Microsoft GH-200 - GitHub Actions Exam

The correct native GitHub Actions method for passing files between jobs in the same workflow is artifact storage. Workflow artifacts are designed to persist files produced by one job so they can be downloaded and used by another job later in the workflow. Job outputs are better suited for passing small values, strings, IDs, flags, or metadata between jobs, not full files. Caching is intended to speed up repeated workflow runs by reusing dependencies or build outputs; it is not the proper mechanism for transferring job-produced files. An external blob store can work architecturally, but it is not the native GitHub Actions method. Therefore, artifact storage is the best and most exam-aligned answer for passing files between jobs.

Searching the GitHub Marketplace for verified actions published by the cloud provider is the quickest and safest approach. Many cloud providers offer verified GitHub Actions that are maintained and optimized to interact with their services. These actions typically come with the correct configurations and best practices, allowing you to get started quickly without reinventing the wheel.

If a workflow depends on an external service that is down, disabling the workflow temporarily will prevent it from running and sending requests to the service, thus avoiding failed requests or unnecessary retries.

If a workflow is causing a negative impact on external services by generating too many requests or incorrect data due to a bug, temporarily disabling the workflow will stop this behavior while the issue is fixed.

The workflow_run event is designed to trigger one workflow based on the execution or completion of another workflow. The deployment workflow can therefore specify the build workflow under on.workflow_run.workflows and use the completed activity type without changing the original build workflow. One important detail is that workflow_run fires when the specified workflow completes regardless of whether it succeeded or failed. If deployment must occur only after a successful build, the deployment job should additionally use a condition such as github.event.workflow_run.conclusion == ' success ' . workflow_dispatch is a manual trigger, repository_dispatch is normally used for externally generated custom events through the GitHub API, and workflow_exec is not a GitHub Actions workflow event. Therefore, workflow_run is the correct trigger.

================

GitHub permits deletion of a workflow run after the run has completed. GitHub ' s current documentation also states that a workflow run may be deleted when it is more than two weeks old. Therefore, option A is a correct general statement and option C is incorrect because GitHub does not impose a 30-day minimum. Deleting a workflow run also does not require repository administrator access; GitHub specifies that write access to the repository is sufficient, which eliminates option D. Option B is too broad because an actively pending or queued workflow run is not generally treated the same as a completed run for immediate deletion. Removing a workflow run also removes its associated logs, which is relevant when managing workflow history or sensitive output. Therefore, option A is the correct selection.

================

The repository_dispatch event allows you to trigger a workflow in response to external activity. It is commonly used when you need to trigger a workflow from outside GitHub, such as from another system or service, by sending a request to the GitHub API. This event provides flexibility to integrate with various external systems and trigger workflows in a GitHub repository.

Self-hosted runners are appropriate when GitHub-hosted runner limits or environment restrictions do not meet the workload requirement. GitHub-hosted runner jobs have a 6-hour job execution limit, while self-hosted runner jobs can run for up to 5 days, so jobs that must run longer than 6 hours require self-hosted runners. Self-hosted runners also let an organization control the machine, operating system, software, hardware, and network location. Therefore, if a workflow job needs to install or access software from the local network, a self-hosted runner is the correct choice. macOS and Windows runners are available as GitHub-hosted options, and GitHub Actions minutes apply to GitHub-hosted usage, not self-hosted runner execution.

While Docker can run on various operating systems, Linux is often the most common and preferred OS for containerized workloads. Docker works well on Linux and is a widely-used platform for running containers.

For container jobs to run on a self-hosted runner, Docker must be installed and properly configured on the runner. Docker is required to build and run containerized workloads in a GitHub Actions workflow.

A. Use the actions/setup-docker action

B. Authenticate to the GitHub Container Registry.

C. Build the container image.

D. Push the image to the GitHub Container Registry

E. Pull the image from the GitHub Container Registry.

The best way for an enterprise to control which GitHub Actions run is by creating a list of approved actions as an enterprise policy. This approach restricts workflows to only use the actions that are explicitly allowed, ensuring security and compliance within the organization.