In a configuration involving both a primary server and one or more secondary servers, Service Desk services must be started in a particular order, and they must be stopped in a particular order. Otherwise, the servers are likely to be out-of-sync, resulting in communication problems. This document explains the correct procedure.
The Services Explained
The CA Service Desk Manager service on the primary server controls the starting and stopping of Service Desk processes such as the daemon manager, domsrvr, webengine, and spelsrvr. It starts and stops the appropriate subset of these processes on the primary itself as well as on all secondary servers in the environment. In order to do this, a communication line between the primary server and each secondary server must exist. It is the CA Service Desk Remote Proctor service on each secondary server that establishes the communication line over which the primary instructs that secondary.
The Start-Up Procedure
To properly start up a Service Desk application environment, the CA Service Desk Remote Proctor service on any and all secondary servers should always be started first. Once the Remote Proctor service is running on all secondary servers in the environment, then the primary server's CA Service Desk Manager service should be started, which will in turn, start up all the appropriate processes on the secondary servers in the environment.
The Shut-Down Procedure
To properly shut down a Service Desk application environment, the primary server's CA Service Desk Manager service should always be stopped first. Stopping the CA Service Desk Manager service on the primary server should cause all related processes on the secondary servers (including domsrvr, webengine, spelsrvr, etc.) to stop. Once the stop completes on the primary, the CA Service Desk Remote Proctor service on each of the secondary servers should be stopped in turn, closing the line of communication between the secondary servers and the primary server.
The Recycle Procedure