Skip to content

Latest commit

 

History

History
323 lines (218 loc) · 18.7 KB

migrating-from-jenkins-with-github-actions-importer.md

File metadata and controls

323 lines (218 loc) · 18.7 KB
title intro versions type topics shortTitle redirect_from
Migrating from Jenkins with GitHub Actions Importer
Learn how to use {% data variables.product.prodname_actions_importer %} to automate the migration of your Jenkins pipelines to {% data variables.product.prodname_actions %}.
fpt ghec ghes
*
*
*
tutorial
Migration
CI
CD
Jenkins migration
/actions/migrating-to-github-actions/automated-migrations/migrating-from-jenkins-with-github-actions-importer

Legal notice

About migrating from Jenkins with GitHub Actions Importer

The instructions below will guide you through configuring your environment to use {% data variables.product.prodname_actions_importer %} to migrate Jenkins pipelines to {% data variables.product.prodname_actions %}.

Prerequisites

  • A Jenkins account or organization with pipelines and jobs that you want to convert to {% data variables.product.prodname_actions %} workflows.
  • Access to create a Jenkins personal API token for your account or organization. {% data reusables.actions.actions-importer-prerequisites %}

Limitations

There are some limitations when migrating from Jenkins to {% data variables.product.prodname_actions %} with {% data variables.product.prodname_actions_importer %}. For example, you must migrate the following constructs manually:

  • Mandatory build tools
  • Scripted pipelines
  • Secrets
  • Self-hosted runners
  • Unknown plugins

For more information on manual migrations, see AUTOTITLE.

Installing the {% data variables.product.prodname_actions_importer %} CLI extension

{% data reusables.actions.installing-actions-importer %}

Configuring credentials

The configure CLI command is used to set required credentials and options for {% data variables.product.prodname_actions_importer %} when working with Jenkins and {% data variables.product.prodname_dotcom %}.

  1. Create a {% data variables.product.prodname_dotcom %} {% data variables.product.pat_v1 %}. For more information, see AUTOTITLE.

    Your token must have the workflow scope.

    After creating the token, copy it and save it in a safe location for later use.

  2. Create a Jenkins API token. For more information, see Authenticating scripted clients in the Jenkins documentation.

    After creating the token, copy it and save it in a safe location for later use.

  3. In your terminal, run the {% data variables.product.prodname_actions_importer %} configure CLI command:

    gh actions-importer configure

    The configure command will prompt you for the following information:

    • For "Which CI providers are you configuring?", use the arrow keys to select Jenkins, press Space to select it, then press Enter.
    • For "{% data variables.product.pat_generic_caps %} for GitHub", enter the value of the {% data variables.product.pat_v1 %} that you created earlier, and press Enter.
    • For "Base url of the GitHub instance", {% ifversion ghes %}enter the URL for {% data variables.location.product_location_enterprise %}, and press Enter.{% else %}press Enter to accept the default value (https://github.com).{% endif %}
    • For "{% data variables.product.pat_generic_caps %} for Jenkins", enter the value for the Jenkins personal API token that you created earlier, and press Enter.
    • For "Username of Jenkins user", enter your Jenkins username and press Enter.
    • For "Base url of the Jenkins instance", enter the URL of your Jenkins instance, and press Enter.

    An example of the configure command is shown below:

    $ gh actions-importer configure
✔ Which CI providers are you configuring?: Jenkins
Enter the following values (leave empty to omit):
✔ {% data variables.product.pat_generic_caps %} for GitHub: ***************
✔ Base url of the GitHub instance: https://github.com
✔ {% data variables.product.pat_generic_caps %} for Jenkins: ***************
✔ Username of Jenkins user: admin
✔ Base url of the Jenkins instance: https://localhost
Environment variables successfully updated.

  4. In your terminal, run the {% data variables.product.prodname_actions_importer %} update CLI command to connect to {% data variables.product.prodname_registry %} {% data variables.product.prodname_container_registry %} and ensure that the container image is updated to the latest version:

    gh actions-importer update

    The output of the command should be similar to below:

    Updating ghcr.io/actions-importer/cli:latest...
ghcr.io/actions-importer/cli:latest up-to-date

Perform an audit of Jenkins

You can use the audit command to get a high-level view of all pipelines in a Jenkins server.

The audit command performs the following steps:

  1. Fetches all of the projects defined in a Jenkins server.
  2. Converts each pipeline to its equivalent {% data variables.product.prodname_actions %} workflow.
  3. Generates a report that summarizes how complete and complex of a migration is possible with {% data variables.product.prodname_actions_importer %}.

Running the audit command

To perform an audit of a Jenkins server, run the following command in your terminal:

gh actions-importer audit jenkins --output-dir tmp/audit

Inspecting the audit results

{% data reusables.actions.gai-inspect-audit %}

Forecast potential build runner usage

You can use the forecast command to forecast potential {% data variables.product.prodname_actions %} usage by computing metrics from completed pipeline runs in your Jenkins server.

Prerequisites for running the forecast command

In order to run the forecast command against a Jenkins instance, you must install the paginated-builds plugin on your Jenkins server. This plugin allows {% data variables.product.prodname_actions_importer %} to efficiently retrieve historical build data for jobs that have a large number of builds. Because Jenkins does not provide a method to retrieve paginated build data, using this plugin prevents timeouts from the Jenkins server that can occur when fetching a large amount of historical data. The paginated-builds plugin is open source, and exposes a REST API endpoint to fetch build data in pages, rather than all at once.

To install the paginated-builds plugin:

  1. On your Jenkins instance, navigate to https://<your-jenkins-instance>/pluginManager/available.
  2. Search for the paginated-builds plugin.
  3. Check the box on the left and select Install without restart.

Running the forecast command

To perform a forecast of potential {% data variables.product.prodname_actions %}, run the following command in your terminal. By default, {% data variables.product.prodname_actions_importer %} includes the previous seven days in the forecast report.

gh actions-importer forecast jenkins --output-dir tmp/forecast

Inspecting the forecast report

The forecast_report.md file in the specified output directory contains the results of the forecast.

Listed below are some key terms that can appear in the forecast report:

  • The job count is the total number of completed jobs.
  • The pipeline count is the number of unique pipelines used.
  • Execution time describes the amount of time a runner spent on a job. This metric can be used to help plan for the cost of {% data variables.product.prodname_dotcom %}-hosted runners.
  • Queue time metrics describe the amount of time a job spent waiting for a runner to be available to execute it.
  • Concurrent jobs metrics describe the amount of jobs running at any given time. This metric can be used to define the number of runners you should configure.

Additionally, these metrics are defined for each queue of runners in Jenkins. This is especially useful if there is a mix of hosted or self-hosted runners, or high or low spec machines, so you can see metrics specific to different types of runners.

Perform a dry-run migration of a Jenkins pipeline

You can use the dry-run command to convert a Jenkins pipeline to its equivalent {% data variables.product.prodname_actions %} workflow.

Running the dry-run command

You can use the dry-run command to convert a Jenkins pipeline to an equivalent {% data variables.product.prodname_actions %} workflow. A dry-run creates the output files in a specified directory, but does not open a pull request to migrate the pipeline.

To perform a dry run of migrating your Jenkins pipelines to {% data variables.product.prodname_actions %}, run the following command in your terminal, replacing my-jenkins-project with the URL of your Jenkins job.

gh actions-importer dry-run jenkins --source-url my-jenkins-project --output-dir tmp/dry-run

Inspecting the converted workflows

You can view the logs of the dry run and the converted workflow files in the specified output directory.

{% data reusables.actions.gai-custom-transformers-rec %}

Perform a production migration of a Jenkins pipeline

You can use the migrate command to convert a Jenkins pipeline and open a pull request with the equivalent {% data variables.product.prodname_actions %} workflow.

Running the migrate command

To migrate a Jenkins pipeline to {% data variables.product.prodname_actions %}, run the following command in your terminal, replacing the target-url value with the URL for your {% data variables.product.github %} repository, and my-jenkins-project with the URL for your Jenkins job.

gh actions-importer migrate jenkins --target-url https://github.com/:owner/:repo --output-dir tmp/migrate --source-url my-jenkins-project

The command's output includes the URL to the pull request that adds the converted workflow to your repository. An example of a successful output is similar to the following:

$ gh actions-importer migrate jenkins --target-url https://github.com/octo-org/octo-repo --output-dir tmp/migrate --source-url http://localhost:8080/job/monas_dev_work/job/monas_freestyle
[2022-08-20 22:08:20] Logs: 'tmp/migrate/log/actions-importer-20220916-014033.log'
[2022-08-20 22:08:20] Pull request: 'https://github.com/octo-org/octo-repo/pull/1'

{% data reusables.actions.gai-inspect-pull-request %}

Reference

This section contains reference information on environment variables, optional arguments, and supported syntax when using {% data variables.product.prodname_actions_importer %} to migrate from Jenkins.

Using environment variables

{% data reusables.actions.gai-config-environment-variables %}

{% data variables.product.prodname_actions_importer %} uses the following environment variables to connect to your Jenkins instance:

  • GITHUB_ACCESS_TOKEN: The {% data variables.product.pat_v1 %} used to create pull requests with a converted workflow (requires repo and workflow scopes).

  • GITHUB_INSTANCE_URL: The URL to the target {% data variables.product.prodname_dotcom %} instance (for example, https://github.com).

  • JENKINS_ACCESS_TOKEN: The Jenkins API token used to view Jenkins resources.

    [!NOTE] This token requires access to all jobs that you want to migrate or audit. In cases where a folder or job does not inherit access control lists from their parent, you must grant explicit permissions or full admin privileges.

  • JENKINS_USERNAME: The username of the user account that created the Jenkins API token.

  • JENKINS_INSTANCE_URL: The URL of the Jenkins instance.

  • JENKINSFILE_ACCESS_TOKEN (Optional) The API token used to retrieve the contents of a Jenkinsfile stored in the build repository. This requires the repo scope. If this is not provided, the GITHUB_ACCESS_TOKEN will be used instead.

These environment variables can be specified in a .env.local file that is loaded by {% data variables.product.prodname_actions_importer %} when it is run.

Using optional arguments

{% data reusables.actions.gai-optional-arguments-intro %}

--source-file-path

You can use the --source-file-path argument with the forecast, dry-run, or migration subcommands.

By default, {% data variables.product.prodname_actions_importer %} fetches pipeline contents from source control. The --source-file-path argument tells {% data variables.product.prodname_actions_importer %} to use the specified source file path instead. You can use this option for Jenkinsfile and multibranch pipelines.

If you would like to supply multiple source files when running the forecast subcommand, you can use pattern matching in the file path value. For example, gh forecast --source-file-path ./tmp/previous_forecast/jobs/*.json supplies {% data variables.product.prodname_actions_importer %} with any source files that match the ./tmp/previous_forecast/jobs/*.json file path.

Jenkinsfile pipeline example

In this example, {% data variables.product.prodname_actions_importer %} uses the specified Jenkinsfile as the source file to perform a dry run.

gh actions-importer dry-run jenkins --output-dir path/to/output/ --source-file-path path/to/Jenkinsfile --source-url :url_to_jenkins_job

--config-file-path

You can use the --config-file-path argument with the audit, dry-run, and migrate subcommands.

By default, {% data variables.product.prodname_actions_importer %} fetches pipeline contents from source control. The --config-file-path argument tells {% data variables.product.prodname_actions_importer %} to use the specified source files instead.

When you use the --config-file-path option with the dry-run or migrate subcommands, {% data variables.product.prodname_actions_importer %} matches the repository slug to the job represented by the --source-url option to select the pipeline. It uses the config-file-path to pull the specified source file.

Audit example

In this example, {% data variables.product.prodname_actions_importer %} uses the specified YAML configuration file to perform an audit.

gh actions-importer audit jenkins --output-dir path/to/output/ --config-file-path path/to/jenkins/config.yml

To audit a Jenkins instance using a config file, the config file must be in the following format, and each repository_slug value must be unique:

source_files:
  - repository_slug: pipeline-name
    path: path/to/Jenkinsfile
  - repository_slug: multi-branch-pipeline-name
    branches:
      - branch: main
        path: path/to/Jenkinsfile
      - branch: node
        path: path/to/Jenkinsfile

Supported syntax for Jenkins pipelines

The following tables show the type of properties {% data variables.product.prodname_actions_importer %} is currently able to convert. For more details about how Jenkins pipeline syntax aligns with {% data variables.product.prodname_actions %}, see AUTOTITLE.

For information about supported Jenkins plugins, see the github/gh-actions-importer repository.

Supported syntax for Freestyle pipelines

Jenkins GitHub Actions Status
docker template jobs.<job_id>.container Supported
build jobs Partially supported
build environment env Partially supported
build triggers on Partially supported
general runners Partially supported

Supported syntax for Jenkinsfile pipelines

Jenkins GitHub Actions Status
docker jobs.<job_id>.container Supported
stage jobs.<job_id> Supported
agent runners Partially supported
environment env Partially supported
stages jobs Partially supported
steps jobs.<job_id>.steps Partially supported
triggers on Partially supported
when jobs.<job_id>.if Partially supported
inputs inputs Unsupported
matrix jobs.<job_id>.strategy.matrix Unsupported
options jobs.<job_id>.strategy Unsupported
parameters inputs Unsupported

Environment variables syntax

{% data variables.product.prodname_actions_importer %} uses the mapping in the table below to convert default Jenkins environment variables to the closest equivalent in {% data variables.product.prodname_actions %}.

Jenkins GitHub Actions
${BUILD_ID} {% raw %}${{ github.run_id }}{% endraw %}
${BUILD_NUMBER} {% raw %}${{ github.run_id }}{% endraw %}
${BUILD_TAG} {% raw %}${{ github.workflow }}-${{ github.run_id }}{% endraw %}
${BUILD_URL} {% raw %}${{ github.server_url }}/${{ github.repository }}/actions/runs/${{ github.run_id }}{% endraw %}
${JENKINS_URL} {% raw %}${{ github.server_url }}{% endraw %}
${JOB_NAME} {% raw %}${{ github.workflow }}{% endraw %}
${WORKSPACE} {% raw %}${{ github.workspace }}{% endraw %}

Legal notice

{% data reusables.actions.actions-importer-legal-notice %}