The dashboard

Pick a time range in the top bar. Every number compares itself with the range before it. Search everything with Ctrl K (⌘ K). Light and dark themes are built in.

Pages#

Page	What you get
Overview	Requests, duration, SQL, error rate, and insights at a glance; request activity, timing breakdown, services, recent requests, latest insights
Requests	Filter by method, status, duration, service, or text; export to CSV; a side panel with summary, headers, response, tags, and timing
Traces	One request as layers: browser, network, routing, handler, each service, each database, render. Plus bottlenecks and the slowest spans
Database	Slow queries over time, time per operation, top slow queries, table hot spots
Rendering	Render time per route, SSR, hydration, image loading, layout shift, and the React components that render too often
Errors	Exceptions over time, top types, errors by service, and the stack trace and request of each one. Browser errors too
Logs	Every log line from every service, filtered by level, service, or text
Metrics	CPU, memory, GC, request times, and your own counters per service, as charts; split by any attribute
Services	Health of each service, database, and outside host, and a map of who calls whom
Insights	Every N+1, duplicate query, slow query, and render problem, ranked, with how to fix it
Settings	How each connected service is set up, and your dashboard preferences

One trace#

Open any trace for the full detail: the timeline, SQL, HTTP calls, logs, and errors, across every service it touched.

Time#

Database and outgoing times are wall time: spans that overlap count once, so a request never shows more database time than server time. When queries ran at the same time, "all queries added up" is shown next to it.

How far back#

Running	Ranges
In your app	Up to 24 hours (newest 2,000 traces)
The hub	Up to 7 days
The cloud	Up to 30 days, computed on the server

Ranges longer than the host keeps are hidden.