Self-hosted runners reference

Find information about setting up and using self-hosted runners.

本文内容

Requirements for self-hosted runner machines

You can use a machine as a self-hosted runner as long as it meets these requirements:

  • You can install and run the self-hosted runner application on the machine. See Supported operating systems and Supported processor architectures.
  • The machine can communicate with GitHub Actions.
  • The machine has enough hardware resources for the type of workflows you plan to run. The self-hosted runner application itself only requires minimal resources.
  • If you want to run workflows that use Docker container actions or service containers, you must use a Linux machine and Docker must be installed.

Supported operating systems

Linux

  • Red Hat Enterprise Linux 8 or later
  • CentOS 8 or later
  • Oracle Linux 8 or later
  • Fedora 29 or later
  • Debian 10 or later
  • Ubuntu 20.04 or later
  • Linux Mint 20 or later
  • openSUSE 15.2 or later
  • SUSE Enterprise Linux (SLES) 15 SP2 or later

Windows

  • Windows 10 64-bit
  • Windows 11 64-bit
  • Windows Server 2016 64-bit
  • Windows Server 2019 64-bit
  • Windows Server 2022 64-bit

macOS

  • macOS 11.0 (Big Sur) or later

Supported processor architectures

  • x64 - Linux, macOS, Windows.
  • ARM64 - Linux, macOS.
  • ARM32 - Linux.

Routing precedence for self-hosted runners

When routing a job to a self-hosted runner, GitHub looks for a runner that matches the job's runs-on labels and groups:

  • If GitHub finds an online and idle runner that matches the job's runs-on labels and groups, the job is then assigned and sent to the runner.
    • If the runner doesn't pick up the assigned job within 60 seconds, the job is re-queued so that a new runner can accept it.
  • If GitHub doesn't find an online and idle runner that matches the job's runs-on labels and groups, then the job will remain queued until a runner comes online.
  • If the job remains queued for more than 24 hours, the job will fail.

Autoscaling

You can automatically increase or decrease the number of self-hosted runners in your environment in response to the webhook events you receive with a particular label.

Supported autoscaling solutions

The actions/actions-runner-controller (ARC) project is a Kubernetes-based runner autoscaler. GitHub recommends ARC if the team deploying it has expert Kubernetes knowledge and experience.

For more information, see 关于 Actions Runner Controller and 关于对 Actions Runner Controller 的支持.

Ephemeral runners for autoscaling

GitHub recommends implementing autoscaling with ephemeral self-hosted runners; autoscaling with persistent self-hosted runners is not recommended. In certain cases, GitHub cannot guarantee that jobs are not assigned to persistent runners while they are shut down. With ephemeral runners, this can be guaranteed because GitHub only assigns one job to a runner.

This approach allows you to manage your runners as ephemeral systems, since you can use automation to provide a clean environment for each job. This helps limit the exposure of any sensitive resources from previous jobs, and also helps mitigate the risk of a compromised runner receiving new jobs.

警告

The runner application log files for ephemeral runners must be forwarded to an external log storage solution for troubleshooting and diagnostic purposes. While it is not required for ephemeral runners to be deployed, GitHub recommends ensuring runner logs are forwarded and preserved externally before deploying an ephemeral runner autoscaling solution in a production environment. For more information, see 对自托管运行程序进行监视和故障排除.

To add an ephemeral runner to your environment, include the --ephemeral parameter when registering your runner using config.sh. For example:

./config.sh --url https://github.com/octo-org --token example-token --ephemeral

The GitHub Actions service will then automatically de-register the runner after it has processed one job. You can then create your own automation that wipes the runner after it has been de-registered.

注意

If a job is labeled for a certain type of runner, but none matching that type are available, the job does not immediately fail at the time of queueing. Instead, the job will remain queued until the 24 hour timeout period expires.

Alternatively, you can create ephemeral, just-in-time runners using the REST API. For more information, see 自托管运行器的 REST API 终结点.

Runner software updates on self-hosted runners

By default, self-hosted runners will automatically perform a software update whenever a new version of the runner software is available. If you use ephemeral runners in containers then this can lead to repeated software updates when a new runner version is released. Turning off automatic updates allows you to update the runner version on the container image directly on your own schedule.

To turn off automatic software updates and install software updates yourself, specify the --disableupdate flag when registering your runner using config.sh. For example:

./config.sh --url https://github.com/YOUR-ORGANIZATION --token EXAMPLE-TOKEN --disableupdate

If you disable automatic updates, you must still update your runner version regularly. New functionality in GitHub Actions requires changes in both the GitHub Actions service and the runner software. The runner may not be able to correctly process jobs that take advantage of new features in GitHub Actions without a software update.

If you disable automatic updates, you will be required to update your runner version within 30 days of a new version being made available. You may want to subscribe to notifications for releases in the actions/runner repository. For more information, see 配置通知.

For instructions on how to install the latest runner version, see the installation instructions for the latest release.

警告

Any updates released for the software, including major, minor or patch releases, are considered as an available update. If you do not perform a software update within 30 days, the GitHub Actions service will not queue jobs to your runner. In addition, if a critical security update is required, the GitHub Actions service will not queue jobs to your runner until it has been updated.

Webhooks for autoscaling

You can create your own autoscaling environment by using payloads received from the workflow_job webhook. This webhook is available at the repository, organization, and enterprise levels, and the payload for this event contains an action key that corresponds to the stages of a workflow job's life-cycle; for example when jobs are queued, in_progress, and completed. You must then create your own scaling automation in response to these webhook payloads.

Authentication requirements

You can register and delete repository and organization self-hosted runners using the API. To authenticate to the API, your autoscaling implementation can use an access token or a GitHub app.

Your access token will require the following scope:

To authenticate using a GitHub App, it must be assigned the following permissions:

  • For repositories, assign the administration permission.
  • For organizations, assign the organization_self_hosted_runners permission.

You can register and delete enterprise self-hosted runners using the API. To authenticate to the API, your autoscaling implementation can use an access token.

Your access token will require the manage_runners:enterprise scope.

Communication

Self-hosted runners connect to 你的 GitHub Enterprise Server 实例 to receive job assignments and download new versions of the runner application.

GitHub Actions 运行器应用程序是开源的。 可以参与 runner 存储库并在其中提交问题。 When a new version is released, the runner application will automatically update within 24 hours.

Requirements for communication with 你的 GitHub Enterprise Server 实例

  • The self-hosted runner application must be running on the host machine to accept and run GitHub Actions jobs.
  • GitHub Enterprise Server must accept inbound connections from your runners over HTTP(S) at 你的 GitHub Enterprise Server 实例's hostname and API subdomain, and your runners must allow outbound connections over HTTP(S) to 你的 GitHub Enterprise Server 实例's hostname and API subdomain.
  • For caching to work, the runner must be able to communicate with, and directly download content from, blob storage.

Communication with GitHub.com

Self-hosted runners do not need to connect to GitHub.com unless you have enabled automatic access to GitHub.com actions for GitHub Enterprise Server. For more information, see 关于在企业中使用操作.

If you want your runner to connect to GitHub.com, the host machine must be able to make outbound HTTP connections over port 80, or HTTPS connections over port 443. To ensure connectivity over HTTPS, configure TLS for GitHub Enterprise Server. See 配置 TLS.

If you have enabled automatic access to GitHub.com actions, then the self-hosted runner will connect directly to GitHub.com to download actions. You must ensure that the machine has the appropriate network access to communicate with the GitHub URLs listed below.

Shell
github.com
api.github.com
codeload.github.com
pkg.actions.githubusercontent.com

You can use the REST API to get meta information about GitHub, including the IP addresses and domain details for GitHub services. The actions_inbound section of the API supports both fully qualified and wildcard domains. Fully qualified domains specify a complete domain name (e.g., example.github.com), while wildcard domains use a * to represent multiple possible subdomains (e.g., *.github.com). An example of the self-hosted runner requirements using wildcard domains has been listed below. For more information, see 元数据的 REST API 终结点.

Shell
github.com
*.github.com
*.githubusercontent.com
ghcr.io

注意

列出的一些域名使用 CNAME 记录配置。 一些防火墙可能要求你为所有 CNAME 记录递归添加规则。 请注意，CNAME 记录将来可能会改变，只有列出的域名将保持不变。