rustfs/.github/workflows/rustfs-performance-test.yml

name: RustFS Performance Test

on:
  workflow_dispatch:
    inputs:
      package_url:
        description: 'Direct .deb URL (nightly/R2). Defaults to the latest nightly deb.'
        required: false
        type: string
      test_method:
        description: 'Benchmark method(s) to run (manual runs only; "all" = GET+PUT+MIXED)'
        type: choice
        options:
          - all
          - get
          - put
          - mixed
        default: 'all'
      object_size:
        description: 'Object size(s) to test (manual runs only; "all" = all 10 sizes)'
        type: choice
        options:
          - all
          - 1KiB
          - 4KiB
          - 16KiB
          - 128KiB
          - 1MiB
          - 4MiB
          - 8MiB
          - 16MiB
          - 32MiB
          - 64MiB
        default: 'all'
      warp_duration:
        description: 'warp duration per round (e.g. 5m, 30s)'
        required: false
        default: '5m'
      warp_concurrency:
        description: 'warp concurrency'
        required: false
        default: '64'
      cleanup_before:
        description: 'Reset the nodes before the test (DESTROYS existing data/config)'
        type: boolean
        default: true
      cleanup_after:
        description: 'Reset the nodes after the test (DESTROYS test data/config)'
        type: boolean
        default: true
  repository_dispatch:
    # Chain entry: dispatched by rustfs-functional-chain.yml (runs on its own
    # pf-testing runner, in parallel with the shared-VM chain).
    types: [rustfs-chain-performance]

permissions:
  contents: read

# Dedicated pf-testing runner/environment: own concurrency group so perf runs
# never block (or are blocked by) the pool-expansion / heal tests.
concurrency:
  group: rustfs-performance-test
  cancel-in-progress: false

defaults:
  run:
    shell: bash

env:
  RUSTFS_ACCESS_KEY: ${{ secrets.RUSTFS_ACCESS_KEY }}
  RUSTFS_SECRET_KEY: ${{ secrets.RUSTFS_SECRET_KEY }}
  # Performance test uses its own node list (4 nodes); the shared
  # RUSTFS_NODES secret is used by the 3-node pool-expansion / heal tests.
  RUSTFS_NODES: ${{ secrets.RUSTFS_PERF_NODES || vars.RUSTFS_PERF_NODES || 'vm000 vm001 vm002 vm003' }}
  RUSTFS_SSH_USER: ${{ secrets.RUSTFS_SSH_USER || vars.RUSTFS_SSH_USER }}
  # Package used by the nightly run (workflow_dispatch inputs are empty for
  # workflow_run events), i.e. the latest nightly deb published by nightly-gnu.yml.
  RUSTFS_NIGHTLY_PACKAGE_URL: ${{ vars.RUSTFS_NIGHTLY_PACKAGE_URL || 'https://dl.rustfs.com/artifacts/rustfs/packages/nightly/rustfs-nightly-latest.deb' }}
  # Fixed benchmark result directory so later steps can read summary.md
  RUSTFS_RESULT_DIR: /tmp/rustfs-perf-results
  # Cross-repo token for uploading reports to rustfs/dashboard (set in repo settings)
  PF_TESTING_GH_TOKEN: ${{ secrets.PF_TESTING_GH_TOKEN }}

jobs:
  performance-test:
    runs-on: pf-testing
    # Requirement: a failing benchmark must not fail the workflow;
    # failures are filed to rustfs/backlog.
    continue-on-error: true
    timeout-minutes: 900
    # Run on manual dispatch, or when the nightly build completed successfully.
    # Skipped when nightly failed.
    if: ${{ github.event_name == 'workflow_dispatch' || github.event_name == 'repository_dispatch' }}
    steps:
      # auto-testing is private: clone it with the dedicated PF token (not
      # GITHUB_TOKEN) and retry transient GitHub/network failures.
      - name: Checkout auto-testing scripts (with retry)
        env:
          GH_TOKEN: ${{ secrets.PF_TESTING_GH_TOKEN }}
        run: |
          set -euo pipefail
          rm -rf auto-testing
          for attempt in 1 2 3 4 5; do
            if gh repo clone rustfs/auto-testing auto-testing -- --depth 1 --quiet; then
              echo "auto-testing cloned (attempt ${attempt})"
              exit 0
            fi
            rm -rf auto-testing
            echo "clone attempt ${attempt} failed; retrying in $((attempt * 15))s" >&2
            sleep $((attempt * 15))
          done
          echo "ERROR: unable to clone rustfs/auto-testing after 5 attempts" >&2
          exit 1

      - name: Show environment
        run: |
          uname -a
          jq --version
          warp --version || true
          df -h /data | tail -1

      - name: Reset test environment (before)
        if: ${{ inputs.cleanup_before != 'false' }}
        run: |
          chmod +x auto-testing/rustfs_performance_test.sh
          ./auto-testing/rustfs_performance_test.sh --step 1 -y

      - name: Install RustFS package & start cluster (4x4)
        run: |
          ARGS=(--steps "2,3,4" -y)
          if [ -n "${{ inputs.package_url }}" ]; then
            ARGS+=(--package-url "${{ inputs.package_url }}")
          else
            ARGS+=(--package-url "${{ env.RUSTFS_NIGHTLY_PACKAGE_URL }}")
          fi
          ./auto-testing/rustfs_performance_test.sh "${ARGS[@]}"

      - name: Preflight checks
        run: |
          ARGS=(--preflight)
          if [ -n "${{ inputs.package_url }}" ]; then
            ARGS+=(--package-url "${{ inputs.package_url }}")
          else
            ARGS+=(--package-url "${{ env.RUSTFS_NIGHTLY_PACKAGE_URL }}")
          fi
          ./auto-testing/rustfs_performance_test.sh "${ARGS[@]}"

      - name: Run benchmark (GET/PUT/MIXED)
        id: benchmark
        run: |
          # Empty on automatic (workflow_run) runs -> full 30 rounds.
          # Manual dispatch can restrict method(s)/size(s).
          export WARP_METHODS="${{ inputs.test_method }}"
          export WARP_SIZES="${{ inputs.object_size }}"
          ./auto-testing/rustfs_performance_test.sh \
            --step 5 -y \
            --warp-duration "${{ inputs.warp_duration || '5m' }}" \
            --warp-concurrency "${{ inputs.warp_concurrency || '64' }}" \
            --log-file /tmp/rustfs-perf-test.log

      - name: Analyze results
        if: ${{ steps.benchmark.conclusion == 'success' }}
        run: |
          ./auto-testing/rustfs_performance_test.sh --step 6 -y

      - name: Collect RustFS version info
        if: ${{ steps.benchmark.conclusion == 'success' }}
        env:
          VERSION_FILE: /tmp/rustfs-version.txt
        run: |
          set -euo pipefail
          read -r -a NODES <<< "${RUSTFS_NODES}"
          [ "${#NODES[@]}" -gt 0 ] || { echo "RUSTFS_NODES is empty"; exit 1; }
          NODE="${NODES[0]}"
          SSH_USER="${RUSTFS_SSH_USER:-azureuser}"
          {
            echo "Node: ${NODE}"
            echo "Command: rustfs --version"
            echo ""
            ssh -o BatchMode=yes -o ConnectTimeout=10 -o StrictHostKeyChecking=accept-new \
              "${SSH_USER}@${NODE}" 'rustfs --version'
          } > "${VERSION_FILE}"

      - name: Upload report to dashboard (reports/YYYY-MM-DD.md)
        if: ${{ steps.benchmark.conclusion == 'success' }}
        env:
          GH_TOKEN: ${{ env.PF_TESTING_GH_TOKEN }}
          RESULT_DIR: ${{ env.RUSTFS_RESULT_DIR }}
          VERSION_FILE: /tmp/rustfs-version.txt
        run: |
          set -euo pipefail
          if [ -z "${GH_TOKEN:-}" ]; then
            echo "PF_TESTING_GH_TOKEN is not configured; skipping report upload"
            exit 0
          fi
          SUMMARY="${RESULT_DIR}/summary.md"
          [ -f "${SUMMARY}" ] || { echo "summary.md not found at ${SUMMARY}"; exit 1; }
          DATE="$(date -u +%Y-%m-%d)"
          REPORT_PATH="reports/${DATE}.md"
          {
            echo "# RustFS nightly build performance testing report"
            echo ""
            echo "- **Date**: ${DATE}"
            echo "- **Run**: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}"
            echo "- **Trigger**: ${{ github.event_name }}"
            echo "- **Package**: ${{ inputs.package_url || 'nightly (R2 latest)' }}"
            echo ""
            cat "${SUMMARY}"
            echo ""
            echo "## RustFS version"
            echo '```text'
            cat "${VERSION_FILE}"
            echo '```'
          } > /tmp/rustfs-perf-report.md
          CONTENT="$(python3 -c 'import base64; print(base64.b64encode(open("/tmp/rustfs-perf-report.md","rb").read()).decode())')"
          SHA="$(gh api "repos/rustfs/dashboard/contents/${REPORT_PATH}" -q '.sha' 2>/dev/null || true)"
          if [ -n "${SHA}" ]; then
            jq -n --arg msg "report: ${DATE}" --arg content "${CONTENT}" --arg sha "${SHA}" \
              '{message:$msg, content:$content, sha:$sha}' \
              | gh api --method PUT "repos/rustfs/dashboard/contents/${REPORT_PATH}" --input - >/dev/null
            echo "updated ${REPORT_PATH} in rustfs/dashboard"
          else
            jq -n --arg msg "report: ${DATE}" --arg content "${CONTENT}" \
              '{message:$msg, content:$content}' \
              | gh api --method PUT "repos/rustfs/dashboard/contents/${REPORT_PATH}" --input - >/dev/null
            echo "created ${REPORT_PATH} in rustfs/dashboard"
          fi

      - name: File failure issue in rustfs/backlog
        if: ${{ always() && (failure() || steps.benchmark.outcome == 'failure' || steps.benchmark.outcome == 'cancelled') }}
        continue-on-error: true
        env:
          GH_TOKEN: ${{ secrets.PF_TESTING_GH_TOKEN }}
          SUITE: 'performance'
          SUITE_LABEL: 'Performance'
          RUN_URL: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}
          REPORT_FILE: '/tmp/rustfs-perf-report.md'
          LOG_FILE: '/tmp/rustfs-perf-test.log'
        run: |
          set -euo pipefail
          if [ -z "${GH_TOKEN:-}" ]; then
            echo "PF_TESTING_GH_TOKEN is not configured; skipping backlog issue"
            exit 0
          fi
          TITLE="[functional][${SUITE}] ${SUITE_LABEL} suite failed (run ${GITHUB_RUN_ID})"
          EXISTING="$(gh issue list -R rustfs/backlog --state all \
            --search "in:title \"run ${GITHUB_RUN_ID}\"" \
            --json number --jq '.[].number' || true)"
          if [ -n "${EXISTING}" ]; then
            echo "backlog issue already exists for run ${GITHUB_RUN_ID}; skipping"
            exit 0
          fi
          redact() {
            sed -E \
              -e 's/(RUSTFS_(ACCESS_KEY|SECRET_KEY)[=: ]+)[^[:space:]]+/\1[REDACTED]/Ig' \
              -e 's/(Authorization:).*/\1 [REDACTED]/Ig' \
              -e 's/(X-Amz-Signature=)[^&[:space:]]+/\1[REDACTED]/Ig' \
              -e 's/^.*(password|secret|token)[=: ].*/[REDACTED SENSITIVE LINE]/Ig'
          }
          BODY_FILE="$(mktemp)"
          {
            echo "The **${SUITE_LABEL}** functional suite failed."
            echo ""
            echo "- Suite: \`${SUITE}\`"
            echo "- Run: ${RUN_URL}"
            echo "- Trigger: ${GITHUB_EVENT_NAME}"
            echo "- Date: $(date -u +%Y-%m-%d)"
            echo ""
            echo "## Report (errors and symptoms)"
            echo ""
            if [ -s "${REPORT_FILE}" ]; then
              redact < "${REPORT_FILE}"
            elif [ -s "${LOG_FILE:-}" ]; then
              echo "(report file missing; log tail below)"
              echo ""
              tail -n 200 "${LOG_FILE}" | redact
            else
              echo "(no report or log file was produced)"
            fi
          } | head -c 55000 > "${BODY_FILE}"
          gh label create functional-test -R rustfs/backlog --color d73a4a 2>/dev/null || true
          if ! gh issue create -R rustfs/backlog --title "${TITLE}" \
              --body-file "${BODY_FILE}" --label functional-test; then
            gh issue create -R rustfs/backlog --title "${TITLE}" --body-file "${BODY_FILE}"
          fi
          echo "filed backlog issue for suite ${SUITE}"

      - name: Upload test logs & results
        if: always()
        uses: actions/upload-artifact@b7c566a772e6b6bfb58ed0dc250532a479d7789f # v6
        with:
          name: rustfs-perf-test-${{ github.run_id }}
          path: |
            /tmp/rustfs-perf-test*.log
            /tmp/rustfs-perf-results/**
            /tmp/rustfs-version.txt
          if-no-files-found: warn

      - name: Reset test environment (after)
        if: ${{ always() && inputs.cleanup_after != 'false' }}
        run: |
          ./auto-testing/rustfs_performance_test.sh --step 7 -y

      - name: Notify on failure
        if: failure()
        run: |
          echo "RustFS performance test failed"
          echo "Package source: ${{ inputs.package_url || 'nightly (R2 latest)' }}"
          echo "See the uploaded log artifact for details."