Instance returns 202 for HTTP calls intermittentlyIssue Load Balancer returns 202 for HTTP calls intermittently if all requests back to back from the same client is been sent to the same node.CauseIf your integration exceeds the threshold of requests that can be processed by a single thread, then any subsequent transactions start to queue. This might not be a problem if the queueing only happens for short periods of time. However, if there are more than 10 "waiters" for a single session, or the currently executing transaction takes longer than 120 seconds, ServiceNow starts to automatically reject any additional requests to the same session. These requests are returned to the client with HTTP code 202.ResolutionSo if the same request going sequentially to the same node all the time there is a high chance of all threads occupied with the same set of processing and sending back 202 to the client.it is not because of semaphores exhausted it is because of threads unavailability.You should be able to reproduce the issue using POSTMAN, where if you make sequential call's to an instance it uses the same BIGIPServerpool cookie and send call's to the same node, if we clear the BIGIPServerpool cookie from the Postman client then immediately call is sent to a different node.So in order to resolve this issue, you need to make sure that the user-agent (Client) should not preserve BIGIPServerpool cookies, check which client you are using and make sure the BIGServerPool cookie is not preserved.