Skip to content

Commit 0b793e1

Browse files
Clarified behavior of Ready condition
when the Pod doesn't define a readiness probe.
1 parent 0de12d1 commit 0b793e1

File tree

1 file changed

+7
-2
lines changed
  • keps/sig-apps/2879-ready-pods-job-status

1 file changed

+7
-2
lines changed

keps/sig-apps/2879-ready-pods-job-status/README.md

Lines changed: 7 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -263,6 +263,7 @@ create new Pods or tracks finishing Pods.
263263

264264
- [x] Metrics
265265
- Metric name: `job_sync_duration_seconds`, `job_sync_total`.
266+
- Components exposing the metric: `kube-controller-manager`
266267

267268
###### Are there any missing metrics that would be useful to have to improve observability of this feature?
268269

@@ -344,5 +345,9 @@ Pod created.
344345

345346
- Add `Job.status.running`, counting Pods with `Running` phase. The `Running`
346347
phase doesn't take into account preparation work before the worker is ready
347-
to accept connections. The `Ready` condition is configurable through a
348-
readiness probe.
348+
to accept connections. On the other hand, the `Ready` condition is
349+
configurable through a readiness probe. If the Pod doesn't have a readiness
350+
probe configured, the `Ready` condition is equivalent to the `Running` phase.
351+
352+
In other words, `Job.status.active` provides as the same behavior as
353+
`Job.status.running` with the advantage of it being configurable.

0 commit comments

Comments
 (0)