Exploring the Runners in GitHub

Exploring the Runners in GitHub

In GitHub Actions, runners are the servers or environments that execute the tasks defined in a workflow. Runners can be hosted by GitHub or self-hosted by users for custom requirements.

What is a Runner?

A runner is a system that executes the jobs in a GitHub Actions workflow. It checks out the code, runs commands, and reports the results back to GitHub.

Types of Runners

1. GitHub-Hosted Runners

Definition:

Preconfigured virtual machines provided by GitHub.

Cost:

Free for public repositories. Limits apply for private repositories (depends on GitHub plan).

Features:

• Automatically scaled and maintained by GitHub.

• Preinstalled with common tools, libraries, and languages.

• Supports Linux, Windows, and macOS environments.

Available Environments:

• ubuntu-latest (or specific versions like ubuntu-20.04, ubuntu-22.04)

• windows-latest (or specific versions like windows-2019, windows-2022)

• macos-latest (or specific versions like macos-11, macos-12)

Example Usage:

xxxxxxxxxx
5
1
jobs:
2
  build:
3
    runs-on: ubuntu-latest
4
    steps:
5
     - run: echo "Hello, GitHub-hosted runner!"

2. Self-Hosted Runners

Definition:

Runners that you host on your own infrastructure.

Cost:

You provide and maintain the infrastructure.

Features:

• Greater control over the hardware, OS, and installed software.

• Useful for running workflows in a secure or specialized environment.

• Can be configured for specific tasks or higher-performance needs.

Use Cases:

• Workflows requiring custom software or hardware (e.g., GPU).

• Running jobs in a secure, on-premises environment.

• Avoiding limitations of GitHub-hosted runners.

Setup:

• Install the GitHub Actions runner application on your machine.

• Register the runner to a repository, organization, or enterprise.

Example Usage:

xxxxxxxxxx
5
1
jobs:
2
  build:
3
    runs-on: self-hosted
4
    steps:
5
     - run: echo "Running on a self-hosted runner!"

Configuring Runners

1. runs-on

Specifies the type of runner for a job.

• GitHub-Hosted Example:

  xxxxxxxxxx
  1
  1
  runs-on: ubuntu-latest
  

• Self-Hosted Example:

  xxxxxxxxxx
  1
  1
  runs-on: self-hosted
  

• Combination of Labels: You can assign labels to self-hosted runners for targeted selection.

  xxxxxxxxxx
  1
  1
  runs-on: [self-hosted, ubuntu, high-memory]
  

2. Runner Labels

• GitHub-hosted runners have predefined labels (e.g., ubuntu-latest, macos-latest).

• Self-hosted runners can be tagged with custom labels for better organization.

Runner Features and Capabilities

1. Preinstalled Tools

GitHub-hosted runners come with preinstalled software, such as:

• Programming languages: Node.js, Python, Ruby, Go, Java, etc.

• CI/CD tools: Docker, Kubernetes, Terraform, etc.

• Build tools: Gradle, Maven, Make, etc.

• Package managers: npm, pip, gem, etc.

2. Runtime Logs

Runners provide detailed logs for each step, making it easier to debug and optimize workflows.

3. Caching and Artifacts

• Runners can store caches to speed up workflows.

• Artifacts can be saved and transferred between jobs.

Choosing Between GitHub-Hosted and Self-Hosted Runners

Advanced Topics

1. Ephemeral Runners

• Temporary runners that are created for a single job and automatically deleted afterward.

• Useful for secure and resource-efficient workflows.

2. Runner Groups

• Self-hosted runners can be grouped for better access control and organization.

• Runner groups can be scoped to specific repositories or organizations.

3. Custom Docker Containers

For more control over the environment, you can use Docker containers as runners:

xxxxxxxxxx
7
1
jobs:
2
  build:
3
    runs-on: ubuntu-latest
4
    container:
5
      image: node:16
6
    steps:
7
     - run: npm install

Monitoring and Managing Runners

1. GitHub Actions Settings: View and manage self-hosted runners under the "Actions" tab in repository or organization settings.

2. Monitoring Tools:

   • Logs for each job provide insights into runner performance.

   • GitHub API allows programmatic management of runners.

Common Use Cases for Runners

1. CI/CD Pipelines: Compile, test, and deploy applications.

2. Machine Learning: Use self-hosted runners with GPUs for ML workloads.

3. Secure Workflows: Run sensitive jobs in isolated self-hosted environments.

4. Cross-Platform Testing: Test applications on multiple OS environments.

Summary

Runners are a powerful part of GitHub Actions, offering flexibility and scalability for various workflows. Write me through your comments if you would like to know about anything deeper into configuring self-hosted runners or using specific environments. I will incorporate that in another post. Untill then, best of luck and keep reading.