v1.0 · source-available under FSL-1.1 · self-host or cloud

Sentry for React teams —
errors, replays & releases in one place

Remetry tracks your frontend errors with symbolicated stack traces, records session replays around them, watches your API latency, and tells you which deploy made things worse.

Get started GitHub →
remetry — releases
▸ release 1.4.2 deployed 12 min ago ✗ REGRESSION errors +245% · LCP p75 +1.2s vs 1.4.1 TypeError: Cannot read properties of undefined (reading 'map') at CheckoutSummary (src/checkout/Summary.tsx:48:21) ▶ open replay alert sent → #frontend-alerts (slack)
Features

Everything between an exception and a fix

One pipeline from your users' browsers to your dashboard: capture, group, symbolicate, replay, compare against the previous release, alert.

{}

Error tracking

Global handlers plus a React error boundary. Events are grouped into issues by release-stable fingerprints, and stack traces are symbolicated with the source maps you upload from CI.

▶

Session replay

rrweb-based recording, lazy-loaded so it never bloats your bundle. On-error mode keeps a ring buffer and only ships the moments around a crash. Inputs masked by default.

⇅

API monitoring

An opt-in plugin patches fetch/XHR and reports every request. URLs normalise server-side (/orders/123 → /orders/:id) into per-endpoint P75/P95 latency and error rate.

Δ

Release regression detection

Every deploy is diffed against its predecessor — error rate via a traffic-normalised statistical test, LCP p75 via effect-size thresholds — so "errors +245%, LCP +1.2s" gets flagged, not noise.

◉

Alerts

Error spikes fan out to Slack, generic webhooks, and email — deduplicated across replicas, so one incident means one ping, not a pager storm.

⌂

Self-host it all

The entire platform — ingestion, ClickHouse analytics, replay storage, dashboard — runs from one docker compose up on your own box, licensed under FSL-1.1 with an Apache 2.0 future grant.

Get started

Two ways to run Remetry

Start on the hosted cloud in minutes, or own the whole data plane on your own infrastructure. Same code, same features.

Cloud

Remetry Cloud

Sign up, create a project, drop the SDK into your app. We run the ingestion, storage, and dashboard for you.

  • No infrastructure to operate
  • TLS, backups, upgrades handled
  • Project-scoped API keys & team access
Create an account
Self-hosted

Your own box

Everything ships in one compose stack: Go ingestion & worker, ClickHouse, Postgres, Redis, MinIO, and the dashboard.

$ git clone https://github.com/Maknie/remetry $ cd remetry && cp .env.example .env $ docker compose up -d
Deployment docs →
SDK

Instrument your app in two minutes

One small browser SDK. Errors and web-vitals are on by default; API monitoring and session replay are a plugin away.

  1. Install the SDKOne dependency, plugins load lazily.
  2. Initialise with your DSNGrab it from your project settings after signup.
  3. Ship itErrors, vitals, requests and replays start flowing to your dashboard.
$ npm install @remetry/browser import { init, errorsPlugin, performancePlugin, apiPlugin, replayPlugin } from "@remetry/browser"; init({ dsn: "your-project-dsn", endpoint: "https://ingest.remetry.dev", release: "1.4.2", // ties errors to deploys + source maps integrations: [errorsPlugin(), performancePlugin(), apiPlugin(), replayPlugin({ mode: "onError" })], });

FAQ

What's the difference between error tracking and session replay — does Remetry do both?

Error tracking captures the stack trace and context when something breaks; session replay shows what the user did beforehand. Remetry combines both — symbolicated stack traces plus rrweb-based session replays — in one platform.

Is Remetry open source?

Remetry is source-available under FSL-1.1 with a future grant to Apache-2.0 — the source is public to read and self-host, and each release becomes Apache-2.0 two years on. Not OSI open source today, and we don't pretend otherwise.

Can I self-host Remetry, or do I have to use the cloud version?

Both are supported. Remetry is self-hostable via Docker for full data control, and also available as a hosted cloud version.

How does Remetry know which deploy caused a regression?

Remetry includes release regression detection — it flags when a new deploy makes errors or API latency worse, so you can pinpoint the deploy responsible instead of guessing.

Does Remetry show readable stack traces for React, or just minified code?

Remetry symbolicates stack traces, translating minified production errors back to real source locations — built for React teams rather than as a generic JS error tracker.