2

What's going on here?

I'm running nothing on the system, iotop and iftop show the network and hard drive are both idle, and top (sorted by %CPU) shows nothing running. So why is the system only 50% idle? What's the other 50% waiting for? How can I find out?

top - 12:01:05 up 3 days, 15:03,  1 user,  load average: 6.00, 6.01, 6.05
Tasks: 179 total,   1 running, 178 sleeping,   0 stopped,   0 zombie
Cpu(s):  0.7%us,  0.0%sy,  0.0%ni, 49.7%id, 49.7%wa,  0.0%hi,  0.0%si,  0.0%st
Mem:   2053996k total,  1992600k used,    61396k free,    81680k buffers
Swap:  4092924k total,    10740k used,  4082184k free,  1338636k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
 1042 deb       20   0 21468 1412 1000 R    1  0.1   0:00.03 top
    1 root      20   0 24188 1952 1152 S    0  0.1   0:01.44 init
    2 root      20   0     0    0    0 S    0  0.0   0:00.05 kthreadd

Update: dmesg shows the printer driver misbehaving:

[28858.561847] cnijnetprn[1503]: segfault at 29 ip 00007f56cf3480f7 sp 00007fffb964ec30 error 4 in libcnnet.so.1.2.0[7f56cf345000+9000]
[68851.187802] cnijnetprn[9180]: segfault at 29 ip 00007ffe7636a0f7 sp 00007fff9a8b1990 error 4 in libcnnet.so.1.2.0[7ffe76367000+9000]
[155412.107826] cnijnetprn[19966]: segfault at 29 ip 00007fc31de770f7 sp 00007fffc03aa8e0 error 4 in libcnnet.so.1.2.0[7fc31de74000+9000]

and also some issue with cp:

[248041.172067] INFO: task cp:27488 blocked for more than 120 seconds.
[248041.172071] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[248041.172075] cp              D ffffffff81805120     0 27488  27345 0x00000004
[248041.172080]  ffff880078d57a38 0000000000000046 ffff880078d579d8 ffffffff81032a79
[248041.172085]  ffff880078d57fd8 ffff880078d57fd8 ffff880078d57fd8 0000000000012a40
[248041.172090]  ffff88007b818000 ffff880069acc560 ffff880078d57a18 ffff88007f8532c0
[248041.172095] Call Trace:
[248041.172104]  [<ffffffff81032a79>] ? default_spin_lock_flags+0x9/0x10
[248041.172109]  [<ffffffff8110a360>] ? __lock_page+0x70/0x70
[248041.172114]  [<ffffffff815f0ecf>] schedule+0x3f/0x60

I did try copying something to the USB stick that's plugged into the router and mounted onto this computer using mount.cifs. That almost always causes everything to lock up, so I'm guessing that's the problem. I'll reboot and stop using mount.cifs.

Improve this question
1
  • Thanks. I updated my question with some dmesg output. Looks like it's a dodgy cifs mount that was the problem. You should post an answer and I'll accept it.
    – Chris Moore
    Mar 25, 2012 at 22:37

1 Answer 1

Reset to default
3

At the time of that measurement your system was idle 99.4 % of time.

Cpu(s): 0.7%us, 0.0%sy, 0.0%ni, 49.7%id, 49.7%wa, 0.0%hi, 0.0%si, 0.0%st

According to the manpage of top, wa stands for iowait:

wa -- iowait
Amount of time the CPU has been waiting for I/O to complete.

Although the CPU is waiting for I/O to complete, it's not in use and can be considered idle.

The question is: Why is there such a high iowait value, when iotop doesn't show any disk activity? This can be caused for example by disk drives that are about to fail or some other issues with local or remote mounts. I suggest checking your disks with the Disk Utility and looking for I/O related messages in the output of dmesg.

Improve this answer

You must log in to answer this question.

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