Nagios High Load on Rasbian

[SilverFox]

I have 4 Pi's setup as a picluster with running Nagios on the Controller and Looking at the load on the other devices. I am running the standard services with Boinc on all 4.

Nagious is telling me that all the PI's has a high current load:

Current Status:
CRITICAL
(for 0d 17h 51m 19s)
Status Information: CRITICAL - load average: 4.72, 4.72, 4.72

Now there is a setting call Max_concurrent_Connections. I have adjust this setting to several different numbers according to Nagios documentation.

Has anyone been able to setup their PI with Nagios and not get this error?

SilverFox

[bzt]

Hi,

"Max_concurrent_Connections" won't help you with the load, that sets how many computers can connect to your nagios at any given time.

First, are you sure you are really having a huge load? If you ssh to one of your Pi, and issue "uptime" do you see the same values as in nagios? If so, what does "top" say, which process eats up the cpu? Is it your nagios plugin perhaps?

If so, there are tons of cpu load plugins for nagios. Just pick another, possibly one which implements a passive check, that way your Pis won't execute anything unless your Controller asks them (preferably no more frequently than 5-15 mins).

If not, and the load is real and it's not the nagios plugin, then "top" will tell you which application is misbehaving. Change the configuration of that application to solve the issue.

Cheers,
bzt