# What does the VARTA error `concurrency_limit_exceeded` mean?

Reserved for a tenant or deployment with too many calls in flight at once.


**HTTP status:** `429`  ·  **Type:** `rate_limit_error`

## What it means

Reserved for a tenant or deployment with too many calls in flight at once.

## What causes it

Cause not yet documented as a live `/v1` path. The underlying concurrent-call cap exists (`backend/app/security/abuse_guard/limiter.py`) but today is wired only into an internal, non-`/v1` endpoint; a `/v1` caller who trips a concurrency limit currently receives `rate_limit_exceeded` instead.

## What to do

Treat `rate_limit_exceeded` as covering this case for now.