Cloud Native CI/CD with GitLab

Order of Precedence for Variables

When you define a variable at the pipeline, job, group, or instance level, you may have variables with the same name at different levels. In this case, GitLab uses the following order of precedence to determine which variable to use in a job (from highest to lowest):

• Project variables <- This is the highest precedence.
• Group variables.
• Instance variables.
• Variables defined in jobs in the .gitlab-ci.yml file (e.g., $JOB_VARIABLE).
• Variables defined in the pipeline in the .gitlab-ci.yml file (e.g., $PIPELINE_VARIABLE).
• Predefined variables (e.g., $CI_COMMIT_SHA).

We can see an example of this by creating a .gitlab-ci.yml file that defines:
- A variable at the pipeline level
- Another variable at the job level with the same name
- A variable at the job level that overrides predefined variables

cat <$HOME/todo/app/.gitlab-ci.yml && \
cd $HOME/todo/app && \
git add . && \
git commit -m "Test variables order of precedence" && \
git push origin main
image: python:3.12

variables:
  # Define a global variable
  VAR: "I'm a variable defined at the pipeline level"

test: