Dealing with a very strange bug here with the Workforce Management Schedule API returning 429 Too Many Requests when pushing 100 threads through JMeter 5.6.2. The rate limit headers are confusing because I am staying well below the documented 50 req/sec limit per tenant, but the errors start appearing almost immediately after ramping up.
I am simulating a shift change scenario where multiple agents update their schedules concurrently. The endpoint POST /api/v2/wfm/schedules is failing with a 429 status code and the response body indicates a retry-after value of 5 seconds. This is blocking our capacity planning for the new quarter.
Here is the payload I am sending:
{
"schedule_id": "sch_12345",
"user_id": "usr_67890",
"start_time": "2024-10-01T09:00:00Z",
"end_time": "2024-10-01T17:00:00Z",
"status": "published"
}
I have added a constant throughput timer in JMeter to ensure we are not exceeding the limit, but the 429s persist. Is there a hidden rate limit for WFM endpoints specifically? Or is this a bug with the Singapore region load balancers? Any help debugging this would be appreciated. I need to know if we can scale this for 500 concurrent users.