performance.yml

name: Performance & Benchmark Pipeline

# Unified performance pipeline (#701).
# Replaces benchmark.yml, performance-regression.yml, and webhook-benchmark.yml
# with a single matrix-driven workflow using shared composite actions:
#   - setup-rust / setup-perf-env for environment provisioning
#   - compare-benchmarks for baseline comparison

on:
  push:
    branches: [main]
    paths:
      - "src/**"
      - "benchmarks/**"
      - ".github/workflows/performance.yml"
      - ".github/actions/setup-perf-env/**"
      - ".github/actions/compare-benchmarks/**"
  pull_request:
    branches: [main]
    paths:
      - "src/**"
      - "benchmarks/**"
      - ".github/workflows/performance.yml"
  workflow_dispatch:
    inputs:
      suite:
        description: "Benchmark suite to run (all, operator, regression, webhook)"
        required: false
        default: all
      baseline_version:
        description: Baseline version to compare
        required: false
        default: latest
      regression_threshold:
        description: Regression threshold %
        required: false
        default: "10"

env:
  CARGO_TERM_COLOR: always
  RUST_BACKTRACE: 1
  K6_VERSION: "0.54.0"

jobs:
  # ── Resolve which suites to run ───────────────────────────────────────────
  resolve-matrix:
    name: Resolve Benchmark Matrix
    runs-on: ubuntu-latest
    outputs:
      suites: ${{ steps.matrix.outputs.suites }}
    steps:
      - id: matrix
        shell: bash
        run: |
          SUITE="${{ github.event.inputs.suite || 'all' }}"
          case "$SUITE" in
            operator)    SUITES='["operator"]' ;;
            regression)  SUITES='["regression"]' ;;
            webhook)     SUITES='["webhook"]' ;;
            *)           SUITES='["operator","regression","webhook"]' ;;
          esac
          echo "suites=$SUITES" >> "$GITHUB_OUTPUT"

  # ── Build once, share across all suites ─────────────────────────────────────
  build:
    name: Build Operator
    runs-on: ubuntu-latest
    outputs:
      version: ${{ steps.version.outputs.version }}
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Get version
        id: version
        run: |
          if [[ "${{ github.ref }}" == refs/tags/* ]]; then
            echo "version=${GITHUB_REF#refs/tags/}" >> "$GITHUB_OUTPUT"
          else
            echo "version=sha-$(git rev-parse --short HEAD)" >> "$GITHUB_OUTPUT"
          fi
name: Continuous Benchmarking

on:
  push:
    branches: [ main ]
  pull_request:
    branches: [ main ]
  workflow_dispatch:

env:
  CARGO_TERM_COLOR: always
  CLUSTER_NAME: stellar-perf
  OPERATOR_NAMESPACE: stellar-system
  IMAGE_TAG: perf-test

jobs:
  performance:
    name: Operator Performance Benchmarks
    runs-on: ubuntu-latest
    timeout-minutes: 30

    steps:
      - uses: actions/checkout@v4

      - name: Setup Rust
        uses: ./.github/actions/setup-rust
        with:
          cache-key: "perf-build"

      - name: Build release
        run: cargo build --release --locked

      - name: Build Docker image
        run: docker build -t "stellar-operator:${{ steps.version.outputs.version }}" .

      - name: Save Docker image
        run: |
          docker save "stellar-operator:${{ steps.version.outputs.version }}" \
            | gzip > stellar-operator-image.tar.gz

      - name: Upload artifacts
        uses: actions/upload-artifact@v4
        with:
          name: perf-artifacts-${{ steps.version.outputs.version }}
          path: |
            target/release/stellar-operator
            stellar-operator-image.tar.gz
          retention-days: 7

  # ── Matrix: operator / regression / webhook ─────────────────────────────────
  benchmark:
    name: "Benchmark (${{ matrix.suite }})"
    runs-on: ubuntu-latest
    needs: [build, resolve-matrix]
    strategy:
      fail-fast: false
      matrix:
        suite: ${{ fromJSON(needs.resolve-matrix.outputs.suites) }}
    steps:
      - uses: actions/checkout@v4

      - name: Download artifacts
        uses: actions/download-artifact@v4
        with:
          name: perf-artifacts-${{ needs.build.outputs.version }}

      - name: Load Docker image
        if: matrix.suite != 'webhook'
        run: |
          gunzip stellar-operator-image.tar.gz
          docker load < stellar-operator-image.tar

      # ── Operator & regression: kind cluster via setup-perf-env ──────────────
      - name: Setup performance environment
        if: matrix.suite != 'webhook'
        continue-on-error: true
        uses: ./.github/actions/setup-perf-env
        with:
          k6-version: ${{ env.K6_VERSION }}
          cluster-name: perf-${{ matrix.suite }}
          operator-image-tag: ${{ needs.build.outputs.version }}
          worker-nodes: "2"

      # ── Webhook: run server directly (no kind) ──────────────────────────────
      - name: Install k6 (webhook suite)
        if: matrix.suite == 'webhook'
        run: |
          K6_VER="${{ env.K6_VERSION }}"
          curl -fsSL \
            "https://github.com/grafana/k6/releases/download/v${K6_VER}/k6-v${K6_VER}-linux-amd64.tar.gz" \
            | tar xvz
          sudo mv "k6-v${K6_VER}-linux-amd64/k6" /usr/local/bin/
          sudo apt-get update && sudo apt-get install -y jq bc

      - name: Start webhook server
        if: matrix.suite == 'webhook'
        run: |
          chmod +x target/release/stellar-operator
          ./target/release/stellar-operator webhook \
            --bind 0.0.0.0:8443 --log-level info &
          echo "WEBHOOK_PID=$!" >> "$GITHUB_ENV"
          for i in $(seq 1 30); do
            curl -sf http://localhost:8443/health > /dev/null 2>&1 && break
            sleep 2
          done
          curl -sf http://localhost:8443/health \
            || { echo "ERROR: Webhook server failed to start"; exit 1; }

      - name: Determine baseline
        id: baseline
        shell: bash
        run: |
          mkdir -p results
          INPUT_BASELINE="${{ github.event.inputs.baseline_version || 'latest' }}"
          case "${{ matrix.suite }}" in
            webhook)
              DEFAULT="webhook-v0.1.0"
              [[ "$INPUT_BASELINE" != "latest" ]] && BASELINE="benchmarks/baselines/${INPUT_BASELINE}.json" \
                || BASELINE="benchmarks/baselines/webhook-v0.1.0.json"
              CURRENT="results/webhook-benchmark.json"
              ;;
            *)
              if [[ "$INPUT_BASELINE" == "latest" ]]; then
                BASELINE=$(ls -t benchmarks/baselines/v*.json 2>/dev/null | head -1 || echo "benchmarks/baselines/v0.1.0.json")
              else
                BASELINE="benchmarks/baselines/${INPUT_BASELINE}.json"
              fi
              CURRENT="results/benchmark-summary.json"
              ;;
          esac
          echo "baseline_file=$BASELINE" >> "$GITHUB_OUTPUT"
          echo "current_file=$CURRENT" >> "$GITHUB_OUTPUT"
          [[ -f "$BASELINE" ]] && echo "has_baseline=true" >> "$GITHUB_OUTPUT" \
            || echo "has_baseline=false" >> "$GITHUB_OUTPUT"

      - name: Run operator benchmark (k6)
        if: matrix.suite == 'operator'
        continue-on-error: true
        run: |
          k6 run \
            --env BASE_URL=http://localhost:8080 \
            --env K8S_API_URL=http://localhost:8001 \
            --env NAMESPACE=stellar-benchmark \
            --env RUN_ID=${{ github.run_id }} \
            --env VERSION=${{ needs.build.outputs.version }} \
            --env GIT_SHA=${{ github.sha }} \
            --env BASELINE_FILE=${{ steps.baseline.outputs.baseline_file }} \
            --out json=results/k6-output.json \
            benchmarks/k6/operator-load-test.js
          python3 -c "
          import json, os
          raw = 'results/k6-output.json'
          data = []
          if os.path.exists(raw):
              with open(raw) as f:
                  data = [json.loads(l) for l in f if l.strip()]
          summary = {'total_requests': len([d for d in data if d.get('type') == 'Point']),
                     'version': '${{ needs.build.outputs.version }}'}
          with open('results/benchmark-summary.json', 'w') as f:
              json.dump(summary, f, indent=2)
          "

      - name: Run regression benchmark (k6)
        if: matrix.suite == 'regression'
        continue-on-error: true
        run: |
          k6 run \
            --env BASE_URL=http://localhost:8080 \
            --env K8S_API_URL=http://localhost:8001 \
            --env NAMESPACE=stellar-benchmark \
            --env RUN_ID=${{ github.run_id }} \
            --env VERSION=${{ needs.build.outputs.version }} \
            --env GIT_SHA=${{ github.sha }} \
            --env BASELINE_FILE=${{ steps.baseline.outputs.baseline_file }} \
            --out json=results/operator-benchmark-raw.json \
            benchmarks/k6/operator-load-test.js
          python3 -c "
          import json, os
          raw = 'results/operator-benchmark-raw.json'
          data = []
          if os.path.exists(raw):
              with open(raw) as f:
                  data = [json.loads(l) for l in f if l.strip()]
          summary = {'total_requests': len([d for d in data if d.get('type') == 'Point']),
                     'version': '${{ needs.build.outputs.version }}'}
          with open('results/benchmark-summary.json', 'w') as f:
              json.dump(summary, f, indent=2)
          "

      - name: Run webhook benchmark (k6)
        if: matrix.suite == 'webhook'
        continue-on-error: true
        run: |
          k6 run \
            --env WEBHOOK_URL=http://localhost:8443 \
            --env VERSION=${{ needs.build.outputs.version }} \
            --env GIT_SHA=${{ github.sha }} \
            --env RUN_ID=${{ github.run_id }} \
            --env BASELINE_FILE=${{ steps.baseline.outputs.baseline_file }} \
            --out json=results/webhook-benchmark-raw.json \
            benchmarks/k6/webhook-load-test.js

      - name: Stop webhook server
        if: always() && matrix.suite == 'webhook'
        run: '[[ -n "${WEBHOOK_PID:-}" ]] && kill "$WEBHOOK_PID" || true'

      - name: Compare with baseline
        if: steps.baseline.outputs.has_baseline == 'true' && matrix.suite != 'webhook'
        continue-on-error: true
        uses: ./.github/actions/compare-benchmarks
        with:
          current-file: ${{ steps.baseline.outputs.current_file }}
          baseline-file: ${{ steps.baseline.outputs.baseline_file }}
          threshold: ${{ github.event.inputs.regression_threshold || '10' }}
          output-file: results/regression-report-${{ matrix.suite }}.json
          fail-on-regression: "false"

      - name: Collect logs
        if: always() && matrix.suite != 'webhook'
        run: |
          kubectl logs -n stellar-system -l app=stellar-operator \
            --tail=500 > results/operator-logs-${{ matrix.suite }}.txt || true

      - name: Upload results
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: perf-results-${{ matrix.suite }}-${{ needs.build.outputs.version }}
          path: results/
          retention-days: 30

  # ── Aggregate report ────────────────────────────────────────────────────────
  report:
    name: Performance Report
    runs-on: ubuntu-latest
    needs: [build, benchmark]
    if: always() && needs.build.result == 'success'
    permissions:
      pull-requests: write
    steps:
      - uses: actions/checkout@v4

      - name: Download all benchmark results
        uses: actions/download-artifact@v4
        continue-on-error: true
        with:
          pattern: perf-results-*
          merge-multiple: true
          path: results/

      - name: Write job summary
        if: always()
        run: |
          mkdir -p results
          {
            echo "## Performance & Benchmark Pipeline (#701)"
            echo ""
            echo "**Version:** ${{ needs.build.outputs.version }}"
            echo "**Commit:** ${{ github.sha }}"
            echo ""
            for report in results/regression-report-*.json; do
              [[ -f "$report" ]] || continue
              SUITE=$(basename "$report" .json | sed 's/regression-report-//')
              PASSED=$(jq -r '.overall_passed // "unknown"' "$report")
              echo "### $SUITE: $PASSED"
            done
          } >> "$GITHUB_STEP_SUMMARY"

      - name: Post PR comment
        if: github.event_name == 'pull_request'
        continue-on-error: true
        uses: peter-evans/create-or-update-comment@v5
        with:
          issue-number: ${{ github.event.pull_request.number }}
          body: |
            ## Performance & Benchmark Pipeline

            **Version:** ${{ needs.build.outputs.version }}
            **Commit:** ${{ github.sha }}

            Unified matrix pipeline completed for operator, regression, and webhook suites.
            See workflow artifacts for detailed results.
      - name: Build operator binary
        run: cargo build --release --locked --bin stellar-operator

      - name: Setup kind cluster
        uses: ./.github/actions/setup-kind-cluster
        with:
          cluster-name: ${{ env.CLUSTER_NAME }}
          image-tag: ${{ env.IMAGE_TAG }}
          operator-namespace: ${{ env.OPERATOR_NAMESPACE }}
          worker-nodes: "1"
          operator-cpu-request: "500m"
          operator-memory-request: "512Mi"

      - name: Install k6
        run: |
          sudo gpg --no-default-keyring --keyring /usr/share/keyrings/k6-archive-keyring.gpg --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys C5AD17C747E3415A3642D57D77C6C491D6AC1D69
          echo "deb [signed-by=/usr/share/keyrings/k6-archive-keyring.gpg] https://dl.k6.io/deb stable main" | sudo tee /etc/apt/sources.list.d/k6.list
          sudo apt-get update
          sudo apt-get install k6

      - name: Run Benchmarks
        env:
          OPERATOR_NAMESPACE: ${{ env.OPERATOR_NAMESPACE }}
          RUN_ID: ${{ github.run_id }}
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
        run: |
          # Set up port forwarding to operator
          kubectl port-forward -n ${{ env.OPERATOR_NAMESPACE }} deployment/stellar-operator 8080:8080 &
          PF_PID=$!
          
          # Set up proxy for K8s API
          kubectl proxy --port=8001 &
          PROXY_PID=$!
          
          sleep 5
          
          mkdir -p results
          
          # Run k6 benchmarks
          k6 run \
            --env BASE_URL=http://localhost:8080 \
            --env K8S_API_URL=http://localhost:8001 \
            --env NAMESPACE=perf-test \
            --env RUN_ID=${{ github.run_id }} \
            --out json=results/benchmarks.json \
            benchmarks/k6/comprehensive-perf.js
          
          kill $PF_PID $PROXY_PID
          
      - name: Compare with Baseline
        id: compare
        run: |
          python3 benchmarks/scripts/compare_benchmarks.py compare \
            --current results/benchmarks.json \
            --baseline benchmarks/baselines/v0.1.0.json \
            --threshold 15 \
            --output results/comparison.json \
            --fail-on-regression
        continue-on-error: true

      - name: Generate Performance Report
        run: |
          # Use a simple python script to generate an HTML report from the JSON
          python3 benchmarks/scripts/generate_report.py \
            --results results/benchmarks.json \
            --comparison results/comparison.json \
            --output results/report.html

      - name: Upload Performance Results
        uses: actions/upload-artifact@v4
        with:
          name: performance-results-${{ github.run_id }}
          path: results/
          retention-days: 30

      - name: Comment PR with Performance Summary
        if: github.event_name == 'pull_request'
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            const comparison = JSON.parse(fs.readFileSync('results/comparison.json', 'utf8'));
            
            let body = `### 📊 Performance Benchmark Results\n\n`;
            body += `${comparison.summary}\n\n`;
            
            if (comparison.regressions.length > 0) {
              body += `#### ⚠️ Regressions Detected\n\n`;
              body += `| Metric | Current | Baseline | Change |\n`;
              body += `| :--- | :--- | :--- | :--- |\n`;
              comparison.regressions.forEach(r => {
                body += `| ${r.metric} | ${r.current} | ${r.baseline} | ${r.change_pct}% |\n`;
              });
              body += `\n`;
            }
            
            if (comparison.improvements.length > 0) {
              body += `#### ✅ Improvements\n\n`;
              body += `| Metric | Current | Baseline | Change |\n`;
              body += `| :--- | :--- | :--- | :--- |\n`;
              comparison.improvements.forEach(i => {
                body += `| ${i.metric} | ${i.current} | ${i.baseline} | ${i.change_pct}% |\n`;
              });
              body += `\n`;
            }
            
            github.rest.issues.createComment({
              issue_number: context.issue.number,
              owner: context.repo.owner,
              repo: context.repo.repo,
              body: body
            });