Troubleshooting a CI/CD pipeline is a lot harder than troubling shooting local code. When debugging problems locally, it’s easy to get timely feedback about what’s going wrong and where. For CI/CD pipelines, historically this has been more difficult. However, in Gitlab, you can use GitlabRunner to locally simulate CI/CD jobs defined in your .gitlab-ci.yaml. Doing this requires some initial setup, but will often save countless hours and build minutes later down the line. o If you lack a tool like GitlabRunner, then the best way to troubleshoot issues with a CI/CD pipeline is to try incremental changes with rollback strategies for if things break. Alternatively, you could build a dummy pipeline in a test environment and iterate there until things work, then bring in the completed pipeline over to your main project.