Run integration tests with custom containers

0

As a PoC I want to build a pipeline for GitOps of policy management. When a policy is committed to a repo, I want to run an existing Docker container (available on Docker Hub) that acts as a policy engine with a REST API exposed for uploading the policies and then run queries against. What is the easiest way to accomplish this?

  1. Commit policies to Git (preferably CodeCommit)
  2. Spin up a prebuilt Docker container
  3. Run a shell script with curl commands to upload and test the validity of the policies on the Docker container
  4. Fail the pipeline if curl does not return 200.
profile pictureAWS
asked 8 months ago261 views
2 Answers
0

Would something like this be the right approach to run the container inside the build?

version: 0.2

phases:
  install:
    runtime-versions:
      docker: 24

  pre_build:
    commands:
      - docker pull my-docker-image:tag
      - docker run -d --name policy-container my-docker-image:tag

  build:
    commands:
      - sh integration_tests.sh 

  post_build:
    commands:
      - docker stop policy-container
      - docker rm policy-container
profile pictureAWS
answered 8 months ago
  • The latest version is docker: 20 in a buildspec. To avoid limits you need to log in:

    version: 0.2

    env: secrets-manager: DOCKERHUB_PASS: "/dockerhub/credentials:password" DOCKERHUB_USERNAME: "/dockerhub/credentials:username" phases: install: commands: - echo pre_build step... - docker login --username $DOCKERHUB_USERNAME --password $DOCKERHUB_PASS - $(aws ecr get-login --no-include-email --region us-east-1)

    In general, your buildspec is fine.

0

You can use AWS CI/CD stack: CodeCommit, CodeBuild, CodePipeline.

https://aws.amazon.com/codecommit/

https://aws.amazon.com/codebuild/

https://aws.amazon.com/codepipeline/

In a buildspec you can define all the needed commands:

https://docs.aws.amazon.com/codebuild/latest/userguide/build-spec-ref.html

profile picture
answered 8 months ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions