When having a lot of processes, like several thousands, on a system and monitoring them,
this lead to a lot of performance counters being created in Check_MK. The management of
these counters was not very optimal, because not updated counters were only deleted based
on time. So systems with a lot of process creations were handling many non needed counters
during every check interval. This affected CPU and memory usage during checking such systems.
As we get the full process table, we have now changed the logic to automatically delete the
counters of all processes that are not existant anymore. This reduces the number of counters
To the list of all Werks