After two years of near-perfect service, the PSS FlexxGate Media Gateways deployed at a major healthcare provider were discovered to have a large number of hung SIP channels associated with calls that did not properly tear down.  Fortunately the N+1 failover redundancy PSS engineered into the customer's network prevented the problem from affecting any customer calls, but the condition did cause a memory issue that eventually would have rendered the system unresponsive had it not been detected.

While rebooting every few months might have been an easy solution to prevent the memory leak from getting out of hand, PSS considered this to be an unacceptable bug that had to be fixed.  PSS also did not want the customer to bear the burden of the constant monitoring necessary to reproduce and trace a problem that occurred so infrequently.   Immediately PSS Support began continuously monitoring the system online in real time.  This allowed PSS to pinpoint the problem and install a fix within days of the problem being reported.   The outcome: a potentially serious problem was detected and fixed without a single call dropped on the self-service or contact center platforms.