[Nagiosplug-devel] Problem with check-host-alive

[Thomas Guyot-Sionnest]

On 16/03/07 04:57 AM, Akiko Moyo wrote:
>  
> I constantly have false alarms on pings checks when overpass RTA=100ms
> even if they are declared in the checkcomand.cfg file as follows:
>  
> define command{
>         command_name    check-host-alive
>         command_line    $USER1$/check_ping -H $HOSTADDRESS$ -w 300.0,80%
> -c 500.0,100% -p 1
>         }
>  
> I have tested several different values and still same result.
>  
> Your advice and help will be very much appreciated.

Did you restart Nagios after changing the settings? What happens if you
run the command manually from the command line (you can try with a
remote host like www.asus.com.tw to make sure latency goes over 100ms)?
Which version of Nagios-plugins are you using?

A suggestion is to replace check_ping with check_icmp, the latter being
a complete ping implementation that doesn't rely on any external
command. Make sure first that check_icmp is owned by root and setuid
(runs as owner):

chown root:root /usr/local/nagios/libexec/check_icmp
chmod 4755 /usr/local/nagios/libexec/check_icmp

_OR_ is you have local users and don't want them to be able to run
check_icmp:

chown root:nagios /usr/local/nagios/libexec/check_icmp
chmod 4750 /usr/local/nagios/libexec/check_icmp

Thomas