- Looking for some advice on troubleshooting this rate-limiting issue when importing bulk schedules via the WFM API.
- Using JMeter to simulate a high-concurrency load test for schedule updates in the ap-southeast-1 region.
- Target endpoint is POST /api/v2/wfm/userschedulesegments.
- The goal is to push 500 requests per second to see how the API handles concurrent writes during peak shift changes.
- After about 30 seconds of steady throughput, the response time spikes significantly.
- The API starts returning 429 Too Many Requests errors instead of 200 OK.
- The retry-after header suggests waiting, but the exact limit isn’t clear from the standard docs.
- Current JMeter config uses HTTP Request Defaults with basic auth and standard JSON headers.
- Not seeing any 403 Forbidden or 401 Unauthorized issues, just pure rate limiting.
- Wondering if there is a specific endpoint-level cap for WFM schedule imports that differs from general API limits.
- Also checking if the load pattern needs to be flattened to avoid hitting the burst limit.
- Any insights on the exact threshold for this endpoint would be helpful for tuning the test script.
Pretty sure the WFM API enforces strict rate limits that will throttle 500 requests per second regardless of region.
- Implement exponential backoff in JMeter to handle 429 responses gracefully.
- Reduce concurrency to match the documented limits for
userschedulesegmentsto prevent timeout spikes.