Insights

The same rules run in .NET and in Node, so every service gives the same answer. Insights show in the HUD, on each trace, and ranked on the Insights page.

N+1 query#

The same SQL statement runs 5 or more times in one trace, with different parameters. The usual cause: a loop that loads one row at a time.

Fix: load all the rows with one query: a join, IN (...), or your ORM's include.

N+1 HTTP call#

5 or more outgoing HTTP calls in one trace with the same method, host, and path shape. Numbers, GUIDs, and query values in the path count as the same shape, so /orders/1 and /orders/2 match.

Fix: add a batch endpoint, or ask for everything in one call.

Duplicate query#

The same statement runs 2 or more times in one trace with the same parameters. The second run gives you nothing new.

Fix: keep the first result, or cache it for the request.

Slow query#

One query takes 100 ms or more.

Fix: look at the query plan. Add an index, or ask for less.

Database wait#

3 or more slow queries in one trace that overlap in time count as one database wait, not three slow queries. So does a slow session statement on its own (a SET does almost no work, so it waited). The time is wall time: queries that waited together waited once.

Fix: look for locks, a full connection pool, or a busy database.

Errors#

Exceptions on the server and errors in the browser, with the stack and the request that led to them.

Render problems#

From the browser: React components that render too often, slow hydration, slow images, and layout shift. See the Rendering page.

Session statements#

SET, RESET, BEGIN, START TRANSACTION, COMMIT, END, ROLLBACK, SAVEPOINT, RELEASE, DISCARD, and USE are never counted as duplicates or N+1. Apps that run SET LOCAL app.tenant_id before each query repeat them by design.

Change the rules#

Rule	.NET	Node	Default
N+1 repeats	NPlusOneThreshold	nPlusOneThreshold	5
Slow query	SlowQueryMs	slowQueryMs	100