You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When a pipeline run is triggered for a repository where this is already a running pipeline, the new pipeline run is queued (created in pending state), and a polling mechanism is kicked off. When polling detects no pipeline is running anymore, the pending one is started. The design may waste time up to the polling interval (30s) before starting the pending pipeline. We could improve this wait time by signalling the polling mechanism from the end of a running pipeline that we are done soon, and shorten the polling interval temporarily.
The text was updated successfully, but these errors were encountered:
When a pipeline run is triggered for a repository where this is already a running pipeline, the new pipeline run is queued (created in pending state), and a polling mechanism is kicked off. When polling detects no pipeline is running anymore, the pending one is started. The design may waste time up to the polling interval (30s) before starting the pending pipeline. We could improve this wait time by signalling the polling mechanism from the end of a running pipeline that we are done soon, and shorten the polling interval temporarily.
The text was updated successfully, but these errors were encountered: