Pool Health Monitor Generates Lots of Close_wait

Hi All,

Im having an issue with my LTM VS, for an ATM application, where the pool health monitor is generating lots of close_wait connections to the Unix server pool (which are not closed) causing the application to eventually hang after a period of time. Is there any way to monitor a service/port of the application on server without making a TCP connection? It seems that the ATM application (which is asynchronous connection) is not closing the health monitor probes generated from F5 LTM. Appreciate for any recommendation / help?

Thanks! Ramil