dragonfly/.github/workflows/benchmark.yml

name: benchmark-tests

on:
  schedule:
    - cron: "0 9 * * *" # run at 6 AM UTC
  workflow_dispatch:

permissions:
  contents: read

jobs:
  benchmark:
    strategy:
      matrix:
        config:
          - operator:
              apiVersion: "dragonflydb.io/v1alpha1"
              kind: "Dragonfly"
              metadata:
                labels:
                  app.kubernetes.io/name: "dragonfly"
                  app.kubernetes.io/instance: "dragonfly-sample"
                  app.kubernetes.io/part-of: "dragonfly-operator"
                  app.kubernetes.io/managed-by: "kustomize"
                  app.kubernetes.io/created-by: "dragonfly-operator"
                name: "dragonfly-sample"
              spec:
                image: "ghcr.io/dragonflydb/dragonfly:latest"
                args: ["--cache_mode"]
                replicas: 2
                resources:
                  requests:
                    cpu: "2"
                    memory: "2000Mi"
                  limits:
                    cpu: "2"
                    memory: "2000Mi"

    runs-on: ubuntu-latest

    container:
      image: ghcr.io/romange/benchmark-dev:latest
      options: --security-opt seccomp=unconfined

    permissions:
      id-token: write

    steps:
      - name: Setup namespace name
        id: setup
        run: echo "namespace=benchmark-$(date +"%Y-%m-%d-%s")" >> $GITHUB_OUTPUT

      - uses: actions/checkout@v4
        with:
          submodules: true

      - name: Configure AWS credentials
        uses: aws-actions/configure-aws-credentials@v4
        with:
          role-to-assume: ${{ secrets.AWS_ROLE_TO_ASSUME }}
          aws-region: ${{ vars.AWS_REGION }}

      - name: Update kube config
        run: aws eks update-kubeconfig --name "$EKS_CLUSTER_NAME" --region "$AWS_REGION"
        env:
          AWS_REGION: ${{ vars.AWS_REGION }}
          EKS_CLUSTER_NAME: dev

      - name: Scale up
        run: |
          set -x
          aws autoscaling set-desired-capacity --auto-scaling-group-name "$AUTOSCALING_GROUP" --desired-capacity "$DESIRED_CAPACITY"          
        env:
          AUTOSCALING_GROUP: ${{ vars.DEV_EKS_AS_GROUP }}
          DESIRED_CAPACITY: 1

      - name: Install the CRD and Operator
        run: |
          # Install the CRD and Operator
          kubectl apply -f https://raw.githubusercontent.com/dragonflydb/dragonfly-operator/main/manifests/dragonfly-operator.yaml          

      - name: Apply Configuration
        run: |
          set -x
          kubectl create namespace ${{ steps.setup.outputs.namespace }} || true
          echo '${{ toJson(matrix.config.operator) }}' | kubectl apply -n ${{ steps.setup.outputs.namespace }} -f -          

      - name: Wait For Service
        run: |
          set -x
          kubectl wait -n ${{ steps.setup.outputs.namespace }} dragonfly/dragonfly-sample --for=jsonpath='{.status.phase}'=ready --timeout=180s
          kubectl wait -n ${{ steps.setup.outputs.namespace }} pods --selector app=dragonfly-sample --for condition=Ready --timeout=120s
          kubectl describe -n ${{ steps.setup.outputs.namespace }} pod dragonfly-sample-0          

      - name: Run Memtier Benchmark
        shell: bash
        run: |
          kubectl apply -n ${{ steps.setup.outputs.namespace }} -f tools/benchmark/k8s-benchmark-job.yaml          

      - name: Version upgrade
        shell: bash
        run: |
          # benchmark is running, wait for 30 seconds before version upgrade
          sleep 30
          kubectl patch dragonfly dragonfly-sample -n ${{ steps.setup.outputs.namespace }}  --type merge -p '{"spec":{"image":"ghcr.io/dragonflydb/dragonfly-weekly:latest"}}'          

      - name: Wait for Memtier Benchmark fail
        shell: bash
        run: |
          # Memtier benchmark run will fail at some point because old master shutdown on version upgrade
          kubectl wait --for=condition=failed --timeout=120s -n ${{ steps.setup.outputs.namespace }} jobs/memtier-benchmark 2>/dev/null
          kubectl logs -n ${{ steps.setup.outputs.namespace }} -f jobs/memtier-benchmark
          kubectl delete -n ${{ steps.setup.outputs.namespace }} jobs/memtier-benchmark          

      - name: Run Memtier Benchmark again
        shell: bash
        run: |
          kubectl apply -n ${{ steps.setup.outputs.namespace }} -f tools/benchmark/k8s-benchmark-job.yaml

          while true; do
            if kubectl wait --for=condition=complete --timeout=0 -n ${{ steps.setup.outputs.namespace }} jobs/memtier-benchmark 2>/dev/null; then
              job_result=0
              break
            fi

            if kubectl wait --for=condition=failed --timeout=0 -n ${{ steps.setup.outputs.namespace }} jobs/memtier-benchmark 2>/dev/null; then
              job_result=1
              break
            fi

            sleep 3
          done

          kubectl logs -n ${{ steps.setup.outputs.namespace }} -f jobs/memtier-benchmark
          if [[ $job_result -eq 1 ]]; then
              exit 1
          fi          

      - name: Server checks
        run: |
          nohup kubectl port-forward -n ${{ steps.setup.outputs.namespace }} service/dragonfly-sample 6379:6379 &
          pip install -r tools/requirements.txt
          python3 tools/benchmark/post_run_checks.py          

      - name: Get Dragonfly logs
        uses: nick-fields/retry@v3
        if: always()
        with:
          timeout_minutes: 1
          max_attempts: 3
          command: |
            kubectl logs -n ${{ steps.setup.outputs.namespace }} dragonfly-sample-0            

      - name: Get Dragonfly replica logs
        uses: nick-fields/retry@v3
        if: always()
        with:
          timeout_minutes: 1
          max_attempts: 3
          command: |
            kubectl logs -n ${{ steps.setup.outputs.namespace }} dragonfly-sample-1            

      - name: Describe dragonflydb object
        uses: nick-fields/retry@v3
        if: always()
        with:
          timeout_minutes: 1
          max_attempts: 3
          command: |
            kubectl describe dragonflies.dragonflydb.io -n ${{ steps.setup.outputs.namespace }} dragonfly-sample            

      - name: Scale down to zero
        if: always()
        run: |
          set -x
          aws autoscaling set-desired-capacity --auto-scaling-group-name "$AUTOSCALING_GROUP" --desired-capacity 0          
        env:
          AUTOSCALING_GROUP: ${{ vars.DEV_EKS_AS_GROUP }}

      - name: Cleanup
        if: always()
        run: |
          set -x
          kubectl delete namespace ${{ steps.setup.outputs.namespace }}
          kubectl delete namespace dragonfly-operator-system          

      - name: Send notification on failure
        if: failure() && github.ref == 'refs/heads/main'
        shell: bash
        run: |
          job_link="${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}"
          message="Benchmark tests failed.\\n Job Link: ${job_link}\\n"

          curl -s \
            -X POST \
            -H 'Content-Type: application/json' \
            '${{ secrets.GSPACES_BOT_DF_BUILD }}' \
            -d '{"text": "'"${message}"'"}'