When the message "slump died" is reported by a Service Desk process on a secondary SDM server, it will terminate itself and try to restart and reconnect to the slump process located on the primary SDM server. This is normal behavior.
There are a few possible reasons on why such message and event is seen:
- The slump process on the primary server may be extremely busy shown as a prolong CPU spike.
As such, Slump is taking a long time to respond to processes connected to it on a Service Desk secondary server.
- The slump process on the primary server actually terminated\crashed and is no longer running.
- The tcp socket connection between the processes on the SDM secondary and Slump is lost caused by a network outage or hiccup.
In most cases, the problem reported by our customers has been as a result of point 3. A Wireshark trace from both servers taken leading to and during the event is the course of action to trouble shoot further.
The Slump process is "sslump_nxd" seen in Windows Task Manager or in Unix "ps" output.