API Requests

The API Requests tab, on the Observability page, tracks the throughput, errors, and latency of API traffic your applications report to KubeWatch, so you can watch how your own services are performing.

What you'll see

• Requests and request rate over the selected range
• Errors and error rate
• p95 latency
• A by-route breakdown (which endpoints are slow or erroring)
• A requests-over-time trend, with p95 latency plotted alongside it so you can see whether a latency spike lines up with a jump in error volume
• An Error / Latency Correlation by Route report: a statistical correlation score per route between its error rate and p95 latency across the selected window, so you can spot which endpoints tend to fail and slow down together rather than independently

The tab refreshes automatically in the background rather than only on a manual range change.

Getting data in

Report request events to KubeWatch from your application, either through the API requests endpoint (send service, route, method, status code, and latency per request) or via OpenTelemetry (OTLP) HTTP spans. The API requests endpoint is available on every plan; OTLP ingestion requires the Pro plan or above. Once events arrive, the charts and the by-route breakdown populate automatically.