Nexus API returns 400 when trying to add a switchport IP that's already assigned

When trying to add an IP address from an address-lot to a switchport, a 400 (Bad Request) is returned if the IP is already assigned to a switchport. This is misleading, as it indicates there was something structurally/semantically wrong with the request rather than the state of the object being the issue. It would make more sense to return a different response code that aligns better to the situation, e.g. 409 (Conflict).

Command to trigger 400 error (after fd00:101::2/126 is already used):
oxide system networking addr add --rack c7efde80-847a-4ab4-b855-5a0579cfa268 --switch switch0 --port qsfp0 --addr fd00:101::2/126 --lot initial-infra-v6 when fd00:101::2/126

Return:

Error Response: status: 400 Bad Request; headers: {"content-type": "application/json", "x-request-id": "024de8d5-50ee-4055-b9f2-2557d0e31ff1", "content-length": "128", "date": "Tue, 1
0 Mar 2026 20:05:52 GMT"}; value: Error { error_code: Some("InvalidRequest"), message: "address unavailable", request_id: "024de8d5-50ee-4055-b9f2-2557d0e31ff1" }

Start with the oxide system networking addr add API entry point and trace the handling of an address that is already assigned to a switchport. Look for existing API error-response tests around this command. Done means the same conflict returns an appropriate conflict response, such as 409, rather than 400, while other invalid requests retain their current behavior.