On Sat, 15 Oct 2011 04:37:00 +0100
Tiago Marques <tiagomnm@...> wrote:

> On Sat, Oct 15, 2011 at 1:26 AM, Patrick Serru <patrick@...> wrote:
> 
> > On Friday 14 October 2011 13:16:53 Timothy Pearson wrote:
> > > I suspect what you saw may have been a transient Xorg glitch.  If you
> > can,
> > > try to reproduce the problem on a machine with a different graphics card
> > > (and therefore a different Xorg driver).
> > >
> > > Tim
> > ---------------------------------------------------------------------
> > I was about to join the opinion of Tim...
> > So I spaced Ksysgard measures to 5 seconds (bottom left, 20sec for bottom
> > right), and stated TOP to a an issue every 20 seconds. I waited, waited ...
> > But luck at restart time (-: after the siesta). I give you the raw results,
> > leaving you the fineness of the analysis.
> >
> >
> The graphics have no legend and I can't find correspondence on mine. top
> seems fine. As suggested, you can also try htop. What am I missing?

There is one oddity--in the first screen shot, top shows ~50% CPU usage,
but with no process using enough CPU to account for that.  I suspect Patrick
may have just missed capturing a shot which would have shown which
process was responsible.