Probe timing parameters (initialDelay, period, timeout) in Kubernetes - Step-by-Step Execution

Process Flow - Probe timing parameters (initialDelay, period, timeout)

Pod starts

↓

Wait initialDelay seconds

↓

Start first probe

↓

Probe runs with timeout

↓

Wait period seconds

↓

Repeat probe

↩Back to Probe runs with timeout

This flow shows how Kubernetes waits before starting probes, runs them with a timeout, then repeats them periodically.

Execution Sample

Kubernetes

livenessProbe:
  initialDelaySeconds: 5
  periodSeconds: 10
  timeoutSeconds: 3
  httpGet:
    path: /health
    port: 8080

Defines a liveness probe that waits 5 seconds before first check, runs every 10 seconds, and times out after 3 seconds.

Process Table

Step	Time (seconds)	Action	Probe State	Result
1	0	Pod starts	No probe yet	Waiting initialDelay
2	5	Initial delay ends	Start first probe	Probe sent
3	5-8	Probe running	Waiting for response	Success if response before 3s timeout
4	8	Probe finished	Probe success	Wait periodSeconds
5	18	Start next probe	Probe sent	Probe running
6	18-21	Probe running	Waiting for response	Success or failure
7	21	Probe finished	Probe success/failure	Wait periodSeconds
8	31	Start next probe	Probe sent	Repeat cycle
9	...	...	...	...
Exit	N/A	Pod stopped or probe failed	Probe stops	End of lifecycle or restart triggered

💡 Execution stops when pod stops or probe fails triggering restart

Status Tracker

Variable	Start	After Step 2	After Step 4	After Step 5	After Step 7	Final
Time (seconds)	0	5	8	18	21	N/A
Probe State	No probe	Running	Success	Running	Success/Failure	Stopped or Restarted

Key Moments - 3 Insights

Why does Kubernetes wait before starting the first probe?

Kubernetes waits for initialDelaySeconds (see Step 2) to give the container time to start before checking health.

What happens if the probe does not respond within timeoutSeconds?

If the probe times out (no response within timeoutSeconds during Steps 3 or 6), Kubernetes treats it as a failure.

How often does Kubernetes run the probe after the first check?

After the first probe, Kubernetes waits periodSeconds (see Steps 4 and 7) before running the next probe.

Visual Quiz - 3 Questions

Look at the execution table, at what time does the first probe start?

A 0 seconds

B 5 seconds

C 8 seconds

D 10 seconds

💡 Hint

Check Step 2 in the execution_table where initialDelay ends and first probe starts.

At which step does Kubernetes wait for the probe response with a timeout?

A Step 1

B Step 5

C Step 3

D Step 8

💡 Hint

Look at Step 3 where probe runs and waits for response within timeoutSeconds.

If periodSeconds is changed from 10 to 5, how does the timing in the table change?

A Probes run every 5 seconds instead of 10 seconds

B Initial delay changes to 5 seconds

C Timeout changes to 5 seconds

D Probe runs only once

💡 Hint

PeriodSeconds controls how often probes repeat after initialDelay (see Steps 4 and 7).

Concept Snapshot

Probe Timing Parameters in Kubernetes:
- initialDelaySeconds: wait before first probe
- periodSeconds: interval between probes
- timeoutSeconds: max wait for probe response
Probes start after initialDelay, run with timeout, then repeat every periodSeconds.

Full Transcript

In Kubernetes, probes check container health. The initialDelaySeconds tells Kubernetes to wait before starting the first probe, allowing the container to initialize. After this delay, Kubernetes sends the probe and waits up to timeoutSeconds for a response. If the probe succeeds, Kubernetes waits periodSeconds before sending the next probe. This cycle repeats until the pod stops or a probe fails, which may trigger a restart. Understanding these timing parameters helps ensure probes run at the right times without false failures.