Skip to content

github-actions.md

Latest commit

 

History

History
104 lines (84 loc) · 6.86 KB

github-actions.md

File metadata and controls

104 lines (84 loc) · 6.86 KB

system-tests expose an official reusable workflow that allow to create a simple workflow running system-tests

name: System Tests

on:
  pushes:
    branches:
      - main
  pull_request:
    branches:
      - "**"

jobs:
  # first, build the artifact to be tested. This part is related to your repo
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - run: ./build.sh
      - uses: actions/upload-artifact@bbbca2ddaa5d8feaa63e36b76fdaad77386f024f # v7.0.0
        with:
          name: binaries
          path: path/to/your/binary


  system-tests:
    needs:
      - build
    uses:  DataDog/system-tests/.github/workflows/system-tests.yml@main
      secrets: inherit
      permissions:
        contents: read
        id-token: write
      with:
        library: java
        binaries_artifact: binaries
        desired_execution_time: 900
        scenarios_groups: tracer-release
        skip_empty_scenarios: true

Parameters

Name Description Type Required Default
artifact_retention_days How long should artifacts produced by the workflow should be retained number false 14
binaries_artifact Artifact name containing the binaries to test string false empty
display_summary Display a workflow summary containing owners of failed tests boolean false false
desired_execution_time In seconds, system-tests will try to respect this time budget. number false empty
excluded_scenarios Comma-separated list of scenarios not to run string false empty
force_execute Comma-separated list of tests to run even if they are skipped by manifest or decorators string false empty
library Library to test string true —
parametric_job_count How many jobs should be used to run PARAMETRIC scenario number false 1
push_to_test_optimization Push tests results to DataDog Test Optimization. Uses TEST_OPTIMIZATION_API_KEY secret if set, otherwise fetches credentials automatically via dd-sts (recommended) boolean false false
test_optimization_datadog_site DataDog site to use for test optimization string false datadoghq.com
ref system-tests ref to run the tests on (can be any valid branch, tag or SHA in system-tests repo) string false main
scenarios Comma-separated list scenarios to run string false DEFAULT
scenarios_groups Comma-separated list of scenarios groups to run string false empty
skip_empty_scenarios Skip scenarios that contain only xfail or irrelevant tests boolean false false

Permissions

The following permissions are always required:

Permission Reason
contents: read Checkout the repository
id-token: write Required by downstream workflows to fetch Datadog credentials via dd-sts. GitHub validates this upfront, so it must be granted even if test optimization is disabled

Secrets

For some purposes, secrets are used in the workflow:

Name Description
CIRCLECI_TOKEN Only used with _system_tests_dev_mode
DD_API_KEY Some scenarios requires valid API and APP keys
DD_APPLICATION_KEY
DD_API_KEY_2
DD_APP_KEY_2
DD_API_KEY_3
DD_APP_KEY_3
DOCKERHUB_USERNAME and DOCKERHUB_TOKEN If both are set, all docker pull are authenticated, which offer higher rate limit
TEST_OPTIMIZATION_API_KEY The DD_API_KEY to use to push tests runs to DataDog Test Optimization. Deprecated: prefer not setting this and letting dd-sts fetch credentials automatically

You can sends them ,either by using secrets: inherit (doc), or use explicit secret ids

Private parameters

Those parameters are used only by system-tests own CI

Name Description Type Required Default
_system_tests_dev_mode Shall we run system tests in dev mode (library and agent dev binary) boolean false false
_system_tests_library_target_branch_map If system-tests dev mode, JSON map of library to branch (e.g. {"java":"b1","dotnet":"b2"}) string false
_build_buddies_images Shall we build buddies images boolean false false
_build_proxy_image Shall we build proxy image boolean false false
_build_lambda_proxy_image Shall we build the lambda-proxy image boolean false false
_build_python_base_images Shall we build python base images for tests on python tracer boolean false false
_enable_replay_scenarios Enable replay scenarios, should only be used in system-tests CI boolean false false