Hi all,
I am running a load test for our new BYOC edge deployment in AWS us-east-1. The goal is to validate the stability of our SIP trunk configuration when handling high concurrent call volumes. I am using JMeter to simulate 200 concurrent inbound calls hitting our Genesys Cloud edge.
The issue is that after about 150 active sessions, we start seeing a significant number of 408 Request Timeout errors from the upstream carrier. The calls are not failing immediately but are timing out during the initial SIP INVITE handshake. The error logs in the Edge diagnostic dashboard show:
SIP/2.0 408 Request Timeout (via SIP/2.0/TCP carrier-sip.example.com)
My current JMeter configuration uses a constant throughput timer to ramp up the load. I have verified that the network bandwidth is not the bottleneck, as the latency between the edge and the carrier is stable at <50ms. The Genesys Cloud edge logs show that the requests are being sent correctly, but the carrier side is not responding in time.
I have checked the API rate limits for the BYOC configuration endpoint, but this seems to be a media layer issue, not an API issue. I am using the latest Genesys Cloud Edge version (v12.4.0).
Has anyone experienced similar timeout issues during high-concurrency load tests on BYOC edges? Is there a specific configuration for the SIP stack timeout or a limit on concurrent SIP transactions that I might be hitting?
Any advice on how to tune the edge settings or the JMeter script to better simulate real-world load without triggering these timeouts would be appreciated.
Thanks.