Timeouts
Configure how long Pingward waits for a response before considering the test failed.
Timeout Setting
Set the maximum time (in milliseconds) to wait for a response:
| Timeout | Use Case |
|---|---|
| 5,000 (5s) | Fast APIs, health checks |
| 10,000 (10s) | Standard APIs |
| 30,000 (30s) | Slow APIs, complex operations |
| 60,000 (60s) | Very slow operations |
| 120,000 (120s) | Maximum allowed |
What Happens on Timeout
When a request exceeds the timeout:
- The connection is terminated
- The test is marked as failed
- Error category:
Timeout - An issue is created (or updated)
Choosing a Timeout
Too Short
A timeout that's too short causes false positives:
- Normal slow responses are marked as failures
- You get alerted for non-issues
- Alert fatigue sets in
Too Long
A timeout that's too long delays detection:
- Real problems take longer to detect
- Your monitoring is less responsive
- MTTR (Mean Time to Recovery) increases
Best Practice
Set the timeout to 2-3x your expected response time:
- If your API typically responds in 200ms, use 500-1000ms
- If your API typically responds in 2s, use 5-10s
Use a Response Time assertion for stricter performance requirements.
Timeout vs Response Time Assertion
| Aspect | Timeout | Response Time Assertion |
|---|---|---|
| Purpose | Prevent hanging requests | Catch slow responses |
| Failure type | Error (Timeout) | Assertion failure |
| Typical value | 2-3x expected | Expected max |
Example:
- Expected response: 200ms
- Response Time assertion: 500ms (alert if slow)
- Timeout: 10s (failsafe for hung requests)
Network Considerations
The timeout includes:
- DNS resolution
- TCP connection
- TLS handshake
- Request transmission
- Server processing
- Response transmission
If you're testing endpoints in distant regions, account for network latency.