Skip to main content

Why it's Not a Good Idea to Use "Retry" in GitLab CI

GitLab CI's "retry" keyword can be alluring because it seems like a simple method to handle pipeline failures and retries. In the long term, though, it might actually cause more issues and headaches.

Extended Pipeline Duration

One of the major problems with using "retry" is that it can make your pipeline run for a lot longer than it needs to. Every time a job fails and needs to be retried, it uses up time and resources, possibly slowing down the remainder of the pipeline and creating a backlog of other jobs that need to run.

Concealing Underlying Issues

Another problem is that it can conceal the true issues in your process. You might just keep retrying the task until it succeeds rather than addressing the underlying cause of the failure. This can create a false sense of security and make it more difficult to find and address real problems.

Focus on Root Cause

It is preferable to concentrate on resolving the primary reason for the failure rather than simply "retrying." This might entail enhancing the employed code or dependencies, modifying the job settings or configurations, or attending to underlying infrastructure problems.


Even though "retry" may appear to be a simple way to handle pipeline failures, it can ultimately result in more issues and expenses. By concentrating on resolving the primary source of errors and using more targeted retry approaches when necessary, you can guarantee a more effective and efficient GitLab CI pipeline.