On 07/17/2012 05:59 PM, zxq9 wrote:
On 07/18/2012 07:46 AM, Orion Poplawski wrote:
On 07/17/2012 04:17 PM, zxq9 wrote:
On 07/18/2012 06:55 AM, Orion Poplawski wrote:
While staring at top on my kvm server waiting for it to crash, I noticed
that my idle Fedora kvm guests appear to be consuming between 8-12% cpu
while my idle SL6 guests appear to be consuming between 0.3-2% cpu as
shown in the qemu-kvm processes in top.
Does anyone have any insight as to why this would be the case?
That depends heavily on what version of Fedora they are (particularly
Rawhide
instances), and what services they are running.
Well, I have a Fedora 17 instance running now with nothing but kernel
processes and the qemu-kvm process still shows using 7-8% cpu.
There's pretty much no such thing as a F17 instance with nothing but the
kernel running. Can you log in to one of the instances and check what top says
is running (like the top 10 or so processes by CPU use)?
Not at that point, because I'd killed everything but the kernel and systemd :)
The odds that kvm itself is spinning more just to keep F17 alive over anything
else is remote, but the odds that something in F17 defaults to wasteful
behavior is quite high (particularly if you are hosting X with Gnome Shell --
there are any number of things that waste cycles in there).
That said, you may be affected by the Fedora version of the July 1st leap
second bug where processes like qpidd eat 1 core (or equivalent) entirely
while stuck in an infinite loop (this behavior can survive a reboot). If you
see a single process in top eating a significant percentage of its allocated
resources try:
date; date $(date +"%m%d%H%M%C%y.%S"); date
on the misbehaving Fedora instance and see if things change.
There are no busy processes. Load avg is at or less than 0.05.
--
Orion Poplawski
Technical Manager 303-415-9701 x222
NWRA, Boulder Office FAX: 303-415-9702
3380 Mitchell Lane [email protected]
Boulder, CO 80301 http://www.nwra.com
