benchmarks #977
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: benchmarks | |
| on: | |
| workflow_dispatch: | |
| inputs: | |
| runStandalone: | |
| description: 'Run the benchmarks against standalone APM Server with Moxy' | |
| required: false | |
| type: boolean | |
| default: false | |
| profile: | |
| description: 'The system profile used to run the benchmarks' | |
| required: false | |
| type: string | |
| runOnStable: | |
| description: 'Run the benchmarks on the latest stable version' | |
| required: false | |
| type: boolean | |
| default: false | |
| benchmarkAgents: | |
| description: 'Set the number of agents to send data to the APM Server' | |
| required: false | |
| type: string | |
| benchmarkRun: | |
| description: 'Set the expression that matches the benchmark scenarios to run' | |
| required: false | |
| type: string | |
| schedule: | |
| - cron: '0 17 * * *' # Scheduled regular benchmarks. | |
| - cron: '0 5 */5 * *' # Scheduled PGO benchmarks. | |
| env: | |
| PNG_REPORT_FILE: out.png | |
| BENCHMARK_CPU_OUT: default.pgo | |
| BENCHMARK_RESULT: benchmark-result.txt | |
| WORKING_DIRECTORY: testing/benchmark | |
| permissions: | |
| contents: read | |
| jobs: | |
| benchmarks: | |
| runs-on: ubuntu-latest | |
| defaults: | |
| run: | |
| working-directory: ${{ env.WORKING_DIRECTORY }} | |
| permissions: | |
| contents: write | |
| id-token: write | |
| env: | |
| SSH_KEY: ./id_rsa_terraform | |
| TF_VAR_private_key: ./id_rsa_terraform | |
| TF_VAR_public_key: ./id_rsa_terraform.pub | |
| TF_VAR_run_standalone: ${{ inputs.runStandalone || github.event.schedule=='0 5 */5 * *' }} | |
| RUN_STANDALONE: ${{ inputs.runStandalone || github.event.schedule=='0 5 */5 * *' }} | |
| TFVARS_SOURCE: ${{ inputs.profile || 'system-profiles/8GBx1zone.tfvars' }} # // Default to use an 8gb profile | |
| TF_VAR_BUILD_ID: ${{ github.run_id }} | |
| TF_VAR_ENVIRONMENT: ci | |
| TF_VAR_REPO: ${{ github.repository }} | |
| GOBENCH_TAGS: branch=${{ github.head_ref || github.ref }},commit=${{ github.sha }},target_branch=${{ github.base_ref }} | |
| GOBENCH_PASSWORD: ${{ secrets.GOBENCH_PASSWORD }} | |
| GOBENCH_USERNAME: ${{ secrets.GOBENCH_USERNAME }} | |
| GOBENCH_HOST: ${{ secrets.GOBENCH_HOST }} | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - uses: actions/setup-go@v5 | |
| with: | |
| go-version-file: 'go.mod' | |
| - uses: rlespinasse/github-slug-action@aba9f8db6ef36e0733227a62673d6592b1f430ea | |
| - name: Set up env | |
| run: | | |
| SLUGGED_BRANCH_NAME=${{ env.GITHUB_HEAD_REF_SLUG || env.GITHUB_REF_SLUG }} | |
| CREATED_AT=$(date +%s) | |
| echo "TF_VAR_BRANCH=${SLUGGED_BRANCH_NAME}" >> "$GITHUB_ENV" | |
| echo "TF_VAR_CREATED_AT=${CREATED_AT}" >> "$GITHUB_ENV" | |
| echo "USER=benchci-$SLUGGED_BRANCH_NAME-$CREATED_AT" >> "$GITHUB_ENV" | |
| if [ ! -z "${{ inputs.benchmarkAgents }}" ]; then | |
| echo "BENCHMARK_AGENTS=${{ inputs.benchmarkAgents }}" >> "$GITHUB_ENV" | |
| fi | |
| if [ ! -z "${{ inputs.benchmarkRun }}" ]; then | |
| echo "BENCHMARK_RUN=${{ inputs.benchmarkRun }}" >> "$GITHUB_ENV" | |
| fi | |
| - name: Log in to the Elastic Container registry | |
| uses: docker/login-action@9780b0c442fbb1117ed29e0efdff1e18412f7567 # v3.3.0 | |
| with: | |
| registry: ${{ secrets.ELASTIC_DOCKER_REGISTRY }} | |
| username: ${{ secrets.ELASTIC_DOCKER_USERNAME }} | |
| password: ${{ secrets.ELASTIC_DOCKER_PASSWORD }} | |
| - uses: elastic/oblt-actions/google/auth@v1 | |
| - uses: elastic/oblt-actions/aws/auth@v1 | |
| with: | |
| role-duration-seconds: 18000 # 5 hours | |
| - uses: google-github-actions/get-secretmanager-secrets@e5bb06c2ca53b244f978d33348d18317a7f263ce # v2.2.2 | |
| with: | |
| export_to_environment: true | |
| secrets: |- | |
| EC_API_KEY:elastic-observability/elastic-cloud-observability-team-pro-api-key | |
| - uses: hashicorp/setup-terraform@v3 | |
| with: | |
| terraform_version: 1.3.7 | |
| terraform_wrapper: false | |
| - name: Init terraform module | |
| id: init | |
| run: make init | |
| - name: Build apmbench | |
| run: make apmbench $SSH_KEY terraform.tfvars | |
| - name: Build APM Server and Moxy | |
| if: ${{ env.RUN_STANDALONE == 'true' }} | |
| run: | | |
| make apm-server | |
| make moxy | |
| - name: Override docker committed version | |
| if: ${{ ! inputs.runOnStable && env.RUN_STANDALONE == 'false' }} | |
| run: make docker-override-committed-version | |
| - name: Spin up benchmark environment | |
| id: deploy | |
| run: | | |
| make apply | |
| admin_console_url=$(terraform output -raw admin_console_url) | |
| echo "admin_console_url=$admin_console_url" >> "$GITHUB_OUTPUT" | |
| echo "-> infra setup done" | |
| - name: Run benchmarks autotuned | |
| if: ${{ inputs.benchmarkAgents == '' }} | |
| run: make run-benchmark-autotuned | |
| - name: Run benchmarks self tuned | |
| if: ${{ inputs.benchmarkAgents != '' }} | |
| run: make run-benchmark | |
| - name: Cat standalone server logs | |
| if: ${{ env.RUN_STANDALONE == 'true' && failure() }} | |
| run: make cat-apm-server-logs | |
| - name: Index benchmarks result | |
| run: make index-benchmark-results | |
| - name: Download PNG | |
| run: >- | |
| ${{ github.workspace }}/.ci/scripts/download-png-from-kibana.sh | |
| ${{ secrets.KIBANA_BENCH_ENDPOINT }} | |
| ${{ secrets.KIBANA_BENCH_USERNAME }} | |
| ${{ secrets.KIBANA_BENCH_PASSWORD }} | |
| $PNG_REPORT_FILE | |
| - name: Upload PNG | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| name: kibana-png-report | |
| path: ${{ env.WORKING_DIRECTORY }}/${{ env.PNG_REPORT_FILE }} | |
| if-no-files-found: error | |
| - name: Upload PNG to AWS S3 | |
| id: s3-upload-png | |
| env: | |
| AWS_DEFAULT_REGION: us-east-1 | |
| run: | | |
| DEST_NAME="github-run-id-${{ github.run_id }}.png" | |
| aws s3 --debug cp ${{ env.PNG_REPORT_FILE }} s3://elastic-apm-server-benchmark-reports/${DEST_NAME} | |
| echo "png_report_url=https://elastic-apm-server-benchmark-reports.s3.amazonaws.com/${DEST_NAME}" >> "$GITHUB_OUTPUT" | |
| - name: Upload benchmark result | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| name: benchmark-result | |
| path: ${{ env.WORKING_DIRECTORY }}/${{ env.BENCHMARK_RESULT }} | |
| if-no-files-found: error | |
| # The next section injects CPU profile collected by apmbench into the build. | |
| # By copying the profile, uploading it to the artifacts and pushing it | |
| # via a PR to update default.pgo. | |
| - name: Copy CPU profile | |
| run: make cp-cpuprof | |
| - name: Upload CPU profile | |
| uses: actions/upload-artifact@v4 | |
| with: | |
| name: cpu-profile | |
| path: ${{ env.WORKING_DIRECTORY }}/${{ env.BENCHMARK_CPU_OUT }} | |
| if-no-files-found: error | |
| - name: Get token | |
| id: get_token | |
| uses: tibdex/github-app-token@3beb63f4bd073e61482598c45c71c1019b59b73a # v2.1.0 | |
| with: | |
| app_id: ${{ secrets.OBS_AUTOMATION_APP_ID }} | |
| private_key: ${{ secrets.OBS_AUTOMATION_APP_PEM }} | |
| permissions: >- | |
| { | |
| "contents": "write", | |
| "pull_requests": "write" | |
| } | |
| # Required to use a service account, otherwise PRs created by | |
| # GitHub bot won't trigger any CI builds. | |
| # See https://github.com/peter-evans/create-pull-request/issues/48#issuecomment-537478081 | |
| - name: Configure git user | |
| uses: elastic/oblt-actions/git/setup@v1 | |
| with: | |
| github-token: ${{ steps.get_token.outputs.token }} | |
| - name: Import GPG key | |
| uses: crazy-max/ghaction-import-gpg@cb9bde2e2525e640591a934b1fd28eef1dcaf5e5 # v6.2.0 | |
| with: | |
| gpg_private_key: ${{ secrets.APM_SERVER_RELEASE_GPG_PRIVATE_KEY }} | |
| passphrase: ${{ secrets.APM_SERVER_RELEASE_PASSPHRASE }} | |
| git_user_signingkey: true | |
| git_commit_gpgsign: true | |
| - name: Open PGO PR | |
| if: ${{ env.RUN_STANDALONE == 'true' }} | |
| run: ${{ github.workspace }}/.ci/scripts/push-pgo-pr.sh | |
| env: | |
| WORKSPACE_PATH: ${{ github.workspace }} | |
| PROFILE_PATH: ${{ env.WORKING_DIRECTORY }}/${{ env.BENCHMARK_CPU_OUT }} | |
| GITHUB_TOKEN: ${{ steps.get_token.outputs.token }} | |
| WORKFLOW: ${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}/attempts/${{ github.run_attempt }} | |
| # Secrets are rotated daily, if the benchmarks run between the rotation window, then | |
| # there is a high chance things will stop working | |
| # This is trying to reduce the chances of that happening. | |
| # See https://github.com/elastic/observability-test-environments/actions/workflows/cluster-rotate-api-keys.yml | |
| - uses: google-github-actions/get-secretmanager-secrets@e5bb06c2ca53b244f978d33348d18317a7f263ce # v2.2.2 | |
| if: always() | |
| with: | |
| export_to_environment: true | |
| secrets: |- | |
| EC_API_KEY:elastic-observability/elastic-cloud-observability-team-pro-api-key | |
| - name: Tear down benchmark environment | |
| if: always() | |
| run: make init destroy | |
| # Notify failure to Slack only on schedule (nightly run) | |
| - if: failure() && github.event_name == 'schedule' | |
| uses: elastic/oblt-actions/slack/notify-result@v1 | |
| with: | |
| bot-token: ${{ secrets.SLACK_BOT_TOKEN }} | |
| channel-id: "#apm-server" | |
| message: Nightly APM Server benchmarks failed! SDH Duty assignee, please have a look and follow this <https://github.com/elastic/observability-dev/blob/main/docs/apm/apm-server/runbooks/benchmarks.md|Runbook>! | |
| # Notify result to Slack only on schedule (nightly run) | |
| - if: github.event_name == 'schedule' | |
| uses: slackapi/slack-github-action@37ebaef184d7626c5f204ab8d3baff4262dd30f0 # v1.27.0 | |
| env: | |
| SLACK_BOT_TOKEN: ${{ secrets.SLACK_BOT_TOKEN }} | |
| with: | |
| channel-id: "#apm-server" | |
| payload: | | |
| { | |
| "blocks": [ | |
| { | |
| "type": "section", | |
| "text": { | |
| "type": "mrkdwn", | |
| "text": "${{ github.event_name == 'schedule' && 'Nightly' || '' }} APM Server benchmarks succesfully executed!" | |
| }, | |
| "accessory": { | |
| "type": "button", | |
| "style": "primary", | |
| "text": { | |
| "type": "plain_text", | |
| "text": "Workflow Run #${{ github.run_id }}", | |
| "emoji": true | |
| }, | |
| "url": "${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}", | |
| "action_id": "workflow-run-button" | |
| } | |
| }, | |
| { | |
| "type": "image", | |
| "image_url": "${{ steps.s3-upload-png.outputs.png_report_url }}", | |
| "alt_text": "kibana-png-report" | |
| }, | |
| { | |
| "type": "actions", | |
| "elements": [ | |
| { | |
| "type": "button", | |
| "text": { | |
| "type": "plain_text", | |
| "text": "Benchmarks dashboard" | |
| }, | |
| "url": "${{ secrets.KIBANA_BENCH_DASHBOARD }}", | |
| "action_id": "kibana-dashboard-button" | |
| }, | |
| { | |
| "type": "button", | |
| "text": { | |
| "type": "plain_text", | |
| "text": "Elastic Cloud deployment" | |
| }, | |
| "url": "${{ steps.deploy.outputs.admin_console_url }}", | |
| "action_id": "admin-console-button" | |
| } | |
| ] | |
| } | |
| ] | |
| } |