Load average

[jeff vier]

On Tue, 2003-11-18 at 05:39, Gert Lindstrom wrote:
> Could someone please explain these numbers for me and if its ok or
> not.
> What would be the correct setting for a 1Ghz processor with 1 Gig
> memory.
> System Load;CRITICAL;HARD;3;CRITICAL - load average: 6.55, 4.75, 3.19

Were they my systems, I'd be a *little* concerned, if it happened a lot.

What "load" means in Unix (and, hence, Linux) is 'the average number of
processes waiting for processor cycles per the designated time period'. 
The given time periods are 1, 5, and 10 minutes respectively.

The ballpark I use for my Intel systems is that a load of 4 /per
processor/ with "some regularity" (open to interpretation, of course) is
earning it's keep.  If it spikes on a rare occasion to 6 or 6.5, that's
no big deal, as long as it comes back down to normal 99% of the time.

Past about 8 (per processor) the machine is struggling pretty hard, and
I'm looking at either beefing the system up, distributing some of its
workload to another machine, or buying a hardier machine.

At 10, it's pretty much locked up, so, something's wrong.

So, if your system seems happy, and it's not constantly sitting at 6+ on
the first number, I wouldn't worry too much.

This is just my opinion for experience - your mileage may vary.

Oh, and if it's a Sparc or Alpha processor, give it another +1 or +2
across the board (depending, of course, on other system specs)



-------------------------------------------------------
This SF. Net email is sponsored by: GoToMyPC
GoToMyPC is the fast, easy and secure way to access your computer from
any Web browser or wireless device. Click here to Try it Free!
https://www.gotomypc.com/tr/OSDN/AW/Q4_2003/t/g22lp?Target=mm/g22lp.tmpl
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null