Why is this setting causing the edge byoc websocket connection to drop with a 1006 error code? i am running a load test on the us1 environment using jmeter 5.6. the goal is to validate concurrent session capacity for a pure cloud setup. we have configured the edge cluster with 10 nodes. the max connections per node is set to 500. total theoretical capacity is 5000 connections. the test script uses the genicloud websocket plugin. it connects to wss://api.mypurecloud.com/icd/websocket. the authentication token is valid. the initial handshake succeeds for all threads. the problem starts when the concurrent thread count reaches 2000. the websocket connections start dropping randomly. the error log shows close code 1006. no reason phrase is provided. the jmeter listener shows a spike in response time right before the drop. the cpu usage on the edge nodes is low, around 15%. memory usage is also low. there are no firewall blocks. the network latency is stable at 10ms. i have checked the architect flow settings. the flow is simple, just a transfer to a queue. no complex scripting. i have tried increasing the timeout settings in jmeter. i have tried reducing the message frequency. nothing helps. the 1006 errors persist at 2000 threads. i need to know if this is a hard limit for the us1 region or if there is a configuration issue. the documentation says the limit is based on the node count. we have 10 nodes. 2000 connections is 40% of the total capacity. it should not fail. is there a hidden rate limit on the websocket handshake? or is the edge cluster not scaling properly? i have attached the jmeter thread group config. the ramp-up time is 60 seconds. the hold loop is 300 seconds. the error rate jumps from 0% to 40% at the 2000 thread mark. this is blocking our performance sign-off. please advise on the correct configuration for high concurrency on edge byoc.