I’m testing the ability for T-Server to directly manage stuck calls using the method defined in chapter 6 of the Framework 7.1 Management Layer User’s Guide. I’ve applied the config options (notify-idle-tout, cleanup-idle-tout & periodic-check-tout) under the relevant section. Now the problem is that I’ve only done this in our test/lab environment, it’s not production. Naturally enough, there aren’t any calls travelling through the test environment
So the question is, how might I simulate a stuck call ?
If the switch in your environment does not support querying the PBX for the status of the call , these settings will clear calls where no events have been received for the “cleanup-idle-timeout” period.
For example, in our environment (NEC), we have the cleanup-idle-timeout set for 30 minutes* . If we have an agent who is on a call longer than 30 minutes (and does not put the caller on hold), the cleanup will remove the call, even though it is still active on the PBX. Unfortunately, once the call actually terminates, the TServer does not recognize the EventOnHook, and I have to manually send an EventOnHook via the Support Phone. (I have had similar behavior in a Nortel Meridian 1 environment).