Skip to content
Branch: master
Find file Copy path
Find file Copy path
Fetching contributors…
Cannot retrieve contributors at this time
34 lines (27 sloc) 1.48 KB

Docker Pipe is a way to execute each instruction in it's own Docker container while persisting working directory between each of the containers. For example, you can build your application in one container, run some lint tools in another containers and finally deploy your app via CLI from another container.

No need to create huge containers with every single tool pre-installed!

A pipe can be defined the same way as a task with the only difference that instructions should be grouped under the steps field defining a Docker image for each step to be executed in. Here is an example of how we build and validate links for the Cirrus CI documentation that you are reading right now:

  name: Build Site and Validate Links
    - image: squidfunk/mkdocs-material:latest
      build_script: mkdocs build
    - image: raviqqe/liche:latest # links validation tool in a separate container
      validate_script: /liche --document-root=site --recursive site/

Amount of CPU and memory that a pipe has access to can be configured with resources field:

    cpu: 2.5
    memory: 5G
  # ...

!!! tip "Docker Pipes on your own GKE cluster" Check GKE documentation for instructions on how to run Docker Pipes on your own Google Kubernetes Engine cluster.

You can’t perform that action at this time.