The SystemEDGE agent runs high CPU usage and Ehealth is not able to collect the performance data on a consistent basis.
SystemEDGE is a single threaded application, and it will not be able to respond to bulk SNMP requests while it is in the process of updating the MIB.
The following entry in the Sysedge.log is a good indicator the Server is overloaded (CPU and\or disk subsystem) and this is impacting SystemEDGE's ability to refresh the MIB in a timely manner:
query_monitor_entries(): Performance problem detected (monitoring cycle). Elapsed time 185.826648s). SystemEDGE may not be responding to SNMP queries during that time.
This issue is brought about by Server\OS performance issues. Review server configuration and consider increasing CPU and\or memory and review disk subsytem for potential I\O related bottlenecks.