[go: up one dir, main page]

Resolve "Add option to skip scheduled pipelines after downtime"

Daniel Prauserequested to merge
539393-add-option-to-skip-scheduled-pipelines-after-downtime into master

What does this MR do and why?

This MR adds the possibility to set a time window (in seconds) where after a pipeline was scheduled/overdue the pipeline is still allowed to be started.

Imagine the following scenario:

  • A pipeline is scheduled for the 1st of January.
  • A GitLab instance (e.g. a self-hosted one) is down for 2 days.
  • With the here added window, we can say "A pipeline is only allowed to be started, if we are in a timeframe of e.g. one day"

If we have many pipelines that would immediately start after an instance comes back online, we'd have a big performance impact of pipelines that don't need to run anymore.

With the newly given timeframe, we can set a time window per pipeline schedule, so that we don't start pipelines that are too late to be running and set them automatically to skipped. Therefore, we can save a lot of cpu time for pipelines that we don't need to be running anymore.

References

Screenshots or screen recordings

image

How to set up and validate locally

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.

Related to #539393 (closed)

Edited by Daniel Prause

Merge request reports