CODEWITHSUNDEEP
spring-bootdockergitlab-citestcontainers
Giovanni Alluvatti
Giovanni Alluvatti

Reputation: 135

Issue when executing Gitlab CI pipeline job running integration test with Testcontainers

I have a Spring Boot application test that used TestContainers (https://www.testcontainers.org) to run some integration tests against a KafkaContainer. When running locally everything is fine, but when running integration test inside a GitlabCI pipeline the container can't start. Here the job log:

> org.testcontainers.containers.ContainerLaunchException: Container
> startup failed Caused by:
> org.rnorth.ducttape.RetryCountExceededException: Retry limit hit with
> exception Caused by:
> org.testcontainers.containers.ContainerLaunchException: Could not
> create/start container Caused by:
> org.testcontainers.containers.ContainerLaunchException: Timed out
> waiting for container port to open (docker ports: [32776, 32778]
> should be listening)

Here Gitlab CI conf:

image: docker:git
services:
  - docker:dind

stages:
  - test
  - push
  - deploy

variables:
  MAVEN_CLI_OPTS: "--batch-mode -Dmaven.repo.local=/.m2"
  DOCKER_HOST: "tcp://docker:2375"
  DOCKER_DRIVER: overlay2

.test_template: &java_test
  image: maven:3.5.3-jdk-8
  stage: test
  tags:
    - test

 myapp-api_test:
   <<: *java_test
   services:
     - docker:dind
   variables:
     DOCKER_HOST: "tcp://docker:2375"
     DOCKER_DRIVER: overlay2
   script:
    - cd myapp-api
    - mvn $MAVEN_CLI_OPTS package

Here Runner conf:

executor = "docker"
  [runners.docker]
    tls_verify = false
    image = "alpine:latest"
    privileged = true
    disable_entrypoint_overwrite = false
    oom_kill_disable = false
    disable_cache = false
    volumes = ["/cache", "/.m2", 
         "/var/run/docker.sock:/var/run/docker.sock"]
    shm_size = 0
    output_limit = 8192

Any help? Thanks

Upvotes: 3

Views: 1576

Answers (1)

Runar Halse
Runar Halse

Reputation: 3558

I know you've probably figured this out by now, but giving you an answer for future reference in case others find themselves in the same kind of problem.

I was having similar issues, and fixed it by adding a custom waitstrategy and/or custom timeout for my containers. Locally on my machine it would work fine, but on the slower gitlab.com shared runners it would timeout

Wait.forHttp("/health").withStartupTimeout(Duration.ofMinutes(5L))

Upvotes: 0

Related Questions