Why is this setting causing rate limit errors when simulating concurrent chat bot sessions?
Environment: Genesys Cloud US1
Tool: JMeter 5.6
Target: /api/v2/bots
Config: 150 concurrent threads sending POST requests to create bot interactions.
The goal is to validate the bot capacity before the holiday rush. The test script logs in, retrieves a bot ID, and then sends a series of message requests. After about 30 seconds, the response rate drops significantly. The server returns 429 Too Many Requests. The error message indicates the client is sending too many requests.
I have checked the documentation for AI/Bot limits. I see limits for API calls per minute. However, the test stays within the documented limit for standard users. The issue seems specific to the bot interaction endpoint.
- Reduced thread count to 50 concurrent users. The test passes without errors. This confirms it is a concurrency or rate limit issue.
- Added a 2-second pause between requests in the JMeter script. The 429 errors stop, but this does not reflect real-world user behavior where messages are sent instantly.
Is there a specific rate limit for bot message API calls that is lower than the general API limit? How should the JMeter script be configured to handle these limits correctly without slowing down the test?