Enhance Endpoint Descriptor processing and add new ept.xml Elements to change auto-undeploy error processing - adding auto-redeploy.

An endpoint descriptor is automatically undeployed when the number of consecutive errors is greater than the number of connections defined in the ept.xml.  Once the system auto-undeploys the endpoint descriptor, it has to be manually re-deployed (via ZMDES DEPLOY) on every processor in a TPF complex up upon which it was undeployed.  This becomes extremely difficult to manage operationally as more persistent connections are added to a TPF system.

We would like to request the following updates to Endpoint Descriptor processing in TPF:

1. In the persistent connection configuration file, create new configurable element “error threshold” to define the number of connection errors at which point the endpoint definition / persistent connection will be deactivated in the system. 
2. In the persistent connection configuration file, create new configurable element, “error time interval” in seconds, that if “x” connection errors happen within “x” timeframe.  The persistent connection will be deactivated.
3. In the persistent connection configuration file, create new configurable element, “redeploy time interval” in seconds.  After the “redeploy time interval” has passed, system should internally re-activate/re-deploy/mark for use the *.ept.xml (persistent connection configuration).
4. If a client request is made to an endpoint that has a persistent connection configuration in the system and it is not active, issue an HTTP error, however, make it a different error than a “404” to better indicate that a deactivated persistent connection is the reason for the failure.