stuck on sip trunk capacity limits during jmeter stress tests.
running jmeter 5.6 to simulate 200 concurrent inbound sip calls to a genesys cloud sip trunk in us1. the goal is to verify call admission control behavior when queue capacity is reached.
config details:
- sip trunk: standard enterprise tier
- routing: predictive outbound disabled, inbound only
- queue: max wait 60s, no limit on concurrent calls set (default)
- jmeter thread group: 200 threads, ramp-up 10s, loop count 1
after about 50 concurrent calls, new sip invites start failing with “486 Busy Here” from the genesys sip proxy. this happens even though the queue is not full and agents are available.
checking the sip trace logs, i see the 486 coming from the gateway before the architect flow is even invoked. is there a hard limit on concurrent sip sessions per trunk that i am missing? or is this a rate limiting issue on the sip interface?
any pointers on how to increase the concurrent sip session limit or if this is expected behavior for the enterprise tier?