Threshold for processes

Palle L Jensen palleje at gmail.com
Mon Feb 4 17:00:08 CET 2008


Very good explanation, that cleared all questions.

Thanks,
Palle

> -----Original Message-----
> From: nagios-users-bounces at lists.sourceforge.net [mailto:nagios-users-
> bounces at lists.sourceforge.net] On Behalf Of Marc Powell
> Sent: Monday, February 04, 2008 10:51 AM
> To: nagios-users at lists.sourceforge.net
> Subject: Re: [Nagios-users] Threshold for processes
> 
> 
> 
> > -----Original Message-----
> > From: nagios-users-bounces at lists.sourceforge.net [mailto:nagios-users-
> > bounces at lists.sourceforge.net] On Behalf Of Palle L Jensen
> > Sent: Monday, February 04, 2008 9:09 AM
> > To: nagios-users at lists.sourceforge.net
> > Subject: [Nagios-users] Threshold for processes
> >
> 
> 
> >
> > When I check the CPU/Ram/Network utilization it shows very low on both
> > CPU's (2-5%) and below half of the Ram utilization, and only 1/10th of
> the
> > swap file. Network traffic is low as well.
> >
> >
> >
> > Here is my question:
> >
> >
> >
> > If the CPU and Ram is not overloaded, what is the critical part with
> > processes? And what is really the maximum processes that can be run,
> when
> > the CPU show no overload (not even close)? Also is the default
> threshold
> 
> It is entirely possible to have hundreds, thousands or 10's of thousands
> of processes 'running' but in a sleep or otherwise idle state with no
> system impact if you have enough memory to support them. The critical
> part would be the number of processes ready to run but waiting on
> processor time. This is generally indicated by the system's load numbers
> but even that is not a hard-and-fast measure. For example, I have a quad
> processor system with an average load of around 20. That means that
> there are 5 processes per processor running, or waiting to run at any
> given time (or waiting to access IO systems). Because this is not a
> real-time system, it's a mail scanning machine, the few seconds delay
> introduced by the number of processes waiting is acceptable. This
> probably wouldn't be acceptable on desktop or other type of more
> real-time service but even then, priorities can help a lot to maintain a
> high load but an interactively normal system.
> 
> > set in Nagios just a general threshold i.e Warning over 200 and
> Critical
> > over 250 procs. What would make the decision of the Warning/Critical
> > threshold?
> 
> The defaults seem pretty arbitrary to me. You should set them to be what
> you consider normal for the machine and the duties it's performing. For
> example, on the mail system above, it is normal and acceptable to have
> ~360 processes 'running' at any given time. I'd be interested if that
> exceeded 450-500ish. On another system, it's normal to have about 80
> processes running. I'd be concerned if that exceeded 100.
> 
> HTH,
> 
> --
> Marc
> 
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> 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


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
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





More information about the Users mailing list