kubernetes / kubernetes/website

probes: missing documentation of the mandatory random initial delay

Open
#52,218 5 comments 0 reactions 1 assignee Assigned to @ninglonglong View on GitHub
kind/bug language/en priority/backlog sig/node triage/accepted
Dominant language
HTML
Stars
5.4k
Forks
15.7k
Avg merge
4d 18h
Merged PRs (30d)
204

Description

**Problem:**

Probes are always initially delayed by a random duration based on `periodSeconds`. This is not documented. I and and others have had to rediscover it by searching bugs and code to understand what is happening.

**Proposed Solution:**

Document that confusing behaviour.

**Page to Update:**

https://kubernetes.io/docs/concepts/configuration/liveness-readiness-startup-probes/ and/or https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/

**Additional Information:**

The [relevant code](https://github.com/kubernetes/kubernetes/blob/2e83bc4bf31e88b7de81d5341939d5ce2460f46f/pkg/kubelet/prober/worker.go#L148-L155).

A [related bug](https://github.com/kubernetes/kubernetes/issues/62036) against kubernetes was closed without resolution but it clearly shows many people have been and keep being confused about this.

There is also a [different but somewhat related bug](https://github.com/kubernetes/website/issues/48519) against the documentation.

I would send a PR or specific wording proposal but based on [past experience](https://github.com/kubernetes/website/pull/51374), you would reject it.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.