How do I add the gitlab ci content into existing pipelines

[zacktzeng]

Hello, I am trying to integrate an existing gitlab repo to sonarqube. However, that repo is using a custom image we built and for some reason I couldn’t add the additional sonarqube .gitlab-ci.yml content to it.

kipping Git submodules setup
Restoring cache
00:01
Checking cache for scan...
FATAL: file does not exist                         
Failed to extract cache
Executing "step_script" stage of the job script
00:00
$ bootstrap ${CI_RUNNER_SKELETON_KEY_BASE64_TOKEN} ${CI_RUNNER_SKELETON_KEY_ACCOUNT_NAME}
/scripts-5028-2330605/step_script: line 177: bootstrap: command not found
Cleaning up project directory and file based variables

This is some of the error message. Can someone help me with this?

[Colin]

What is the full content of your gitlab-ci.yml file?

[zacktzeng]

This is the existing .gitlab-ci.yml. It has its own image.

image: gcr.io/t/terraform:
services:
- docker:18-dind

variables:
  DOCKER_DRIVER: 
  DOCKER_HOST: 
  PROJECT: 
  REGION: 
  CLUSTER: 
  NAMESPACE: 
  DOCKER_TLS_CERTDIR: 
  TF_PLUGIN_DIR: 
  TF_GET_PLUGINS: 
  FF_USE_LEGACY_KUBERNETES_EXECUTION_STRATEGY: 0

stages:
  - test
  - plan
  - manual-apply
  - manual-destroy

before_script:
  - bootstrap ${CI_RUNNER_SKELETON_KEY_BASE64_TOKEN} ${CI_RUNNER_SKELETON_KEY_ACCOUNT_NAME}
  - terraform-robot $PROJECT $REGION $CLUSTER $NAMESPACE
  - export GOOGLE_APPLICATION_CREDENTIALS=$HOME/.gcloud/terraform_robot_credentials.json

lint:
  stage: test
  script:
    - make init && make test

plan:
  stage: plan
  script:
    - make init && make ci-plan
  artifacts:
    paths:
    - tfplan
    expire_in: 1 month

apply:
  stage: manual-apply
  dependencies:
    - plan
  script:
    - make init && make ci-apply
  when: manual
  only:
    refs:
      - main
  
destroy:
  stage: manual-destroy
  script:
    - make init && make ci-destroy
  allow_failure: true
  dependencies:
    - plan
  when: manual
  only:
    refs:
      - main

[Colin]

And how does it look when you try and add the additional SonarQube content?

[zacktzeng]

image: gcr.io/t/terraform:
services:
- docker:18-dind

variables:
  DOCKER_DRIVER: 
  DOCKER_HOST: 
  PROJECT: 
  REGION: 
  CLUSTER: 
  NAMESPACE: 
  DOCKER_TLS_CERTDIR: 
  TF_PLUGIN_DIR: 
  TF_GET_PLUGINS: 
  FF_USE_LEGACY_KUBERNETES_EXECUTION_STRATEGY: 
  SONAR_USER_HOME: "${CI_PROJECT_DIR}/.sonar"
  GIT_DEPTH: "0"

stages:
  - scan
  - test
  - plan
  - manual-apply
  - manual-destroy

before_script:
  - bootstrap ${CI_RUNNER_SKELETON_KEY_BASE64_TOKEN} ${CI_RUNNER_SKELETON_KEY_ACCOUNT_NAME}
  - terraform-robot $PROJECT $REGION $CLUSTER $NAMESPACE
  - export GOOGLE_APPLICATION_CREDENTIALS=

scan:
  stage: scan
  image: 
    name: sonarsource/sonar-scanner-cli:latest
    entrypoint: [""]
  cache:
    key: "${CI_JOB_NAME}"
    paths:
      - .sonar/cache
  script: 
    - sonar-scanner -Dsonar.projectKey=security_italpha_drawio_AYLQdxtpP_KkG93sb6Ht -Dsonar.qualitygate.wait=true
  allow_failure: true
  only:
    - merge_requests
    - master # or the name of your main branch
    - develop

lint:
  stage: test
  script:
    - make init && make test

plan:
  stage: plan
  script:
    - make init && make ci-plan
  artifacts:
    paths:
    - tfplan
    expire_in: 1 month

apply:
  stage: manual-apply
  dependencies:
    - plan
  script:
    - make init && make ci-apply
  when: manual
  only:
    refs:
      - main
  
destroy:
  stage: manual-destroy
  script:
    - make init && make ci-destroy
  allow_failure: true
  dependencies:
    - plan
  when: manual
  only:
    refs:
      - main

[zacktzeng]

@Colin the error message seems to suggest that the cache file cannot be found. This doesn’t make sense because I’ve included the variables already.

[zacktzeng]

Thanks for the help. This question can be closed now. It turned out that the existing .gitlab-ci.yml file contains a before_script block and the sonar scan job doesn’t support it. So overriding the before_script block in sonar scan job resolved the issue.