title | intro | redirect_from | versions | type | topics | shortTitle | |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Migrating from Jenkins to GitHub Actions | {% data variables.product.prodname_actions %} and Jenkins share multiple similarities, which makes migration to {% data variables.product.prodname_actions %} relatively straightforward. |
|
| tutorial |
| Migrate from Jenkins |
{% data reusables.actions.enterprise-github-hosted-runners %}
Jenkins and {% data variables.product.prodname_actions %} both allow you to create workflows that automatically build, test, publish, release, and deploy code. Jenkins and {% data variables.product.prodname_actions %} share some similarities in workflow configuration:
- Jenkins creates workflows using Declarative Pipelines, which are similar to {% data variables.product.prodname_actions %} workflow files.
- Jenkins uses stages to run a collection of steps, while {% data variables.product.prodname_actions %} uses jobs to group one or more steps or individual commands.
- Jenkins and {% data variables.product.prodname_actions %} support container-based builds. For more information, see AUTOTITLE.
- Steps or tasks can be reused and shared with the community.
For more information, see AUTOTITLE.
- Jenkins has two types of syntax for creating pipelines: Declarative Pipeline and Scripted Pipeline. {% data variables.product.prodname_actions %} uses YAML to create workflows and configuration files. For more information, see AUTOTITLE.
- Jenkins deployments are typically self-hosted, with users maintaining the servers in their own data centers. {% data variables.product.prodname_actions %} offers a hybrid cloud approach by hosting its own runners that you can use to run jobs, while also supporting self-hosted runners. For more information, see AUTOTITLE.
Jenkins lets you send builds to a single build agent, or you can distribute them across multiple agents. You can also classify these agents according to various attributes, such as operating system types.
Similarly, {% data variables.product.prodname_actions %} can send jobs to {% data variables.product.prodname_dotcom %}-hosted or self-hosted runners, and you can use labels to classify runners according to various attributes. For more information, see AUTOTITLE and AUTOTITLE.
Jenkins splits its Declarative Pipelines into multiple sections. Similarly, {% data variables.product.prodname_actions %} organizes its workflows into separate sections. The table below compares Jenkins sections with the {% data variables.product.prodname_actions %} workflow.
Jenkins Directives | {% data variables.product.prodname_actions %} |
---|---|
agent | jobs.<job_id>.runs-on jobs.<job_id>.container |
post | None |
stages | jobs |
steps | jobs.<job_id>.steps |
Jenkins uses directives to manage Declarative Pipelines. These directives define the characteristics of your workflow and how it will execute. The table below demonstrates how these directives map to concepts within {% data variables.product.prodname_actions %}.
Jenkins can run the stages
and steps
in parallel, while {% data variables.product.prodname_actions %} currently only runs jobs in parallel.
Jenkins Parallel | {% data variables.product.prodname_actions %} |
---|---|
parallel | jobs.<job_id>.strategy.max-parallel |
Both {% data variables.product.prodname_actions %} and Jenkins let you use a matrix to define various system combinations.
Jenkins | {% data variables.product.prodname_actions %} |
---|---|
axis | strategy/matrix context |
stages | steps-context |
excludes | None |
Jenkins groups steps
together in stages
. Each of these steps can be a script, function, or command, among others. Similarly, {% data variables.product.prodname_actions %} uses jobs
to execute specific groups of steps
.
Jenkins | {% data variables.product.prodname_actions %} |
---|---|
steps | jobs.<job_id>.steps |
pipeline {agent anytriggers {cron('H/15 * * * 1-5')}}
on: schedule: - cron: '*/15 * * * 1-5'
pipeline {agent anyenvironment {MAVEN_PATH = '/usr/local/maven'}}
jobs: maven-build: env: MAVEN_PATH: '/usr/local/maven'
pipeline {triggers {upstream(upstreamProjects: 'job1,job2',threshold: hudson.model.Result.SUCCESS)}}
jobs: job1: job2: needs: job1job3: needs: [job1, job2]
pipeline {agent nonestages {stage('Run Tests') {matrix {axes {axis {name: 'PLATFORM'values: 'macos', 'linux'}}agent { label "${PLATFORM}" }stages {stage('test') {tools { nodejs "node-20" }steps {dir("scripts/myapp") {sh(script: "npm install -g bats")sh(script: "bats tests")}}}}}}}}
name: demo-workflowon: push: jobs: test: runs-on: {% raw %}${{ matrix.os }}{% endraw %}strategy: fail-fast: falsematrix: os: [macos-latest, ubuntu-latest]steps: - uses: {% data reusables.actions.action-checkout %} - uses: {% data reusables.actions.action-setup-node %}with: node-version: 20 - run: npm install -g bats - run: bats testsworking-directory: ./scripts/myapp