Define and correct behavior for webhook pipeline events and manual jobs
Everyone can contribute. Help move this issue forward while earning points, leveling up and collecting rewards.
Proposal
Originally discussed via this (internal only) ticket:
Currently, webhook pipeline events act in unexpected ways when manual jobs are ran. Often times manual jobs won't affect whether a pipeline is successful or not (Though job dependencies can certainly affect this).
In scenarios where manual jobs are ran after a pipeline succeeds, running a manual job will trigger a webhook pipeline event that is confusing. The problem appears to be that while the manual job is running, the pipeline briefly transitions from either "success" or "failure" to "running" and then back to it's previous state.
The proposal is that these events are suppressed; Job events already accomplish this task and the redundant events should simply be stopped. Once a pipeline reaches it's final state (Success, failure, canceled, etc.) pipeline events should no longer be sent for the pipeline since manual jobs will have no effect on the overall pipeline status.
There may be some more cases to consider here, but the overall proposal is to minimize noise since these events (After a pipeline has finished) are confusing at best.