Having some issues getting my configuration to work when I try to simulate high volume bot interactions. I am running a load test on Genesys Cloud US1 to check bot performance at scale. The goal is to see how the system handles 300 concurrent WebSocket connections initiating bot dialogs. I am using JMeter 5.6 with the WebSocket sampler plugin. The test targets the /api/v2/analytics/conversations/events endpoint for data validation and the standard bot websocket endpoint for interaction.
The issue is immediate. When the thread count goes above 150, I start getting 429 Too Many Requests errors from the REST API calls used to set up the conversation context before the websocket opens. The bot flow itself is simple, just a text response and a data action. The data action calls an external webhook. The webhook is fast, under 100ms. But the Genesys Cloud API gateway seems to be throttling the setup requests.
Here is the error I see in the JMeter response:
HTTP/1.1 429 Too Many Requests
Retry-After: 5
Content-Type: application/json
{
"message": "Rate limit exceeded. Please retry after 5 seconds."
}
I checked the API documentation for rate limits on bot interactions. It says the limit is 100 requests per second for conversation updates. My JMeter test is configured to send requests sequentially per thread, but with 300 threads, the aggregate rate is high. I am not sure if the rate limit applies per user or per application. The websocket connections stay open, but the REST API calls for logging and context fail.
I need to know if there is a way to increase the rate limit for bot interactions. Or if I should change my test approach. The current setup makes it impossible to test true concurrency. I am using the standard Genesys Cloud API keys. The environment is a production-like staging instance. Any advice on how to structure the JMeter test to avoid hitting these limits would be helpful. I want to measure the bot response time, not the API rate limit error rate.