If you are looking on the Elastifile storage nodes, under the Google instance monitoring, you might be surprise that the CPU is 100% all the time.
In fact, you can find that the the instance is pretty idle:
root@elastifile-elfs-0971f003 ~# top
top - 08:32:28 up 17 days, 17:30, 1 user, load average: 0.00, 0.06, 0.11
Tasks: 147 total, 1 running, 146 sleeping, 0 stopped, 0 zombie
%Cpu0 : 1.3 us, 0.0 sy, 0.0 ni, 98.3 id, 0.3 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu1 : 1.3 us, 0.3 sy, 0.0 ni, 98.0 id, 0.3 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu2 : 0.7 us, 0.0 sy, 0.0 ni, 99.3 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
%Cpu3 : 0.7 us, 0.0 sy, 0.0 ni, 99.3 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st
KiB Mem : 38973236 total, 25733068 free, 9938460 used, 3301708 buff/cache
KiB Swap: 0 total, 0 free, 0 used. 26507708 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
17264 root 20 0 1147864 236660 109500 S 3.0 0.6 2090:25 python
1703 root 20 0 32.8g 11.1g 2.3g S 2.3 30.0 2896:48 elfs
1275 root 0 -20 0 0 0 S 0.3 0.0 4:26.57 kworker/2:1H
1 root 20 0 191132 4048 2588 S 0.0 0.0 0:18.78 systemd
2 root 20 0 0 0 0 S 0.0 0.0 0:00.26 kthreadd
The reason for this behavior is that Elastifile uses a Kernel configuration to enable idle=poll, so it will be reported to Google as fully-utilized:
root@elastifile-elfs-0971f003 ~# cat /etc/sysconfig/grub
GRUB_TIMEOUT=0
GRUB_DISTRIBUTOR="$(sed 's, release .*$,,g' /etc/system-release)"
GRUB_DEFAULT=saved
GRUB_DISABLE_SUBMENU=true
GRUB_TERMINAL="serial console"
GRUB_SERIAL_COMMAND="serial --speed=38400"
GRUB_CMDLINE_LINUX="crashkernel=auto console=ttyS0,38400n8 intel_idle.max_cstate=0 processor.max_cstate=0 idle=poll"
GRUB_DISABLE_RECOVERY="true"
In order to monitor the real CPUs consumption, you can login to the Elastifile UI.
From the left menu, choose the 'Services' option.
