fix(watch): not_running is the new queued (#221)

## Which problem is this PR solving? We are seeing premature buildevents terminations because we see all jobs succeeded or failed, and ignore the not_running. ## Short description of the changes Treat the new not_running status the same as "queued". ## How to verify https://app.circleci.com/pipelines/github/honeycombio/hound/80521/workflows/d0f8b476-2b41-4aeb-93b0-428c3c46ed9c/jobs/1341656?invite=true#step-102-5277_75 jobs like this should no longer show "succeeded" when there are still jobs left "not_running" we should stop seeing async traces like https://ui-dogfood.honeycomb.io/prod/environments/prod/datasets/build-events/result/Hne5hzeBSEu/trace/bP2CUXc6L59?source=query&span=d0f8b476-2b41-4aeb-93b0-428c3c46ed9c --------- Co-authored-by: Jamie Danielson <[email protected]>
honeycombio · Oct 16, 2024 · ef81eda · ef81eda
1 parent 8a9106e
commit ef81eda
Showing 1 changed file with 3 additions and 0 deletions.
diff --git a/cmd_watch.go b/cmd_watch.go
@@ -251,6 +251,9 @@ func evalWorkflow(client *circleci.Client, wfID string, jobName string) (anyRunn
 			// it's not configured to run this build (because of a tag or something)
 			anyBlocked = true
 			continue
+		case "not_running":
+			// not_running is the same as queued
+			fallthrough
 		case "queued":
 			// queued means a job is due to start running soon, so we consider it running
 			// already.