On 09/28/2010 12:53 PM, Mike wrote:
> Thanks for all the help.  I tried but couldn't see a definite spike anyways.
> I used mpstat to see the irq% for each CPU, and although it shows me spikes
> to 100% for one CPU (not always the same one is spiking irq%) the intr/s
> isn't actually spiking.
> 
> It`s like suddently there aren't more interrupts, but the IRQ% time for the
> CPU shoots up.
> 
> What can make it do that?

Now we're back to my original assertion that I believe what you're
seeing is related to the way % cpu time is calculated statistically via
a timer interrupt.  I would bet that you have a steady amount of cycles
dedicated to servicing the interrupt handler but every 10 minutes there
are a confluence of factors that affect the calculation.

You might be able to use a tool like cyclictest [1] to see if there
really is some inordinate amount of scheduling latency every 10 minutes
as opposed to some sampling anomaly.

[1] https://rt.wiki.kernel.org/index.php/Cyclictest

-- 
Shaun Ruffell
Digium, Inc. | Linux Kernel Developer
445 Jan Davis Drive NW - Huntsville, AL 35806 - USA
Check us out at: www.digium.com & www.asterisk.org

-- 
_____________________________________________________________________
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
               http://www.asterisk.org/hello

asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
   http://lists.digium.com/mailman/listinfo/asterisk-users

Reply via email to