Record any production request across 10+ microservices.
Replay it on your laptop β byte-for-byte identical.
Debug in 5 minutes what used to take 5 hours.
Logs tell you WHAT happened. Traces tell you WHEN. But nobody tells you WHY β until now.
"Works on my machine." The bug only happens in production with that exact request, that exact DB state, that exact timing. Your local env is useless.
50,000 log lines. 10 services. 3 engineers staring at Datadog for 4 hours. You see WHAT failed, but the actual state that caused it? Gone forever.
Jaeger shows you Service A called Service B in 23ms. Cool. But what was the request body? What did the DB return? Traces don't capture state.
Add the probe to your Express app. Every HTTP request, DB query, random value, and timer is captured at I/O boundaries β with <1% CPU overhead.
βDownload the recording. Run it locally. All I/O is mocked with recorded values. The app executes exactly as it did in production β deterministically.
βStep through the timeline. See every DB query result, every API response, every timestamp. Find the bug in 5 minutes β not 5 hours.
| Tool | Record | Replay | Distributed | Production-Safe |
|---|---|---|---|---|
| Datadog | β | β | β | β |
| Sentry | β | β | β | β |
| rr (Mozilla) | β | β | β | β |
| Replay.io | β | β | β | β |
| ERGENEKON β‘ | β | β | β | β |
From production HTTP request to byte-perfect local replay β
five decoupled packages, zero shared state.
Full Pro access for 90 days β no credit card required. After trial, choose the plan that fits.
Install the probe. Record your first request. Replay it locally. See the difference in 2 minutes.
Open source under BSL-1.1 Β· View source on GitHub