On Mon, 2003-03-31 at 01:48, Sam Tilders wrote:
> Folks...
>
> I use check_procs to check if certain things are running, radiusd
> authentication server, nfs daemons, anti virus scanning daemon, etc.
>
> Sometimes I find that I get timeouts calling the check_procs plugin via
> check_nrpe and yet nrpe is quite responsive.
>
> I've narrowed it down to times when the process list is quite large, the
> machine is not necessarily overloaded. Unfortunately Linux shows threads in
> the process list so the process list can be quite long even when there's not
> a lot going on.
>
> Do people have any thoughts on how else to check if a process is running? In
> the context of making the check_procs plugin work better of course.
Check_procs uses the same 'ps' as you use on the command line. I suppose
the testing loop might be eating up CPU cycles. How many proceeses are
we talking about? Is the a lag when you type 'ps -ef' on the comand
line?
--
Karl
This post was automatically imported from historical nagios-devel mailing list archives
Original poster: [email protected]