Load average is the standard unix load metric. Its the average number of
runnable jobs in the scheduler's run queue taken over a specified time period.
This can indeed go above 1 per cpu. It is different than "the number of cycles
used for running processes". There is a "load" figure on the summ
do you have access to CVS? if not, i'll set that up for you. i think
you should drop the new solaris.c into CVS.
-matt
On Fri, 2004-02-13 at 00:48, Martin Knoblauch wrote:
> --- "Adesanya, Adeyemi" <[EMAIL PROTECTED]> wrote:
> >
> > Hi Martin.
> >
> > I would be happy if you took care of the
We have found that a dual-P3 server running gmetad can handle roughly
1000 hosts before it starts to fail. 1000 hosts here means pure-gmond
host data, of course, not picking up summaries of other grids.
In fact, we just hit the case where adding an additional (128-node)
monitored cluster made
--- "Adesanya, Adeyemi" <[EMAIL PROTECTED]> wrote:
>
> Hi Martin.
>
> I would be happy if you took care of the various solaris.c patches.
> Yes, that was a nice fix by Robert!!
>
>
> Yemi
>
OK, "solaris.c" seems to be in my court now :-) I am preparing a
version that works with 2.5.7 (re