Massive load average spikes tracing the cause

[Tyler Earman]

Please CC me if you would.


Basically, I'm running a machine that will have massive load average
spikes (up into the 5-10 range on a 2 CPU machine) at seemingly random
intervals.  I've tried most normal tracing conventions but now I'm
going to have to delve into the kernel to find out exactly what's
going on here.

The kernel does softlock occasionally on this event, but the CPU usage
stays very low (1-10%).

I'm guessing this is an I/O lockup somewhere, but the question is /where/.

So, in short, how would I go about hooking into the kernel directly to
try and find these problems?  Is there a way to watch just about
everything going on when these events happen?

Thanks for the help,
Tyler
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/