Screen Recording API 503 during JMeter spike

Configuration is broken for some reason when hitting the screen recording endpoint under load. Running a JMeter test with 200 concurrent threads triggers immediate 503 Service Unavailable errors when attempting to start screen recordings.

“The recording service scales automatically based on concurrent active sessions.”

Docs say it auto-scales, but the response time spikes to 30s before failing. Is there a hard limit on simultaneous recording starts per org? Using Genesys Cloud EU.

Check your request payload structure and authentication scope. The 503 often indicates the backend queue is saturated, not a hard org limit. Ensure your JMeter script respects the 100 requests/minute rate limit for recording metadata calls. Implementing jitter in your thread start times usually prevents this bottleneck.

Note: Verify the recordings permission is granted to the service account used in the test.