Trying to understand the rate limiting behavior on the Outbound Campaign API endpoints during high-concurrency load tests.
Background
Running JMeter tests to validate campaign creation throughput. The goal is to simulate 50 concurrent users creating outbound campaigns via the /api/v2/outbound/campaigns endpoint. Environment is Genesys Cloud US-East. Using the latest SDK version 2.5.0 for payload generation.
Issue
When ramping up to 20 threads, requests succeed. At 25 threads, hitting consistent 429 Too Many Requests errors. The response body includes a Retry-After header, but the value is inconsistent. Sometimes it’s 1 second, other times 5 seconds. The error response payload is empty, which makes debugging tricky. No X-RateLimit headers are present in the response, making it hard to predict the limit.
Troubleshooting
- Verified payload structure against the Swagger spec. All required fields are present.
- Checked account-level API quotas. No custom limits applied.
- Reduced thread count to 10. Errors stop. Increased to 15. Errors appear sporadically.
- Tried adding a
Retry-Afterdelay in JMeter. Helps, but throughput drops significantly. Not viable for our scale requirements.
Is there a documented per-endpoint limit for campaign creation? Or is this a global API throttle? Need to adjust the load test strategy accordingly.