Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

on my linux appliance based on 3.0.0-14 kernel I got:

RPN:/tmp# ps axuf | grep migration
root         6 92.9  0.0      0     0 ?        S    Apr23 2788:33  \_ [migration/0]
root         7 99.7  0.0      0     0 ?        S    Apr23 2993:20  \_ [migration/1]

my top is

RPN:/tmp# top -b -n1
top - 12:03:41 up 2 days,  2:18,  5 users,  load average: 25.76, 25.26, 24.73
Tasks: 171 total,   1 running, 168 sleeping,   0 stopped,   2 zombie
Cpu(s): 14.0%us, 12.6%sy,  0.8%ni, 72.0%id,  0.3%wa,  0.0%hi,  0.3%si,  0.0%st
Mem:   1543032k total,  1264728k used,   278304k free,    25308k buffers
Swap:        0k total,        0k used,        0k free,   183168k cached

My question: why processes "migration/N" take so much CPU?

share|improve this question
    
Do you have any start-up tasks? – evan.bovie May 19 '12 at 5:35

Compare CPU usage between.

top -b -n1 | grep migration
and
ps axuf | grep migration

Some 3.X.X.X kernel's have been known to report incorrect PS value's

share|improve this answer

This appears to be a known bug. This openVZ forum post suggests to set the proc/sys/kernel/sched_cpulimit_nr_balance to 0.

# echo 0 > /proc/sys/kernel/sched_cpulimit_nr_balance

Which is a workaround you can use until a patch is released.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .