Latency

Slow responses, dripping streams, and delayed behavior.

GET /drip

Starts responding immediately but streams the body in small chunks with a configurable delay between each chunk. Unlike /slow, the response headers arrive right away — only the body trickles.

bytes Total bytes to stream. Range: 1–100000. Default: 1000.

interval Milliseconds between chunks. Range: 10–5000. Default: 200.

chunk Bytes per chunk. Range: 1–4096. Default: 100.

details

GET /slow

Sleeps for the specified number of milliseconds, then returns 200 with a JSON body confirming the delay.

ms Delay in milliseconds. Range: 0–15000. Default: 1000.

details

Endpoints that take time. Useful for testing read timeouts, retry logic, and client-side concurrency limits.

/slow delays before sending anything — the connection idles until the full wait elapses, then a complete response arrives at once. /drip is different: headers arrive immediately, but the body trickles in chunks. Use /drip to test streaming parsers, chunked-transfer handling, and read-deadline behavior.

Both endpoints are capped to keep total duration under the platform’s 30-second wall-clock limit. Extreme /drip combinations (many tiny chunks with large intervals) can still approach that ceiling — keep total stream time under ~25 seconds.