Parent-child pipelines: Child job stuck in 'pending' state without a reason

Problem to solve

When setting up a Parent-child pipeline in the UI, child job(s) can get stuck in 'pending' state, without any obvious reason and without triggering the downstream pipeline/job.

The missing failure reason can cause frustration when setting up a new (or editing already existing) pipeline-chain.

Intended users

Devon (DevOps Engineer)

Further details

This behavior can be observed in both Self-managed (12.8.x) and GitLab.com.

Cases observed so far:

Proposal

Expected behavior would be to notify the user, why the child job is stuck in pending, (similarly to when there are no runners available).

Expl.

  • "The child pipeline file <PATH_TO_FILE> cannot be found."
  • "There is a syntax error in the child-pipeline file <PATH_TO_FILE>"
    • Another option would be to allow the user to validate particular '.yml' or '.yaml' file proactively. Button name in the UI could be e.g "CI/CD validate"
  • "The downstream project is missing or was renamed."

Implementing such a notification/validation would improve user experience with the tool.

Permissions and Security

Whoever can set up a parent-child pipeline should be able to see the reason for a pending child job.

Documentation

https://docs.gitlab.com/ee/ci/parent_child_pipelines.html

Availability & Testing

What does success look like, and how can we measure that?

What is the type of buyer?

Is this a cross-stage feature?

Links / references

Edited by 🤖 GitLab Bot 🤖