ci: add workflow to deploy kepler on equinix

.github/clean-equinix-runner/action.yaml

@@ -0,0 +1,24 @@
+name: Clean Equinix Runner
+description: Deletes a given Equinix runner
+inputs:
+  runner_name:
+    description: Name of the runner to be deleted
+    required: true
+  project_id:
+    description: Project ID of the Equinix project
+    required: true
+  api_token:
+    description: API token of the Equinix project
+    required: true
+runs:
+  using: composite
+  steps:
+    - name: Delete runner
+      # Using custom equinix metal sweeper action instead of official one,
+      # as the official action does not support deleting a specific runner.
+      # See: https://github.com/sustainable-computing-io/metal-sweeper-action/pull/1
+      uses: sustainable-computing-io/metal-sweeper-action@main
+      with:
+        authToken: ${{ inputs.api_token }}
+        projectID: ${{ inputs.project_id }}
+        runnerName: ${{ inputs.runner_name }}

.github/create-equinix-runner/action.yaml

@@ -0,0 +1,34 @@
+name: Create Equinix Runner
+description: Creates a new Equinix runner for a given OS
+inputs:
+  os_name:
+    description: Name of the OS to be requested for the runner
+    required: true
+  github_token:
+    description: GitHub token
+    required: true
+  metal_auth_token:
+    description: Equinix API token
+    required: true
+  metal_project_id:
+    description: Equinix project ID
+    required: true
+
+runs:
+  using: composite
+  steps:
+    - name: Create runner
+      # Using custom equinix metal runner action instead of official one,
+      # as the official action does not support passing organization name parameter.
+      # This is necessary for workflows triggered by events like `issue_comment`, where
+      # the `github.organization` context is not available.
+      # See: https://github.com/sustainable-computing-io/metal-runner-action/pull/1
+      uses: sustainable-computing-io/metal-runner-action@main
+      with:
+        github_token: ${{ inputs.github_token }}
+        metal_auth_token: ${{ inputs.metal_auth_token }}
+        metal_project_id: ${{ inputs.metal_project_id }}
+        metro: da
+        plan: c3.small.x86
+        os: ${{ inputs.os_name }}
+        organization: sustainable-computing-io

.github/workflows/k8s-equinix.yaml

@@ -0,0 +1,189 @@
+name: Build and Deploy K8s on Equinix
+
+on: #yamllint disable-line rule:truthy
+  pull_request:
+    branches: [reboot]
+  # In order to trigger this workflow on comment, GH accepts the workflow to be available in the default branch only.
+  # issue_comment:
+  #   types: [created]
+
+permissions:
+  pull-requests: write
+  contents: write
+
+jobs:
+  create-runner:
+    # TODO: Uncomment once we enable workflow on issue_comment
+    # if: github.event.issue.pull_request && contains(github.event.comment.body, '/test-equinix')
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout source
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.event.issue.pull_request.head.sha }}
+
+      - name: Create Equinix runner
+        uses: ./.github/create-equinix-runner
+        with:
+          github_token: ${{ secrets.GH_SELF_HOSTED_RUNNER_TOKEN }}
+          metal_auth_token: ${{ secrets.EQUINIX_API_TOKEN }}
+          metal_project_id: ${{ secrets.EQUINIX_PROJECT_ID }}
+          os_name: ubuntu_22_04
+
+  build-and-deploy:
+    needs: create-runner
+    runs-on: self-hosted
+    env:
+      GOCACHE: /home/ghrunner/.cache/go-build
+      HOME: /home/ghrunner
+    outputs:
+      runner_name: ${{ runner.name }}
+    steps:
+      - name: Install Docker
+        shell: bash
+        run: |
+          sudo apt-get update
+          sudo apt-get install -y apt-transport-https ca-certificates curl software-properties-common
+          curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /usr/share/keyrings/docker-archive-keyring.gpg
+          echo "deb [arch=$(dpkg --print-architecture) signed-by=/usr/share/keyrings/docker-archive-keyring.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list
+          sudo apt-get update
+          sudo apt-get install -y docker-ce docker-ce-cli containerd.io
+          sudo usermod -aG docker root
+
+      - name: Verify Docker installation
+        shell: bash
+        run: |
+          docker ps
+          docker --version
+
+      - name: Install Kind
+        shell: bash
+        run: |
+          curl -Lo ./kind https://kind.sigs.k8s.io/dl/v0.23.0/kind-linux-amd64
+          chmod +x ./kind
+          sudo mv ./kind /usr/local/bin/kind
+
+      - name: Verify Kind installation
+        shell: bash
+        run: |
+          kind version
+
+      - name: Install Kubectl
+        shell: bash
+        run: |
+          curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl"
+          chmod +x ./kubectl
+          sudo mv ./kubectl /usr/local/bin/kubectl
+
+      - name: Verify Kubectl installation
+        shell: bash
+        run: |
+          kubectl version --client
+
+      - name: Checkout source
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.event.issue.pull_request.head.sha }}
+          fetch-depth: 0
+
+      - name: Setup Go
+        uses: actions/setup-go@v5.4.0
+        with:
+          go-version-file: go.mod
+          cache: false
+
+      - name: Build image
+        shell: bash
+        run: |
+          make image
+        env:
+          IMG_BASE: localhost:5001
+          VERSION: dev
+
+      - name: Setup Kind cluster
+        shell: bash
+        run: |
+          make cluster-up
+
+      - name: Push image
+        shell: bash
+        run: |
+          make push
+        env:
+          IMG_BASE: localhost:5001
+          VERSION: dev
+
+      - name: Deploy Kepler
+        shell: bash
+        run: |
+          make deploy
+        env:
+          IMG_BASE: localhost:5001
+          VERSION: dev
+
+      - name: Verify Kepler deployment
+        shell: bash
+        run: |
+          kubectl rollout status daemonset/kepler -n kepler --timeout=5m
+
+      # TODO: Move this once we add validator tool to the repo
+      - name: Validate metric endpoint
+        id: validate
+        shell: bash
+        run: |
+          kubectl port-forward service/kepler 28282:28282 -n kepler &
+          sleep 20 # sleep for 20 seconds to give the service time to start
+
+          HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" http://localhost:28282/metrics)
+          [[ $HTTP_STATUS -ne 200 ]] && echo "HTTP status code is not 200" && exit 1
+
+          curl -s -o /tmp/metrics.txt http://localhost:28282/metrics
+
+          for metric in kepler_process_cpu_watts \
+            kepler_node_cpu_info \
+            kepler_process_cpu_joules_total \
+            kepler_container_cpu_joules_total \
+            kepler_container_cpu_watts; do
+            echo "Checking metric: $metric"
+            if ! grep -q "$metric" /tmp/metrics.txt; then
+              echo "Metric $metric not found"
+              exit 1
+            fi
+          done
+
+      - name: Run must gather
+        if: failure()
+        shell: bash
+        run: |
+          echo "::group::Get pods in kepler namespace"
+          kubectl get pods -n kepler || true
+          echo "::endgroup::"
+
+          echo "::group::Get pods in monitoring namespace"
+          kubectl get pods -n monitoring || true
+          echo "::endgroup::"
+
+          echo "::group::Get logs for kepler daemonset"
+          kubectl logs daemonset/kepler -n kepler || true
+          echo "::endgroup::"
+
+          echo "::group::Fetch metrics from localhost:28282"
+          curl -s http://localhost:28282/metrics || true
+          echo "::endgroup::"
+
+  cleanup-runner:
+    needs: build-and-deploy
+    runs-on: ubuntu-latest
+    if: always() # Run even if the previous job fails
+    steps:
+      - name: Checkout source
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.event.issue.pull_request.head.sha }}
+
+      - name: Clean Equinix runner
+        uses: ./.github/clean-equinix-runner
+        with:
+          runner_name: ${{ needs.build-and-deploy.outputs.runner_name }}
+          project_id: ${{ secrets.EQUINIX_PROJECT_ID }}
+          api_token: ${{ secrets.EQUINIX_API_TOKEN }}