When monitoring CPU utilization on the NetScaler, you will periodically notice Management CPU (CPU 0) reporting 100% Utilization. This will most likely appear cyclic (recurring at a fairly regular interval).

This is normal behavior. 

The NetScaler's management CPU performs periodic tasks to compress and roll-over log files, amongst other things.  These processes are designed to take 100% CPU utilization on the Management CPU.

You can confirm this by using the "top" command from the shell.  When doing so during a high utilization event, you will see lines similar to the following taking up most of the CPU:

  PID  PPID STAT      TIME %CPU   VSZ   RSS %MEM COMMAND
24295 24283 RN     0:05.41 85.1 15040  2900  0.0 tar -czvf newnslog.104.tar.gz newnslog.104 (bsdtar)

High utilization of the management CPU will not affect enduser traffic.

Note: You will also see NSPPE taking up 100% CPU on the other CPU cores, when viewed with the top command.  This is also by design. The packet engines (NSPPE) perform their own kernel scheduling services internally and their utilization cannot be monitored with the top command.

---

Problem Cause

The management CPU is tasked with performing some periodic functions such as compress the log files that cause high CPU utilization which is by design.

Periodically, management CPU utilization will report being at 100%

For information on troubleshooting high management CPU that is constant see CTX226784