Grouped errors
Events collapse into issues ranked by severity and users affected, with stack traces, breadcrumbs and devices.
Self-hosted monitoring
Know what broke, who it hit, and where to fix it.
Self-hosted error monitoring, performance and tracing for web, iOS, Android and any backend. One small Node process, one SQLite file.
Everything you need to ship with confidence
Release stability
Crash-free session rate per release against your targets, and errors introduced by each release.
Tracing
Page loads, requests and backend spans in one waterfall. Accepts OpenTelemetry.
Feature flags
See which errors only happen with a flag or variant turned on.
GitHub, Jira, webhooks
Open a ticket from an error, jump to the exact line, alert Slack or Teams.
Built for agents
An MCP server lets IDE agents read, fix and resolve errors.
Every detail on one page
- Stack trace resolved through your source maps, linked to the line in GitHub
- Breadcrumbs, devices, releases, release stages and feature flags
- Comments and a full activity log for your team
- Snooze until it happens again, or resolve and get told if it regresses
Performance next to your errors
- p50, p75 and p95 per operation for the last hour, day or week
- Monitors flag operations that cross a threshold
- Distributed traces from the browser to your backend
Run it on your own server today
One Node process and one SQLite file. Your error data never leaves your infrastructure.