Sentry
The error-tracking standard. Captures frontend and backend exceptions with full context. First tool teams add for production observability.
Mindmap
The plain-English version
Sentry captures errors and exceptions from running applications, with full stack traces, breadcrumbs (recent user actions), source maps for minified frontend code, and grouping that collapses thousands of similar errors into one issue. Backend, frontend, mobile — all in one dashboard.
The problem it solves
Without an error tracker, you find out about errors when users complain. Sentry catches them as they happen, deduplicates aggressively, and gives you the context to fix them. Free tier is generous; the value-per-dollar is unusually good.
Alternatives
| Alternative | Type | When it wins |
|---|---|---|
| Datadog | APM | Comprehensive observability — metrics, logs, APM, RUM, all under one expensive roof. |
| Prometheus | metrics | The open-source metrics standard. Pull-based scraping, time-series database, the basis of most cloud-native observability. |
| ELK Stack | log mgmt | Elasticsearch + Logstash + Kibana — the open-source log management trio. Now also "Elastic Stack" with Beats. |
Deep links
The words you'll hear
- Issue
- A group of similar errors. Sentry deduplicates by stack trace fingerprint.
- Event
- One occurrence of an error.
- Breadcrumb
- A trail of recent activity before the error (clicks, requests, console logs).
- Source map
- Maps minified production JS back to the original source. Required for usable frontend stack traces.
- Release
- A version of your app. Lets Sentry mark issues as new in this release, or regressions.
- Performance / Tracing
- Sentry's APM-light. Spans within a request, slow query detection.
Bad vs. good prompt for Sentry
Why it works: Specifies the SDK package (Next-specific), source maps (the frontend nuance everyone forgets), the release pattern, and sampling — getting these right turns Sentry from noisy to useful.
What bites real teams
Minified stack traces are useless. Configure source-map upload in build.
Breadcrumbs can capture user input and request bodies. Configure scrubbing for emails, tokens, PII.
Performance tracing at 100% sample rate is expensive and noisy. 5–10% is usually plenty.