time `json:"expiresAt"`
 }
 
+type WorkVolumeClaimTemplate struct {
+	StorageClassName string                              `json:"storageClassName"`
+	AccessModes      []corev1.PersistentVolumeAccessMode `json:"accessModes"`
+	Resources        corev1.ResourceRequirements         `json:"resources"`
+}
+
+func (w *WorkVolumeClaimTemplate) validate() error {
+	if w.AccessModes == nil || len(w.AccessModes) == 0 {
+		return errors.New("Access mode should have at least one mode specified")
+	}
+
+	for _, accessMode := range w.AccessModes {
+		switch accessMode {
+		case corev1.ReadWriteOnce, corev1.ReadWriteMany:
+		default:
+			return fmt.Errorf("Access mode %v is not supported", accessMode)
+		}
+	}
+	return nil
+}
+
+func (w *WorkVolumeClaimTemplate) V1Volume() corev1.Volume {
+	return corev1.Volume{
+		Name: "work",
+		VolumeSource: corev1.VolumeSource{
+			Ephemeral: &corev1.EphemeralVolumeSource{
+				VolumeClaimTemplate: &corev1.PersistentVolumeClaimTemplate{
+					Spec: corev1.PersistentVolumeClaimSpec{
+						AccessModes:      w.AccessModes,
+						StorageClassName: &w.StorageClassName,
+						Resources:        w.Resources,
+					},
+				},
+			},
+		},
+	}
+}
+
+func (w *WorkVolumeClaimTemplate) V1VolumeMount(mountPath string) corev1.VolumeMount {
+	return corev1.VolumeMount{
+		MountPath: mountPath,
+		Name:      "work",
+	}
+}
+
 // +kubebuilder:object:root=true
 // +kubebuilder:subresource:status
 // +kubebuilder:printcolumn:JSONPath=".spec.enterprise",name=Enterprise,type=string
 // +kubebuilder:printcolumn:JSONPath=".spec.organization",name=Organization,type=string
 // +kubebuilder:printcolumn:JSONPath=".spec.repository",name=Repository,type=string
+// +kubebuilder:printcolumn:JSONPath=".spec.group",name=Group,type=string
 // +kubebuilder:printcolumn:JSONPath=".spec.labels",name=Labels,type=string
 // +kubebuilder:printcolumn:JSONPath=".status.phase",name=Status,type=string
+// +kubebuilder:printcolumn:JSONPath=".status.message",name=Message,type=string
 // +kubebuilder:printcolumn:name="Age",type="date",JSONPath=".metadata.creationTimestamp"
 
 // Runner is the Schema for the runners API

api/v1alpha1/runner_webhook.go

@@ -66,15 +66,7 @@ func (r *Runner) ValidateDelete() error {
 
 // Validate validates resource spec.
 func (r *Runner) Validate() error {
-	var (
-		errList field.ErrorList
-		err     error
-	)
-
-	err = r.Spec.ValidateRepository()
-	if err != nil {
-		errList = append(errList, field.Invalid(field.NewPath("spec", "repository"), r.Spec.Repository, err.Error()))
-	}
+	errList := r.Spec.Validate(field.NewPath("spec"))
 
 	if len(errList) > 0 {
 		return apierrors.NewInvalid(r.GroupVersionKind().GroupKind(), r.Name, errList)

api/v1alpha1/runnerdeployment_types.go

@@ -33,7 +33,7 @@ type RunnerDeploymentSpec struct {
 
 	// EffectiveTime is the time the upstream controller requested to sync Replicas.
 	// It is usually populated by the webhook-based autoscaler via HRA.
-	// The value is inherited to RunnerRepicaSet(s) and used to prevent ephemeral runners from unnecessarily recreated.
+	// The value is inherited to RunnerReplicaSet(s) and used to prevent ephemeral runners from unnecessarily recreated.
 	//
 	// +optional
 	// +nullable

api/v1alpha1/runnerdeployment_webhook.go

@@ -66,15 +66,7 @@ func (r *RunnerDeployment) ValidateDelete() error {
 
 // Validate validates resource spec.
 func (r *RunnerDeployment) Validate() error {
-	var (
-		errList field.ErrorList
-		err     error
-	)
-
-	err = r.Spec.Template.Spec.ValidateRepository()
-	if err != nil {
-		errList = append(errList, field.Invalid(field.NewPath("spec", "template", "spec", "repository"), r.Spec.Template.Spec.Repository, err.Error()))
-	}
+	errList := r.Spec.Template.Spec.Validate(field.NewPath("spec", "template", "spec"))
 
 	if len(errList) > 0 {
 		return apierrors.NewInvalid(r.GroupVersionKind().GroupKind(), r.Name, errList)

api/v1alpha1/runnerreplicaset_webhook.go

@@ -66,15 +66,7 @@ func (r *RunnerReplicaSet) ValidateDelete() error {
 
 // Validate validates resource spec.
 func (r *RunnerReplicaSet) Validate() error {
-	var (
-		errList field.ErrorList
-		err     error
-	)
-
-	err = r.Spec.Template.Spec.ValidateRepository()
-	if err != nil {
-		errList = append(errList, field.Invalid(field.NewPath("spec", "template", "spec", "repository"), r.Spec.Template.Spec.Repository, err.Error()))
-	}
+	errList := r.Spec.Template.Spec.Validate(field.NewPath("spec", "template", "spec"))
 
 	if len(errList) > 0 {
 		return apierrors.NewInvalid(r.GroupVersionKind().GroupKind(), r.Name, errList)

api/v1alpha1/runnerset_types.go

@@ -33,6 +33,12 @@ type RunnerSetSpec struct {
 	// +nullable
 	EffectiveTime *metav1.Time `json:"effectiveTime,omitempty"`
 
+	// +optional
+	ServiceAccountName string `json:"serviceAccountName,omitempty"`
+
+	// +optional
+	WorkVolumeClaimTemplate *WorkVolumeClaimTemplate `json:"workVolumeClaimTemplate,omitempty"`
+
 	appsv1.StatefulSetSpec `json:",inline"`
 }
 

api/v1alpha1/zz_generated.deepcopy.go

@@ -90,6 +90,22 @@ func (in *CheckRunSpec) DeepCopy() *CheckRunSpec {
 	return out
 }
 
+// DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.
+func (in *GitHubAPICredentialsFrom) DeepCopyInto(out *GitHubAPICredentialsFrom) {
+	*out = *in
+	out.SecretRef = in.SecretRef
+}
+
+// DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new GitHubAPICredentialsFrom.
+func (in *GitHubAPICredentialsFrom) DeepCopy() *GitHubAPICredentialsFrom {
+	if in == nil {
+		return nil
+	}
+	out := new(GitHubAPICredentialsFrom)
+	in.DeepCopyInto(out)
+	return out
+}
+
 // DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.
 func (in *GitHubEventScaleUpTriggerSpec) DeepCopyInto(out *GitHubEventScaleUpTriggerSpec) {
 	*out = *in
@@ -108,6 +124,11 @@ func (in *GitHubEventScaleUpTriggerSpec) DeepCopyInto(out *GitHubEventScaleUpTri
 		*out = new(PushSpec)
 		**out = **in
 	}
+	if in.WorkflowJob != nil {
+		in, out := &in.WorkflowJob, &out.WorkflowJob
+		*out = new(WorkflowJobSpec)
+		**out = **in
+	}
 }
 
 // DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new GitHubEventScaleUpTriggerSpec.
@@ -226,6 +247,11 @@ func (in *HorizontalRunnerAutoscalerSpec) DeepCopyInto(out *HorizontalRunnerAuto
 			(*in)[i].DeepCopyInto(&(*out)[i])
 		}
 	}
+	if in.GitHubAPICredentialsFrom != nil {
+		in, out := &in.GitHubAPICredentialsFrom, &out.GitHubAPICredentialsFrom
+		*out = new(GitHubAPICredentialsFrom)
+		**out = **in
+	}
 }
 
 // DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new HorizontalRunnerAutoscalerSpec.
@@ -420,6 +446,11 @@ func (in *RunnerConfig) DeepCopyInto(out *RunnerConfig) {
 		*out = new(string)
 		**out = **in
 	}
+	if in.GitHubAPICredentialsFrom != nil {
+		in, out := &in.GitHubAPICredentialsFrom, &out.GitHubAPICredentialsFrom
+		*out = new(GitHubAPICredentialsFrom)
+		**out = **in
+	}
 }
 
 // DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new RunnerConfig.
@@ -733,10 +764,13 @@ func (in *RunnerPodSpec) DeepCopyInto(out *RunnerPodSpec) {
 	}
 	if in.DnsConfig != nil {
 		in, out := &in.DnsConfig, &out.DnsConfig
-		*out = make([]v1.PodDNSConfig, len(*in))
-		for i := range *in {
-			(*in)[i].DeepCopyInto(&(*out)[i])
-		}
+		*out = new(v1.PodDNSConfig)
+		(*in).DeepCopyInto(*out)
+	}
+	if in.WorkVolumeClaimTemplate != nil {
+		in, out := &in.WorkVolumeClaimTemplate, &out.WorkVolumeClaimTemplate
+		*out = new(WorkVolumeClaimTemplate)
+		(*in).DeepCopyInto(*out)
 	}
 }
 
@@ -936,6 +970,11 @@ func (in *RunnerSetSpec) DeepCopyInto(out *RunnerSetSpec) {
 		in, out := &in.EffectiveTime, &out.EffectiveTime
 		*out = (*in).DeepCopy()
 	}
+	if in.WorkVolumeClaimTemplate != nil {
+		in, out := &in.WorkVolumeClaimTemplate, &out.WorkVolumeClaimTemplate
+		*out = new(WorkVolumeClaimTemplate)
+		(*in).DeepCopyInto(*out)
+	}
 	in.StatefulSetSpec.DeepCopyInto(&out.StatefulSetSpec)
 }
 
@@ -1122,3 +1161,54 @@ func (in *ScheduledOverride) DeepCopy() *ScheduledOverride {
 	in.DeepCopyInto(out)
 	return out
 }
+
+// DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.
+func (in *SecretReference) DeepCopyInto(out *SecretReference) {
+	*out = *in
+}
+
+// DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new SecretReference.
+func (in *SecretReference) DeepCopy() *SecretReference {
+	if in == nil {
+		return nil
+	}
+	out := new(SecretReference)
+	in.DeepCopyInto(out)
+	return out
+}
+
+// DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.
+func (in *WorkVolumeClaimTemplate) DeepCopyInto(out *WorkVolumeClaimTemplate) {
+	*out = *in
+	if in.AccessModes != nil {
+		in, out := &in.AccessModes, &out.AccessModes
+		*out = make([]v1.PersistentVolumeAccessMode, len(*in))
+		copy(*out, *in)
+	}
+	in.Resources.DeepCopyInto(&out.Resources)
+}
+
+// DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new WorkVolumeClaimTemplate.
+func (in *WorkVolumeClaimTemplate) DeepCopy() *WorkVolumeClaimTemplate {
+	if in == nil {
+		return nil
+	}
+	out := new(WorkVolumeClaimTemplate)
+	in.DeepCopyInto(out)
+	return out
+}
+
+// DeepCopyInto is an autogenerated deepcopy function, copying the receiver, writing into out. in must be non-nil.
+func (in *WorkflowJobSpec) DeepCopyInto(out *WorkflowJobSpec) {
+	*out = *in
+}
+
+// DeepCopy is an autogenerated deepcopy function, copying the receiver, creating a new WorkflowJobSpec.
+func (in *WorkflowJobSpec) DeepCopy() *WorkflowJobSpec {
+	if in == nil {
+		return nil
+	}
+	out := new(WorkflowJobSpec)
+	in.DeepCopyInto(out)
+	return out
+}

build/version.go

@@ -0,0 +1,4 @@
+package build
+
+// This is overridden at build-time using go-build ldflags. dev is the fallback value
+var Version = "NA"

charts/actions-runner-controller/Chart.yaml

@@ -15,10 +15,10 @@ 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: 0.17.2
+version: 0.20.2
 
 # Used as the default manager tag value when no tag property is provided in the values.yaml
-appVersion: 0.22.2
+appVersion: 0.25.2
 
 home: https://github.com/actions-runner-controller/actions-runner-controller
 

charts/actions-runner-controller/README.md

@@ -4,107 +4,109 @@ All additional docs are kept in the `docs/` folder, this README is solely for do
 
 ## Values
 
-**_The values are documented as of HEAD, to review the configuration options for your chart version ensure you view this file at the relevent [tag](https://github.com/actions-runner-controller/actions-runner-controller/tags)_**
+**_The values are documented as of HEAD, to review the configuration options for your chart version ensure you view this file at the relevant [tag](https://github.com/actions-runner-controller/actions-runner-controller/tags)_**
 
-> _Default values are the defaults set in the charts values.yaml, some properties have default configurations in the code for when the property is omitted or invalid_
+> _Default values are the defaults set in the charts `values.yaml`, some properties have default configurations in the code for when the property is omitted or invalid_
 
-| Key                                                      | Description                                                                                                                | Default                                                              |
-|----------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------|
-| `labels`                                                 | Set labels to apply to all resources in the chart                                                                          |                                                                      |
-| `replicaCount`                                           | Set the number of controller pods                                                                                          | 1                                                                    |
-| `syncPeriod`                                             | Set the period in which the controler reconciles the desired runners count                                                 | 10m                                                                  |
-| `enableLeaderElection`                                   | Enable election configuration                                                                                              | true                                                                 |
-| `leaderElectionId`                                       | Set the election ID for the controller group                                                                               |                                                                      |
-| `githubEnterpriseServerURL`                              | Set the URL for a self-hosted GitHub Enterprise Server                                                                     |                                                                      |
-| `githubURL`                                              | Override GitHub URL to be used for GitHub API calls                                                                        |                                                                      |
-| `githubUploadURL`                                        | Override GitHub Upload URL to be used for GitHub API calls                                                                 |                                                                      |
-| `runnerGithubURL`                                        | Override GitHub URL to be used by runners during registration                                                              |                                                                      |
-| `logLevel`                                               | Set the log level of the controller container                                                                              |                                                                      |
-| `additionalVolumes`                                      | Set additional volumes to add to the manager container                                                                     |                                                                      |
-| `additionalVolumeMounts`                                 | Set additional volume mounts to add to the manager container                                                               |                                                                      |
-| `authSecret.create`                                      | Deploy the controller auth secret                                                                                          | false                                                                |
-| `authSecret.name`                                        | Set the name of the auth secret                                                                                            | controller-manager                                                   |
-| `authSecret.annotations`                                 | Set annotations for the auth Secret                                                                                        |                                                                      |
-| `authSecret.github_app_id`                               | The ID of your GitHub App. **This can't be set at the same time as `authSecret.github_token`**                             |                                                                      |
-| `authSecret.github_app_installation_id`                  | The ID of your GitHub App installation. **This can't be set at the same time as `authSecret.github_token`**                |                                                                      |
-| `authSecret.github_app_private_key`                      | The multiline string of your GitHub App's private key. **This can't be set at the same time as `authSecret.github_token`** |                                                                      |
-| `authSecret.github_token`                                | Your chosen GitHub PAT token. **This can't be set at the same time as the `authSecret.github_app_*`**                      |                                                                      |
-| `authSecret.github_basicauth_username`                     | Username for GitHub basic auth to use instead of PAT or GitHub APP in case it's running behind a proxy API                 |                                                                      |
-| `authSecret.github_basicauth_password`                     | Password for GitHub basic auth to use instead of PAT or GitHub APP in case it's running behind a proxy API                 |                                                                      |
-| `dockerRegistryMirror`                                   | The default Docker Registry Mirror used by runners.                                                                        |                                                                      |
-| `hostNetwork`                                            | The "hostNetwork" of the controller container                                                                              | false                                                                |
-| `image.repository`                                       | The "repository/image" of the controller container                                                                         | summerwind/actions-runner-controller                                 |
-| `image.tag`                                              | The tag of the controller container                                                                                        |                                                                      |
-| `image.actionsRunnerRepositoryAndTag`                    | The "repository/image" of the actions runner container                                                                     | summerwind/actions-runner:latest                                     |
-| `image.actionsRunnerImagePullSecrets`                    | Optional image pull secrets to be included in the runner pod's ImagePullSecrets                                            |                                                                      |
-| `image.dindSidecarRepositoryAndTag`                      | The "repository/image" of the dind sidecar container                                                                       | docker:dind                                                          |
-| `image.pullPolicy`                                       | The pull policy of the controller image                                                                                    | IfNotPresent                                                         |
-| `metrics.serviceMonitor`                                 | Deploy serviceMonitor kind for for use with prometheus-operator CRDs                                                       | false                                                                |
-| `metrics.serviceAnnotations`                             | Set annotations for the provisioned metrics service resource                                                               |                                                                      |
-| `metrics.port`                                           | Set port of metrics service                                                                                                | 8443                                                                 |
-| `metrics.proxy.enabled`                                  | Deploy kube-rbac-proxy container in controller pod                                                                         | true                                                                 |
-| `metrics.proxy.image.repository`                         | The "repository/image" of the kube-proxy container                                                                         | quay.io/brancz/kube-rbac-proxy                                       |
-| `metrics.proxy.image.tag`                                | The tag of the kube-proxy image to use when pulling the container                                                          | v0.10.0                                                              |
-| `metrics.serviceMonitorLabels`                           | Set labels to apply to ServiceMonitor resources                                                                            |                                                                      |
-| `imagePullSecrets`                                       | Specifies the secret to be used when pulling the controller pod containers                                                 |                                                                      |
-| `fullnameOverride`                                       | Override the full resource names	                                                                                        |                                                                      |
-| `nameOverride`                                           | Override the resource name prefix	                                                                                        |                                                                      |
-| `serviceAccount.annotations`                              | Set annotations to the service account                                                                                     |                                                                      |
-| `serviceAccount.create`                                  | Deploy the controller pod under a service account                                                                          | true                                                                 |
-| `podAnnotations`                                         | Set annotations for the controller pod                                                                                     |                                                                      |
-| `podLabels`                                              | Set labels for the controller pod                                                                                          |                                                                      |
-| `serviceAccount.name`                                    | Set the name of the service account                                                                                        |                                                                      |
-| `securityContext`                                        | Set the security context for each container in the controller pod                                                          |                                                                      |
-| `podSecurityContext`                                     | Set the security context to controller pod                                                                                 |                                                                      |
-| `service.annotations`                                    | Set annotations for the provisioned webhook service resource                                                               |                                                                      |
-| `service.port`                                           | Set controller service ports                                                                                                |                                                                      |
-| `service.type`                                           | Set controller service type                                                                                               |                                                                      |
-| `topologySpreadConstraints`                              | Set the controller pod topologySpreadConstraints                                                                           |                                                                      |
-| `nodeSelector`                                           | Set the controller pod nodeSelector                                                                                        |                                                                      |
-| `resources`                                              | Set the controller pod resources                                                                                           |                                                                      |
-| `affinity`                                               | Set the controller pod affinity rules                                                                                      |                                                                      |
-| `podDisruptionBudget.enabled`                            | Enables a PDB to ensure HA of controller pods                                                                              |      false                                                           |
-| `podDisruptionBudget.minAvailable`                       | Minimum number of pods that must be available after eviction                                                               |                                                                      |
-| `podDisruptionBudget.maxUnavailable`                     | Maximum number of pods that can be unavailable after eviction. Kubernetes 1.7+ required.                                   |                                                                      |
-| `tolerations`                                            | Set the controller pod tolerations                                                                                         |                                                                      |
-| `env`                                                    | Set environment variables for the controller container                                                                     |                                                                      |
-| `priorityClassName`                                      | Set the controller pod priorityClassName                                                                                   |                                                                      |
-| `scope.watchNamespace`                                   | Tells the controller and the github webhook server which namespace to watch if `scope.singleNamespace` is true             | `Release.Namespace` (the default namespace of the helm chart).       |
-| `scope.singleNamespace`                                  | Limit the controller to watch a single namespace                                                                           | false                                                                |
-| `certManagerEnabled`                                     | Enable cert-manager. If disabled you must set admissionWebHooks.caBundle and create TLS secrets manually                   | true                                                                 |
-| `admissionWebHooks.caBundle`                             | Base64-encoded PEM bundle containing the CA that signed the webhook's serving certificate                                  |                                                                      |
-| `githubWebhookServer.logLevel`                           | Set the log level of the githubWebhookServer container                                                                     |                                                                      |
-| `githubWebhookServer.replicaCount`                       | Set the number of webhook server pods                                                                                      | 1                                                                    |
-| `githubWebhookServer.useRunnerGroupsVisibility`          | Enable supporting runner groups with custom visibility. This will incur in extra API calls and may blow up your budget. Currently, you also need to set `githubWebhookServer.secret.enabled` to enable this feature. | false                                                                |
-| `githubWebhookServer.syncPeriod`                         | Set the period in which the controller reconciles the resources                                                            | 10m                                                                  |
-| `githubWebhookServer.enabled`                            | Deploy the webhook server pod                                                                                              | false                                                                |
-| `githubWebhookServer.secret.enabled`                      | Passes the webhook hook secret to the github-webhook-server                                                                             | false                                                                |
-| `githubWebhookServer.secret.create`                      | Deploy the webhook hook secret                                                                                             | false                                                                |
-| `githubWebhookServer.secret.name`                        | Set the name of the webhook hook secret                                                                                    | github-webhook-server                                                |
-| `githubWebhookServer.secret.github_webhook_secret_token` | Set the webhook secret token value                                                                                         |                                                                      |
-| `githubWebhookServer.imagePullSecrets`                   | Specifies the secret to be used when pulling the githubWebhookServer pod containers                                        |                                                                      |
-| `githubWebhookServer.nameOverride`                        | Override the resource name prefix	                                                                                        |                                                                      |
-| `githubWebhookServer.fullnameOverride`                    | Override the full resource names	                                                                                        |                                                                      |
-| `githubWebhookServer.serviceAccount.create`              | Deploy the githubWebhookServer under a service account                                                                     | true                                                                 |
-| `githubWebhookServer.serviceAccount.annotations`         | Set annotations for the service account                                                                                    |                                                                      |
-| `githubWebhookServer.serviceAccount.name`                | Set the service account name                                                                                               |                                                                      |
-| `githubWebhookServer.podAnnotations`                     | Set annotations for the githubWebhookServer pod                                                                            |                                                                      |
-| `githubWebhookServer.podLabels`                          | Set labels for the githubWebhookServer pod                                                                                 |                                                                      |
-| `githubWebhookServer.podSecurityContext`                 | Set the security context to githubWebhookServer pod                                                                        |                                                                      |
-| `githubWebhookServer.securityContext`                    | Set the security context for each container in the githubWebhookServer pod                                                 |                                                                      |
-| `githubWebhookServer.resources`                          | Set the githubWebhookServer pod resources                                                                                  |                                                                      |
-| `githubWebhookServer.topologySpreadConstraints`          | Set the githubWebhookServer pod topologySpreadConstraints                                                                  |                                                                      |
-| `githubWebhookServer.nodeSelector`                       | Set the githubWebhookServer pod nodeSelector                                                                               |                                                                      |
-| `githubWebhookServer.tolerations`                        | Set the githubWebhookServer pod tolerations                                                                                |                                                                      |
-| `githubWebhookServer.affinity`                           | Set the githubWebhookServer pod affinity rules                                                                             |                                                                      |
-| `githubWebhookServer.priorityClassName`                  | Set the githubWebhookServer pod priorityClassName                                                                          |                                                                      |
-| `githubWebhookServer.service.type`                       | Set githubWebhookServer service type                                                                                       |                                                                      |
-| `githubWebhookServer.service.ports`                      | Set githubWebhookServer service ports                                                                                      | `[{"port":80, "targetPort:"http", "protocol":"TCP", "name":"http"}]` |
-| `githubWebhookServer.ingress.enabled`                    | Deploy an ingress kind for the githubWebhookServer                                                                         | false                                                                |
-| `githubWebhookServer.ingress.annotations`                | Set annotations for the ingress kind                                                                                       |                                                                      |
-| `githubWebhookServer.ingress.hosts`                      | Set hosts configuration for ingress                                                                                        | `[{"host": "chart-example.local", "paths": []}]`                     |
-| `githubWebhookServer.ingress.tls`                        | Set tls configuration for ingress                                                                                          |                                                                      |
-| `githubWebhookServer.ingress.ingressClassName`           | Set ingress class name                                                                                                     |                                                                      |
-| `githubWebhookServer.podDisruptionBudget.enabled`        | Enables a PDB to ensure HA of githubwebhook pods                                                                           |      false                                                           |
-| `githubWebhookServer.podDisruptionBudget.minAvailable`   | Minimum number of pods that must be available after eviction                                                               |                                                                      |
-| `githubWebhookServer.podDisruptionBudget.maxUnavailable` | Maximum number of pods that can be unavailable after eviction. Kubernetes 1.7+ required.                                   |                                                                      |
+| Key                                                      | Description                                                                                                                | Default                                                                                         |
+|----------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------|
+| `labels`                                                 | Set labels to apply to all resources in the chart                                                                          |                                                                                                 |
+| `replicaCount`                                           | Set the number of controller pods                                                                                          | 1                                                                                               |
+| `webhookPort`                                            | Set the containerPort for the webhook Pod                                                                                  | 9443                                                                                            |
+| `syncPeriod`                                             | Set the period in which the controler reconciles the desired runners count                                                 | 10m                                                                                             |
+| `enableLeaderElection`                                   | Enable election configuration                                                                                              | true                                                                                            |
+| `leaderElectionId`                                       | Set the election ID for the controller group                                                                               |                                                                                                 |
+| `githubEnterpriseServerURL`                              | Set the URL for a self-hosted GitHub Enterprise Server                                                                     |                                                                                                 |
+| `githubURL`                                              | Override GitHub URL to be used for GitHub API calls                                                                        |                                                                                                 |
+| `githubUploadURL`                                        | Override GitHub Upload URL to be used for GitHub API calls                                                                 |                                                                                                 |
+| `runnerGithubURL`                                        | Override GitHub URL to be used by runners during registration                                                              |                                                                                                 |
+| `logLevel`                                               | Set the log level of the controller container                                                                              |                                                                                                 |
+| `additionalVolumes`                                      | Set additional volumes to add to the manager container                                                                     |                                                                                                 |
+| `additionalVolumeMounts`                                 | Set additional volume mounts to add to the manager container                                                               |                                                                                                 |
+| `authSecret.create`                                      | Deploy the controller auth secret                                                                                          | false                                                                                           |
+| `authSecret.name`                                        | Set the name of the auth secret                                                                                            | controller-manager                                                                              |
+| `authSecret.annotations`                                 | Set annotations for the auth Secret                                                                                        |                                                                                                 |
+| `authSecret.github_app_id`                               | The ID of your GitHub App. **This can't be set at the same time as `authSecret.github_token`**                             |                                                                                                 |
+| `authSecret.github_app_installation_id`                  | The ID of your GitHub App installation. **This can't be set at the same time as `authSecret.github_token`**                |                                                                                                 |
+| `authSecret.github_app_private_key`                      | The multiline string of your GitHub App's private key. **This can't be set at the same time as `authSecret.github_token`** |                                                                                                 |
+| `authSecret.github_token`                                | Your chosen GitHub PAT token. **This can't be set at the same time as the `authSecret.github_app_*`**                      |                                                                                                 |
+| `authSecret.github_basicauth_username`                   | Username for GitHub basic auth to use instead of PAT or GitHub APP in case it's running behind a proxy API                 |                                                                                                 |
+| `authSecret.github_basicauth_password`                   | Password for GitHub basic auth to use instead of PAT or GitHub APP in case it's running behind a proxy API                 |                                                                                                 |
+| `dockerRegistryMirror`                                   | The default Docker Registry Mirror used by runners.                                                                        |                                                                                                 |
+| `hostNetwork`                                            | The "hostNetwork" of the controller container                                                                              | false                                                                                           |
+| `image.repository`                                       | The "repository/image" of the controller container                                                                         | summerwind/actions-runner-controller                                                            |
+| `image.tag`                                              | The tag of the controller container                                                                                        |                                                                                                 |
+| `image.actionsRunnerRepositoryAndTag`                    | The "repository/image" of the actions runner container                                                                     | summerwind/actions-runner:latest                                                                |
+| `image.actionsRunnerImagePullSecrets`                    | Optional image pull secrets to be included in the runner pod's ImagePullSecrets                                            |                                                                                                 |
+| `image.dindSidecarRepositoryAndTag`                      | The "repository/image" of the dind sidecar container                                                                       | docker:dind                                                                                     |
+| `image.pullPolicy`                                       | The pull policy of the controller image                                                                                    | IfNotPresent                                                                                    |
+| `metrics.serviceMonitor`                                 | Deploy serviceMonitor kind for for use with prometheus-operator CRDs                                                       | false                                                                                           |
+| `metrics.serviceAnnotations`                             | Set annotations for the provisioned metrics service resource                                                               |                                                                                                 |
+| `metrics.port`                                           | Set port of metrics service                                                                                                | 8443                                                                                            |
+| `metrics.proxy.enabled`                                  | Deploy kube-rbac-proxy container in controller pod                                                                         | true                                                                                            |
+| `metrics.proxy.image.repository`                         | The "repository/image" of the kube-proxy container                                                                         | quay.io/brancz/kube-rbac-proxy                                                                  |
+| `metrics.proxy.image.tag`                                | The tag of the kube-proxy image to use when pulling the container                                                          | v0.10.0                                                                                         |
+| `metrics.serviceMonitorLabels`                           | Set labels to apply to ServiceMonitor resources                                                                            |                                                                                                 |
+| `imagePullSecrets`                                       | Specifies the secret to be used when pulling the controller pod containers                                                 |                                                                                                 |
+| `fullnameOverride`                                       | Override the full resource names	                                                                                        |                                                                                                 |
+| `nameOverride`                                           | Override the resource name prefix	                                                                                        |                                                                                                 |
+| `serviceAccount.annotations`                             | Set annotations to the service account                                                                                     |                                                                                                 |
+| `serviceAccount.create`                                  | Deploy the controller pod under a service account                                                                          | true                                                                                            |
+| `podAnnotations`                                         | Set annotations for the controller pod                                                                                     |                                                                                                 |
+| `podLabels`                                              | Set labels for the controller pod                                                                                          |                                                                                                 |
+| `serviceAccount.name`                                    | Set the name of the service account                                                                                        |                                                                                                 |
+| `securityContext`                                        | Set the security context for each container in the controller pod                                                          |                                                                                                 |
+| `podSecurityContext`                                     | Set the security context to controller pod                                                                                 |                                                                                                 |
+| `service.annotations`                                    | Set annotations for the provisioned webhook service resource                                                               |                                                                                                 |
+| `service.port`                                           | Set controller service ports                                                                                               |                                                                                                 |
+| `service.type`                                           | Set controller service type                                                                                                |                                                                                                 |
+| `topologySpreadConstraints`                              | Set the controller pod topologySpreadConstraints                                                                           |                                                                                                 |
+| `nodeSelector`                                           | Set the controller pod nodeSelector                                                                                        |                                                                                                 |
+| `resources`                                              | Set the controller pod resources                                                                                           |                                                                                                 |
+| `affinity`                                               | Set the controller pod affinity rules                                                                                      |                                                                                                 |
+| `podDisruptionBudget.enabled`                            | Enables a PDB to ensure HA of controller pods                                                                              |      false                                                                                      |
+| `podDisruptionBudget.minAvailable`                       | Minimum number of pods that must be available after eviction                                                               |                                                                                                 |
+| `podDisruptionBudget.maxUnavailable`                     | Maximum number of pods that can be unavailable after eviction. Kubernetes 1.7+ required.                                   |                                                                                                 |
+| `tolerations`                                            | Set the controller pod tolerations                                                                                         |                                                                                                 |
+| `env`                                                    | Set environment variables for the controller container                                                                     |                                                                                                 |
+| `priorityClassName`                                      | Set the controller pod priorityClassName                                                                                   |                                                                                                 |
+| `scope.watchNamespace`                                   | Tells the controller and the github webhook server which namespace to watch if `scope.singleNamespace` is true             | `Release.Namespace` (the default namespace of the helm chart).                                  |
+| `scope.singleNamespace`                                  | Limit the controller to watch a single namespace                                                                           | false                                                                                           |
+| `certManagerEnabled`                                     | Enable cert-manager. If disabled you must set admissionWebHooks.caBundle and create TLS secrets manually                   | true                                                                                            |
+| `runner.statusUpdateHook.enabled`                        | Use custom RBAC for runners (role, role binding and service account), this will enable reporting runner statuses           | false                                                                                           |
+| `admissionWebHooks.caBundle`                             | Base64-encoded PEM bundle containing the CA that signed the webhook's serving certificate                                  |                                                                                                 |
+| `githubWebhookServer.logLevel`                           | Set the log level of the githubWebhookServer container                                                                     |                                                                                                 |
+| `githubWebhookServer.replicaCount`                       | Set the number of webhook server pods                                                                                      | 1                                                                                               |
+| `githubWebhookServer.useRunnerGroupsVisibility`          | Enable supporting runner groups with custom visibility. This will incur in extra API calls and may blow up your budget. Currently, you also need to set `githubWebhookServer.secret.enabled` to enable this feature. | false |
+| `githubWebhookServer.enabled`                            | Deploy the webhook server pod                                                                                              | false                                                                                           |
+| `githubWebhookServer.queueLimit`                         | Set the queue size limit in the githubWebhookServer                                                                        |                                                                                                 |
+| `githubWebhookServer.secret.enabled`                     | Passes the webhook hook secret to the github-webhook-server                                                                | false                                                                                           |
+| `githubWebhookServer.secret.create`                      | Deploy the webhook hook secret                                                                                             | false                                                                                           |
+| `githubWebhookServer.secret.name`                        | Set the name of the webhook hook secret                                                                                    | github-webhook-server                                                                           |
+| `githubWebhookServer.secret.github_webhook_secret_token` | Set the webhook secret token value                                                                                         |                                                                                                 |
+| `githubWebhookServer.imagePullSecrets`                   | Specifies the secret to be used when pulling the githubWebhookServer pod containers                                        |                                                                                                 |
+| `githubWebhookServer.nameOverride`                       | Override the resource name prefix	                                                                                        |                                                                                                 |
+| `githubWebhookServer.fullnameOverride`                   | Override the full resource names	                                                                                        |                                                                                                 |
+| `githubWebhookServer.serviceAccount.create`              | Deploy the githubWebhookServer under a service account                                                                     | true                                                                                            |
+| `githubWebhookServer.serviceAccount.annotations`         | Set annotations for the service account                                                                                    |                                                                                                 |
+| `githubWebhookServer.serviceAccount.name`                | Set the service account name                                                                                               |                                                                                                 |
+| `githubWebhookServer.podAnnotations`                     | Set annotations for the githubWebhookServer pod                                                                            |                                                                                                 |
+| `githubWebhookServer.podLabels`                          | Set labels for the githubWebhookServer pod                                                                                 |                                                                                                 |
+| `githubWebhookServer.podSecurityContext`                 | Set the security context to githubWebhookServer pod                                                                        |                                                                                                 |
+| `githubWebhookServer.securityContext`                    | Set the security context for each container in the githubWebhookServer pod                                                 |                                                                                                 |
+| `githubWebhookServer.resources`                          | Set the githubWebhookServer pod resources                                                                                  |                                                                                                 |
+| `githubWebhookServer.topologySpreadConstraints`          | Set the githubWebhookServer pod topologySpreadConstraints                                                                  |                                                                                                 |
+| `githubWebhookServer.nodeSelector`                       | Set the githubWebhookServer pod nodeSelector                                                                               |                                                                                                 |
+| `githubWebhookServer.tolerations`                        | Set the githubWebhookServer pod tolerations                                                                                |                                                                                                 |
+| `githubWebhookServer.affinity`                           | Set the githubWebhookServer pod affinity rules                                                                             |                                                                                                 |
+| `githubWebhookServer.priorityClassName`                  | Set the githubWebhookServer pod priorityClassName                                                                          |                                                                                                 |
+| `githubWebhookServer.service.type`                       | Set githubWebhookServer service type                                                                                       |                                                                                                 |
+| `githubWebhookServer.service.ports`                      | Set githubWebhookServer service ports                                                                                      | `[{"port":80, "targetPort:"http", "protocol":"TCP", "name":"http"}]`                            |
+| `githubWebhookServer.ingress.enabled`                    | Deploy an ingress kind for the githubWebhookServer                                                                         | false                                                                                           |
+| `githubWebhookServer.ingress.annotations`                | Set annotations for the ingress kind                                                                                       |                                                                                                 |
+| `githubWebhookServer.ingress.hosts`                      | Set hosts configuration for ingress                                                                                        | `[{"host": "chart-example.local", "paths": []}]`                                                |
+| `githubWebhookServer.ingress.tls`                        | Set tls configuration for ingress                                                                                          |                                                                                                 |
+| `githubWebhookServer.ingress.ingressClassName`           | Set ingress class name                                                                                                     |                                                                                                 |
+| `githubWebhookServer.podDisruptionBudget.enabled`        | Enables a PDB to ensure HA of githubwebhook pods                                                                           | false                                                                                           |
+| `githubWebhookServer.podDisruptionBudget.minAvailable`   | Minimum number of pods that must be available after eviction                                                               |                                                                                                 |
+| `githubWebhookServer.podDisruptionBudget.maxUnavailable` | Maximum number of pods that can be unavailable after eviction. Kubernetes 1.7+ required.                                   |                                                                                                 |

charts/actions-runner-controller/crds/actions.summerwind.dev_horizontalrunnerautoscalers.yaml

@@ -61,6 +61,16 @@ spec:
                         type: integer
                     type: object
                   type: array
+                githubAPICredentialsFrom:
+                  properties:
+                    secretRef:
+                      properties:
+                        name:
+                          type: string
+                      required:
+                        - name
+                      type: object
+                  type: object
                 maxReplicas:
                   description: MaxReplicas is the maximum number of replicas the deployment is allowed to scale
                   type: integer
@@ -92,7 +102,7 @@ spec:
                         description: ScaleUpThreshold is the percentage of busy runners greater than which will trigger the hpa to scale runners up.
                         type: string
                       type:
-                        description: Type is the type of metric to be used for autoscaling. The only supported Type is TotalNumberOfQueuedAndInProgressWorkflowRuns
+                        description: Type is the type of metric to be used for autoscaling. It can be TotalNumberOfQueuedAndInProgressWorkflowRuns or PercentageRunnersBusy.
                         type: string
                     type: object
                   type: array
@@ -141,6 +151,7 @@ spec:
                               status:
                                 type: string
                               types:
+                                description: 'One of: created, rerequested, or completed'
                                 items:
                                   type: string
                                 type: array
@@ -160,13 +171,16 @@ spec:
                           push:
                             description: PushSpec is the condition for triggering scale-up on push event Also see https://docs.github.com/en/actions/reference/events-that-trigger-workflows#push
                             type: object
+                          workflowJob:
+                            description: https://docs.github.com/en/developers/webhooks-and-events/webhooks/webhook-events-and-payloads#workflow_job
+                            type: object
                         type: object
                     type: object
                   type: array
                 scheduledOverrides:
                   description: ScheduledOverrides is the list of ScheduledOverride. It can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule. The earlier a scheduled override is, the higher it is prioritized.
                   items:
-                    description: ScheduledOverride can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule. A schedule can optionally be recurring, so that the correspoding override happens every day, week, month, or year.
+                    description: ScheduledOverride can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule. A schedule can optionally be recurring, so that the corresponding override happens every day, week, month, or year.
                     properties:
                       endTime:
                         description: EndTime is the time at which the first override ends.

charts/actions-runner-controller/docs/UPGRADING.md

@@ -22,7 +22,7 @@ Due to the above you can't just do a `helm upgrade` to release the latest versio
 
 ```shell
 # REMEMBER TO UPDATE THE CHART_VERSION TO RELEVANT CHART VERISON!!!!
-CHART_VERSION=0.17.0
+CHART_VERSION=0.18.0
 
 curl -L https://github.com/actions-runner-controller/actions-runner-controller/releases/download/actions-runner-controller-${CHART_VERSION}/actions-runner-controller-${CHART_VERSION}.tgz | tar zxv --strip 1 actions-runner-controller/crds
 

charts/actions-runner-controller/templates/controller.metrics.serviceMonitor.yaml

@@ -8,6 +8,7 @@ metadata:
     {{- toYaml . | nindent 4 }}
   {{- end }}
   name: {{ include "actions-runner-controller.serviceMonitorName" . }}
+  namespace: {{ .Release.Namespace }}
 spec:
   endpoints:
     - path: /metrics

charts/actions-runner-controller/templates/deployment.yaml

@@ -44,7 +44,9 @@ spec:
         {{- if .Values.leaderElectionId }}
         - "--leader-election-id={{ .Values.leaderElectionId }}"
         {{- end }}
+        - "--port={{ .Values.webhookPort }}"
         - "--sync-period={{ .Values.syncPeriod }}"
+        - "--default-scale-down-delay={{ .Values.defaultScaleDownDelay }}"
         - "--docker-image={{ .Values.image.dindSidecarRepositoryAndTag }}"
         - "--runner-image={{ .Values.image.actionsRunnerRepositoryAndTag }}"
         {{- range .Values.image.actionsRunnerImagePullSecrets }}
@@ -56,15 +58,15 @@ spec:
         {{- if .Values.scope.singleNamespace }}
         - "--watch-namespace={{ default .Release.Namespace .Values.scope.watchNamespace }}"
         {{- end }}
-        {{- if .Values.githubAPICacheDuration }}
-        - "--github-api-cache-duration={{ .Values.githubAPICacheDuration }}"
-        {{- end }}
         {{- if .Values.logLevel }}
         - "--log-level={{ .Values.logLevel }}"
         {{- end }}
         {{- if .Values.runnerGithubURL  }}
         - "--runner-github-url={{ .Values.runnerGithubURL }}"
         {{- end }}
+        {{- if .Values.runner.statusUpdateHook.enabled }}
+        - "--runner-status-update-hook"
+        {{- end }}
         command:
         - "/manager"
         env:
@@ -116,15 +118,19 @@ spec:
               name: {{ include "actions-runner-controller.secretName" . }}
               optional: true
         {{- end }}
+        {{- if kindIs "slice" .Values.env }}
+        {{- toYaml .Values.env | nindent 8 }}
+        {{- else }}
         {{- range $key, $val := .Values.env }}
         - name: {{ $key }}
           value: {{ $val | quote }}
         {{- end }}
+        {{- end }}
         image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default (cat "v" .Chart.AppVersion | replace " " "") }}"
         name: manager
         imagePullPolicy: {{ .Values.image.pullPolicy }}
         ports:
-        - containerPort: 9443
+        - containerPort: {{ .Values.webhookPort }}
           name: webhook-server
           protocol: TCP
         {{- if not .Values.metrics.proxy.enabled }}

charts/actions-runner-controller/templates/githubwebhook.deployment.yaml

@@ -39,7 +39,6 @@ spec:
         {{- $metricsHost := .Values.metrics.proxy.enabled | ternary "127.0.0.1" "0.0.0.0" }}
         {{- $metricsPort := .Values.metrics.proxy.enabled | ternary "8080" .Values.metrics.port }}
         - "--metrics-addr={{ $metricsHost }}:{{ $metricsPort }}"
-        - "--sync-period={{ .Values.githubWebhookServer.syncPeriod }}"
         {{- if .Values.githubWebhookServer.logLevel }}
         - "--log-level={{ .Values.githubWebhookServer.logLevel }}"
         {{- end }}
@@ -49,6 +48,9 @@ spec:
         {{- if .Values.runnerGithubURL  }}
         - "--runner-github-url={{ .Values.runnerGithubURL }}"
         {{- end }}
+        {{- if .Values.githubWebhookServer.queueLimit }}
+        - "--queue-limit={{ .Values.githubWebhookServer.queueLimit }}"
+        {{- end }}
         command:
         - "/github-webhook-server"
         env:

charts/actions-runner-controller/templates/githubwebhook.ingress.yaml

@@ -1,16 +1,11 @@
 {{- if .Values.githubWebhookServer.ingress.enabled -}}
 {{- $fullName := include "actions-runner-controller-github-webhook-server.fullname" . -}}
 {{- $svcPort := (index .Values.githubWebhookServer.service.ports 0).port -}}
-{{- if .Capabilities.APIVersions.Has "networking.k8s.io/v1" }}
 apiVersion: networking.k8s.io/v1
-{{- else if .Capabilities.APIVersions.Has "networking.k8s.io/v1beta1" }}
-apiVersion: networking.k8s.io/v1beta1
-{{- else if .Capabilities.APIVersions.Has "extensions/v1beta1" }}
-apiVersion: extensions/v1beta1
-{{- end }}
 kind: Ingress
 metadata:
   name: {{ $fullName }}
+  namespace: {{ .Release.Namespace }}
   labels:
     {{- include "actions-runner-controller.labels" . | nindent 4 }}
   {{- with .Values.githubWebhookServer.ingress.annotations }}
@@ -36,21 +31,17 @@ spec:
     - host: {{ .host | quote }}
       http:
         paths:
+          {{- if .extraPaths }}
+          {{- toYaml .extraPaths | nindent 10 }}
+          {{- end }}
           {{- range .paths }}
           - path: {{ .path }}
-            {{- if $.Capabilities.APIVersions.Has "networking.k8s.io/v1" }}
             pathType: {{ .pathType }}
-            {{- end }}
             backend:
-              {{- if $.Capabilities.APIVersions.Has "networking.k8s.io/v1" }}
               service:
                name: {{ $fullName }}
                port:
                  number: {{ $svcPort }}
-              {{- else }}
-              serviceName: {{ $fullName }}
-              servicePort: {{ $svcPort }}
-              {{- end }}
           {{- end }}
     {{- end }}
   {{- end }}

charts/actions-runner-controller/templates/githubwebhook.secrets.yaml

@@ -12,5 +12,17 @@ data:
 {{- if .Values.githubWebhookServer.secret.github_webhook_secret_token }}
   github_webhook_secret_token: {{ .Values.githubWebhookServer.secret.github_webhook_secret_token | toString | b64enc }}
 {{- end }}
+{{- if .Values.githubWebhookServer.secret.github_app_id }}
+  github_app_id: {{ .Values.githubWebhookServer.secret.github_app_id | toString | b64enc }}
+{{- end }}
+{{- if .Values.githubWebhookServer.secret.github_app_installation_id }}
+  github_app_installation_id: {{ .Values.githubWebhookServer.secret.github_app_installation_id | toString | b64enc }}
+{{- end }}
+{{- if .Values.githubWebhookServer.secret.github_app_private_key }}
+  github_app_private_key: {{ .Values.githubWebhookServer.secret.github_app_private_key | toString | b64enc }}
+{{- end }}
+{{- if .Values.githubWebhookServer.secret.github_token }}
+  github_token: {{ .Values.githubWebhookServer.secret.github_token | toString | b64enc }}
+{{- end }}
 {{- end }}
 {{- end }}

charts/actions-runner-controller/templates/githubwebhook.serviceMonitor.yaml

@@ -8,6 +8,7 @@ metadata:
     {{- toYaml . | nindent 4 }}
   {{- end }}
   name: {{ include "actions-runner-controller-github-webhook-server.serviceMonitorName" . }}
+  namespace: {{ .Release.Namespace }}
 spec:
   endpoints:
     - path: /metrics

charts/actions-runner-controller/templates/manager_role.yaml

@@ -195,6 +195,28 @@ rules:
   verbs:
   - create
   - patch
+- apiGroups:
+  - ""
+  resources:
+  - persistentvolumeclaims
+  verbs:
+  - delete
+  - get
+  - list
+  - patch
+  - update
+  - watch
+- apiGroups:
+  - ""
+  resources:
+  - persistentvolumes
+  verbs:
+  - delete
+  - get
+  - list
+  - patch
+  - update
+  - watch
 - apiGroups:
   - coordination.k8s.io
   resources:
@@ -228,3 +250,72 @@ rules:
   - patch
   - update
   - watch
+- apiGroups:
+  - ""
+  resources:
+  - secrets
+  verbs:
+  - get
+  - list
+  - watch
+{{- if .Values.runner.statusUpdateHook.enabled }}
+- apiGroups:
+  - ""
+  resources:
+  - serviceaccounts
+  verbs:
+  - create
+  - delete
+  - get
+- apiGroups:
+  - rbac.authorization.k8s.io
+  resources:
+  - rolebindings
+  verbs:
+  - create
+  - delete
+  - get
+- apiGroups:
+  - rbac.authorization.k8s.io
+  resources:
+  - roles
+  verbs:
+  - create
+  - delete
+  - get
+{{- end }}
+{{- if .Values.rbac.allowGrantingKubernetesContainerModePermissions }}
+{{/* These permissions are required by ARC to create RBAC resources for the runner pod to use the kubernetes container mode. */}}
+{{/* See https://github.com/actions-runner-controller/actions-runner-controller/pull/1268/files#r917331632 */}}
+- apiGroups:
+  - ""
+  resources:
+  - pods/exec
+  verbs:
+  - create
+  - get
+- apiGroups:
+  - ""
+  resources:
+  - pods/log
+  verbs:
+  - get
+  - list
+  - watch
+- apiGroups:
+  - "batch"
+  resources:
+  - jobs
+  verbs:
+  - get
+  - list
+  - create
+  - delete
+- apiGroups:
+  - ""
+  resources:
+  - secrets
+  verbs:
+  - create
+  - delete
+{{- end }}

charts/actions-runner-controller/templates/webhook_service.yaml

@@ -13,7 +13,7 @@ spec:
   type: {{ .Values.service.type }}
   ports:
     - port: 443
-      targetPort: 9443
+      targetPort: {{ .Values.webhookPort }}
       protocol: TCP
       name: https
   selector:

charts/actions-runner-controller/values.yaml

@@ -6,19 +6,15 @@ labels: {}
 
 replicaCount: 1
 
-syncPeriod: 10m
+webhookPort: 9443
+syncPeriod: 1m
+defaultScaleDownDelay: 10m
 
 enableLeaderElection: true
 # Specifies the controller id for leader election.
 # Must be unique if more than one controller installed onto the same namespace.
 #leaderElectionId: "actions-runner-controller"
 
-# DEPRECATED: This has been removed as unnecessary in #1192
-# The controller tries its best not to repeat the duplicate GitHub API call
-# within this duration.
-# Defaults to syncPeriod - 10s.
-#githubAPICacheDuration: 30s
-
 # The URL of your GitHub Enterprise server, if you're using one.
 #githubEnterpriseServerURL: https://github.example.com
 
@@ -65,6 +61,18 @@ imagePullSecrets: []
 nameOverride: ""
 fullnameOverride: ""
 
+runner:
+  statusUpdateHook:
+    enabled: false
+
+rbac:
+  {}
+  # # This allows ARC to dynamically create a ServiceAccount and a Role for each Runner pod that uses "kubernetes" container mode,
+  # # by extending ARC's manager role to have the same permissions required by the pod runs the runner agent in "kubernetes" container mode.
+  # # Without this, Kubernetes blocks ARC to create the role to prevent a priviledge escalation.
+  # # See https://github.com/actions-runner-controller/actions-runner-controller/pull/1268/files#r917327010
+  # allowGrantingKubernetesContainerModePermissions: true
+
 serviceAccount:
   # Specifies whether a service account should be created
   create: true
@@ -107,7 +115,7 @@ metrics:
     enabled: true
     image:
       repository: quay.io/brancz/kube-rbac-proxy
-      tag: v0.11.0
+      tag: v0.13.0
 
 resources:
   {}
@@ -141,10 +149,20 @@ priorityClassName: ""
 
 env:
   {}
+  # specify additional environment variables for the controller pod.
+  # It's possible to specify either key vale pairs e.g.:
   # http_proxy: "proxy.com:8080"
   # https_proxy: "proxy.com:8080"
   # no_proxy: ""
 
+  # or a list of complete environment variable definitions e.g.:
+  # - name: GITHUB_APP_INSTALLATION_ID
+  #   valueFrom:
+  #     secretKeyRef:
+  #       key: some_key_in_the_secret
+  #       name: some-secret-name
+  #       optional: true
+
 ## specify additional volumes to mount in the manager container, this can be used
 ## to specify additional storage of material or to inject files from ConfigMaps
 ## into the running container
@@ -173,7 +191,6 @@ admissionWebHooks:
 githubWebhookServer:
   enabled: false
   replicaCount: 1
-  syncPeriod: 10m
   useRunnerGroupsVisibility: false
   secret:
     enabled: false
@@ -181,6 +198,13 @@ githubWebhookServer:
     name: "github-webhook-server"
     ### GitHub Webhook Configuration
     github_webhook_secret_token: ""
+    ### GitHub Apps Configuration
+    ## NOTE: IDs MUST be strings, use quotes
+    #github_app_id: ""
+    #github_app_installation_id: ""
+    #github_app_private_key: |
+    ### GitHub PAT Configuration
+    #github_token: ""
   imagePullSecrets: []
   nameOverride: ""
   fullnameOverride: ""
@@ -222,6 +246,20 @@ githubWebhookServer:
         paths: []
         #  - path: /*
         #    pathType: ImplementationSpecific
+        # Extra paths that are not automatically connected to the server. This is useful when working with annotation based services.
+        extraPaths: []
+        # - path: /*
+        #   backend:
+        #     serviceName: ssl-redirect
+        #     servicePort: use-annotation
+        ## for Kubernetes >=1.19 (when "networking.k8s.io/v1" is used)
+        # - path: /*
+        #   pathType: Prefix
+        #   backend:
+        #     service:
+        #       name: ssl-redirect
+        #       port:
+        #         name: use-annotation
     tls: []
     #  - secretName: chart-example-tls
     #    hosts:
@@ -232,3 +270,4 @@ githubWebhookServer:
     enabled: false
     # minAvailable: 1
     # maxUnavailable: 3
+  # queueLimit: 100

cmd/githubwebhookserver/main.go

@@ -69,9 +69,8 @@ func main() {
 
 		watchNamespace string
 
-		enableLeaderElection bool
-		syncPeriod           time.Duration
-		logLevel             string
+		logLevel   string
+		queueLimit int
 
 		ghClient *github.Client
 	)
@@ -88,10 +87,8 @@ func main() {
 	flag.StringVar(&webhookAddr, "webhook-addr", ":8000", "The address the metric endpoint binds to.")
 	flag.StringVar(&metricsAddr, "metrics-addr", ":8080", "The address the metric endpoint binds to.")
 	flag.StringVar(&watchNamespace, "watch-namespace", "", "The namespace to watch for HorizontalRunnerAutoscaler's to scale on Webhook. Set to empty for letting it watch for all namespaces.")
-	flag.BoolVar(&enableLeaderElection, "enable-leader-election", false,
-		"Enable leader election for controller manager. Enabling this will ensure there is only one active controller manager.")
-	flag.DurationVar(&syncPeriod, "sync-period", 10*time.Minute, "Determines the minimum frequency at which K8s resources managed by this controller are reconciled. When you use autoscaling, set to a lower value like 10 minute, because this corresponds to the minimum time to react on demand change")
 	flag.StringVar(&logLevel, "log-level", logging.LogLevelDebug, `The verbosity of the logging. Valid values are "debug", "info", "warn", "error". Defaults to "debug".`)
+	flag.IntVar(&queueLimit, "queue-limit", controllers.DefaultQueueLimit, `The maximum length of the scale operation queue. The scale opration is enqueued per every matching webhook event, and the server returns a 500 HTTP status when the queue was already full on enqueue attempt.`)
 	flag.StringVar(&webhookSecretToken, "github-webhook-secret-token", "", "The personal access token of GitHub.")
 	flag.StringVar(&c.Token, "github-token", c.Token, "The personal access token of GitHub.")
 	flag.Int64Var(&c.AppID, "github-app-id", c.AppID, "The application ID of GitHub App.")
@@ -142,10 +139,10 @@ func main() {
 		setupLog.Info("GitHub client is not initialized. Runner groups with custom visibility are not supported. If needed, please provide GitHub authentication. This will incur in extra GitHub API calls")
 	}
 
+	syncPeriod := 10 * time.Minute
 	mgr, err := ctrl.NewManager(ctrl.GetConfigOrDie(), ctrl.Options{
 		Scheme:             scheme,
 		SyncPeriod:         &syncPeriod,
-		LeaderElection:     enableLeaderElection,
 		Namespace:          watchNamespace,
 		MetricsBindAddress: metricsAddr,
 		Port:               9443,
@@ -164,6 +161,7 @@ func main() {
 		SecretKeyBytes: []byte(webhookSecretToken),
 		Namespace:      watchNamespace,
 		GitHubClient:   ghClient,
+		QueueLimit:     queueLimit,
 	}
 
 	if err = hraGitHubWebhook.SetupWithManager(mgr); err != nil {

config/crd/bases/actions.summerwind.dev_horizontalrunnerautoscalers.yaml

@@ -61,6 +61,16 @@ spec:
                         type: integer
                     type: object
                   type: array
+                githubAPICredentialsFrom:
+                  properties:
+                    secretRef:
+                      properties:
+                        name:
+                          type: string
+                      required:
+                        - name
+                      type: object
+                  type: object
                 maxReplicas:
                   description: MaxReplicas is the maximum number of replicas the deployment is allowed to scale
                   type: integer
@@ -92,7 +102,7 @@ spec:
                         description: ScaleUpThreshold is the percentage of busy runners greater than which will trigger the hpa to scale runners up.
                         type: string
                       type:
-                        description: Type is the type of metric to be used for autoscaling. The only supported Type is TotalNumberOfQueuedAndInProgressWorkflowRuns
+                        description: Type is the type of metric to be used for autoscaling. It can be TotalNumberOfQueuedAndInProgressWorkflowRuns or PercentageRunnersBusy.
                         type: string
                     type: object
                   type: array
@@ -141,6 +151,7 @@ spec:
                               status:
                                 type: string
                               types:
+                                description: 'One of: created, rerequested, or completed'
                                 items:
                                   type: string
                                 type: array
@@ -160,13 +171,16 @@ spec:
                           push:
                             description: PushSpec is the condition for triggering scale-up on push event Also see https://docs.github.com/en/actions/reference/events-that-trigger-workflows#push
                             type: object
+                          workflowJob:
+                            description: https://docs.github.com/en/developers/webhooks-and-events/webhooks/webhook-events-and-payloads#workflow_job
+                            type: object
                         type: object
                     type: object
                   type: array
                 scheduledOverrides:
                   description: ScheduledOverrides is the list of ScheduledOverride. It can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule. The earlier a scheduled override is, the higher it is prioritized.
                   items:
-                    description: ScheduledOverride can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule. A schedule can optionally be recurring, so that the correspoding override happens every day, week, month, or year.
+                    description: ScheduledOverride can be used to override a few fields of HorizontalRunnerAutoscalerSpec on schedule. A schedule can optionally be recurring, so that the corresponding override happens every day, week, month, or year.
                     properties:
                       endTime:
                         description: EndTime is the time at which the first override ends.

config/crd/patches/webhook_in_runners.yaml

@@ -8,6 +8,7 @@ spec:
   conversion:
     strategy: Webhook
     webhook:
+      conversionReviewVersions: ["v1","v1beta1"]
       clientConfig:
         # this is "\n" used as a placeholder, otherwise it will be rejected by the apiserver for being blank,
         # but we're going to set it later using the cert-manager (or potentially a patch if not using cert-manager)

config/default/kustomization.yaml

@@ -22,8 +22,6 @@ bases:
 - ../certmanager
 # [PROMETHEUS] To enable prometheus monitor, uncomment all sections with 'PROMETHEUS'.
 #- ../prometheus
-# [GH_WEBHOOK_SERVER] To enable the GitHub webhook server, uncomment all sections with 'GH_WEBHOOK_SERVER'.
-#- ../github-webhook-server
 
 patchesStrategicMerge:
 # Protect the /metrics endpoint by putting it behind auth.
@@ -46,10 +44,6 @@ patchesStrategicMerge:
 # 'CERTMANAGER' needs to be enabled to use ca injection
 - webhookcainjection_patch.yaml
 
-# [GH_WEBHOOK_SERVER] To enable the GitHub webhook server, uncomment all sections with 'GH_WEBHOOK_SERVER'.
-# Protect the GitHub webhook server metrics endpoint by putting it behind auth.
-# - gh-webhook-server-auth-proxy-patch.yaml
-
 # the following config is for teaching kustomize how to do var substitution
 vars:
 # [CERTMANAGER] To enable cert-manager, uncomment all sections with 'CERTMANAGER' prefix.

config/github-webhook-server/kustomization.yaml

@@ -2,11 +2,14 @@ apiVersion: kustomize.config.k8s.io/v1beta1
 kind: Kustomization
 
 images:
-  - name: controller
-    newName: summerwind/actions-runner-controller
-    newTag: latest
+- name: controller
+  newName: summerwind/actions-runner-controller
+  newTag: latest
 
 resources:
-  - deployment.yaml
-  - rbac.yaml
-  - service.yaml
+- deployment.yaml
+- rbac.yaml
+- service.yaml
+
+patchesStrategicMerge:
+- gh-webhook-server-auth-proxy-patch.yaml

config/rbac/role.yaml

@@ -202,6 +202,29 @@ rules:
   verbs:
   - create
   - patch
+- apiGroups:
+  - ""
+  resources:
+  - persistentvolumeclaims
+  verbs:
+  - create
+  - delete
+  - get
+  - list
+  - patch
+  - update
+  - watch
+- apiGroups:
+  - ""
+  resources:
+  - persistentvolumes
+  verbs:
+  - delete
+  - get
+  - list
+  - patch
+  - update
+  - watch
 - apiGroups:
   - ""
   resources:
@@ -226,3 +249,36 @@ rules:
   - patch
   - update
   - watch
+- apiGroups:
+  - ""
+  resources:
+  - secrets
+  verbs:
+  - delete
+  - get
+  - list
+  - watch
+- apiGroups:
+  - ""
+  resources:
+  - serviceaccounts
+  verbs:
+  - create
+  - delete
+  - get
+- apiGroups:
+  - rbac.authorization.k8s.io
+  resources:
+  - rolebindings
+  verbs:
+  - create
+  - delete
+  - get
+- apiGroups:
+  - rbac.authorization.k8s.io
+  resources:
+  - roles
+  verbs:
+  - create
+  - delete
+  - get

controllers/autoscaling.go

@@ -9,7 +9,11 @@ import (
 	"strings"
 
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
-	"github.com/google/go-github/v39/github"
+	prometheus_metrics "github.com/actions-runner-controller/actions-runner-controller/controllers/metrics"
+	arcgithub "github.com/actions-runner-controller/actions-runner-controller/github"
+	"github.com/google/go-github/v47/github"
+	corev1 "k8s.io/api/core/v1"
+	"sigs.k8s.io/controller-runtime/pkg/client"
 )
 
 const (
@@ -19,7 +23,7 @@ const (
 	defaultScaleDownFactor    = 0.7
 )
 
-func (r *HorizontalRunnerAutoscalerReconciler) suggestDesiredReplicas(st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler) (*int, error) {
+func (r *HorizontalRunnerAutoscalerReconciler) suggestDesiredReplicas(ghc *arcgithub.Client, st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler) (*int, error) {
 	if hra.Spec.MinReplicas == nil {
 		return nil, fmt.Errorf("horizontalrunnerautoscaler %s/%s is missing minReplicas", hra.Namespace, hra.Name)
 	} else if hra.Spec.MaxReplicas == nil {
@@ -29,10 +33,8 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestDesiredReplicas(st scaleTa
 	metrics := hra.Spec.Metrics
 	numMetrics := len(metrics)
 	if numMetrics == 0 {
-		if len(hra.Spec.ScaleUpTriggers) == 0 {
-			return r.suggestReplicasByQueuedAndInProgressWorkflowRuns(st, hra, nil)
-		}
-
+		// We don't default to anything since ARC 0.23.0
+		// See https://github.com/actions-runner-controller/actions-runner-controller/issues/728
 		return nil, nil
 	} else if numMetrics > 2 {
 		return nil, fmt.Errorf("too many autoscaling metrics configured: It must be 0 to 2, but got %d", numMetrics)
@@ -48,9 +50,9 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestDesiredReplicas(st scaleTa
 
 	switch primaryMetricType {
 	case v1alpha1.AutoscalingMetricTypeTotalNumberOfQueuedAndInProgressWorkflowRuns:
-		suggested, err = r.suggestReplicasByQueuedAndInProgressWorkflowRuns(st, hra, &primaryMetric)
+		suggested, err = r.suggestReplicasByQueuedAndInProgressWorkflowRuns(ghc, st, hra, &primaryMetric)
 	case v1alpha1.AutoscalingMetricTypePercentageRunnersBusy:
-		suggested, err = r.suggestReplicasByPercentageRunnersBusy(st, hra, primaryMetric)
+		suggested, err = r.suggestReplicasByPercentageRunnersBusy(ghc, st, hra, primaryMetric)
 	default:
 		return nil, fmt.Errorf("validating autoscaling metrics: unsupported metric type %q", primaryMetric)
 	}
@@ -83,11 +85,10 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestDesiredReplicas(st scaleTa
 		)
 	}
 
-	return r.suggestReplicasByQueuedAndInProgressWorkflowRuns(st, hra, &fallbackMetric)
+	return r.suggestReplicasByQueuedAndInProgressWorkflowRuns(ghc, st, hra, &fallbackMetric)
 }
 
-func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgressWorkflowRuns(st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler, metrics *v1alpha1.MetricSpec) (*int, error) {
-
+func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgressWorkflowRuns(ghc *arcgithub.Client, st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler, metrics *v1alpha1.MetricSpec) (*int, error) {
 	var repos [][]string
 	repoID := st.repo
 	if repoID == "" {
@@ -126,7 +127,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgr
 		opt := github.ListWorkflowJobsOptions{ListOptions: github.ListOptions{PerPage: 50}}
 		var allJobs []*github.WorkflowJob
 		for {
-			jobs, resp, err := r.GitHubClient.Actions.ListWorkflowJobs(context.TODO(), user, repoName, runID, &opt)
+			jobs, resp, err := ghc.Actions.ListWorkflowJobs(context.TODO(), user, repoName, runID, &opt)
 			if err != nil {
 				r.Log.Error(err, "Error listing workflow jobs")
 				return //err
@@ -140,7 +141,29 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgr
 		if len(allJobs) == 0 {
 			fallback_cb()
 		} else {
+		JOB:
 			for _, job := range allJobs {
+				runnerLabels := make(map[string]struct{}, len(st.labels))
+				for _, l := range st.labels {
+					runnerLabels[l] = struct{}{}
+				}
+
+				if len(job.Labels) == 0 {
+					// This shouldn't usually happen
+					r.Log.Info("Detected job with no labels, which is not supported by ARC. Skipping anyway.", "labels", job.Labels, "run_id", job.GetRunID(), "job_id", job.GetID())
+					continue JOB
+				}
+
+				for _, l := range job.Labels {
+					if l == "self-hosted" {
+						continue
+					}
+
+					if _, ok := runnerLabels[l]; !ok {
+						continue JOB
+					}
+				}
+
 				switch job.GetStatus() {
 				case "completed":
 					// We add a case for `completed` so it is not counted in `unknown`.
@@ -162,7 +185,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgr
 
 	for _, repo := range repos {
 		user, repoName := repo[0], repo[1]
-		workflowRuns, err := r.GitHubClient.ListRepositoryWorkflowRuns(context.TODO(), user, repoName)
+		workflowRuns, err := ghc.ListRepositoryWorkflowRuns(context.TODO(), user, repoName)
 		if err != nil {
 			return nil, err
 		}
@@ -189,6 +212,20 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgr
 
 	necessaryReplicas := queued + inProgress
 
+	prometheus_metrics.SetHorizontalRunnerAutoscalerQueuedAndInProgressWorkflowRuns(
+		hra.ObjectMeta,
+		st.enterprise,
+		st.org,
+		st.repo,
+		st.kind,
+		st.st,
+		necessaryReplicas,
+		completed,
+		inProgress,
+		queued,
+		unknown,
+	)
+
 	r.Log.V(1).Info(
 		fmt.Sprintf("Suggested desired replicas of %d by TotalNumberOfQueuedAndInProgressWorkflowRuns", necessaryReplicas),
 		"workflow_runs_completed", completed,
@@ -204,7 +241,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByQueuedAndInProgr
 	return &necessaryReplicas, nil
 }
 
-func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByPercentageRunnersBusy(st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler, metrics v1alpha1.MetricSpec) (*int, error) {
+func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByPercentageRunnersBusy(ghc *arcgithub.Client, st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler, metrics v1alpha1.MetricSpec) (*int, error) {
 	ctx := context.Background()
 	scaleUpThreshold := defaultScaleUpThreshold
 	scaleDownThreshold := defaultScaleDownThreshold
@@ -273,7 +310,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByPercentageRunner
 	)
 
 	// ListRunners will return all runners managed by GitHub - not restricted to ns
-	runners, err := r.GitHubClient.ListRunners(
+	runners, err := ghc.ListRunners(
 		ctx,
 		enterprise,
 		organization,
@@ -294,22 +331,52 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByPercentageRunner
 		numRunners           int
 		numRunnersRegistered int
 		numRunnersBusy       int
+		numTerminatingBusy   int
 	)
 
 	numRunners = len(runnerMap)
 
+	busyTerminatingRunnerPods := map[string]struct{}{}
+
+	kindLabel := LabelKeyRunnerDeploymentName
+	if hra.Spec.ScaleTargetRef.Kind == "RunnerSet" {
+		kindLabel = LabelKeyRunnerSetName
+	}
+
+	var runnerPodList corev1.PodList
+	if err := r.Client.List(ctx, &runnerPodList, client.InNamespace(hra.Namespace), client.MatchingLabels(map[string]string{
+		kindLabel: hra.Spec.ScaleTargetRef.Name,
+	})); err != nil {
+		return nil, err
+	}
+
+	for _, p := range runnerPodList.Items {
+		if p.Annotations[AnnotationKeyUnregistrationFailureMessage] != "" {
+			busyTerminatingRunnerPods[p.Name] = struct{}{}
+		}
+	}
+
 	for _, runner := range runners {
 		if _, ok := runnerMap[*runner.Name]; ok {
 			numRunnersRegistered++
 
 			if runner.GetBusy() {
 				numRunnersBusy++
+			} else if _, ok := busyTerminatingRunnerPods[*runner.Name]; ok {
+				numTerminatingBusy++
 			}
+
+			delete(busyTerminatingRunnerPods, *runner.Name)
 		}
 	}
 
+	// Remaining busyTerminatingRunnerPods are runners that were not on the ListRunners API response yet
+	for range busyTerminatingRunnerPods {
+		numTerminatingBusy++
+	}
+
 	var desiredReplicas int
-	fractionBusy := float64(numRunnersBusy) / float64(desiredReplicasBefore)
+	fractionBusy := float64(numRunnersBusy+numTerminatingBusy) / float64(desiredReplicasBefore)
 	if fractionBusy >= scaleUpThreshold {
 		if scaleUpAdjustment > 0 {
 			desiredReplicas = desiredReplicasBefore + scaleUpAdjustment
@@ -330,6 +397,19 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByPercentageRunner
 	//
 	// - num_runners can be as twice as large as replicas_desired_before while
 	//   the runnerdeployment controller is replacing RunnerReplicaSet for runner update.
+	prometheus_metrics.SetHorizontalRunnerAutoscalerPercentageRunnersBusy(
+		hra.ObjectMeta,
+		st.enterprise,
+		st.org,
+		st.repo,
+		st.kind,
+		st.st,
+		desiredReplicas,
+		numRunners,
+		numRunnersRegistered,
+		numRunnersBusy,
+		numTerminatingBusy,
+	)
 
 	r.Log.V(1).Info(
 		fmt.Sprintf("Suggested desired replicas of %d by PercentageRunnersBusy", desiredReplicas),
@@ -338,6 +418,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) suggestReplicasByPercentageRunner
 		"num_runners", numRunners,
 		"num_runners_registered", numRunnersRegistered,
 		"num_runners_busy", numRunnersBusy,
+		"num_terminating_busy", numTerminatingBusy,
 		"namespace", hra.Namespace,
 		"kind", st.kind,
 		"name", st.st,

controllers/autoscaling_test.go

@@ -41,8 +41,12 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 
 	metav1Now := metav1.Now()
 	testcases := []struct {
-		repo      string
-		org       string
+		description string
+
+		repo   string
+		org    string
+		labels []string
+
 		fixed     *int
 		max       *int
 		min       *int
@@ -68,6 +72,19 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"status":"in_progress"}, {"status":"in_progress"}]}"`,
 			want:                     3,
 		},
+		// Explicitly speified the default `self-hosted` label which is ignored by the simulator,
+		// as we assume that GitHub Actions automatically associates the `self-hosted` label to every self-hosted runner.
+		// 3 demanded, max at 3
+		{
+			repo:                     "test/valid",
+			labels:                   []string{"self-hosted"},
+			min:                      intPtr(2),
+			max:                      intPtr(3),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"status":"queued"}, {"status":"in_progress"}, {"status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"status":"in_progress"}, {"status":"in_progress"}]}"`,
+			want:                     3,
+		},
 		// 2 demanded, max at 3, currently 3, delay scaling down due to grace period
 		{
 			repo:                     "test/valid",
@@ -152,9 +169,40 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 			want:                     3,
 		},
 
-		// Job-level autoscaling
-		// 5 requested from 3 workflows
 		{
+			description:              "Job-level autoscaling with no explicit runner label (runners have implicit self-hosted, requested self-hosted, 5 jobs from 3 workflows)",
+			repo:                     "test/valid",
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted"]}, {"status":"queued", "labels":["self-hosted"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted"]}, {"status":"completed", "labels":["self-hosted"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted"]}, {"status":"queued", "labels":["self-hosted"]}]}`,
+			},
+			want: 5,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling with no explicit runner label (runners have implicit self-hosted, requested self-hosted+custom, 0 jobs from 3 workflows)",
+			repo:                     "test/valid",
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 2,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling with no label (runners have implicit self-hosted, jobs had no labels, 0 jobs from 3 workflows)",
 			repo:                     "test/valid",
 			min:                      intPtr(2),
 			max:                      intPtr(10),
@@ -166,6 +214,91 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 				2: `{"jobs": [{"status": "in_progress"}, {"status":"completed"}]}`,
 				3: `{"jobs": [{"status": "in_progress"}, {"status":"queued"}]}`,
 			},
+			want: 2,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling with default runner label (runners have self-hosted only, requested self-hosted+custom, 0 jobs from 3 workflows)",
+			repo:                     "test/valid",
+			labels:                   []string{"self-hosted"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 2,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling with custom runner label (runners have custom2, requested self-hosted+custom, 0 jobs from 5 workflows",
+			repo:                     "test/valid",
+			labels:                   []string{"custom2"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 2,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling with default runner label (runners have self-hosted, requested managed-runner-label, 0 jobs from 3 runs)",
+			repo:                     "test/valid",
+			labels:                   []string{"self-hosted"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["managed-runner-label"]}, {"status":"queued", "labels":["managed-runner-label"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["managed-runner-label"]}, {"status":"completed", "labels":["managed-runner-label"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["managed-runner-label"]}, {"status":"queued", "labels":["managed-runner-label"]}]}`,
+			},
+			want: 2,
+		},
+
+		{
+			description:              "Job-level autoscaling with default + custom runner label (runners have self-hosted+custom, requested self-hosted+custom, 5 jobs from 3 workflows)",
+			repo:                     "test/valid",
+			labels:                   []string{"self-hosted", "custom"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 5,
+		},
+
+		{
+			description:              "Job-level autoscaling with custom runner label (runners have custom, requested self-hosted+custom, 5 jobs from 3 workflows)",
+			repo:                     "test/valid",
+			labels:                   []string{"custom"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
 			want: 5,
 		},
 	}
@@ -181,7 +314,12 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 		_ = clientgoscheme.AddToScheme(scheme)
 		_ = v1alpha1.AddToScheme(scheme)
 
-		t.Run(fmt.Sprintf("case %d", i), func(t *testing.T) {
+		testName := fmt.Sprintf("case %d", i)
+		if tc.description != "" {
+			testName = tc.description
+		}
+
+		t.Run(testName, func(t *testing.T) {
 			server := fake.NewServer(
 				fake.WithListRepositoryWorkflowRunsResponse(200, tc.workflowRuns, tc.workflowRuns_queued, tc.workflowRuns_in_progress),
 				fake.WithListWorkflowJobsResponse(200, tc.workflowJobs),
@@ -191,9 +329,9 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 			client := newGithubClient(server)
 
 			h := &HorizontalRunnerAutoscalerReconciler{
-				Log:          log,
-				GitHubClient: client,
-				Scheme:       scheme,
+				Log:                   log,
+				Scheme:                scheme,
+				DefaultScaleDownDelay: DefaultScaleDownDelay,
 			}
 
 			rd := v1alpha1.RunnerDeployment{
@@ -206,6 +344,7 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 						Spec: v1alpha1.RunnerSpec{
 							RunnerConfig: v1alpha1.RunnerConfig{
 								Repository: tc.repo,
+								Labels:     tc.labels,
 							},
 						},
 					},
@@ -220,6 +359,11 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 				Spec: v1alpha1.HorizontalRunnerAutoscalerSpec{
 					MaxReplicas: tc.max,
 					MinReplicas: tc.min,
+					Metrics: []v1alpha1.MetricSpec{
+						{
+							Type: "TotalNumberOfQueuedAndInProgressWorkflowRuns",
+						},
+					},
 				},
 				Status: v1alpha1.HorizontalRunnerAutoscalerStatus{
 					DesiredReplicas:            tc.sReplicas,
@@ -234,7 +378,7 @@ func TestDetermineDesiredReplicas_RepositoryRunner(t *testing.T) {
 
 			st := h.scaleTargetFromRD(context.Background(), rd)
 
-			got, err := h.computeReplicasWithCache(log, metav1Now.Time, st, hra, minReplicas)
+			got, err := h.computeReplicasWithCache(client, log, metav1Now.Time, st, hra, minReplicas)
 			if err != nil {
 				if tc.err == "" {
 					t.Fatalf("unexpected error: expected none, got %v", err)
@@ -258,8 +402,12 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 
 	metav1Now := metav1.Now()
 	testcases := []struct {
-		repos     []string
-		org       string
+		description string
+
+		repos  []string
+		org    string
+		labels []string
+
 		fixed     *int
 		max       *int
 		min       *int
@@ -399,9 +547,43 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 			err:                      "validating autoscaling metrics: spec.autoscaling.metrics[].repositoryNames is required and must have one more more entries for organizational runner deployment",
 		},
 
-		// Job-level autoscaling
-		// 5 requested from 3 workflows
 		{
+			description:              "Job-level autoscaling (runners have implicit self-hosted, requested self-hosted, 5 jobs from 3 runs)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted"]}, {"status":"queued", "labels":["self-hosted"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted"]}, {"status":"completed", "labels":["self-hosted"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted"]}, {"status":"queued", "labels":["self-hosted"]}]}`,
+			},
+			want: 5,
+		},
+
+		{
+			description:              "Job-level autoscaling (runners have explicit self-hosted, requested self-hosted, 5 jobs from 3 runs)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			labels:                   []string{"self-hosted"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted"]}, {"status":"queued", "labels":["self-hosted"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted"]}, {"status":"completed", "labels":["self-hosted"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted"]}, {"status":"queued", "labels":["self-hosted"]}]}`,
+			},
+			want: 5,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling (jobs lack labels, 0 requested from 3 workflows)",
 			org:                      "test",
 			repos:                    []string{"valid"},
 			min:                      intPtr(2),
@@ -414,8 +596,97 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 				2: `{"jobs": [{"status": "in_progress"}, {"status":"completed"}]}`,
 				3: `{"jobs": [{"status": "in_progress"}, {"status":"queued"}]}`,
 			},
+			want: 2,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling (runners have valid and implicit self-hosted, requested self-hosted+custom, 0 jobs from 3 runs)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 2,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling (runners have self-hosted, requested self-hosted+custom, 0 jobs from 3 workflows)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			labels:                   []string{"self-hosted"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 2,
+		},
+
+		{
+			description:              "Job-level autoscaling (runners have custom, requested self-hosted+custom, 5 requested from 3 workflows)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			labels:                   []string{"custom"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
 			want: 5,
 		},
+
+		{
+			description:              "Job-level autoscaling (runners have custom, requested custom, 5 requested from 3 workflows)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			labels:                   []string{"custom"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["custom"]}, {"status":"queued", "labels":["custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["custom"]}, {"status":"completed", "labels":["custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["custom"]}, {"status":"queued", "labels":["custom"]}]}`,
+			},
+			want: 5,
+		},
+
+		{
+			description:              "Skipped job-level autoscaling (specified custom2, 0 requested from 3 workflows)",
+			org:                      "test",
+			repos:                    []string{"valid"},
+			labels:                   []string{"custom2"},
+			min:                      intPtr(2),
+			max:                      intPtr(10),
+			workflowRuns:             `{"total_count": 4, "workflow_runs":[{"id": 1, "status":"queued"}, {"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowRuns_queued:      `{"total_count": 1, "workflow_runs":[{"id": 1, "status":"queued"}]}"`,
+			workflowRuns_in_progress: `{"total_count": 2, "workflow_runs":[{"id": 2, "status":"in_progress"}, {"id": 3, "status":"in_progress"}, {"status":"completed"}]}"`,
+			workflowJobs: map[int]string{
+				1: `{"jobs": [{"status":"queued", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+				2: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"completed", "labels":["self-hosted", "custom"]}]}`,
+				3: `{"jobs": [{"status": "in_progress", "labels":["self-hosted", "custom"]}, {"status":"queued", "labels":["self-hosted", "custom"]}]}`,
+			},
+			want: 2,
+		},
 	}
 
 	for i := range testcases {
@@ -429,7 +700,12 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 		_ = clientgoscheme.AddToScheme(scheme)
 		_ = v1alpha1.AddToScheme(scheme)
 
-		t.Run(fmt.Sprintf("case %d", i), func(t *testing.T) {
+		testName := fmt.Sprintf("case %d", i)
+		if tc.description != "" {
+			testName = tc.description
+		}
+
+		t.Run(testName, func(t *testing.T) {
 			t.Helper()
 
 			server := fake.NewServer(
@@ -441,9 +717,9 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 			client := newGithubClient(server)
 
 			h := &HorizontalRunnerAutoscalerReconciler{
-				Log:          log,
-				Scheme:       scheme,
-				GitHubClient: client,
+				Log:                   log,
+				Scheme:                scheme,
+				DefaultScaleDownDelay: DefaultScaleDownDelay,
 			}
 
 			rd := v1alpha1.RunnerDeployment{
@@ -465,6 +741,7 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 						Spec: v1alpha1.RunnerSpec{
 							RunnerConfig: v1alpha1.RunnerConfig{
 								Organization: tc.org,
+								Labels:       tc.labels,
 							},
 						},
 					},
@@ -502,7 +779,7 @@ func TestDetermineDesiredReplicas_OrganizationalRunner(t *testing.T) {
 
 			st := h.scaleTargetFromRD(context.Background(), rd)
 
-			got, err := h.computeReplicasWithCache(log, metav1Now.Time, st, hra, minReplicas)
+			got, err := h.computeReplicasWithCache(client, log, metav1Now.Time, st, hra, minReplicas)
 			if err != nil {
 				if tc.err == "" {
 					t.Fatalf("unexpected error: expected none, got %v", err)

controllers/constants.go

@@ -4,17 +4,22 @@ import "time"
 
 const (
 	LabelKeyRunnerSetName = "runnerset-name"
+	LabelKeyRunner        = "actions-runner"
 )
 
 const (
 	// This names requires at least one slash to work.
 	// See https://github.com/google/knative-gcp/issues/378
-	runnerPodFinalizerName = "actions.summerwind.dev/runner-pod"
+	runnerPodFinalizerName             = "actions.summerwind.dev/runner-pod"
+	runnerLinkedResourcesFinalizerName = "actions.summerwind.dev/linked-resources"
 
 	annotationKeyPrefix = "actions-runner/"
 
 	AnnotationKeyLastRegistrationCheckTime = "actions-runner-controller/last-registration-check-time"
 
+	// AnnotationKeyUnregistrationFailureMessage is the annotation that is added onto the pod once it failed to be unregistered from GitHub due to e.g. 422 error
+	AnnotationKeyUnregistrationFailureMessage = annotationKeyPrefix + "unregistration-failure-message"
+
 	// AnnotationKeyUnregistrationCompleteTimestamp is the annotation that is added onto the pod once the previously started unregistration process has been completed.
 	AnnotationKeyUnregistrationCompleteTimestamp = annotationKeyPrefix + "unregistration-complete-timestamp"
 
@@ -47,8 +52,6 @@ const (
 	// A pod that is timed out can be terminated if needed.
 	registrationTimeout = 10 * time.Minute
 
-	defaultRegistrationCheckInterval = time.Minute
-
 	// DefaultRunnerPodRecreationDelayAfterWebhookScale is the delay until syncing the runners with the desired replicas
 	// after a webhook-based scale up.
 	// This is used to prevent ARC from recreating completed runner pods that are deleted soon without being used at all.
@@ -63,4 +66,7 @@ const (
 
 	EnvVarRunnerName  = "RUNNER_NAME"
 	EnvVarRunnerToken = "RUNNER_TOKEN"
+
+	// defaultHookPath is path to the hook script used when the "containerMode: kubernetes" is specified
+	defaultRunnerHookPath = "/runner/k8s/index.js"
 )

controllers/horizontal_runner_autoscaler_batch_scale.go

@@ -0,0 +1,207 @@
+package controllers
+
+import (
+	"context"
+	"fmt"
+	"sync"
+	"time"
+
+	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
+	"github.com/go-logr/logr"
+	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
+	"k8s.io/apimachinery/pkg/types"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+)
+
+type batchScaler struct {
+	Ctx      context.Context
+	Client   client.Client
+	Log      logr.Logger
+	interval time.Duration
+
+	queue       chan *ScaleTarget
+	workerStart sync.Once
+}
+
+func newBatchScaler(ctx context.Context, client client.Client, log logr.Logger) *batchScaler {
+	return &batchScaler{
+		Ctx:      ctx,
+		Client:   client,
+		Log:      log,
+		interval: 3 * time.Second,
+	}
+}
+
+type batchScaleOperation struct {
+	namespacedName types.NamespacedName
+	scaleOps       []scaleOperation
+}
+
+type scaleOperation struct {
+	trigger v1alpha1.ScaleUpTrigger
+	log     logr.Logger
+}
+
+// Add the scale target to the unbounded queue, blocking until the target is successfully added to the queue.
+// All the targets in the queue are dequeued every 3 seconds, grouped by the HRA, and applied.
+// In a happy path, batchScaler update each HRA only once, even though the HRA had two or more associated webhook events in the 3 seconds interval,
+// which results in less K8s API calls and less HRA update conflicts in case your ARC installation receives a lot of webhook events
+func (s *batchScaler) Add(st *ScaleTarget) {
+	if st == nil {
+		return
+	}
+
+	s.workerStart.Do(func() {
+		var expBackoff = []time.Duration{time.Second, 2 * time.Second, 4 * time.Second, 8 * time.Second, 16 * time.Second}
+
+		s.queue = make(chan *ScaleTarget)
+
+		log := s.Log
+
+		go func() {
+			log.Info("Starting batch worker")
+			defer log.Info("Stopped batch worker")
+
+			for {
+				select {
+				case <-s.Ctx.Done():
+					return
+				default:
+				}
+
+				log.V(2).Info("Batch worker is dequeueing operations")
+
+				batches := map[types.NamespacedName]batchScaleOperation{}
+				after := time.After(s.interval)
+				var ops uint
+
+			batch:
+				for {
+					select {
+					case <-after:
+						after = nil
+						break batch
+					case st := <-s.queue:
+						nsName := types.NamespacedName{
+							Namespace: st.HorizontalRunnerAutoscaler.Namespace,
+							Name:      st.HorizontalRunnerAutoscaler.Name,
+						}
+						b, ok := batches[nsName]
+						if !ok {
+							b = batchScaleOperation{
+								namespacedName: nsName,
+							}
+						}
+						b.scaleOps = append(b.scaleOps, scaleOperation{
+							log:     *st.log,
+							trigger: st.ScaleUpTrigger,
+						})
+						batches[nsName] = b
+						ops++
+					}
+				}
+
+				log.V(2).Info("Batch worker dequeued operations", "ops", ops, "batches", len(batches))
+
+			retry:
+				for i := 0; ; i++ {
+					failed := map[types.NamespacedName]batchScaleOperation{}
+
+					for nsName, b := range batches {
+						b := b
+						if err := s.batchScale(context.Background(), b); err != nil {
+							log.V(2).Info("Failed to scale due to error", "error", err)
+							failed[nsName] = b
+						} else {
+							log.V(2).Info("Successfully ran batch scale", "hra", b.namespacedName)
+						}
+					}
+
+					if len(failed) == 0 {
+						break retry
+					}
+
+					batches = failed
+
+					delay := 16 * time.Second
+					if i < len(expBackoff) {
+						delay = expBackoff[i]
+					}
+					time.Sleep(delay)
+				}
+			}
+		}()
+	})
+
+	s.queue <- st
+}
+
+func (s *batchScaler) batchScale(ctx context.Context, batch batchScaleOperation) error {
+	var hra v1alpha1.HorizontalRunnerAutoscaler
+
+	if err := s.Client.Get(ctx, batch.namespacedName, &hra); err != nil {
+		return err
+	}
+
+	copy := hra.DeepCopy()
+
+	copy.Spec.CapacityReservations = getValidCapacityReservations(copy)
+
+	var added, completed int
+
+	for _, scale := range batch.scaleOps {
+		amount := 1
+
+		if scale.trigger.Amount != 0 {
+			amount = scale.trigger.Amount
+		}
+
+		scale.log.V(2).Info("Adding capacity reservation", "amount", amount)
+
+		if amount > 0 {
+			now := time.Now()
+			copy.Spec.CapacityReservations = append(copy.Spec.CapacityReservations, v1alpha1.CapacityReservation{
+				EffectiveTime:  metav1.Time{Time: now},
+				ExpirationTime: metav1.Time{Time: now.Add(scale.trigger.Duration.Duration)},
+				Replicas:       amount,
+			})
+
+			added += amount
+		} else if amount < 0 {
+			var reservations []v1alpha1.CapacityReservation
+
+			var found bool
+
+			for _, r := range copy.Spec.CapacityReservations {
+				if !found && r.Replicas+amount == 0 {
+					found = true
+				} else {
+					reservations = append(reservations, r)
+				}
+			}
+
+			copy.Spec.CapacityReservations = reservations
+
+			completed += amount
+		}
+	}
+
+	before := len(hra.Spec.CapacityReservations)
+	expired := before - len(copy.Spec.CapacityReservations)
+	after := len(copy.Spec.CapacityReservations)
+
+	s.Log.V(1).Info(
+		fmt.Sprintf("Updating hra %s for capacityReservations update", hra.Name),
+		"before", before,
+		"expired", expired,
+		"added", added,
+		"completed", completed,
+		"after", after,
+	)
+
+	if err := s.Client.Update(ctx, copy); err != nil {
+		return fmt.Errorf("updating horizontalrunnerautoscaler to add capacity reservation: %w", err)
+	}
+
+	return nil
+}

controllers/horizontal_runner_autoscaler_webhook.go

@@ -23,14 +23,14 @@ import (
 	"io/ioutil"
 	"net/http"
 	"strings"
+	"sync"
 	"time"
 
-	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/apimachinery/pkg/types"
 	"sigs.k8s.io/controller-runtime/pkg/reconcile"
 
 	"github.com/go-logr/logr"
-	gogithub "github.com/google/go-github/v39/github"
+	gogithub "github.com/google/go-github/v47/github"
 	"k8s.io/apimachinery/pkg/runtime"
 	"k8s.io/client-go/tools/record"
 	ctrl "sigs.k8s.io/controller-runtime"
@@ -46,6 +46,8 @@ const (
 
 	keyPrefixEnterprise = "enterprises/"
 	keyRunnerGroup      = "/group/"
+
+	DefaultQueueLimit = 100
 )
 
 // HorizontalRunnerAutoscalerGitHubWebhook autoscales a HorizontalRunnerAutoscaler and the RunnerDeployment on each
@@ -68,6 +70,15 @@ type HorizontalRunnerAutoscalerGitHubWebhook struct {
 	// Set to empty for letting it watch for all namespaces.
 	Namespace string
 	Name      string
+
+	// QueueLimit is the maximum length of the bounded queue of scale targets and their associated operations
+	// A scale target is enqueued on each retrieval of each eligible webhook event, so that it is processed asynchronously.
+	QueueLimit int
+
+	worker      *worker
+	workerInit  sync.Once
+	workerStart sync.Once
+	batchCh     chan *ScaleTarget
 }
 
 func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) Reconcile(_ context.Context, request reconcile.Request) (reconcile.Result, error) {
@@ -242,18 +253,23 @@ func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) Handle(w http.Respons
 				enterpriseSlug,
 				labels,
 			)
-
-			if target != nil {
-				if e.GetAction() == "queued" {
-					target.Amount = 1
-				} else if e.GetAction() == "completed" {
-					// A nagative amount is processed in the tryScale func as a scale-down request,
-					// that erasese the oldest CapacityReservation with the same amount.
-					// If the first CapacityReservation was with Replicas=1, this negative scale target erases that,
-					// so that the resulting desired replicas decreases by 1.
-					target.Amount = -1
-				}
+			if target == nil {
+				break
 			}
+
+			if e.GetAction() == "queued" {
+				target.Amount = 1
+				break
+			} else if e.GetAction() == "completed" && e.GetWorkflowJob().GetConclusion() != "skipped" {
+				// A nagative amount is processed in the tryScale func as a scale-down request,
+				// that erasese the oldest CapacityReservation with the same amount.
+				// If the first CapacityReservation was with Replicas=1, this negative scale target erases that,
+				// so that the resulting desired replicas decreases by 1.
+				target.Amount = -1
+				break
+			}
+			// If the conclusion is "skipped", we will ignore it and fallthrough to the default case.
+			fallthrough
 		default:
 			ok = true
 
@@ -307,9 +323,19 @@ func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) Handle(w http.Respons
 		return
 	}
 
-	if err := autoscaler.tryScale(context.TODO(), target); err != nil {
-		log.Error(err, "could not scale up")
+	autoscaler.workerInit.Do(func() {
+		batchScaler := newBatchScaler(context.Background(), autoscaler.Client, autoscaler.Log)
 
+		queueLimit := autoscaler.QueueLimit
+		if queueLimit == 0 {
+			queueLimit = DefaultQueueLimit
+		}
+		autoscaler.worker = newWorker(context.Background(), queueLimit, batchScaler.Add)
+	})
+
+	target.log = &log
+	if ok := autoscaler.worker.Add(target); !ok {
+		log.Error(err, "Could not scale up due to queue full")
 		return
 	}
 
@@ -378,6 +404,8 @@ func matchTriggerConditionAgainstEvent(types []string, eventAction *string) bool
 type ScaleTarget struct {
 	v1alpha1.HorizontalRunnerAutoscaler
 	v1alpha1.ScaleUpTrigger
+
+	log *logr.Logger
 }
 
 func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) searchScaleTargets(hras []v1alpha1.HorizontalRunnerAutoscaler, f func(v1alpha1.ScaleUpTrigger) bool) []ScaleTarget {
@@ -496,6 +524,7 @@ func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) getScaleUpTargetWithF
 	if autoscaler.GitHubClient != nil {
 		simu := &simulator.Simulator{
 			Client: autoscaler.GitHubClient,
+			Log:    log,
 		}
 		// Get available organization runner groups and enterprise runner groups for a repository
 		// These are the sum of runner groups with repository access = All repositories and runner groups
@@ -663,16 +692,29 @@ HRA:
 
 		if len(hra.Spec.ScaleUpTriggers) > 1 {
 			autoscaler.Log.V(1).Info("Skipping this HRA as it has too many ScaleUpTriggers to be used in workflow_job based scaling", "hra", hra.Name)
+			continue
+		}
+
+		if len(hra.Spec.ScaleUpTriggers) == 0 {
+			autoscaler.Log.V(1).Info("Skipping this HRA as it has no ScaleUpTriggers configured", "hra", hra.Name)
+			continue
+		}
+
+		scaleUpTrigger := hra.Spec.ScaleUpTriggers[0]
+
+		if scaleUpTrigger.GitHubEvent == nil {
+			autoscaler.Log.V(1).Info("Skipping this HRA as it has no `githubEvent` scale trigger configured", "hra", hra.Name)
 
 			continue
 		}
 
-		var duration metav1.Duration
+		if scaleUpTrigger.GitHubEvent.WorkflowJob == nil {
+			autoscaler.Log.V(1).Info("Skipping this HRA as it has no `githubEvent.workflowJob` scale trigger configured", "hra", hra.Name)
 
-		if len(hra.Spec.ScaleUpTriggers) > 0 {
-			duration = hra.Spec.ScaleUpTriggers[0].Duration
+			continue
 		}
 
+		duration := scaleUpTrigger.Duration
 		if duration.Duration <= 0 {
 			// Try to release the reserved capacity after at least 10 minutes by default,
 			// we won't end up in the reserved capacity remained forever in case GitHub somehow stopped sending us "completed" workflow_job events.
@@ -752,63 +794,6 @@ HRA:
 	return nil, nil
 }
 
-func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) tryScale(ctx context.Context, target *ScaleTarget) error {
-	if target == nil {
-		return nil
-	}
-
-	copy := target.HorizontalRunnerAutoscaler.DeepCopy()
-
-	amount := 1
-
-	if target.ScaleUpTrigger.Amount != 0 {
-		amount = target.ScaleUpTrigger.Amount
-	}
-
-	capacityReservations := getValidCapacityReservations(copy)
-
-	if amount > 0 {
-		now := time.Now()
-		copy.Spec.CapacityReservations = append(capacityReservations, v1alpha1.CapacityReservation{
-			EffectiveTime:  metav1.Time{Time: now},
-			ExpirationTime: metav1.Time{Time: now.Add(target.ScaleUpTrigger.Duration.Duration)},
-			Replicas:       amount,
-		})
-	} else if amount < 0 {
-		var reservations []v1alpha1.CapacityReservation
-
-		var found bool
-
-		for _, r := range capacityReservations {
-			if !found && r.Replicas+amount == 0 {
-				found = true
-			} else {
-				reservations = append(reservations, r)
-			}
-		}
-
-		copy.Spec.CapacityReservations = reservations
-	}
-
-	before := len(target.HorizontalRunnerAutoscaler.Spec.CapacityReservations)
-	expired := before - len(capacityReservations)
-	after := len(copy.Spec.CapacityReservations)
-
-	autoscaler.Log.V(1).Info(
-		fmt.Sprintf("Patching hra %s for capacityReservations update", target.HorizontalRunnerAutoscaler.Name),
-		"before", before,
-		"expired", expired,
-		"amount", amount,
-		"after", after,
-	)
-
-	if err := autoscaler.Client.Patch(ctx, copy, client.MergeFrom(&target.HorizontalRunnerAutoscaler)); err != nil {
-		return fmt.Errorf("patching horizontalrunnerautoscaler to add capacity reservation: %w", err)
-	}
-
-	return nil
-}
-
 func getValidCapacityReservations(autoscaler *v1alpha1.HorizontalRunnerAutoscaler) []v1alpha1.CapacityReservation {
 	var capacityReservations []v1alpha1.CapacityReservation
 

controllers/horizontal_runner_autoscaler_webhook_on_check_run.go

@@ -3,7 +3,7 @@ package controllers
 import (
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
 	"github.com/actions-runner-controller/actions-runner-controller/pkg/actionsglob"
-	"github.com/google/go-github/v39/github"
+	"github.com/google/go-github/v47/github"
 )
 
 func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) MatchCheckRunEvent(event *github.CheckRunEvent) func(scaleUpTrigger v1alpha1.ScaleUpTrigger) bool {

controllers/horizontal_runner_autoscaler_webhook_on_pull_request.go

@@ -2,7 +2,7 @@ package controllers
 
 import (
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
-	"github.com/google/go-github/v39/github"
+	"github.com/google/go-github/v47/github"
 )
 
 func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) MatchPullRequestEvent(event *github.PullRequestEvent) func(scaleUpTrigger v1alpha1.ScaleUpTrigger) bool {

controllers/horizontal_runner_autoscaler_webhook_on_push.go

@@ -2,7 +2,7 @@ package controllers
 
 import (
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
-	"github.com/google/go-github/v39/github"
+	"github.com/google/go-github/v47/github"
 )
 
 func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) MatchPushEvent(event *github.PushEvent) func(scaleUpTrigger v1alpha1.ScaleUpTrigger) bool {
@@ -15,10 +15,6 @@ func (autoscaler *HorizontalRunnerAutoscalerGitHubWebhook) MatchPushEvent(event
 
 		push := g.Push
 
-		if push == nil {
-			return false
-		}
-
-		return true
+		return push != nil
 	}
 }

controllers/horizontal_runner_autoscaler_webhook_test.go

@@ -15,7 +15,7 @@ import (
 
 	actionsv1alpha1 "github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
 	"github.com/go-logr/logr"
-	"github.com/google/go-github/v39/github"
+	"github.com/google/go-github/v47/github"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/apimachinery/pkg/runtime"
 	clientgoscheme "k8s.io/client-go/kubernetes/scheme"
@@ -138,6 +138,13 @@ func TestWebhookWorkflowJob(t *testing.T) {
 				ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
 					Name: "test-name",
 				},
+				ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+					{
+						GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+							WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+						},
+					},
+				},
 			},
 		}
 
@@ -177,6 +184,13 @@ func TestWebhookWorkflowJob(t *testing.T) {
 				ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
 					Name: "test-name",
 				},
+				ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+					{
+						GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+							WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+						},
+					},
+				},
 			},
 		}
 
@@ -217,6 +231,13 @@ func TestWebhookWorkflowJob(t *testing.T) {
 				ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
 					Name: "test-name",
 				},
+				ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+					{
+						GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+							WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+						},
+					},
+				},
 			},
 		}
 
@@ -277,6 +298,13 @@ func TestWebhookWorkflowJobWithSelfHostedLabel(t *testing.T) {
 				ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
 					Name: "test-name",
 				},
+				ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+					{
+						GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+							WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+						},
+					},
+				},
 			},
 		}
 
@@ -316,6 +344,13 @@ func TestWebhookWorkflowJobWithSelfHostedLabel(t *testing.T) {
 				ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
 					Name: "test-name",
 				},
+				ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+					{
+						GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+							WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+						},
+					},
+				},
 			},
 		}
 
@@ -356,6 +391,13 @@ func TestWebhookWorkflowJobWithSelfHostedLabel(t *testing.T) {
 				ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
 					Name: "test-name",
 				},
+				ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+					{
+						GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+							WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+						},
+					},
+				},
 			},
 		}
 

controllers/horizontal_runner_autoscaler_webhook_worker.go

@@ -0,0 +1,55 @@
+package controllers
+
+import (
+	"context"
+)
+
+// worker is a worker that has a non-blocking bounded queue of scale targets, dequeues scale target and executes the scale operation one by one.
+type worker struct {
+	scaleTargetQueue chan *ScaleTarget
+	work             func(*ScaleTarget)
+	done             chan struct{}
+}
+
+func newWorker(ctx context.Context, queueLimit int, work func(*ScaleTarget)) *worker {
+	w := &worker{
+		scaleTargetQueue: make(chan *ScaleTarget, queueLimit),
+		work:             work,
+		done:             make(chan struct{}),
+	}
+
+	go func() {
+		defer close(w.done)
+
+		for {
+			select {
+			case <-ctx.Done():
+				return
+			case t := <-w.scaleTargetQueue:
+				work(t)
+			}
+		}
+	}()
+
+	return w
+}
+
+// Add the scale target to the bounded queue, returning the result as a bool value. It returns true on successful enqueue, and returns false otherwise.
+// When returned false, the queue is already full so the enqueue operation must be retried later.
+// If the enqueue was triggered by an external source and there's no intermediate queue that we can use,
+// you must instruct the source to resend the original request later.
+// In case you're building a webhook server around this worker, this means that you must return a http error to the webhook server,
+// so that (hopefully) the sender can resend the webhook event later, or at least the human operator can notice or be notified about the
+// webhook develiery failure so that a manual retry can be done later.
+func (w *worker) Add(st *ScaleTarget) bool {
+	select {
+	case w.scaleTargetQueue <- st:
+		return true
+	default:
+		return false
+	}
+}
+
+func (w *worker) Done() chan struct{} {
+	return w.done
+}

controllers/horizontal_runner_autoscaler_webhook_worker_test.go

@@ -0,0 +1,36 @@
+package controllers
+
+import (
+	"context"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+)
+
+func TestWorker_Add(t *testing.T) {
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	w := newWorker(ctx, 2, func(st *ScaleTarget) {})
+	require.True(t, w.Add(&ScaleTarget{}))
+	require.True(t, w.Add(&ScaleTarget{}))
+	require.False(t, w.Add(&ScaleTarget{}))
+}
+
+func TestWorker_Work(t *testing.T) {
+	ctx, cancel := context.WithCancel(context.Background())
+	defer cancel()
+
+	var count int
+
+	w := newWorker(ctx, 1, func(st *ScaleTarget) {
+		count++
+		cancel()
+	})
+	require.True(t, w.Add(&ScaleTarget{}))
+	require.False(t, w.Add(&ScaleTarget{}))
+
+	<-w.Done()
+
+	require.Equal(t, count, 1)
+}

controllers/horizontalrunnerautoscaler_controller.go

@@ -24,7 +24,6 @@ import (
 
 	corev1 "k8s.io/api/core/v1"
 
-	"github.com/actions-runner-controller/actions-runner-controller/github"
 	"github.com/go-logr/logr"
 	kerrors "k8s.io/apimachinery/pkg/api/errors"
 	"k8s.io/apimachinery/pkg/types"
@@ -38,6 +37,7 @@ import (
 
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
 	"github.com/actions-runner-controller/actions-runner-controller/controllers/metrics"
+	arcgithub "github.com/actions-runner-controller/actions-runner-controller/github"
 )
 
 const (
@@ -47,13 +47,12 @@ const (
 // HorizontalRunnerAutoscalerReconciler reconciles a HorizontalRunnerAutoscaler object
 type HorizontalRunnerAutoscalerReconciler struct {
 	client.Client
-	GitHubClient *github.Client
-	Log          logr.Logger
-	Recorder     record.EventRecorder
-	Scheme       *runtime.Scheme
-
-	CacheDuration time.Duration
-	Name          string
+	GitHubClient          *MultiGitHubClient
+	Log                   logr.Logger
+	Recorder              record.EventRecorder
+	Scheme                *runtime.Scheme
+	DefaultScaleDownDelay time.Duration
+	Name                  string
 }
 
 const defaultReplicas = 1
@@ -73,6 +72,8 @@ func (r *HorizontalRunnerAutoscalerReconciler) Reconcile(ctx context.Context, re
 	}
 
 	if !hra.ObjectMeta.DeletionTimestamp.IsZero() {
+		r.GitHubClient.DeinitForHRA(&hra)
+
 		return ctrl.Result{}, nil
 	}
 
@@ -159,6 +160,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) Reconcile(ctx context.Context, re
 			org:        rs.Spec.Organization,
 			repo:       rs.Spec.Repository,
 			replicas:   replicas,
+			labels:     rs.Spec.RunnerConfig.Labels,
 			getRunnerMap: func() (map[string]struct{}, error) {
 				// return the list of runners in namespace. Horizontal Runner Autoscaler should only be responsible for scaling resources in its own ns.
 				var runnerPodList corev1.PodList
@@ -251,6 +253,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) scaleTargetFromRD(ctx context.Con
 		org:        rd.Spec.Template.Spec.Organization,
 		repo:       rd.Spec.Template.Spec.Repository,
 		replicas:   rd.Spec.Replicas,
+		labels:     rd.Spec.Template.Spec.RunnerConfig.Labels,
 		getRunnerMap: func() (map[string]struct{}, error) {
 			// return the list of runners in namespace. Horizontal Runner Autoscaler should only be responsible for scaling resources in its own ns.
 			var runnerList v1alpha1.RunnerList
@@ -293,6 +296,7 @@ type scaleTarget struct {
 	st, kind              string
 	enterprise, repo, org string
 	replicas              *int
+	labels                []string
 
 	getRunnerMap func() (map[string]struct{}, error)
 }
@@ -307,7 +311,12 @@ func (r *HorizontalRunnerAutoscalerReconciler) reconcile(ctx context.Context, re
 		return ctrl.Result{}, err
 	}
 
-	newDesiredReplicas, err := r.computeReplicasWithCache(log, now, st, hra, minReplicas)
+	ghc, err := r.GitHubClient.InitForHRA(context.Background(), &hra)
+	if err != nil {
+		return ctrl.Result{}, err
+	}
+
+	newDesiredReplicas, err := r.computeReplicasWithCache(ghc, log, now, st, hra, minReplicas)
 	if err != nil {
 		r.Recorder.Event(&hra, corev1.EventTypeNormal, "RunnerAutoscalingFailure", err.Error())
 
@@ -458,10 +467,10 @@ func (r *HorizontalRunnerAutoscalerReconciler) getMinReplicas(log logr.Logger, n
 	return minReplicas, active, upcoming, nil
 }
 
-func (r *HorizontalRunnerAutoscalerReconciler) computeReplicasWithCache(log logr.Logger, now time.Time, st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler, minReplicas int) (int, error) {
+func (r *HorizontalRunnerAutoscalerReconciler) computeReplicasWithCache(ghc *arcgithub.Client, log logr.Logger, now time.Time, st scaleTarget, hra v1alpha1.HorizontalRunnerAutoscaler, minReplicas int) (int, error) {
 	var suggestedReplicas int
 
-	v, err := r.suggestDesiredReplicas(st, hra)
+	v, err := r.suggestDesiredReplicas(ghc, st, hra)
 	if err != nil {
 		return 0, err
 	}
@@ -497,7 +506,7 @@ func (r *HorizontalRunnerAutoscalerReconciler) computeReplicasWithCache(log logr
 	if hra.Spec.ScaleDownDelaySecondsAfterScaleUp != nil {
 		scaleDownDelay = time.Duration(*hra.Spec.ScaleDownDelaySecondsAfterScaleUp) * time.Second
 	} else {
-		scaleDownDelay = DefaultScaleDownDelay
+		scaleDownDelay = r.DefaultScaleDownDelay
 	}
 
 	var scaleDownDelayUntil *time.Time

controllers/integration_test.go

@@ -8,7 +8,7 @@ import (
 	"time"
 
 	github2 "github.com/actions-runner-controller/actions-runner-controller/github"
-	"github.com/google/go-github/v39/github"
+	"github.com/google/go-github/v47/github"
 
 	"github.com/actions-runner-controller/actions-runner-controller/github/fake"
 
@@ -99,12 +99,14 @@ func SetupIntegrationTest(ctx2 context.Context) *testEnvironment {
 			return fmt.Sprintf("%s%s", ns.Name, name)
 		}
 
+		multiClient := NewMultiGitHubClient(mgr.GetClient(), env.ghClient)
+
 		runnerController := &RunnerReconciler{
 			Client:                      mgr.GetClient(),
 			Scheme:                      scheme.Scheme,
 			Log:                         logf.Log,
 			Recorder:                    mgr.GetEventRecorderFor("runnerreplicaset-controller"),
-			GitHubClient:                env.ghClient,
+			GitHubClient:                multiClient,
 			RunnerImage:                 "example/runner:test",
 			DockerImage:                 "example/docker:test",
 			Name:                        controllerName("runner"),
@@ -116,12 +118,11 @@ func SetupIntegrationTest(ctx2 context.Context) *testEnvironment {
 		Expect(err).NotTo(HaveOccurred(), "failed to setup runner controller")
 
 		replicasetController := &RunnerReplicaSetReconciler{
-			Client:       mgr.GetClient(),
-			Scheme:       scheme.Scheme,
-			Log:          logf.Log,
-			Recorder:     mgr.GetEventRecorderFor("runnerreplicaset-controller"),
-			GitHubClient: env.ghClient,
-			Name:         controllerName("runnerreplicaset"),
+			Client:   mgr.GetClient(),
+			Scheme:   scheme.Scheme,
+			Log:      logf.Log,
+			Recorder: mgr.GetEventRecorderFor("runnerreplicaset-controller"),
+			Name:     controllerName("runnerreplicaset"),
 		}
 		err = replicasetController.SetupWithManager(mgr)
 		Expect(err).NotTo(HaveOccurred(), "failed to setup runnerreplicaset controller")
@@ -137,13 +138,12 @@ func SetupIntegrationTest(ctx2 context.Context) *testEnvironment {
 		Expect(err).NotTo(HaveOccurred(), "failed to setup runnerdeployment controller")
 
 		autoscalerController := &HorizontalRunnerAutoscalerReconciler{
-			Client:        mgr.GetClient(),
-			Scheme:        scheme.Scheme,
-			Log:           logf.Log,
-			GitHubClient:  env.ghClient,
-			Recorder:      mgr.GetEventRecorderFor("horizontalrunnerautoscaler-controller"),
-			CacheDuration: 1 * time.Second,
-			Name:          controllerName("horizontalrunnerautoscaler"),
+			Client:       mgr.GetClient(),
+			Scheme:       scheme.Scheme,
+			Log:          logf.Log,
+			GitHubClient: multiClient,
+			Recorder:     mgr.GetEventRecorderFor("horizontalrunnerautoscaler-controller"),
+			Name:         controllerName("horizontalrunnerautoscaler"),
 		}
 		err = autoscalerController.SetupWithManager(mgr)
 		Expect(err).NotTo(HaveOccurred(), "failed to setup autoscaler controller")
@@ -537,6 +537,106 @@ var _ = Context("INTEGRATION: Inside of a new namespace", func() {
 			}
 		})
 
+		It("should create and scale organization's repository runners on workflow_job event", func() {
+			name := "example-runnerdeploy"
+
+			{
+				rd := &actionsv1alpha1.RunnerDeployment{
+					ObjectMeta: metav1.ObjectMeta{
+						Name:      name,
+						Namespace: ns.Name,
+					},
+					Spec: actionsv1alpha1.RunnerDeploymentSpec{
+						Replicas: intPtr(1),
+						Selector: &metav1.LabelSelector{
+							MatchLabels: map[string]string{
+								"foo": "bar",
+							},
+						},
+						Template: actionsv1alpha1.RunnerTemplate{
+							ObjectMeta: metav1.ObjectMeta{
+								Labels: map[string]string{
+									"foo": "bar",
+								},
+							},
+							Spec: actionsv1alpha1.RunnerSpec{
+								RunnerConfig: actionsv1alpha1.RunnerConfig{
+									Repository: "test/valid",
+									Image:      "bar",
+									Group:      "baz",
+								},
+								RunnerPodSpec: actionsv1alpha1.RunnerPodSpec{
+									Env: []corev1.EnvVar{
+										{Name: "FOO", Value: "FOOVALUE"},
+									},
+								},
+							},
+						},
+					},
+				}
+
+				ExpectCreate(ctx, rd, "test RunnerDeployment")
+				ExpectRunnerSetsCountEventuallyEquals(ctx, ns.Name, 1)
+				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 1)
+				env.ExpectRegisteredNumberCountEventuallyEquals(1, "count of fake list runners")
+			}
+
+			// Scale-up to 1 replica via ScaleUpTriggers.GitHubEvent.WorkflowJob based scaling
+			{
+				hra := &actionsv1alpha1.HorizontalRunnerAutoscaler{
+					ObjectMeta: metav1.ObjectMeta{
+						Name:      name,
+						Namespace: ns.Name,
+					},
+					Spec: actionsv1alpha1.HorizontalRunnerAutoscalerSpec{
+						ScaleTargetRef: actionsv1alpha1.ScaleTargetRef{
+							Name: name,
+						},
+						MinReplicas:                       intPtr(1),
+						MaxReplicas:                       intPtr(5),
+						ScaleDownDelaySecondsAfterScaleUp: intPtr(1),
+						ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
+							{
+								GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+									WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+								},
+								Amount:   1,
+								Duration: metav1.Duration{Duration: time.Minute},
+							},
+						},
+					},
+				}
+
+				ExpectCreate(ctx, hra, "test HorizontalRunnerAutoscaler")
+
+				ExpectRunnerSetsCountEventuallyEquals(ctx, ns.Name, 1)
+				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 1)
+				env.ExpectRegisteredNumberCountEventuallyEquals(1, "count of fake list runners")
+			}
+
+			// Scale-up to 2 replicas on first workflow_job.queued webhook event
+			{
+				env.SendWorkflowJobEvent("test", "valid", "queued", []string{"self-hosted"})
+				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 2, "runners after first webhook event")
+				env.ExpectRegisteredNumberCountEventuallyEquals(2, "count of fake list runners")
+			}
+
+			// Scale-up to 3 replicas on second workflow_job.queued webhook event
+			{
+				env.SendWorkflowJobEvent("test", "valid", "queued", []string{"self-hosted"})
+				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 3, "runners after second webhook event")
+				env.ExpectRegisteredNumberCountEventuallyEquals(3, "count of fake list runners")
+			}
+
+			// Do not scale-up on third workflow_job.queued webhook event
+			// repo "example" doesn't match our Spec
+			{
+				env.SendWorkflowJobEvent("test", "example", "queued", []string{"self-hosted"})
+				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 3, "runners after third webhook event")
+				env.ExpectRegisteredNumberCountEventuallyEquals(3, "count of fake list runners")
+			}
+		})
+
 		It("should create and scale organization's repository runners only on check_run event", func() {
 			name := "example-runnerdeploy"
 
@@ -581,9 +681,7 @@ var _ = Context("INTEGRATION: Inside of a new namespace", func() {
 				env.ExpectRegisteredNumberCountEventuallyEquals(1, "count of fake list runners")
 			}
 
-			// Scale-up to 3 replicas by the default TotalNumberOfQueuedAndInProgressWorkflowRuns-based scaling
-			// See workflowRunsFor3Replicas_queued and workflowRunsFor3Replicas_in_progress for GitHub List-Runners API responses
-			// used while testing.
+			// Scale-up to 1 replica via ScaleUpTriggers.GitHubEvent.CheckRun based scaling
 			{
 				hra := &actionsv1alpha1.HorizontalRunnerAutoscaler{
 					ObjectMeta: metav1.ObjectMeta{
@@ -1130,9 +1228,11 @@ var _ = Context("INTEGRATION: Inside of a new namespace", func() {
 						ScaleDownDelaySecondsAfterScaleUp: intPtr(1),
 						ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
 							{
-								GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{},
-								Amount:      1,
-								Duration:    metav1.Duration{Duration: time.Minute},
+								GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+									WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+								},
+								Amount:   1,
+								Duration: metav1.Duration{Duration: time.Minute},
 							},
 						},
 					},
@@ -1150,7 +1250,7 @@ var _ = Context("INTEGRATION: Inside of a new namespace", func() {
 
 			// Scale-up to 2 replicas on first workflow_job webhook event
 			{
-				env.SendWorkflowJobEvent("test", "valid", "pending", "created", []string{"self-hosted"})
+				env.SendWorkflowJobEvent("test", "valid", "queued", []string{"self-hosted"})
 				ExpectRunnerSetsCountEventuallyEquals(ctx, ns.Name, 1, "runner sets after webhook")
 				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 2, "runners after first webhook event")
 				env.ExpectRegisteredNumberCountEventuallyEquals(2, "count of fake list runners")
@@ -1212,9 +1312,11 @@ var _ = Context("INTEGRATION: Inside of a new namespace", func() {
 						ScaleDownDelaySecondsAfterScaleUp: intPtr(1),
 						ScaleUpTriggers: []actionsv1alpha1.ScaleUpTrigger{
 							{
-								GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{},
-								Amount:      1,
-								Duration:    metav1.Duration{Duration: time.Minute},
+								GitHubEvent: &actionsv1alpha1.GitHubEventScaleUpTriggerSpec{
+									WorkflowJob: &actionsv1alpha1.WorkflowJobSpec{},
+								},
+								Amount:   1,
+								Duration: metav1.Duration{Duration: time.Minute},
 							},
 						},
 					},
@@ -1232,7 +1334,7 @@ var _ = Context("INTEGRATION: Inside of a new namespace", func() {
 
 			// Scale-up to 2 replicas on first workflow_job webhook event
 			{
-				env.SendWorkflowJobEvent("test", "valid", "pending", "created", []string{"custom-label"})
+				env.SendWorkflowJobEvent("test", "valid", "queued", []string{"custom-label"})
 				ExpectRunnerSetsCountEventuallyEquals(ctx, ns.Name, 1, "runner sets after webhook")
 				ExpectRunnerSetsManagedReplicasCountEventuallyEquals(ctx, ns.Name, 2, "runners after first webhook event")
 				env.ExpectRegisteredNumberCountEventuallyEquals(2, "count of fake list runners")
@@ -1265,7 +1367,7 @@ func (env *testEnvironment) ExpectRegisteredNumberCountEventuallyEquals(want int
 
 			return len(rs)
 		},
-		time.Second*5, time.Millisecond*500).Should(Equal(want), optionalDescriptions...)
+		time.Second*10, time.Millisecond*500).Should(Equal(want), optionalDescriptions...)
 }
 
 func (env *testEnvironment) SendOrgPullRequestEvent(org, repo, branch, action string) {
@@ -1313,6 +1415,30 @@ func (env *testEnvironment) SendOrgCheckRunEvent(org, repo, status, action strin
 	ExpectWithOffset(1, resp.StatusCode).To(Equal(200))
 }
 
+func (env *testEnvironment) SendWorkflowJobEvent(org, repo, statusAndAction string, labels []string) {
+	resp, err := sendWebhook(env.webhookServer, "workflow_job", &github.WorkflowJobEvent{
+		WorkflowJob: &github.WorkflowJob{
+			Status: &statusAndAction,
+			Labels: labels,
+		},
+		Org: &github.Organization{
+			Login: github.String(org),
+		},
+		Repo: &github.Repository{
+			Name: github.String(repo),
+			Owner: &github.User{
+				Login: github.String(org),
+				Type:  github.String("Organization"),
+			},
+		},
+		Action: github.String(statusAndAction),
+	})
+
+	ExpectWithOffset(1, err).NotTo(HaveOccurred(), "failed to send workflow_job event")
+
+	ExpectWithOffset(1, resp.StatusCode).To(Equal(200))
+}
+
 func (env *testEnvironment) SendUserPullRequestEvent(owner, repo, branch, action string) {
 	resp, err := sendWebhook(env.webhookServer, "pull_request", &github.PullRequestEvent{
 		PullRequest: &github.PullRequest{
@@ -1355,28 +1481,6 @@ func (env *testEnvironment) SendUserCheckRunEvent(owner, repo, status, action st
 	ExpectWithOffset(1, resp.StatusCode).To(Equal(200))
 }
 
-func (env *testEnvironment) SendWorkflowJobEvent(owner, repo, status, action string, labels []string) {
-	resp, err := sendWebhook(env.webhookServer, "workflow_job", &github.WorkflowJobEvent{
-		Org: &github.Organization{
-			Name: github.String(owner),
-		},
-		WorkflowJob: &github.WorkflowJob{
-			Labels: labels,
-		},
-		Action: github.String("queued"),
-		Repo: &github.Repository{
-			Name: github.String(repo),
-			Owner: &github.User{
-				Login: github.String(owner),
-				Type:  github.String("Organization"),
-			},
-		},
-	})
-
-	ExpectWithOffset(1, err).NotTo(HaveOccurred(), "failed to send check_run event")
-
-	ExpectWithOffset(1, resp.StatusCode).To(Equal(200))
-}
 func (env *testEnvironment) SyncRunnerRegistrations() {
 	var runnerList actionsv1alpha1.RunnerList
 

controllers/metrics/horizontalrunnerautoscaler.go

@@ -7,8 +7,13 @@ import (
 )
 
 const (
-	hraName      = "horizontalrunnerautoscaler"
-	hraNamespace = "namespace"
+	hraName        = "horizontalrunnerautoscaler"
+	hraNamespace   = "namespace"
+	stEnterprise   = "enterprise"
+	stOrganization = "organization"
+	stRepository   = "repository"
+	stKind         = "kind"
+	stName         = "name"
 )
 
 var (
@@ -16,6 +21,16 @@ var (
 		horizontalRunnerAutoscalerMinReplicas,
 		horizontalRunnerAutoscalerMaxReplicas,
 		horizontalRunnerAutoscalerDesiredReplicas,
+		horizontalRunnerAutoscalerReplicasDesired,
+		horizontalRunnerAutoscalerRunners,
+		horizontalRunnerAutoscalerRunnersRegistered,
+		horizontalRunnerAutoscalerRunnersBusy,
+		horizontalRunnerAutoscalerTerminatingBusy,
+		horizontalRunnerAutoscalerNecessaryReplicas,
+		horizontalRunnerAutoscalerWorkflowRunsCompleted,
+		horizontalRunnerAutoscalerWorkflowRunsInProgress,
+		horizontalRunnerAutoscalerWorkflowRunsQueued,
+		horizontalRunnerAutoscalerWorkflowRunsUnknown,
 	}
 )
 
@@ -41,6 +56,78 @@ var (
 		},
 		[]string{hraName, hraNamespace},
 	)
+	// PercentageRunnersBusy
+	horizontalRunnerAutoscalerReplicasDesired = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_replicas_desired",
+			Help: "replicas_desired of PercentageRunnersBusy",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerRunners = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_runners",
+			Help: "num_runners of PercentageRunnersBusy",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerRunnersRegistered = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_runners_registered",
+			Help: "num_runners_registered of PercentageRunnersBusy",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerRunnersBusy = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_runners_busy",
+			Help: "num_runners_busy of PercentageRunnersBusy",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerTerminatingBusy = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_terminating_busy",
+			Help: "num_terminating_busy of PercentageRunnersBusy",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	// QueuedAndInProgressWorkflowRuns
+	horizontalRunnerAutoscalerNecessaryReplicas = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_necessary_replicas",
+			Help: "necessary_replicas of QueuedAndInProgressWorkflowRuns",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerWorkflowRunsCompleted = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_workflow_runs_completed",
+			Help: "workflow_runs_completed of QueuedAndInProgressWorkflowRuns",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerWorkflowRunsInProgress = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_workflow_runs_in_progress",
+			Help: "workflow_runs_in_progress of QueuedAndInProgressWorkflowRuns",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerWorkflowRunsQueued = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_workflow_runs_queued",
+			Help: "workflow_runs_queued of QueuedAndInProgressWorkflowRuns",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
+	horizontalRunnerAutoscalerWorkflowRunsUnknown = prometheus.NewGaugeVec(
+		prometheus.GaugeOpts{
+			Name: "horizontalrunnerautoscaler_workflow_runs_unknown",
+			Help: "workflow_runs_unknown of QueuedAndInProgressWorkflowRuns",
+		},
+		[]string{hraName, hraNamespace, stEnterprise, stOrganization, stRepository, stKind, stName},
+	)
 )
 
 func SetHorizontalRunnerAutoscalerSpec(o metav1.ObjectMeta, spec v1alpha1.HorizontalRunnerAutoscalerSpec) {
@@ -65,3 +152,61 @@ func SetHorizontalRunnerAutoscalerStatus(o metav1.ObjectMeta, status v1alpha1.Ho
 		horizontalRunnerAutoscalerDesiredReplicas.With(labels).Set(float64(*status.DesiredReplicas))
 	}
 }
+
+func SetHorizontalRunnerAutoscalerPercentageRunnersBusy(
+	o metav1.ObjectMeta,
+	enterprise string,
+	organization string,
+	repository string,
+	kind string,
+	name string,
+	desiredReplicas int,
+	numRunners int,
+	numRunnersRegistered int,
+	numRunnersBusy int,
+	numTerminatingBusy int,
+) {
+	labels := prometheus.Labels{
+		hraName:        o.Name,
+		hraNamespace:   o.Namespace,
+		stEnterprise:   enterprise,
+		stOrganization: organization,
+		stRepository:   repository,
+		stKind:         kind,
+		stName:         name,
+	}
+	horizontalRunnerAutoscalerReplicasDesired.With(labels).Set(float64(desiredReplicas))
+	horizontalRunnerAutoscalerRunners.With(labels).Set(float64(numRunners))
+	horizontalRunnerAutoscalerRunnersRegistered.With(labels).Set(float64(numRunnersRegistered))
+	horizontalRunnerAutoscalerRunnersBusy.With(labels).Set(float64(numRunnersBusy))
+	horizontalRunnerAutoscalerTerminatingBusy.With(labels).Set(float64(numTerminatingBusy))
+}
+
+func SetHorizontalRunnerAutoscalerQueuedAndInProgressWorkflowRuns(
+	o metav1.ObjectMeta,
+	enterprise string,
+	organization string,
+	repository string,
+	kind string,
+	name string,
+	necessaryReplicas int,
+	workflowRunsCompleted int,
+	workflowRunsInProgress int,
+	workflowRunsQueued int,
+	workflowRunsUnknown int,
+) {
+	labels := prometheus.Labels{
+		hraName:        o.Name,
+		hraNamespace:   o.Namespace,
+		stEnterprise:   enterprise,
+		stOrganization: organization,
+		stRepository:   repository,
+		stKind:         kind,
+		stName:         name,
+	}
+	horizontalRunnerAutoscalerNecessaryReplicas.With(labels).Set(float64(necessaryReplicas))
+	horizontalRunnerAutoscalerWorkflowRunsCompleted.With(labels).Set(float64(workflowRunsCompleted))
+	horizontalRunnerAutoscalerWorkflowRunsInProgress.With(labels).Set(float64(workflowRunsInProgress))
+	horizontalRunnerAutoscalerWorkflowRunsQueued.With(labels).Set(float64(workflowRunsQueued))
+	horizontalRunnerAutoscalerWorkflowRunsUnknown.With(labels).Set(float64(workflowRunsUnknown))
+}

controllers/multi_githubclient.go

@@ -0,0 +1,358 @@
+package controllers
+
+import (
+	"context"
+	"crypto/sha1"
+	"encoding/hex"
+	"fmt"
+	"sort"
+	"strconv"
+	"sync"
+
+	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
+	"github.com/actions-runner-controller/actions-runner-controller/github"
+	corev1 "k8s.io/api/core/v1"
+	"k8s.io/apimachinery/pkg/types"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+)
+
+const (
+	// The api creds scret annotation is added by the runner controller or the runnerset controller according to runner.spec.githubAPICredentialsFrom.secretRef.name,
+	// so that the runner pod controller can share the same GitHub API credentials and the instance of the GitHub API client with the upstream controllers.
+	annotationKeyGitHubAPICredsSecret = annotationKeyPrefix + "github-api-creds-secret"
+)
+
+type runnerOwnerRef struct {
+	// kind is either StatefulSet or Runner, and populated via the owner reference in the runner pod controller or via the reconcilation target's kind in
+	// runnerset and runner controllers.
+	kind     string
+	ns, name string
+}
+
+type secretRef struct {
+	ns, name string
+}
+
+// savedClient is the each cache entry that contains the client for the specific set of credentials,
+// like a PAT or a pair of key and cert.
+// the `hash` is a part of the savedClient not the key because we are going to keep only the client for the latest creds
+// in case the operator updated the k8s secret containing the credentials.
+type savedClient struct {
+	hash string
+
+	// refs is the map of all the objects that references this client, used for reference counting to gc
+	// the client if unneeded.
+	refs map[runnerOwnerRef]struct{}
+
+	*github.Client
+}
+
+type resourceReader interface {
+	Get(context.Context, types.NamespacedName, client.Object) error
+}
+
+type MultiGitHubClient struct {
+	mu sync.Mutex
+
+	client resourceReader
+
+	githubClient *github.Client
+
+	// The saved client is freed once all its dependents disappear, or the contents of the secret changed.
+	// We track dependents via a golang map embedded within the savedClient struct. Each dependent is checked on their respective Kubernetes finalizer,
+	// so that we won't miss any dependent's termination.
+	// The change is the secret is determined using the hash of its contents.
+	clients map[secretRef]savedClient
+}
+
+func NewMultiGitHubClient(client resourceReader, githubClient *github.Client) *MultiGitHubClient {
+	return &MultiGitHubClient{
+		client:       client,
+		githubClient: githubClient,
+		clients:      map[secretRef]savedClient{},
+	}
+}
+
+// Init sets up and return the *github.Client for the object.
+// In case the object (like RunnerDeployment) does not request a custom client, it returns the default client.
+func (c *MultiGitHubClient) InitForRunnerPod(ctx context.Context, pod *corev1.Pod) (*github.Client, error) {
+	// These 3 default values are used only when the user created the pod directly, not via Runner, RunnerReplicaSet, RunnerDeploment, or RunnerSet resources.
+	ref := refFromRunnerPod(pod)
+	secretName := pod.Annotations[annotationKeyGitHubAPICredsSecret]
+
+	// kind can be any of Pod, Runner, RunnerReplicaSet, RunnerDeployment, or RunnerSet depending on which custom resource the user directly created.
+	return c.initClientWithSecretName(ctx, pod.Namespace, secretName, ref)
+}
+
+// Init sets up and return the *github.Client for the object.
+// In case the object (like RunnerDeployment) does not request a custom client, it returns the default client.
+func (c *MultiGitHubClient) InitForRunner(ctx context.Context, r *v1alpha1.Runner) (*github.Client, error) {
+	var secretName string
+	if r.Spec.GitHubAPICredentialsFrom != nil {
+		secretName = r.Spec.GitHubAPICredentialsFrom.SecretRef.Name
+	}
+
+	// These 3 default values are used only when the user created the runner resource directly, not via RunnerReplicaSet, RunnerDeploment, or RunnerSet resources.
+	ref := refFromRunner(r)
+	if ref.ns != r.Namespace {
+		return nil, fmt.Errorf("referencing github api creds secret from owner in another namespace is not supported yet")
+	}
+
+	// kind can be any of Runner, RunnerReplicaSet, or RunnerDeployment depending on which custom resource the user directly created.
+	return c.initClientWithSecretName(ctx, r.Namespace, secretName, ref)
+}
+
+// Init sets up and return the *github.Client for the object.
+// In case the object (like RunnerDeployment) does not request a custom client, it returns the default client.
+func (c *MultiGitHubClient) InitForRunnerSet(ctx context.Context, rs *v1alpha1.RunnerSet) (*github.Client, error) {
+	ref := refFromRunnerSet(rs)
+
+	var secretName string
+	if rs.Spec.GitHubAPICredentialsFrom != nil {
+		secretName = rs.Spec.GitHubAPICredentialsFrom.SecretRef.Name
+	}
+
+	return c.initClientWithSecretName(ctx, rs.Namespace, secretName, ref)
+}
+
+// Init sets up and return the *github.Client for the object.
+// In case the object (like RunnerDeployment) does not request a custom client, it returns the default client.
+func (c *MultiGitHubClient) InitForHRA(ctx context.Context, hra *v1alpha1.HorizontalRunnerAutoscaler) (*github.Client, error) {
+	ref := refFromHorizontalRunnerAutoscaler(hra)
+
+	var secretName string
+	if hra.Spec.GitHubAPICredentialsFrom != nil {
+		secretName = hra.Spec.GitHubAPICredentialsFrom.SecretRef.Name
+	}
+
+	return c.initClientWithSecretName(ctx, hra.Namespace, secretName, ref)
+}
+
+func (c *MultiGitHubClient) DeinitForRunnerPod(p *corev1.Pod) {
+	secretName := p.Annotations[annotationKeyGitHubAPICredsSecret]
+	c.derefClient(p.Namespace, secretName, refFromRunnerPod(p))
+}
+
+func (c *MultiGitHubClient) DeinitForRunner(r *v1alpha1.Runner) {
+	var secretName string
+	if r.Spec.GitHubAPICredentialsFrom != nil {
+		secretName = r.Spec.GitHubAPICredentialsFrom.SecretRef.Name
+	}
+
+	c.derefClient(r.Namespace, secretName, refFromRunner(r))
+}
+
+func (c *MultiGitHubClient) DeinitForRunnerSet(rs *v1alpha1.RunnerSet) {
+	var secretName string
+	if rs.Spec.GitHubAPICredentialsFrom != nil {
+		secretName = rs.Spec.GitHubAPICredentialsFrom.SecretRef.Name
+	}
+
+	c.derefClient(rs.Namespace, secretName, refFromRunnerSet(rs))
+}
+
+func (c *MultiGitHubClient) deinitClientForRunnerReplicaSet(rs *v1alpha1.RunnerReplicaSet) {
+	c.derefClient(rs.Namespace, rs.Spec.Template.Spec.GitHubAPICredentialsFrom.SecretRef.Name, refFromRunnerReplicaSet(rs))
+}
+
+func (c *MultiGitHubClient) deinitClientForRunnerDeployment(rd *v1alpha1.RunnerDeployment) {
+	c.derefClient(rd.Namespace, rd.Spec.Template.Spec.GitHubAPICredentialsFrom.SecretRef.Name, refFromRunnerDeployment(rd))
+}
+
+func (c *MultiGitHubClient) DeinitForHRA(hra *v1alpha1.HorizontalRunnerAutoscaler) {
+	var secretName string
+	if hra.Spec.GitHubAPICredentialsFrom != nil {
+		secretName = hra.Spec.GitHubAPICredentialsFrom.SecretRef.Name
+	}
+
+	c.derefClient(hra.Namespace, secretName, refFromHorizontalRunnerAutoscaler(hra))
+}
+
+func (c *MultiGitHubClient) initClientForSecret(secret *corev1.Secret, dependent *runnerOwnerRef) (*savedClient, error) {
+	secRef := secretRef{
+		ns:   secret.Namespace,
+		name: secret.Name,
+	}
+
+	cliRef := c.clients[secRef]
+
+	var ks []string
+
+	for k := range secret.Data {
+		ks = append(ks, k)
+	}
+
+	sort.SliceStable(ks, func(i, j int) bool { return ks[i] < ks[j] })
+
+	hash := sha1.New()
+	for _, k := range ks {
+		hash.Write(secret.Data[k])
+	}
+	hashStr := hex.EncodeToString(hash.Sum(nil))
+
+	if cliRef.hash != hashStr {
+		delete(c.clients, secRef)
+
+		conf, err := secretDataToGitHubClientConfig(secret.Data)
+		if err != nil {
+			return nil, err
+		}
+
+		// Fallback to the controller-wide setting if EnterpriseURL is not set and the original client is an enterprise client.
+		if conf.EnterpriseURL == "" && c.githubClient.IsEnterprise {
+			conf.EnterpriseURL = c.githubClient.GithubBaseURL
+		}
+
+		cli, err := conf.NewClient()
+		if err != nil {
+			return nil, err
+		}
+
+		cliRef = savedClient{
+			hash:   hashStr,
+			refs:   map[runnerOwnerRef]struct{}{},
+			Client: cli,
+		}
+
+		c.clients[secRef] = cliRef
+	}
+
+	if dependent != nil {
+		c.clients[secRef].refs[*dependent] = struct{}{}
+	}
+
+	return &cliRef, nil
+}
+
+func (c *MultiGitHubClient) initClientWithSecretName(ctx context.Context, ns, secretName string, runRef *runnerOwnerRef) (*github.Client, error) {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+
+	if secretName == "" {
+		return c.githubClient, nil
+	}
+
+	secRef := secretRef{
+		ns:   ns,
+		name: secretName,
+	}
+
+	if _, ok := c.clients[secRef]; !ok {
+		c.clients[secRef] = savedClient{}
+	}
+
+	var sec corev1.Secret
+	if err := c.client.Get(ctx, types.NamespacedName{Namespace: ns, Name: secretName}, &sec); err != nil {
+		return nil, err
+	}
+
+	savedClient, err := c.initClientForSecret(&sec, runRef)
+	if err != nil {
+		return nil, err
+	}
+
+	return savedClient.Client, nil
+}
+
+func (c *MultiGitHubClient) derefClient(ns, secretName string, dependent *runnerOwnerRef) {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+
+	secRef := secretRef{
+		ns:   ns,
+		name: secretName,
+	}
+
+	if dependent != nil {
+		delete(c.clients[secRef].refs, *dependent)
+	}
+
+	cliRef := c.clients[secRef]
+
+	if dependent == nil || len(cliRef.refs) == 0 {
+		delete(c.clients, secRef)
+	}
+}
+
+func secretDataToGitHubClientConfig(data map[string][]byte) (*github.Config, error) {
+	var (
+		conf github.Config
+
+		err error
+	)
+
+	conf.URL = string(data["github_url"])
+
+	conf.UploadURL = string(data["github_upload_url"])
+
+	conf.EnterpriseURL = string(data["github_enterprise_url"])
+
+	conf.RunnerGitHubURL = string(data["github_runner_url"])
+
+	conf.Token = string(data["github_token"])
+
+	appID := string(data["github_app_id"])
+
+	conf.AppID, err = strconv.ParseInt(appID, 10, 64)
+	if err != nil {
+		return nil, err
+	}
+
+	instID := string(data["github_app_installation_id"])
+
+	conf.AppInstallationID, err = strconv.ParseInt(instID, 10, 64)
+	if err != nil {
+		return nil, err
+	}
+
+	conf.AppPrivateKey = string(data["github_app_private_key"])
+
+	return &conf, nil
+}
+
+func refFromRunnerDeployment(rd *v1alpha1.RunnerDeployment) *runnerOwnerRef {
+	return &runnerOwnerRef{
+		kind: rd.Kind,
+		ns:   rd.Namespace,
+		name: rd.Name,
+	}
+}
+
+func refFromRunnerReplicaSet(rs *v1alpha1.RunnerReplicaSet) *runnerOwnerRef {
+	return &runnerOwnerRef{
+		kind: rs.Kind,
+		ns:   rs.Namespace,
+		name: rs.Name,
+	}
+}
+
+func refFromRunner(r *v1alpha1.Runner) *runnerOwnerRef {
+	return &runnerOwnerRef{
+		kind: r.Kind,
+		ns:   r.Namespace,
+		name: r.Name,
+	}
+}
+
+func refFromRunnerPod(po *corev1.Pod) *runnerOwnerRef {
+	return &runnerOwnerRef{
+		kind: po.Kind,
+		ns:   po.Namespace,
+		name: po.Name,
+	}
+}
+func refFromRunnerSet(rs *v1alpha1.RunnerSet) *runnerOwnerRef {
+	return &runnerOwnerRef{
+		kind: rs.Kind,
+		ns:   rs.Namespace,
+		name: rs.Name,
+	}
+}
+
+func refFromHorizontalRunnerAutoscaler(hra *v1alpha1.HorizontalRunnerAutoscaler) *runnerOwnerRef {
+	return &runnerOwnerRef{
+		kind: hra.Kind,
+		ns:   hra.Namespace,
+		name: hra.Name,
+	}
+}

controllers/persistent_volume_claim_controller.go

@@ -0,0 +1,74 @@
+/*
+Copyright 2022 The actions-runner-controller authors.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package controllers
+
+import (
+	"context"
+
+	"github.com/go-logr/logr"
+
+	"k8s.io/apimachinery/pkg/runtime"
+	"k8s.io/client-go/tools/record"
+	ctrl "sigs.k8s.io/controller-runtime"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+
+	corev1 "k8s.io/api/core/v1"
+)
+
+// RunnerPersistentVolumeClaimReconciler reconciles a PersistentVolume object
+type RunnerPersistentVolumeClaimReconciler struct {
+	client.Client
+	Log      logr.Logger
+	Recorder record.EventRecorder
+	Scheme   *runtime.Scheme
+	Name     string
+}
+
+// +kubebuilder:rbac:groups=core,resources=persistentvolumeclaims,verbs=get;list;watch;update;patch;delete
+// +kubebuilder:rbac:groups=core,resources=persistentvolumes,verbs=get;list;watch;update;patch;delete
+// +kubebuilder:rbac:groups=core,resources=events,verbs=create;patch
+
+func (r *RunnerPersistentVolumeClaimReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
+	log := r.Log.WithValues("pvc", req.NamespacedName)
+
+	var pvc corev1.PersistentVolumeClaim
+	if err := r.Get(ctx, req.NamespacedName, &pvc); err != nil {
+		return ctrl.Result{}, client.IgnoreNotFound(err)
+	}
+
+	res, err := syncPVC(ctx, r.Client, log, req.Namespace, &pvc)
+
+	if res == nil {
+		res = &ctrl.Result{}
+	}
+
+	return *res, err
+}
+
+func (r *RunnerPersistentVolumeClaimReconciler) SetupWithManager(mgr ctrl.Manager) error {
+	name := "runnerpersistentvolumeclaim-controller"
+	if r.Name != "" {
+		name = r.Name
+	}
+
+	r.Recorder = mgr.GetEventRecorderFor(name)
+
+	return ctrl.NewControllerManagedBy(mgr).
+		For(&corev1.PersistentVolumeClaim{}).
+		Named(name).
+		Complete(r)
+}

controllers/persistent_volume_controller.go

@@ -0,0 +1,72 @@
+/*
+Copyright 2022 The actions-runner-controller authors.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License.
+*/
+
+package controllers
+
+import (
+	"context"
+
+	"github.com/go-logr/logr"
+
+	"k8s.io/apimachinery/pkg/runtime"
+	"k8s.io/client-go/tools/record"
+	ctrl "sigs.k8s.io/controller-runtime"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+
+	corev1 "k8s.io/api/core/v1"
+)
+
+// RunnerPersistentVolumeReconciler reconciles a PersistentVolume object
+type RunnerPersistentVolumeReconciler struct {
+	client.Client
+	Log      logr.Logger
+	Recorder record.EventRecorder
+	Scheme   *runtime.Scheme
+	Name     string
+}
+
+// +kubebuilder:rbac:groups=core,resources=persistentvolumes,verbs=get;list;watch;update;patch;delete
+// +kubebuilder:rbac:groups=core,resources=events,verbs=create;patch
+
+func (r *RunnerPersistentVolumeReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
+	log := r.Log.WithValues("pv", req.NamespacedName)
+
+	var pv corev1.PersistentVolume
+	if err := r.Get(ctx, req.NamespacedName, &pv); err != nil {
+		return ctrl.Result{}, client.IgnoreNotFound(err)
+	}
+
+	res, err := syncPV(ctx, r.Client, log, req.Namespace, &pv)
+	if res == nil {
+		res = &ctrl.Result{}
+	}
+
+	return *res, err
+}
+
+func (r *RunnerPersistentVolumeReconciler) SetupWithManager(mgr ctrl.Manager) error {
+	name := "runnerpersistentvolume-controller"
+	if r.Name != "" {
+		name = r.Name
+	}
+
+	r.Recorder = mgr.GetEventRecorderFor(name)
+
+	return ctrl.NewControllerManagedBy(mgr).
+		For(&corev1.PersistentVolume{}).
+		Named(name).
+		Complete(r)
+}

controllers/pod_runner_token_injector.go

@@ -6,7 +6,6 @@ import (
 	"net/http"
 	"time"
 
-	"github.com/actions-runner-controller/actions-runner-controller/github"
 	"github.com/go-logr/logr"
 	"gomodules.xyz/jsonpatch/v2"
 	admissionv1 "k8s.io/api/admission/v1"
@@ -29,7 +28,7 @@ type PodRunnerTokenInjector struct {
 	Name         string
 	Log          logr.Logger
 	Recorder     record.EventRecorder
-	GitHubClient *github.Client
+	GitHubClient *MultiGitHubClient
 	decoder      *admission.Decoder
 }
 
@@ -66,7 +65,12 @@ func (t *PodRunnerTokenInjector) Handle(ctx context.Context, req admission.Reque
 		return newEmptyResponse()
 	}
 
-	rt, err := t.GitHubClient.GetRegistrationToken(context.Background(), enterprise, org, repo, pod.Name)
+	ghc, err := t.GitHubClient.InitForRunnerPod(ctx, &pod)
+	if err != nil {
+		return admission.Errored(http.StatusInternalServerError, err)
+	}
+
+	rt, err := ghc.GetRegistrationToken(context.Background(), enterprise, org, repo, pod.Name)
 	if err != nil {
 		t.Log.Error(err, "Failed to get new registration token")
 		return admission.Errored(http.StatusInternalServerError, err)
@@ -78,9 +82,7 @@ func (t *PodRunnerTokenInjector) Handle(ctx context.Context, req admission.Reque
 
 	updated.Annotations[AnnotationKeyTokenExpirationDate] = ts
 
-	if pod.Spec.RestartPolicy != corev1.RestartPolicyOnFailure {
-		updated.Spec.RestartPolicy = corev1.RestartPolicyOnFailure
-	}
+	forceRunnerPodRestartPolicyNever(updated)
 
 	buf, err := json.Marshal(updated)
 	if err != nil {

controllers/runner_controller.go

@@ -18,7 +18,10 @@ package controllers
 
 import (
 	"context"
+	"errors"
 	"fmt"
+	"reflect"
+	"strconv"
 	"strings"
 	"time"
 
@@ -33,10 +36,10 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/reconcile"
 
 	corev1 "k8s.io/api/core/v1"
+	rbacv1 "k8s.io/api/rbac/v1"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
-	"github.com/actions-runner-controller/actions-runner-controller/github"
 )
 
 const (
@@ -47,15 +50,13 @@ const (
 
 	retryDelayOnGitHubAPIRateLimitError = 30 * time.Second
 
-	// This is an annotation internal to actions-runner-controller and can change in backward-incompatible ways
-	annotationKeyRegistrationOnly = "actions-runner-controller/registration-only"
-
-	EnvVarOrg                        = "RUNNER_ORG"
-	EnvVarRepo                       = "RUNNER_REPO"
-	EnvVarEnterprise                 = "RUNNER_ENTERPRISE"
-	EnvVarEphemeral                  = "RUNNER_EPHEMERAL"
-	EnvVarRunnerFeatureFlagEphemeral = "RUNNER_FEATURE_FLAG_EPHEMERAL"
-	EnvVarTrue                       = "true"
+	EnvVarOrg        = "RUNNER_ORG"
+	EnvVarRepo       = "RUNNER_REPO"
+	EnvVarGroup      = "RUNNER_GROUP"
+	EnvVarLabels     = "RUNNER_LABELS"
+	EnvVarEnterprise = "RUNNER_ENTERPRISE"
+	EnvVarEphemeral  = "RUNNER_EPHEMERAL"
+	EnvVarTrue       = "true"
 )
 
 // RunnerReconciler reconciles a Runner object
@@ -64,7 +65,7 @@ type RunnerReconciler struct {
 	Log                         logr.Logger
 	Recorder                    record.EventRecorder
 	Scheme                      *runtime.Scheme
-	GitHubClient                *github.Client
+	GitHubClient                *MultiGitHubClient
 	RunnerImage                 string
 	RunnerImagePullSecrets      []string
 	DockerImage                 string
@@ -72,16 +73,20 @@ type RunnerReconciler struct {
 	Name                        string
 	RegistrationRecheckInterval time.Duration
 	RegistrationRecheckJitter   time.Duration
-
-	UnregistrationRetryDelay time.Duration
+	UseRunnerStatusUpdateHook   bool
+	UnregistrationRetryDelay    time.Duration
 }
 
 // +kubebuilder:rbac:groups=actions.summerwind.dev,resources=runners,verbs=get;list;watch;create;update;patch;delete
 // +kubebuilder:rbac:groups=actions.summerwind.dev,resources=runners/finalizers,verbs=get;list;watch;create;update;patch;delete
 // +kubebuilder:rbac:groups=actions.summerwind.dev,resources=runners/status,verbs=get;update;patch
 // +kubebuilder:rbac:groups=core,resources=pods,verbs=get;list;watch;create;update;patch;delete
+// +kubebuilder:rbac:groups=core,resources=secrets,verbs=get;list;watch;delete
 // +kubebuilder:rbac:groups=core,resources=pods/finalizers,verbs=get;list;watch;create;update;patch;delete
 // +kubebuilder:rbac:groups=core,resources=events,verbs=create;patch
+// +kubebuilder:rbac:groups=core,resources=serviceaccounts,verbs=create;delete;get
+// +kubebuilder:rbac:groups=rbac.authorization.k8s.io,resources=roles,verbs=create;delete;get
+// +kubebuilder:rbac:groups=rbac.authorization.k8s.io,resources=rolebindings,verbs=create;delete;get
 
 func (r *RunnerReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
 	log := r.Log.WithValues("runner", req.NamespacedName)
@@ -106,15 +111,19 @@ func (r *RunnerReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctr
 			return ctrl.Result{}, nil
 		}
 	} else {
+		// Request to remove a runner. DeletionTimestamp was set in the runner - we need to unregister runner
 		var pod corev1.Pod
 		if err := r.Get(ctx, req.NamespacedName, &pod); err != nil {
 			if !kerrors.IsNotFound(err) {
 				log.Info(fmt.Sprintf("Retrying soon as we failed to get runner pod: %v", err))
 				return ctrl.Result{Requeue: true}, nil
 			}
+			// Pod was not found
+			return r.processRunnerDeletion(runner, ctx, log, nil)
 		}
 
-		// Request to remove a runner. DeletionTimestamp was set in the runner - we need to unregister runner
+		r.GitHubClient.DeinitForRunner(&runner)
+
 		return r.processRunnerDeletion(runner, ctx, log, &pod)
 	}
 
@@ -134,7 +143,7 @@ func (r *RunnerReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctr
 
 	ready := runnerPodReady(&pod)
 
-	if runner.Status.Phase != phase || runner.Status.Ready != ready {
+	if (runner.Status.Phase != phase || runner.Status.Ready != ready) && !r.UseRunnerStatusUpdateHook || runner.Status.Phase == "" && r.UseRunnerStatusUpdateHook {
 		if pod.Status.Phase == corev1.PodRunning {
 			// Seeing this message, you can expect the runner to become `Running` soon.
 			log.V(1).Info(
@@ -187,7 +196,7 @@ func runnerContainerExitCode(pod *corev1.Pod) *int32 {
 func runnerPodOrContainerIsStopped(pod *corev1.Pod) bool {
 	// If pod has ended up succeeded we need to restart it
 	// Happens e.g. when dind is in runner and run completes
-	stopped := pod.Status.Phase == corev1.PodSucceeded
+	stopped := pod.Status.Phase == corev1.PodSucceeded || pod.Status.Phase == corev1.PodFailed
 
 	if !stopped {
 		if pod.Status.Phase == corev1.PodRunning {
@@ -196,7 +205,7 @@ func runnerPodOrContainerIsStopped(pod *corev1.Pod) bool {
 					continue
 				}
 
-				if status.State.Terminated != nil && status.State.Terminated.ExitCode == 0 {
+				if status.State.Terminated != nil {
 					stopped = true
 				}
 			}
@@ -206,6 +215,24 @@ func runnerPodOrContainerIsStopped(pod *corev1.Pod) bool {
 	return stopped
 }
 
+func ephemeralRunnerContainerStatus(pod *corev1.Pod) *corev1.ContainerStatus {
+	if getRunnerEnv(pod, "RUNNER_EPHEMERAL") != "true" {
+		return nil
+	}
+
+	for _, status := range pod.Status.ContainerStatuses {
+		if status.Name != containerName {
+			continue
+		}
+
+		status := status
+
+		return &status
+	}
+
+	return nil
+}
+
 func (r *RunnerReconciler) processRunnerDeletion(runner v1alpha1.Runner, ctx context.Context, log logr.Logger, pod *corev1.Pod) (reconcile.Result, error) {
 	finalizers, removed := removeFinalizer(runner.ObjectMeta.Finalizers, finalizerName)
 
@@ -237,6 +264,96 @@ func (r *RunnerReconciler) processRunnerCreation(ctx context.Context, runner v1a
 		return ctrl.Result{}, err
 	}
 
+	needsServiceAccount := runner.Spec.ServiceAccountName == "" && (r.UseRunnerStatusUpdateHook || runner.Spec.ContainerMode == "kubernetes")
+	if needsServiceAccount {
+		serviceAccount := &corev1.ServiceAccount{
+			ObjectMeta: metav1.ObjectMeta{
+				Name:      runner.ObjectMeta.Name,
+				Namespace: runner.ObjectMeta.Namespace,
+			},
+		}
+		if res := r.createObject(ctx, serviceAccount, serviceAccount.ObjectMeta, &runner, log); res != nil {
+			return *res, nil
+		}
+
+		rules := []rbacv1.PolicyRule{}
+
+		if r.UseRunnerStatusUpdateHook {
+			rules = append(rules, []rbacv1.PolicyRule{
+				{
+					APIGroups:     []string{"actions.summerwind.dev"},
+					Resources:     []string{"runners/status"},
+					Verbs:         []string{"get", "update", "patch"},
+					ResourceNames: []string{runner.ObjectMeta.Name},
+				},
+			}...)
+		}
+
+		if runner.Spec.ContainerMode == "kubernetes" {
+			// Permissions based on https://github.com/actions/runner-container-hooks/blob/main/packages/k8s/README.md
+			rules = append(rules, []rbacv1.PolicyRule{
+				{
+					APIGroups: []string{""},
+					Resources: []string{"pods"},
+					Verbs:     []string{"get", "list", "create", "delete"},
+				},
+				{
+					APIGroups: []string{""},
+					Resources: []string{"pods/exec"},
+					Verbs:     []string{"get", "create"},
+				},
+				{
+					APIGroups: []string{""},
+					Resources: []string{"pods/log"},
+					Verbs:     []string{"get", "list", "watch"},
+				},
+				{
+					APIGroups: []string{"batch"},
+					Resources: []string{"jobs"},
+					Verbs:     []string{"get", "list", "create", "delete"},
+				},
+				{
+					APIGroups: []string{""},
+					Resources: []string{"secrets"},
+					Verbs:     []string{"get", "list", "create", "delete"},
+				},
+			}...)
+		}
+
+		role := &rbacv1.Role{
+			ObjectMeta: metav1.ObjectMeta{
+				Name:      runner.ObjectMeta.Name,
+				Namespace: runner.ObjectMeta.Namespace,
+			},
+			Rules: rules,
+		}
+		if res := r.createObject(ctx, role, role.ObjectMeta, &runner, log); res != nil {
+			return *res, nil
+		}
+
+		roleBinding := &rbacv1.RoleBinding{
+			ObjectMeta: metav1.ObjectMeta{
+				Name:      runner.ObjectMeta.Name,
+				Namespace: runner.ObjectMeta.Namespace,
+			},
+			RoleRef: rbacv1.RoleRef{
+				APIGroup: "rbac.authorization.k8s.io",
+				Kind:     "Role",
+				Name:     runner.ObjectMeta.Name,
+			},
+			Subjects: []rbacv1.Subject{
+				{
+					Kind:      "ServiceAccount",
+					Name:      runner.ObjectMeta.Name,
+					Namespace: runner.ObjectMeta.Namespace,
+				},
+			},
+		}
+		if res := r.createObject(ctx, roleBinding, roleBinding.ObjectMeta, &runner, log); res != nil {
+			return *res, nil
+		}
+	}
+
 	if err := r.Create(ctx, &newPod); err != nil {
 		if kerrors.IsAlreadyExists(err) {
 			// Gracefully handle pod-already-exists errors due to informer cache delay.
@@ -259,6 +376,27 @@ func (r *RunnerReconciler) processRunnerCreation(ctx context.Context, runner v1a
 	return ctrl.Result{}, nil
 }
 
+func (r *RunnerReconciler) createObject(ctx context.Context, obj client.Object, meta metav1.ObjectMeta, runner *v1alpha1.Runner, log logr.Logger) *ctrl.Result {
+	kind := strings.Split(reflect.TypeOf(obj).String(), ".")[1]
+	if err := ctrl.SetControllerReference(runner, obj, r.Scheme); err != nil {
+		log.Error(err, fmt.Sprintf("Could not add owner reference to %s %s. %s", kind, meta.Name, err.Error()))
+		return &ctrl.Result{Requeue: true}
+	}
+	if err := r.Create(ctx, obj); err != nil {
+		if kerrors.IsAlreadyExists(err) {
+			log.Info(fmt.Sprintf("Failed to create %s %s as it already exists. Reusing existing %s", kind, meta.Name, kind))
+			r.Recorder.Event(runner, corev1.EventTypeNormal, fmt.Sprintf("%sReused", kind), fmt.Sprintf("Reused %s '%s'", kind, meta.Name))
+			return nil
+		}
+
+		log.Error(err, fmt.Sprintf("Retrying as failed to create %s %s resource", kind, meta.Name))
+		return &ctrl.Result{Requeue: true}
+	}
+	r.Recorder.Event(runner, corev1.EventTypeNormal, fmt.Sprintf("%sCreated", kind), fmt.Sprintf("Created %s '%s'", kind, meta.Name))
+	log.Info(fmt.Sprintf("Created %s", kind), "name", meta.Name)
+	return nil
+}
+
 func (r *RunnerReconciler) updateRegistrationToken(ctx context.Context, runner v1alpha1.Runner) (bool, error) {
 	if runner.IsRegisterable() {
 		return false, nil
@@ -266,7 +404,12 @@ func (r *RunnerReconciler) updateRegistrationToken(ctx context.Context, runner v
 
 	log := r.Log.WithValues("runner", runner.Name)
 
-	rt, err := r.GitHubClient.GetRegistrationToken(ctx, runner.Spec.Enterprise, runner.Spec.Organization, runner.Spec.Repository, runner.Name)
+	ghc, err := r.GitHubClient.InitForRunner(ctx, &runner)
+	if err != nil {
+		return false, err
+	}
+
+	rt, err := ghc.GetRegistrationToken(ctx, runner.Spec.Enterprise, runner.Spec.Organization, runner.Spec.Repository, runner.Name)
 	if err != nil {
 		// An error can be a permanent, permission issue like the below:
 		//    POST https://api.github.com/enterprises/YOUR_ENTERPRISE/actions/runners/registration-token: 403 Resource not accessible by integration []
@@ -306,6 +449,11 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 		labels[k] = v
 	}
 
+	ghc, err := r.GitHubClient.InitForRunner(context.Background(), &runner)
+	if err != nil {
+		return corev1.Pod{}, err
+	}
+
 	// This implies that...
 	//
 	// (1) We recreate the runner pod whenever the runner has changes in:
@@ -329,7 +477,7 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 		filterLabels(runner.ObjectMeta.Labels, LabelKeyRunnerTemplateHash),
 		runner.ObjectMeta.Annotations,
 		runner.Spec,
-		r.GitHubClient.GithubBaseURL,
+		ghc.GithubBaseURL,
 		// Token change should trigger replacement.
 		// We need to include this explicitly here because
 		// runner.Spec does not contain the possibly updated token stored in the
@@ -348,31 +496,66 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 
 	if len(runner.Spec.Containers) == 0 {
 		template.Spec.Containers = append(template.Spec.Containers, corev1.Container{
-			Name:            "runner",
-			ImagePullPolicy: runner.Spec.ImagePullPolicy,
-			EnvFrom:         runner.Spec.EnvFrom,
-			Env:             runner.Spec.Env,
-			Resources:       runner.Spec.Resources,
+			Name: "runner",
 		})
 
 		if (runner.Spec.DockerEnabled == nil || *runner.Spec.DockerEnabled) && (runner.Spec.DockerdWithinRunnerContainer == nil || !*runner.Spec.DockerdWithinRunnerContainer) {
 			template.Spec.Containers = append(template.Spec.Containers, corev1.Container{
-				Name:         "docker",
-				VolumeMounts: runner.Spec.DockerVolumeMounts,
-				Resources:    runner.Spec.DockerdContainerResources,
-				Env:          runner.Spec.DockerEnv,
+				Name: "docker",
 			})
 		}
 	} else {
 		template.Spec.Containers = runner.Spec.Containers
 	}
 
+	for i, c := range template.Spec.Containers {
+		switch c.Name {
+		case "runner":
+			if c.ImagePullPolicy == "" {
+				template.Spec.Containers[i].ImagePullPolicy = runner.Spec.ImagePullPolicy
+			}
+			if len(c.EnvFrom) == 0 {
+				template.Spec.Containers[i].EnvFrom = runner.Spec.EnvFrom
+			}
+			if len(c.Env) == 0 {
+				template.Spec.Containers[i].Env = runner.Spec.Env
+			}
+			if len(c.Resources.Requests) == 0 {
+				template.Spec.Containers[i].Resources.Requests = runner.Spec.Resources.Requests
+			}
+			if len(c.Resources.Limits) == 0 {
+				template.Spec.Containers[i].Resources.Limits = runner.Spec.Resources.Limits
+			}
+		case "docker":
+			if len(c.VolumeMounts) == 0 {
+				template.Spec.Containers[i].VolumeMounts = runner.Spec.DockerVolumeMounts
+			}
+			if len(c.Resources.Limits) == 0 {
+				template.Spec.Containers[i].Resources.Limits = runner.Spec.DockerdContainerResources.Limits
+			}
+			if len(c.Resources.Requests) == 0 {
+				template.Spec.Containers[i].Resources.Requests = runner.Spec.DockerdContainerResources.Requests
+			}
+			if len(c.Env) == 0 {
+				template.Spec.Containers[i].Env = runner.Spec.DockerEnv
+			}
+		}
+	}
+
 	template.Spec.SecurityContext = runner.Spec.SecurityContext
 	template.Spec.EnableServiceLinks = runner.Spec.EnableServiceLinks
 
-	registrationOnly := metav1.HasAnnotation(runner.ObjectMeta, annotationKeyRegistrationOnly)
+	if runner.Spec.ContainerMode == "kubernetes" {
+		workDir := runner.Spec.WorkDir
+		if workDir == "" {
+			workDir = "/runner/_work"
+		}
+		if err := applyWorkVolumeClaimTemplateToPod(&template, runner.Spec.WorkVolumeClaimTemplate, workDir); err != nil {
+			return corev1.Pod{}, err
+		}
+	}
 
-	pod, err := newRunnerPod(runner.Name, template, runner.Spec.RunnerConfig, r.RunnerImage, r.RunnerImagePullSecrets, r.DockerImage, r.DockerRegistryMirror, r.GitHubClient.GithubBaseURL, registrationOnly)
+	pod, err := newRunnerPodWithContainerMode(runner.Spec.ContainerMode, template, runner.Spec.RunnerConfig, r.RunnerImage, r.RunnerImagePullSecrets, r.DockerImage, r.DockerRegistryMirror, ghc.GithubBaseURL, r.UseRunnerStatusUpdateHook)
 	if err != nil {
 		return pod, err
 	}
@@ -384,6 +567,9 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 		// if operater provides a work volume mount, use that
 		isPresent, _ := workVolumeMountPresent(runnerSpec.VolumeMounts)
 		if isPresent {
+			if runnerSpec.ContainerMode == "kubernetes" {
+				return pod, errors.New("volume mount \"work\" should be specified by workVolumeClaimTemplate in container mode kubernetes")
+			}
 			// remove work volume since it will be provided from runnerSpec.Volumes
 			// if we don't remove it here we would get a duplicate key error, i.e. two volumes named work
 			_, index := workVolumeMountPresent(pod.Spec.Containers[0].VolumeMounts)
@@ -397,6 +583,9 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 		// if operator provides a work volume. use that
 		isPresent, _ := workVolumePresent(runnerSpec.Volumes)
 		if isPresent {
+			if runnerSpec.ContainerMode == "kubernetes" {
+				return pod, errors.New("volume \"work\" should be specified by workVolumeClaimTemplate in container mode kubernetes")
+			}
 			_, index := workVolumePresent(pod.Spec.Volumes)
 
 			// remove work volume since it will be provided from runnerSpec.Volumes
@@ -406,6 +595,7 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 
 		pod.Spec.Volumes = append(pod.Spec.Volumes, runnerSpec.Volumes...)
 	}
+
 	if len(runnerSpec.InitContainers) != 0 {
 		pod.Spec.InitContainers = append(pod.Spec.InitContainers, runnerSpec.InitContainers...)
 	}
@@ -413,9 +603,13 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 	if runnerSpec.NodeSelector != nil {
 		pod.Spec.NodeSelector = runnerSpec.NodeSelector
 	}
+
 	if runnerSpec.ServiceAccountName != "" {
 		pod.Spec.ServiceAccountName = runnerSpec.ServiceAccountName
+	} else if r.UseRunnerStatusUpdateHook || runner.Spec.ContainerMode == "kubernetes" {
+		pod.Spec.ServiceAccountName = runner.ObjectMeta.Name
 	}
+
 	if runnerSpec.AutomountServiceAccountToken != nil {
 		pod.Spec.AutomountServiceAccountToken = runnerSpec.AutomountServiceAccountToken
 	}
@@ -436,6 +630,10 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 		pod.Spec.Tolerations = runnerSpec.Tolerations
 	}
 
+	if runnerSpec.PriorityClassName != "" {
+		pod.Spec.PriorityClassName = runnerSpec.PriorityClassName
+	}
+
 	if len(runnerSpec.TopologySpreadConstraints) != 0 {
 		pod.Spec.TopologySpreadConstraints = runnerSpec.TopologySpreadConstraints
 	}
@@ -452,6 +650,10 @@ func (r *RunnerReconciler) newPod(runner v1alpha1.Runner) (corev1.Pod, error) {
 		pod.Spec.HostAliases = runnerSpec.HostAliases
 	}
 
+	if runnerSpec.DnsConfig != nil {
+		pod.Spec.DNSConfig = runnerSpec.DnsConfig
+	}
+
 	if runnerSpec.RuntimeClassName != nil {
 		pod.Spec.RuntimeClassName = runnerSpec.RuntimeClassName
 	}
@@ -482,7 +684,45 @@ func mutatePod(pod *corev1.Pod, token string) *corev1.Pod {
 	return updated
 }
 
-func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.RunnerConfig, defaultRunnerImage string, defaultRunnerImagePullSecrets []string, defaultDockerImage, defaultDockerRegistryMirror string, githubBaseURL string, registrationOnly bool) (corev1.Pod, error) {
+func runnerHookEnvs(pod *corev1.Pod) ([]corev1.EnvVar, error) {
+	isRequireSameNode, err := isRequireSameNode(pod)
+	if err != nil {
+		return nil, err
+	}
+
+	return []corev1.EnvVar{
+		{
+			Name:  "ACTIONS_RUNNER_CONTAINER_HOOKS",
+			Value: defaultRunnerHookPath,
+		},
+		{
+			Name:  "ACTIONS_RUNNER_REQUIRE_JOB_CONTAINER",
+			Value: "true",
+		},
+		{
+			Name: "ACTIONS_RUNNER_POD_NAME",
+			ValueFrom: &corev1.EnvVarSource{
+				FieldRef: &corev1.ObjectFieldSelector{
+					FieldPath: "metadata.name",
+				},
+			},
+		},
+		{
+			Name: "ACTIONS_RUNNER_JOB_NAMESPACE",
+			ValueFrom: &corev1.EnvVarSource{
+				FieldRef: &corev1.ObjectFieldSelector{
+					FieldPath: "metadata.namespace",
+				},
+			},
+		},
+		corev1.EnvVar{
+			Name:  "ACTIONS_RUNNER_REQUIRE_SAME_NODE",
+			Value: strconv.FormatBool(isRequireSameNode),
+		},
+	}, nil
+}
+
+func newRunnerPodWithContainerMode(containerMode string, template corev1.Pod, runnerSpec v1alpha1.RunnerConfig, defaultRunnerImage string, defaultRunnerImagePullSecrets []string, defaultDockerImage, defaultDockerRegistryMirror string, githubBaseURL string, useRunnerStatusUpdateHook bool) (corev1.Pod, error) {
 	var (
 		privileged                bool = true
 		dockerdInRunner           bool = runnerSpec.DockerdWithinRunnerContainer != nil && *runnerSpec.DockerdWithinRunnerContainer
@@ -491,11 +731,20 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 		dockerdInRunnerPrivileged bool = dockerdInRunner
 	)
 
+	if containerMode == "kubernetes" {
+		dockerdInRunner = false
+		dockerEnabled = false
+		dockerdInRunnerPrivileged = false
+	}
+
 	template = *template.DeepCopy()
 
 	// This label selector is used by default when rd.Spec.Selector is empty.
-	template.ObjectMeta.Labels = CloneAndAddLabel(template.ObjectMeta.Labels, LabelKeyRunnerSetName, runnerName)
+	template.ObjectMeta.Labels = CloneAndAddLabel(template.ObjectMeta.Labels, LabelKeyRunner, "")
 	template.ObjectMeta.Labels = CloneAndAddLabel(template.ObjectMeta.Labels, LabelKeyPodMutation, LabelValuePodMutation)
+	if runnerSpec.GitHubAPICredentialsFrom != nil {
+		template.ObjectMeta.Annotations = CloneAndAddLabel(template.ObjectMeta.Annotations, annotationKeyGitHubAPICredsSecret, runnerSpec.GitHubAPICredentialsFrom.SecretRef.Name)
+	}
 
 	workDir := runnerSpec.WorkDir
 	if workDir == "" {
@@ -525,11 +774,11 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 			Value: runnerSpec.Enterprise,
 		},
 		{
-			Name:  "RUNNER_LABELS",
+			Name:  EnvVarLabels,
 			Value: strings.Join(runnerSpec.Labels, ","),
 		},
 		{
-			Name:  "RUNNER_GROUP",
+			Name:  EnvVarGroup,
 			Value: runnerSpec.Group,
 		},
 		{
@@ -552,14 +801,10 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 			Name:  EnvVarEphemeral,
 			Value: fmt.Sprintf("%v", ephemeral),
 		},
-	}
-
-	if registrationOnly {
-		env = append(env, corev1.EnvVar{
-			Name:  "RUNNER_REGISTRATION_ONLY",
-			Value: "true",
+		{
+			Name:  "RUNNER_STATUS_UPDATE_HOOK",
+			Value: fmt.Sprintf("%v", useRunnerStatusUpdateHook),
 		},
-		)
 	}
 
 	var seLinuxOptions *corev1.SELinuxOptions
@@ -585,6 +830,17 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 		}
 	}
 
+	if containerMode == "kubernetes" {
+		if dockerdContainer != nil {
+			template.Spec.Containers = append(template.Spec.Containers[:dockerdContainerIndex], template.Spec.Containers[dockerdContainerIndex+1:]...)
+		}
+		if runnerContainerIndex < runnerContainerIndex {
+			runnerContainerIndex--
+		}
+		dockerdContainer = nil
+		dockerdContainerIndex = -1
+	}
+
 	if runnerContainer == nil {
 		runnerContainerIndex = -1
 		runnerContainer = &corev1.Container{
@@ -615,18 +871,26 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 	}
 
 	runnerContainer.Env = append(runnerContainer.Env, env...)
+	if containerMode == "kubernetes" {
+		hookEnvs, err := runnerHookEnvs(&template)
+		if err != nil {
+			return corev1.Pod{}, err
+		}
+		runnerContainer.Env = append(runnerContainer.Env, hookEnvs...)
+	}
 
 	if runnerContainer.SecurityContext == nil {
 		runnerContainer.SecurityContext = &corev1.SecurityContext{}
 	}
-	// Runner need to run privileged if it contains DinD
-	runnerContainer.SecurityContext.Privileged = &dockerdInRunnerPrivileged
+
+	if runnerContainer.SecurityContext.Privileged == nil {
+		// Runner need to run privileged if it contains DinD
+		runnerContainer.SecurityContext.Privileged = &dockerdInRunnerPrivileged
+	}
 
 	pod := template.DeepCopy()
 
-	if pod.Spec.RestartPolicy == "" {
-		pod.Spec.RestartPolicy = "OnFailure"
-	}
+	forceRunnerPodRestartPolicyNever(pod)
 
 	if mtu := runnerSpec.DockerMTU; mtu != nil && dockerdInRunner {
 		runnerContainer.Env = append(runnerContainer.Env, []corev1.EnvVar{
@@ -704,13 +968,18 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 			)
 		}
 
-		pod.Spec.Volumes = append(pod.Spec.Volumes,
-			corev1.Volume{
-				Name: "work",
-				VolumeSource: corev1.VolumeSource{
-					EmptyDir: &corev1.EmptyDirVolumeSource{},
+		if ok, _ := workVolumePresent(pod.Spec.Volumes); !ok {
+			pod.Spec.Volumes = append(pod.Spec.Volumes,
+				corev1.Volume{
+					Name: "work",
+					VolumeSource: corev1.VolumeSource{
+						EmptyDir: &corev1.EmptyDirVolumeSource{},
+					},
 				},
-			},
+			)
+		}
+
+		pod.Spec.Volumes = append(pod.Spec.Volumes,
 			corev1.Volume{
 				Name: "certs-client",
 				VolumeSource: corev1.VolumeSource{
@@ -719,11 +988,16 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 			},
 		)
 
+		if ok, _ := workVolumeMountPresent(runnerContainer.VolumeMounts); !ok {
+			runnerContainer.VolumeMounts = append(runnerContainer.VolumeMounts,
+				corev1.VolumeMount{
+					Name:      "work",
+					MountPath: workDir,
+				},
+			)
+		}
+
 		runnerContainer.VolumeMounts = append(runnerContainer.VolumeMounts,
-			corev1.VolumeMount{
-				Name:      "work",
-				MountPath: workDir,
-			},
 			corev1.VolumeMount{
 				Name:      "certs-client",
 				MountPath: "/certs/client",
@@ -825,15 +1099,13 @@ func newRunnerPod(runnerName string, template corev1.Pod, runnerSpec v1alpha1.Ru
 		}
 	}
 
-	// TODO Remove this once we remove RUNNER_FEATURE_FLAG_EPHEMERAL from runner's entrypoint.sh
-	// and make --ephemeral the default option.
-	if getRunnerEnv(pod, EnvVarRunnerFeatureFlagEphemeral) == "" {
-		setRunnerEnv(pod, EnvVarRunnerFeatureFlagEphemeral, EnvVarTrue)
-	}
-
 	return *pod, nil
 }
 
+func newRunnerPod(template corev1.Pod, runnerSpec v1alpha1.RunnerConfig, defaultRunnerImage string, defaultRunnerImagePullSecrets []string, defaultDockerImage, defaultDockerRegistryMirror string, githubBaseURL string, useRunnerStatusUpdateHookEphemeralRole bool) (corev1.Pod, error) {
+	return newRunnerPodWithContainerMode("", template, runnerSpec, defaultRunnerImage, defaultRunnerImagePullSecrets, defaultDockerImage, defaultDockerRegistryMirror, githubBaseURL, useRunnerStatusUpdateHookEphemeralRole)
+}
+
 func (r *RunnerReconciler) SetupWithManager(mgr ctrl.Manager) error {
 	name := "runner-controller"
 	if r.Name != "" {
@@ -896,3 +1168,71 @@ func workVolumeMountPresent(items []corev1.VolumeMount) (bool, int) {
 	}
 	return false, 0
 }
+
+func applyWorkVolumeClaimTemplateToPod(pod *corev1.Pod, workVolumeClaimTemplate *v1alpha1.WorkVolumeClaimTemplate, workDir string) error {
+	if workVolumeClaimTemplate == nil {
+		return errors.New("work volume claim template must be specified in container mode kubernetes")
+	}
+	for i := range pod.Spec.Volumes {
+		if pod.Spec.Volumes[i].Name == "work" {
+			return fmt.Errorf("Work volume should not be specified in container mode kubernetes. workVolumeClaimTemplate field should be used instead.")
+		}
+	}
+	pod.Spec.Volumes = append(pod.Spec.Volumes, workVolumeClaimTemplate.V1Volume())
+
+	var runnerContainer *corev1.Container
+	for i := range pod.Spec.Containers {
+		if pod.Spec.Containers[i].Name == "runner" {
+			runnerContainer = &pod.Spec.Containers[i]
+			break
+		}
+	}
+
+	if runnerContainer == nil {
+		return fmt.Errorf("runner container is not present when applying work volume claim template")
+	}
+
+	if isPresent, _ := workVolumeMountPresent(runnerContainer.VolumeMounts); isPresent {
+		return fmt.Errorf("volume mount \"work\" should not be present on the runner container in container mode kubernetes")
+	}
+
+	runnerContainer.VolumeMounts = append(runnerContainer.VolumeMounts, workVolumeClaimTemplate.V1VolumeMount(workDir))
+
+	return nil
+}
+
+// isRequireSameNode specifies for the runner in kubernetes mode wether it should
+// schedule jobs to the same node where the runner is
+//
+// This function should only be called in containerMode: kubernetes
+func isRequireSameNode(pod *corev1.Pod) (bool, error) {
+	isPresent, index := workVolumePresent(pod.Spec.Volumes)
+	if !isPresent {
+		return true, errors.New("internal error: work volume mount must exist in containerMode: kubernetes")
+	}
+
+	if pod.Spec.Volumes[index].Ephemeral == nil || pod.Spec.Volumes[index].Ephemeral.VolumeClaimTemplate == nil {
+		return true, errors.New("containerMode: kubernetes should have pod.Spec.Volumes[].Ephemeral.VolumeClaimTemplate set")
+	}
+
+	for _, accessMode := range pod.Spec.Volumes[index].Ephemeral.VolumeClaimTemplate.Spec.AccessModes {
+		switch accessMode {
+		case corev1.ReadWriteOnce:
+			return true, nil
+		case corev1.ReadWriteMany:
+		default:
+			return true, errors.New("actions-runner-controller supports ReadWriteOnce and ReadWriteMany modes only")
+		}
+	}
+	return false, nil
+}
+
+func overwriteRunnerEnv(runner *v1alpha1.Runner, key string, value string) {
+	for i := range runner.Spec.Env {
+		if runner.Spec.Env[i].Name == key {
+			runner.Spec.Env[i].Value = value
+			return
+		}
+	}
+	runner.Spec.Env = append(runner.Spec.Env, corev1.EnvVar{Name: key, Value: value})
+}

controllers/runner_graceful_stop.go

@@ -9,7 +9,7 @@ import (
 
 	"github.com/actions-runner-controller/actions-runner-controller/github"
 	"github.com/go-logr/logr"
-	gogithub "github.com/google/go-github/v39/github"
+	gogithub "github.com/google/go-github/v47/github"
 	corev1 "k8s.io/api/core/v1"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	ctrl "sigs.k8s.io/controller-runtime"
@@ -113,10 +113,28 @@ func ensureRunnerUnregistration(ctx context.Context, retryDelay time.Duration, l
 		// Happens e.g. when dind is in runner and run completes
 		log.Info("Runner pod has been stopped with a successful status.")
 	} else if pod != nil && pod.Annotations[AnnotationKeyRunnerCompletionWaitStartTimestamp] != "" {
-		log.Info("Runner pod is annotated to wait for completion")
+		ct := ephemeralRunnerContainerStatus(pod)
+		if ct == nil {
+			log.Info("Runner pod is annotated to wait for completion, and the runner container is not ephemeral")
 
-		return &ctrl.Result{RequeueAfter: retryDelay}, nil
-	} else if ok, err := unregisterRunner(ctx, ghClient, enterprise, organization, repository, runner, *runnerID); err != nil {
+			return &ctrl.Result{RequeueAfter: retryDelay}, nil
+		}
+
+		lts := ct.LastTerminationState.Terminated
+		if lts == nil {
+			log.Info("Runner pod is annotated to wait for completion, and the runner container is not restarting")
+
+			return &ctrl.Result{RequeueAfter: retryDelay}, nil
+		}
+
+		// Prevent runner pod from stucking in Terminating.
+		// See https://github.com/actions-runner-controller/actions-runner-controller/issues/1369
+		log.Info("Deleting runner pod anyway because it has stopped prematurely. This may leave a dangling runner resource in GitHub Actions",
+			"lastState.exitCode", lts.ExitCode,
+			"lastState.message", lts.Message,
+			"pod.phase", pod.Status.Phase,
+		)
+	} else if ok, err := unregisterRunner(ctx, ghClient, enterprise, organization, repository, *runnerID); err != nil {
 		if errors.Is(err, &gogithub.RateLimitError{}) {
 			// We log the underlying error when we failed calling GitHub API to list or unregisters,
 			// or the runner is still busy.
@@ -133,7 +151,10 @@ func ensureRunnerUnregistration(ctx context.Context, retryDelay time.Duration, l
 
 		log.V(1).Info("Failed to unregister runner before deleting the pod.", "error", err)
 
-		var runnerBusy bool
+		var (
+			runnerBusy                         bool
+			runnerUnregistrationFailureMessage string
+		)
 
 		errRes := &gogithub.ErrorResponse{}
 		if errors.As(err, &errRes) {
@@ -155,6 +176,7 @@ func ensureRunnerUnregistration(ctx context.Context, retryDelay time.Duration, l
 			}
 
 			runnerBusy = errRes.Response.StatusCode == 422
+			runnerUnregistrationFailureMessage = errRes.Message
 
 			if runnerBusy && code != nil {
 				log.V(2).Info("Runner container has already stopped but the unregistration attempt failed. "+
@@ -169,13 +191,18 @@ func ensureRunnerUnregistration(ctx context.Context, retryDelay time.Duration, l
 		}
 
 		if runnerBusy {
+			_, err := annotatePodOnce(ctx, c, log, pod, AnnotationKeyUnregistrationFailureMessage, runnerUnregistrationFailureMessage)
+			if err != nil {
+				return &ctrl.Result{}, err
+			}
+
 			// We want to prevent spamming the deletion attemps but returning ctrl.Result with RequeueAfter doesn't
 			// work as the reconcilation can happen earlier due to pod status update.
 			// For ephemeral runners, we can expect it to stop and unregister itself on completion.
 			// So we can just wait for the completion without actively retrying unregistration.
 			ephemeral := getRunnerEnv(pod, EnvVarEphemeral)
 			if ephemeral == "true" {
-				pod, err = annotatePodOnce(ctx, c, log, pod, AnnotationKeyRunnerCompletionWaitStartTimestamp, time.Now().Format(time.RFC3339))
+				_, err = annotatePodOnce(ctx, c, log, pod, AnnotationKeyRunnerCompletionWaitStartTimestamp, time.Now().Format(time.RFC3339))
 				if err != nil {
 					return &ctrl.Result{}, err
 				}
@@ -352,7 +379,7 @@ func setRunnerEnv(pod *corev1.Pod, key, value string) {
 // There isn't a single right grace period that works for everyone.
 // The longer the grace period is, the earlier a cluster resource shortage can occur due to throttoled runner pod deletions,
 // while the shorter the grace period is, the more likely you may encounter the race issue.
-func unregisterRunner(ctx context.Context, client *github.Client, enterprise, org, repo, name string, id int64) (bool, error) {
+func unregisterRunner(ctx context.Context, client *github.Client, enterprise, org, repo string, id int64) (bool, error) {
 	// For the record, historically ARC did not try to call RemoveRunner on a busy runner, but it's no longer true.
 	// The reason ARC did so was to let a runner running a job to not stop prematurely.
 	//

controllers/runner_pod.go

@@ -0,0 +1,22 @@
+package controllers
+
+import corev1 "k8s.io/api/core/v1"
+
+// Force the runner pod managed by either RunnerDeployment and RunnerSet to have restartPolicy=Never.
+// See https://github.com/actions-runner-controller/actions-runner-controller/issues/1369 for more context.
+//
+// This is to prevent runner pods from stucking in Terminating when a K8s node disappeared along with the runnr pod and the runner container within it.
+//
+// Previously, we used restartPolicy of OnFailure, it turned wrong later, and therefore we now set Never.
+//
+// When the restartPolicy is OnFailure and the node disappeared, runner pods on the node seem to stuck in state.terminated==nil, state.waiting!=nil, and state.lastTerminationState!=nil,
+// and will ever become Running.
+// It's probably due to that the node onto which the pods have been scheduled will ever come back, hence the container restart attempt swill ever succeed,
+// the pods stuck waiting for successful restarts forever.
+//
+// By forcing runner pods to never restart, we hope there will be no chances of pods being stuck waiting.
+func forceRunnerPodRestartPolicyNever(pod *corev1.Pod) {
+	if pod.Spec.RestartPolicy != corev1.RestartPolicyNever {
+		pod.Spec.RestartPolicy = corev1.RestartPolicyNever
+	}
+}

controllers/runner_pod_controller.go

@@ -20,6 +20,7 @@ import (
 	"context"
 	"errors"
 	"fmt"
+	"sync"
 	"time"
 
 	"github.com/go-logr/logr"
@@ -31,8 +32,6 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/client"
 
 	corev1 "k8s.io/api/core/v1"
-
-	"github.com/actions-runner-controller/actions-runner-controller/github"
 )
 
 // RunnerPodReconciler reconciles a Runner object
@@ -41,7 +40,7 @@ type RunnerPodReconciler struct {
 	Log                         logr.Logger
 	Recorder                    record.EventRecorder
 	Scheme                      *runtime.Scheme
-	GitHubClient                *github.Client
+	GitHubClient                *MultiGitHubClient
 	Name                        string
 	RegistrationRecheckInterval time.Duration
 	RegistrationRecheckJitter   time.Duration
@@ -50,6 +49,7 @@ type RunnerPodReconciler struct {
 }
 
 // +kubebuilder:rbac:groups=core,resources=pods,verbs=get;list;watch;update;patch;delete
+// +kubebuilder:rbac:groups=core,resources=secrets,verbs=get;list;watch
 // +kubebuilder:rbac:groups=core,resources=events,verbs=create;patch
 
 func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (ctrl.Result, error) {
@@ -60,8 +60,11 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 		return ctrl.Result{}, client.IgnoreNotFound(err)
 	}
 
-	_, isRunnerPod := runnerPod.Labels[LabelKeyRunnerSetName]
-	if !isRunnerPod {
+	_, isRunnerPod := runnerPod.Labels[LabelKeyRunner]
+	_, isRunnerSetPod := runnerPod.Labels[LabelKeyRunnerSetName]
+	_, isRunnerDeploymentPod := runnerPod.Labels[LabelKeyRunnerDeploymentName]
+
+	if !isRunnerPod && !isRunnerSetPod && !isRunnerDeploymentPod {
 		return ctrl.Result{}, nil
 	}
 
@@ -77,6 +80,7 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 	}
 
 	var enterprise, org, repo string
+	var isContainerMode bool
 
 	for _, e := range envvars {
 		switch e.Name {
@@ -86,13 +90,25 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 			org = e.Value
 		case EnvVarRepo:
 			repo = e.Value
+		case "ACTIONS_RUNNER_CONTAINER_HOOKS":
+			isContainerMode = true
 		}
 	}
 
+	ghc, err := r.GitHubClient.InitForRunnerPod(ctx, &runnerPod)
+	if err != nil {
+		return ctrl.Result{}, err
+	}
+
 	if runnerPod.ObjectMeta.DeletionTimestamp.IsZero() {
 		finalizers, added := addFinalizer(runnerPod.ObjectMeta.Finalizers, runnerPodFinalizerName)
 
-		if added {
+		var cleanupFinalizersAdded bool
+		if isContainerMode {
+			finalizers, cleanupFinalizersAdded = addFinalizer(finalizers, runnerLinkedResourcesFinalizerName)
+		}
+
+		if added || cleanupFinalizersAdded {
 			newRunner := runnerPod.DeepCopy()
 			newRunner.ObjectMeta.Finalizers = finalizers
 
@@ -108,13 +124,34 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 	} else {
 		log.V(2).Info("Seen deletion-timestamp is already set")
 
+		if finalizers, removed := removeFinalizer(runnerPod.ObjectMeta.Finalizers, runnerLinkedResourcesFinalizerName); removed {
+			if err := r.cleanupRunnerLinkedPods(ctx, &runnerPod, log); err != nil {
+				log.Info("Runner-linked pods clean up that has failed due to an error. If this persists, please manually remove the runner-linked pods to unblock ARC", "err", err.Error())
+				return ctrl.Result{Requeue: true, RequeueAfter: 30 * time.Second}, nil
+			}
+			if err := r.cleanupRunnerLinkedSecrets(ctx, &runnerPod, log); err != nil {
+				log.Info("Runner-linked secrets clean up that has failed due to an error. If this persists, please manually remove the runner-linked secrets to unblock ARC", "err", err.Error())
+				return ctrl.Result{Requeue: true, RequeueAfter: 30 * time.Second}, nil
+			}
+			patchedPod := runnerPod.DeepCopy()
+			patchedPod.ObjectMeta.Finalizers = finalizers
+
+			if err := r.Patch(ctx, patchedPod, client.MergeFrom(&runnerPod)); err != nil {
+				log.Error(err, "Failed to update runner for finalizer linked resources removal")
+				return ctrl.Result{}, err
+			}
+
+			// Otherwise the subsequent patch request can revive the removed finalizer and it will trigger a unnecessary reconcilation
+			runnerPod = *patchedPod
+		}
+
 		finalizers, removed := removeFinalizer(runnerPod.ObjectMeta.Finalizers, runnerPodFinalizerName)
 
 		if removed {
 			// In a standard scenario, the upstream controller, like runnerset-controller, ensures this runner to be gracefully stopped before the deletion timestamp is set.
 			// But for the case that the user manually deleted it for whatever reason,
 			// we have to ensure it to gracefully stop now.
-			updatedPod, res, err := tickRunnerGracefulStop(ctx, r.unregistrationRetryDelay(), log, r.GitHubClient, r.Client, enterprise, org, repo, runnerPod.Name, &runnerPod)
+			updatedPod, res, err := tickRunnerGracefulStop(ctx, r.unregistrationRetryDelay(), log, ghc, r.Client, enterprise, org, repo, runnerPod.Name, &runnerPod)
 			if res != nil {
 				return *res, err
 			}
@@ -130,6 +167,8 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 
 			log.V(2).Info("Removed finalizer")
 
+			r.GitHubClient.DeinitForRunnerPod(updatedPod)
+
 			return ctrl.Result{}, nil
 		}
 
@@ -168,7 +207,7 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 		return ctrl.Result{}, nil
 	}
 
-	po, res, err := ensureRunnerPodRegistered(ctx, log, r.GitHubClient, r.Client, enterprise, org, repo, runnerPod.Name, &runnerPod)
+	po, res, err := ensureRunnerPodRegistered(ctx, log, ghc, r.Client, enterprise, org, repo, runnerPod.Name, &runnerPod)
 	if res != nil {
 		return *res, err
 	}
@@ -182,7 +221,7 @@ func (r *RunnerPodReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 		//
 		// In a standard scenario, ARC starts the unregistration process before marking the pod for deletion at all,
 		// so that it isn't subject to terminationGracePeriod and can safely take hours to finish it's work.
-		_, res, err := tickRunnerGracefulStop(ctx, r.unregistrationRetryDelay(), log, r.GitHubClient, r.Client, enterprise, org, repo, runnerPod.Name, &runnerPod)
+		_, res, err := tickRunnerGracefulStop(ctx, r.unregistrationRetryDelay(), log, ghc, r.Client, enterprise, org, repo, runnerPod.Name, &runnerPod)
 		if res != nil {
 			return *res, err
 		}
@@ -222,3 +261,93 @@ func (r *RunnerPodReconciler) SetupWithManager(mgr ctrl.Manager) error {
 		Named(name).
 		Complete(r)
 }
+
+func (r *RunnerPodReconciler) cleanupRunnerLinkedPods(ctx context.Context, pod *corev1.Pod, log logr.Logger) error {
+	var runnerLinkedPodList corev1.PodList
+	if err := r.List(ctx, &runnerLinkedPodList, client.InNamespace(pod.Namespace), client.MatchingLabels(
+		map[string]string{
+			"runner-pod": pod.ObjectMeta.Name,
+		},
+	)); err != nil {
+		return fmt.Errorf("failed to list runner-linked pods: %w", err)
+	}
+
+	var (
+		wg   sync.WaitGroup
+		errs []error
+	)
+	for _, p := range runnerLinkedPodList.Items {
+		if !p.ObjectMeta.DeletionTimestamp.IsZero() {
+			continue
+		}
+
+		p := p
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			if err := r.Delete(ctx, &p); err != nil {
+				if kerrors.IsNotFound(err) || kerrors.IsGone(err) {
+					return
+				}
+				errs = append(errs, fmt.Errorf("delete pod %q error: %v", p.ObjectMeta.Name, err))
+			}
+		}()
+	}
+
+	wg.Wait()
+
+	if len(errs) > 0 {
+		for _, err := range errs {
+			log.Error(err, "failed to remove runner-linked pod")
+		}
+		return errors.New("failed to remove some runner linked pods")
+	}
+
+	return nil
+}
+
+func (r *RunnerPodReconciler) cleanupRunnerLinkedSecrets(ctx context.Context, pod *corev1.Pod, log logr.Logger) error {
+	log.V(2).Info("Listing runner-linked secrets to be deleted", "ns", pod.Namespace)
+
+	var runnerLinkedSecretList corev1.SecretList
+	if err := r.List(ctx, &runnerLinkedSecretList, client.InNamespace(pod.Namespace), client.MatchingLabels(
+		map[string]string{
+			"runner-pod": pod.ObjectMeta.Name,
+		},
+	)); err != nil {
+		return fmt.Errorf("failed to list runner-linked secrets: %w", err)
+	}
+
+	var (
+		wg   sync.WaitGroup
+		errs []error
+	)
+	for _, s := range runnerLinkedSecretList.Items {
+		if !s.ObjectMeta.DeletionTimestamp.IsZero() {
+			continue
+		}
+
+		s := s
+		wg.Add(1)
+		go func() {
+			defer wg.Done()
+			if err := r.Delete(ctx, &s); err != nil {
+				if kerrors.IsNotFound(err) || kerrors.IsGone(err) {
+					return
+				}
+				errs = append(errs, fmt.Errorf("delete secret %q error: %v", s.ObjectMeta.Name, err))
+			}
+		}()
+	}
+
+	wg.Wait()
+
+	if len(errs) > 0 {
+		for _, err := range errs {
+			log.Error(err, "failed to remove runner-linked secret")
+		}
+		return errors.New("failed to remove some runner linked secrets")
+	}
+
+	return nil
+}

controllers/runnerdeployment_controller.go

@@ -165,6 +165,8 @@ func (r *RunnerDeploymentReconciler) Reconcile(ctx context.Context, req ctrl.Req
 			return ctrl.Result{}, err
 		}
 
+		log.V(1).Info("Updated runnerreplicaset due to selector change")
+
 		// At this point, we are already sure that there's no need to create a new replicaset
 		// as the runner template hash is not changed.
 		//
@@ -179,7 +181,17 @@ func (r *RunnerDeploymentReconciler) Reconcile(ctx context.Context, req ctrl.Req
 	newDesiredReplicas := getIntOrDefault(desiredRS.Spec.Replicas, defaultReplicas)
 
 	// Please add more conditions that we can in-place update the newest runnerreplicaset without disruption
-	if currentDesiredReplicas != newDesiredReplicas {
+	//
+	// If we missed taking the EffectiveTime diff into account, you might end up experiencing scale-ups being delayed scale-down.
+	// See https://github.com/actions-runner-controller/actions-runner-controller/pull/1477#issuecomment-1164154496
+	var et1, et2 time.Time
+	if newestSet.Spec.EffectiveTime != nil {
+		et1 = newestSet.Spec.EffectiveTime.Time
+	}
+	if rd.Spec.EffectiveTime != nil {
+		et2 = rd.Spec.EffectiveTime.Time
+	}
+	if currentDesiredReplicas != newDesiredReplicas || et1 != et2 {
 		newestSet.Spec.Replicas = &newDesiredReplicas
 		newestSet.Spec.EffectiveTime = rd.Spec.EffectiveTime
 
@@ -189,6 +201,13 @@ func (r *RunnerDeploymentReconciler) Reconcile(ctx context.Context, req ctrl.Req
 			return ctrl.Result{}, err
 		}
 
+		log.V(1).Info("Updated runnerreplicaset due to spec change",
+			"currentDesiredReplicas", currentDesiredReplicas,
+			"newDesiredReplicas", newDesiredReplicas,
+			"currentEffectiveTime", newestSet.Spec.EffectiveTime,
+			"newEffectiveTime", rd.Spec.EffectiveTime,
+		)
+
 		return ctrl.Result{}, err
 	}
 
@@ -421,9 +440,7 @@ func getSelector(rd *v1alpha1.RunnerDeployment) *metav1.LabelSelector {
 func newRunnerReplicaSet(rd *v1alpha1.RunnerDeployment, commonRunnerLabels []string, scheme *runtime.Scheme) (*v1alpha1.RunnerReplicaSet, error) {
 	newRSTemplate := *rd.Spec.Template.DeepCopy()
 
-	for _, l := range commonRunnerLabels {
-		newRSTemplate.Spec.Labels = append(newRSTemplate.Spec.Labels, l)
-	}
+	newRSTemplate.Spec.Labels = append(newRSTemplate.Spec.Labels, commonRunnerLabels...)
 
 	templateHash := ComputeHash(&newRSTemplate)
 

controllers/runnerreplicaset_controller.go

@@ -32,17 +32,15 @@ import (
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 
 	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
-	"github.com/actions-runner-controller/actions-runner-controller/github"
 )
 
 // RunnerReplicaSetReconciler reconciles a Runner object
 type RunnerReplicaSetReconciler struct {
 	client.Client
-	Log          logr.Logger
-	Recorder     record.EventRecorder
-	Scheme       *runtime.Scheme
-	GitHubClient *github.Client
-	Name         string
+	Log      logr.Logger
+	Recorder record.EventRecorder
+	Scheme   *runtime.Scheme
+	Name     string
 }
 
 const (
@@ -205,7 +203,3 @@ func (r *RunnerReplicaSetReconciler) SetupWithManager(mgr ctrl.Manager) error {
 		Named(name).
 		Complete(r)
 }
-
-func registrationOnlyRunnerNameFor(rsName string) string {
-	return rsName + "-registration-only"
-}

controllers/runnerreplicaset_controller_test.go

@@ -52,15 +52,13 @@ func SetupTest(ctx2 context.Context) *corev1.Namespace {
 
 		runnersList = fake.NewRunnersList()
 		server = runnersList.GetServer()
-		ghClient := newGithubClient(server)
 
 		controller := &RunnerReplicaSetReconciler{
-			Client:       mgr.GetClient(),
-			Scheme:       scheme.Scheme,
-			Log:          logf.Log,
-			Recorder:     mgr.GetEventRecorderFor("runnerreplicaset-controller"),
-			GitHubClient: ghClient,
-			Name:         "runnerreplicaset-" + ns.Name,
+			Client:   mgr.GetClient(),
+			Scheme:   scheme.Scheme,
+			Log:      logf.Log,
+			Recorder: mgr.GetEventRecorderFor("runnerreplicaset-controller"),
+			Name:     "runnerreplicaset-" + ns.Name,
 		}
 		err = controller.SetupWithManager(mgr)
 		Expect(err).NotTo(HaveOccurred(), "failed to setup controller")

controllers/runnerset_controller.go

@@ -45,12 +45,13 @@ type RunnerSetReconciler struct {
 	Recorder record.EventRecorder
 	Scheme   *runtime.Scheme
 
-	CommonRunnerLabels     []string
-	GitHubBaseURL          string
-	RunnerImage            string
-	RunnerImagePullSecrets []string
-	DockerImage            string
-	DockerRegistryMirror   string
+	CommonRunnerLabels        []string
+	GitHubClient              *MultiGitHubClient
+	RunnerImage               string
+	RunnerImagePullSecrets    []string
+	DockerImage               string
+	DockerRegistryMirror      string
+	UseRunnerStatusUpdateHook bool
 }
 
 // +kubebuilder:rbac:groups=actions.summerwind.dev,resources=runnersets,verbs=get;list;watch;create;update;patch;delete
@@ -58,6 +59,7 @@ type RunnerSetReconciler struct {
 // +kubebuilder:rbac:groups=actions.summerwind.dev,resources=runnersets/status,verbs=get;update;patch
 // +kubebuilder:rbac:groups=apps,resources=statefulsets,verbs=get;list;watch;create;update;patch;delete
 // +kubebuilder:rbac:groups=apps,resources=statefulsets/status,verbs=get;update;patch
+// +kubebuilder:rbac:groups=core,resources=persistentvolumeclaims,verbs=get;list;watch;create;update;patch;delete
 // +kubebuilder:rbac:groups=core,resources=events,verbs=create;patch
 // +kubebuilder:rbac:groups=coordination.k8s.io,resources=leases,verbs=get;list;create;update
 
@@ -79,6 +81,8 @@ func (r *RunnerSetReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 	}
 
 	if !runnerSet.ObjectMeta.DeletionTimestamp.IsZero() {
+		r.GitHubClient.DeinitForRunnerSet(runnerSet)
+
 		return ctrl.Result{}, nil
 	}
 
@@ -96,7 +100,7 @@ func (r *RunnerSetReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 		return ctrl.Result{}, nil
 	}
 
-	desiredStatefulSet, err := r.newStatefulSet(runnerSet)
+	desiredStatefulSet, err := r.newStatefulSet(ctx, runnerSet)
 	if err != nil {
 		r.Recorder.Event(runnerSet, corev1.EventTypeNormal, "RunnerAutoscalingFailure", err.Error())
 
@@ -129,6 +133,12 @@ func (r *RunnerSetReconciler) Reconcile(ctx context.Context, req ctrl.Request) (
 		owners = append(owners, &ss)
 	}
 
+	if res, err := syncVolumes(ctx, r.Client, log, req.Namespace, runnerSet, statefulsets); err != nil {
+		return ctrl.Result{}, err
+	} else if res != nil {
+		return *res, nil
+	}
+
 	res, err := syncRunnerPodsOwners(ctx, r.Client, log, effectiveTime, newDesiredReplicas, func() client.Object { return create.DeepCopy() }, ephemeral, owners)
 	if err != nil || res == nil {
 		return ctrl.Result{}, err
@@ -178,19 +188,50 @@ func getRunnerSetSelector(runnerSet *v1alpha1.RunnerSet) *metav1.LabelSelector {
 var LabelKeyPodMutation = "actions-runner-controller/inject-registration-token"
 var LabelValuePodMutation = "true"
 
-func (r *RunnerSetReconciler) newStatefulSet(runnerSet *v1alpha1.RunnerSet) (*appsv1.StatefulSet, error) {
+func (r *RunnerSetReconciler) newStatefulSet(ctx context.Context, runnerSet *v1alpha1.RunnerSet) (*appsv1.StatefulSet, error) {
 	runnerSetWithOverrides := *runnerSet.Spec.DeepCopy()
 
-	for _, l := range r.CommonRunnerLabels {
-		runnerSetWithOverrides.Labels = append(runnerSetWithOverrides.Labels, l)
-	}
+	runnerSetWithOverrides.Labels = append(runnerSetWithOverrides.Labels, r.CommonRunnerLabels...)
 
 	template := corev1.Pod{
 		ObjectMeta: runnerSetWithOverrides.StatefulSetSpec.Template.ObjectMeta,
 		Spec:       runnerSetWithOverrides.StatefulSetSpec.Template.Spec,
 	}
 
-	pod, err := newRunnerPod(runnerSet.Name, template, runnerSet.Spec.RunnerConfig, r.RunnerImage, r.RunnerImagePullSecrets, r.DockerImage, r.DockerRegistryMirror, r.GitHubBaseURL, false)
+	if runnerSet.Spec.RunnerConfig.ContainerMode == "kubernetes" {
+		found := false
+		for i := range template.Spec.Containers {
+			if template.Spec.Containers[i].Name == containerName {
+				found = true
+			}
+		}
+		if !found {
+			template.Spec.Containers = append(template.Spec.Containers, corev1.Container{
+				Name: "runner",
+			})
+		}
+
+		workDir := runnerSet.Spec.RunnerConfig.WorkDir
+		if workDir == "" {
+			workDir = "/runner/_work"
+		}
+		if err := applyWorkVolumeClaimTemplateToPod(&template, runnerSet.Spec.WorkVolumeClaimTemplate, workDir); err != nil {
+			return nil, err
+		}
+
+		template.Spec.ServiceAccountName = runnerSet.Spec.ServiceAccountName
+	}
+
+	template.ObjectMeta.Labels = CloneAndAddLabel(template.ObjectMeta.Labels, LabelKeyRunnerSetName, runnerSet.Name)
+
+	ghc, err := r.GitHubClient.InitForRunnerSet(ctx, runnerSet)
+	if err != nil {
+		return nil, err
+	}
+
+	githubBaseURL := ghc.GithubBaseURL
+
+	pod, err := newRunnerPodWithContainerMode(runnerSet.Spec.RunnerConfig.ContainerMode, template, runnerSet.Spec.RunnerConfig, r.RunnerImage, r.RunnerImagePullSecrets, r.DockerImage, r.DockerRegistryMirror, githubBaseURL, r.UseRunnerStatusUpdateHook)
 	if err != nil {
 		return nil, err
 	}

controllers/schedule_test.go

@@ -605,3 +605,13 @@ func parseAndMatchRecurringPeriod(now time.Time, start, end, frequency, until st
 
 	return MatchSchedule(now, startTime, endTime, RecurrenceRule{Frequency: frequency, UntilTime: untilTime})
 }
+
+func FuzzMatchSchedule(f *testing.F) {
+	start := time.Now()
+	end := time.Now()
+	now := time.Now()
+	f.Fuzz(func(t *testing.T, freq string) {
+		// Verify that it never panics
+		_, _, _ = MatchSchedule(now, start, end, RecurrenceRule{Frequency: freq})
+	})
+}

controllers/sync_volumes.go

@@ -0,0 +1,185 @@
+package controllers
+
+import (
+	"context"
+	"fmt"
+	"time"
+
+	"github.com/actions-runner-controller/actions-runner-controller/api/v1alpha1"
+	"github.com/go-logr/logr"
+	appsv1 "k8s.io/api/apps/v1"
+	corev1 "k8s.io/api/core/v1"
+	kerrors "k8s.io/apimachinery/pkg/api/errors"
+	"k8s.io/apimachinery/pkg/types"
+	ctrl "sigs.k8s.io/controller-runtime"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+)
+
+const (
+	labelKeyCleanup               = "pending-cleanup"
+	labelKeyRunnerStatefulSetName = "runner-statefulset-name"
+)
+
+func syncVolumes(ctx context.Context, c client.Client, log logr.Logger, ns string, runnerSet *v1alpha1.RunnerSet, statefulsets []appsv1.StatefulSet) (*ctrl.Result, error) {
+	log = log.WithValues("ns", ns)
+
+	for _, t := range runnerSet.Spec.StatefulSetSpec.VolumeClaimTemplates {
+		for _, sts := range statefulsets {
+			pvcName := fmt.Sprintf("%s-%s-0", t.Name, sts.Name)
+
+			var pvc corev1.PersistentVolumeClaim
+			if err := c.Get(ctx, types.NamespacedName{Namespace: ns, Name: pvcName}, &pvc); err != nil {
+				if !kerrors.IsNotFound(err) {
+					return nil, err
+				}
+				continue
+			}
+
+			// TODO move this to statefulset reconciler so that we spam this less,
+			// by starting the loop only after the statefulset got deletionTimestamp set.
+			// Perhaps you can just wrap this in a finalizer here.
+			if pvc.Labels[labelKeyRunnerStatefulSetName] == "" {
+				updated := pvc.DeepCopy()
+				updated.Labels[labelKeyRunnerStatefulSetName] = sts.Name
+				if err := c.Update(ctx, updated); err != nil {
+					return nil, err
+				}
+				log.V(1).Info("Added runner-statefulset-name label to PVC", "sts", sts.Name, "pvc", pvcName)
+			}
+		}
+	}
+
+	// PVs are not namespaced hence we don't need client.InNamespace(ns).
+	// If we added that, c.List will silently return zero items.
+	//
+	// This `List` needs to be done in a dedicated reconciler that is registered to the manager via the `For` func.
+	// Otherwise the List func might return outdated contents(I saw status.phase being Bound even after K8s updated it to Released, and it lasted minutes).
+	//
+	// cleanupLabels := map[string]string{
+	// 	labelKeyCleanup: runnerSet.Name,
+	// }
+	// pvList := &corev1.PersistentVolumeList{}
+	// if err := c.List(ctx, pvList, client.MatchingLabels(cleanupLabels)); err != nil {
+	// 	log.Info("retrying pv listing", "ns", ns, "err", err)
+	// 	return nil, err
+	// }
+
+	return nil, nil
+}
+
+func syncPVC(ctx context.Context, c client.Client, log logr.Logger, ns string, pvc *corev1.PersistentVolumeClaim) (*ctrl.Result, error) {
+	stsName := pvc.Labels[labelKeyRunnerStatefulSetName]
+	if stsName == "" {
+		return nil, nil
+	}
+
+	log.V(2).Info("Reconciling runner PVC")
+
+	// TODO: Probably we'd better remove PVCs related to the RunnetSet that is nowhere now?
+	// Otherwise, a bunch of continuously recreated StatefulSet
+	// can leave dangling PVCs forever, which might stress the cluster.
+
+	var sts appsv1.StatefulSet
+	if err := c.Get(ctx, types.NamespacedName{Namespace: ns, Name: stsName}, &sts); err != nil {
+		if !kerrors.IsNotFound(err) {
+			return nil, err
+		}
+	} else {
+		// We assume that the statefulset is shortly terminated, hence retry forever until it gets removed.
+		retry := 10 * time.Second
+		log.V(1).Info("Retrying sync until statefulset gets removed", "requeueAfter", retry)
+		return &ctrl.Result{RequeueAfter: retry}, nil
+	}
+
+	log = log.WithValues("sts", stsName)
+
+	pvName := pvc.Spec.VolumeName
+
+	if pvName != "" {
+		// If we deleted PVC before unsetting pv.spec.claimRef,
+		// K8s seems to revive the claimRef :thinking:
+		// So we need to mark PV for claimRef unset first, and delete PVC, and finally unset claimRef on PV.
+
+		var pv corev1.PersistentVolume
+		if err := c.Get(ctx, types.NamespacedName{Namespace: ns, Name: pvName}, &pv); err != nil {
+			if !kerrors.IsNotFound(err) {
+				return nil, err
+			}
+			return nil, nil
+		}
+
+		pvCopy := pv.DeepCopy()
+		if pvCopy.Labels == nil {
+			pvCopy.Labels = map[string]string{}
+		}
+		pvCopy.Labels[labelKeyCleanup] = stsName
+
+		log.V(2).Info("Scheduling to unset PV's claimRef", "pv", pv.Name)
+
+		// Apparently K8s doesn't reconcile PV immediately after PVC deletion.
+		// So we start a relatively busy loop of PV reconcilation slightly before the PVC deletion,
+		// so that PV can be unbound as soon as possible after the PVC got deleted.
+		if err := c.Update(ctx, pvCopy); err != nil {
+			return nil, err
+		}
+
+		log.Info("Updated PV to unset claimRef")
+
+		// At this point, the PV is still Bound
+
+		log.V(2).Info("Deleting unused PVC")
+
+		if err := c.Delete(ctx, pvc); err != nil {
+			return nil, err
+		}
+
+		log.Info("Deleted unused PVC")
+
+		// At this point, the PV is still "Bound", but we are ready to unset pv.spec.claimRef in pv controller.
+		// Once the pv controller unsets claimRef, the PV becomes "Released", hence available for reuse by another eligible PVC.
+	}
+
+	return nil, nil
+}
+
+func syncPV(ctx context.Context, c client.Client, log logr.Logger, ns string, pv *corev1.PersistentVolume) (*ctrl.Result, error) {
+	if pv.Spec.ClaimRef == nil {
+		return nil, nil
+	}
+
+	log.V(2).Info("Reconciling PV")
+
+	if pv.Labels[labelKeyCleanup] == "" {
+		// We assume that the pvc is shortly terminated, hence retry forever until it gets removed.
+		retry := 10 * time.Second
+		log.V(2).Info("Retrying sync to see if this PV needs to be managed by ARC", "requeueAfter", retry)
+		return &ctrl.Result{RequeueAfter: retry}, nil
+	}
+
+	log.V(2).Info("checking pv phase", "phase", pv.Status.Phase)
+
+	if pv.Status.Phase != corev1.VolumeReleased {
+		// We assume that the pvc is shortly terminated, hence retry forever until it gets removed.
+		retry := 10 * time.Second
+		log.V(1).Info("Retrying sync until pvc gets released", "requeueAfter", retry)
+		return &ctrl.Result{RequeueAfter: retry}, nil
+	}
+
+	// At this point, the PV is still Released
+
+	pvCopy := pv.DeepCopy()
+	delete(pvCopy.Labels, labelKeyCleanup)
+	pvCopy.Spec.ClaimRef = nil
+	log.V(2).Info("Unsetting PV's claimRef", "pv", pv.Name)
+	if err := c.Update(ctx, pvCopy); err != nil {
+		return nil, err
+	}
+
+	log.Info("PV should be Available now")
+
+	// At this point, the PV becomes Available, if it's reclaim policy is "Retain".
+	// I have not yet tested it with "Delete" but perhaps it's deleted automatically after the update?
+	// https://kubernetes.io/docs/concepts/storage/persistent-volumes/#retain
+
+	return nil, nil
+}

controllers/testresourcereader.go

@@ -0,0 +1,31 @@
+package controllers
+
+import (
+	"context"
+	"errors"
+	"reflect"
+
+	kerrors "k8s.io/apimachinery/pkg/api/errors"
+	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
+	"k8s.io/apimachinery/pkg/types"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+)
+
+type testResourceReader struct {
+	objects map[types.NamespacedName]client.Object
+}
+
+func (r *testResourceReader) Get(_ context.Context, nsName types.NamespacedName, obj client.Object) error {
+	ret, ok := r.objects[nsName]
+	if !ok {
+		return &kerrors.StatusError{ErrStatus: metav1.Status{Reason: metav1.StatusReasonNotFound}}
+	}
+	v := reflect.ValueOf(obj)
+	if v.Kind() != reflect.Ptr {
+		return errors.New("obj must be a pointer")
+	}
+
+	v.Elem().Set(reflect.ValueOf(ret).Elem())
+
+	return nil
+}

controllers/testresourcereader_test.go

@@ -0,0 +1,35 @@
+package controllers
+
+import (
+	"context"
+	"testing"
+
+	"github.com/stretchr/testify/require"
+	corev1 "k8s.io/api/core/v1"
+	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
+	"k8s.io/apimachinery/pkg/types"
+	"sigs.k8s.io/controller-runtime/pkg/client"
+)
+
+func TestResourceReader(t *testing.T) {
+	rr := &testResourceReader{
+		objects: map[types.NamespacedName]client.Object{
+			{Namespace: "default", Name: "sec1"}: &corev1.Secret{
+				ObjectMeta: metav1.ObjectMeta{
+					Namespace: "default",
+					Name:      "sec1",
+				},
+				Data: map[string][]byte{
+					"foo": []byte("bar"),
+				},
+			},
+		},
+	}
+
+	var sec corev1.Secret
+
+	err := rr.Get(context.Background(), types.NamespacedName{Namespace: "default", Name: "sec1"}, &sec)
+	require.NoError(t, err)
+
+	require.Equal(t, []byte("bar"), sec.Data["foo"])
+}