So I’m seeing a very odd bug with the Agent Scripting API while trying to validate our deployment pipeline under load. We are using JMeter 5.6 to simulate 200 concurrent script updates via POST /api/v2/scripts.
The first 50 requests succeed with a 201 Created status. After that, we start seeing intermittent 400 Bad Request errors. The error payload is not very helpful:
{
“code”: “bad_request”,
“message”: “Invalid node configuration”,
“details”: “Node type mismatch in flow definition”
}
We are using the exact same JSON payload for all requests. The node types are explicitly defined as SetVariable and SystemMessage. There is no schema drift in the payload.
This happens only when the requests are fired within a 1-second window. If we add a 500ms delay between requests, the issue disappears. This suggests a race condition on the server side or a strict rate limit that returns a misleading error code.
Is there a known limit on script creation throughput? We need to deploy 500 scripts in under 10 minutes for our staging environment.
Thanks.