Skip to content

server_at_capacity

HTTP 503 · code server_at_capacity · limit MaxConcurrentRequests

The server is handling as many requests as it is configured to allow.

Sent before routing when MaxConcurrentRequests is set and reached. Retry-After: 1 is set so clients back off instead of retrying in a tight loop.

Retry after the Retry-After interval, with back-off.

Add capacity (more instances), or raise MaxConcurrentRequests if the machine has headroom.

{
"type": "https://photon.agenticmarket.dev/errors/server_at_capacity",
"title": "the server is at its concurrency limit, retry shortly",
"status": 503,
"code": "server_at_capacity",
"limit": "MaxConcurrentRequests"
}

Match on the code field rather than on title, which is meant for people and may change. See all error codes.