Nagios Support Forum

Afternoon.

Im coming from https://support.nagios.com/forum/viewto ... =7&t=50429
Same version 2.2.1 and OS Debian 8 and 9

where the answer is:

This actually is fixing an issue with the previous version.

If you are sending 5 packets and waiting several seconds for each packet, a timeout of 10 should exit BEFORE it can continue to check all 5 times, which is what it is doing.

I dont fully approved of the answer, i do understand it but i think it should be changed. Let me explain.

If i take the pure ping as an example. This is returning after 14seg, the 4 sec it took to do the pings and then the 10sec it took from the -W variable for timeout.

If i take the check_icmp as an example.

time /opt/wocu/embedded/nagios/libexec/check_icmp -H 8.8.8.1 -w 1000,100% -c 3000,100% -p 5 -t 10
CRITICAL - 8.8.8.1: rta nan, lost 100%|rta=0.000ms;1000.000;3000.000;0; pl=100%;100;100;; rtmax=0.000ms;;;; rtmin=0.000ms;;;;

real 0m10.004s
user 0m0.008s
sys 0m0.008s

It actually returns what is expected. 5 packets in a timeout of 10.

NOW, lets see the check_ping.
This is where im lost.
1) Why this behavior ONLY applies when you're using more than 1 packet. If i use -p 1 this is the reply 2) Why is taking 11 seg more to reply after the initial pings. Shouldnt it take similar to the time /bin/ping takes?
3) Since this is a monitoring tool, shouldn't the timeout specify to the entire command not only to the package count, similar to what check_icmp does?

Any clarification is appreciated.
Regards.

This actually is fixing an issue with the previous version.

If you are sending 5 packets and waiting several seconds for each packet, a timeout of 10 should exit BEFORE it can continue to check all 5 times, which is what it is doing.