need some help troubleshooting predictive routing campaign setup. running a load test on Genesys Cloud US1 with jmeter 5.6. trying to validate api throughput for campaign creation and agent assignment.
the issue starts when i push more than 10 concurrent threads to /api/v2/outbound/campaigns. immediately getting http 429 too many requests. then after a few seconds the jmeter sampler returns http 503 service unavailable. the error body says “internal server error” but no specific trace id.
i am using the standard oauth2 client credentials flow for authentication. the token refresh is working fine. i have checked the architect flow and the campaign is configured with a simple voice interaction. no complex scripts. just dialing and answering.
the environment is a standard org. no custom connectors. i have increased the jmeter ramp-up time to 60 seconds to be gentle but still hitting the limit. the api docs say the limit is 100 requests per minute per client id. i am well below that. only 10 threads.
also noticed that the websocket connection for agent status updates drops with code 1006 when the campaign is active. this happens even if i stop the jmeter test. the agents go offline in the admin console.
i have tried using different client ids but same result. is there a hidden rate limit for predictive outbound campaigns? or is this a bug in the us1 region?
looking for any insights on how to structure the jmeter test to avoid this. maybe i need to add delays between requests? or is the api just unstable right now?
any help would be appreciated. this is blocking our capacity planning for the new quarter. we need to know if we can run multiple campaigns simultaneously without hitting these errors.
thanks in advance.