GitLab Pipeline trigger action

GitHub | GitHub Marketplace

This GitHub action triggers and waits for a GitLab pipeline to complete.

You can for example use this action in your GitHub workflow to trigger a deployment pipeline on a private GitLab server after a successful build pipeline and wait for the deployment (with possible End2End tests) to finish, so you would get a notification if the deployment failed.

sequenceDiagram
    participant GITHUB as GitHub Actions
    participant ACTION as Action
    participant GITLAB as GitLab CI

    GITHUB->>ACTION: Run action
    ACTION->>GITLAB: Trigger a new pipeline

loop Every 15 seconds
    ACTION->>GITLAB: Poll pipeline status
    GITLAB-->>ACTION: Current pipeline status
end

    opt download_artifacts is enabled
        ACTION->>GITLAB: Poll for artifacts
        GITLAB-->>ACTION: Download artifacts
        ACTION->>GITHUB: Save artifacts locally
    end

    ACTION-->>GITHUB: Final pipeline status

Loading

Inputs

host

The GitLab host to trigger the pipeline on. Default gitlab.com.

id

Required The ID or path of the project owned by the authenticated user. You will find the Project ID in the General Settings of your GitLab project.

ref

Required The branch or tag to run the pipeline on.

trigger_token

Required The GitLab pipeline trigger token to trigger the pipeline.

access_token

The GitLab pipeline access token to access the pipeline via the API. You need the read_api and read_repository scopes with Reporter role for this token.

For public projects you don't need to provide an access token.

Note: Required when download_artifacts is enabled.

variables

A map of key-valued strings containing the pipeline variables. For example: { VAR1: "value1", VAR2: "value2" }. The value has to be valid JSON. If not set the default is {}.

download_artifacts

Whether to download artifacts generated by the pipeline. Default false.

Note: Requires access_token to be set.

download_artifacts_on_failure

Whether to download artifacts even when the pipeline fails. Default true.

Note: Requires both access_token and download_artifacts to be enabled.

fail_if_no_artifacts

Whether to fail the action if no artifacts are found. Default false.

Note: Requires download_artifacts to be enabled. Useful for ensuring pipelines generate expected artifacts.

download_job_logs

Whether to download job console output. Default false.

Note: Requires access_token to be set. Downloads console output from all pipeline jobs independently of artifacts.

download_path

Path where to save downloaded artifacts and logs. Defaults to ./downloads.

verbose

Whether to enable verbose logging. When disabled, only generic messages are shown without potentially sensitive information like URLs, IDs, or names. Default false.

Outputs

status

The last status of the pipeline. See GitLab project pipelines for more information about which status values there are.

web_url

The URL of the pipeline, for example https://gitlab.com/foo/bar/pipelines/47.

artifacts_downloaded

Whether artifacts were successfully downloaded. Only set when download_artifacts is enabled.

Example usage

Basic usage

uses: digital-blueprint/gitlab-pipeline-trigger-action@v1
with:
    host: 'gitlab.example.com'
    trigger_token: ${{ secrets.DEPLOY_TRIGGER_TOKEN }}
    access_token: ${{ secrets.DEPLOY_ACCESS_TOKEN }}
    id: '123'
    ref: 'main'
    variables: '{"VAR1":"value1","VAR2":"value2"}'

With artifact downloading

uses: digital-blueprint/gitlab-pipeline-trigger-action@v1
with:
    host: 'gitlab.example.com'
    trigger_token: ${{ secrets.DEPLOY_TRIGGER_TOKEN }}
    access_token: ${{ secrets.DEPLOY_ACCESS_TOKEN }}
    id: '123'
    ref: 'main'
    variables: '{"VAR1":"value1","VAR2":"value2"}'
    download_artifacts: 'true'
    download_path: './build-artifacts'

With job logs

uses: digital-blueprint/gitlab-pipeline-trigger-action@v1
with:
    host: 'gitlab.example.com'
    trigger_token: ${{ secrets.DEPLOY_TRIGGER_TOKEN }}
    access_token: ${{ secrets.DEPLOY_ACCESS_TOKEN }}
    id: '123'
    ref: 'main'
    variables: '{"VAR1":"value1","VAR2":"value2"}'
    download_job_logs: 'true'
    download_path: './pipeline-logs'

Downloading artifacts and job logs

When download_artifacts is enabled and access_token is provided, the action will download artifacts based on the pipeline status:

• Pipeline succeeds: Artifacts are downloaded automatically.
• Pipeline fails: The artifacts are downloaded if download_artifacts_on_failure is enabled. Default true.

Artifact Validation:

• If fail_if_no_artifacts is enabled, the action will fail if no artifacts are found.
• This helps ensure pipelines generate expected outputs and catch configuration issues early.

Job Logs:

• If download_job_logs is enabled, console output from all jobs will be downloaded.
• Logs are saved as job.log files in each job's directory.
• This provides complete visibility into pipeline execution for debugging.
• Job log downloading is independent of artifact downloading.

The action will:

1. Fetch all jobs from the completed pipeline;
2. Identify jobs that have artifacts;
3. Download artifacts from each job if download_artifacts is enabled;
4. Extract the artifacts to the specified download_path;
5. Organize artifacts by job (each job gets its own subdirectory);
6. Download job console logs if download_job_logs is enabled;
7. Save logs as job.log files in each job's directory.

The artifacts and logs are organized as follows:

download_path/
├── job_123_job_name_1/
│   ├── artifacts/       # Artifacts subdirectory (if download_artifacts enabled)
│   │   ├── artifact1.txt
│   │   └── artifact2.txt
│   └── job.log          # Console output (if download_job_logs enabled)
├── job_124_job_name_2/
│   ├── artifacts/       # Artifacts subdirectory (if download_artifacts enabled)
│   │   ├── build/
│   │   └── dist/
│   └── job.log          # Console output (if download_job_logs enabled)
└── ...