Difference beetween echo return and nagios shown

[djulien]

Hi everyone

I try to understand a mistake :

I use an old pluggin called check_win_users (https://exchange.nagios.org/directory/P ... ck/details) to check the number of users on my RDP server.

if I execute the following command :
# su nagios /usr/lib/nagios/plugins/check_win_users.sh 172.16.0.12 NAGIOS 30 25 && echo $?

I've got this result :
4 users(s).
0

And this is perfectly correct.

But in Nagios, I've this :
"OK" -1 user(s).

Why does Nagios push "-1 user(s)." and not the result of the command as it does ?

The problem is only on the value of the result of the check commend, the rest ("user(s)") is in the echo like it does.

The following command works :

under root user or nagios user
expr $(snmpwalk 172.16.0.12 -v2c -c NAGIOS -Oqv 1.3.6.1.2.1.25.4.2.1.2 |grep winlogon.exe |wc -l) - 1

return 4



Here is the script :
#!/bin/bash

STATE_OK=0
STATE_WARNING=1
STATE_CRITICAL=2
STATE_UNKNOWN=3
STATE_DEPENDENT=4

nbUsers=`expr $(snmpwalk $1 -v2c -c $2 -Oqv 1.3.6.1.2.1.25.4.2.1.2 |grep winlogon.exe |wc -l) - 1`

if [[ $nbUsers -gt $3 ]]
then
echo "$nbUsers utilisateur(s)."
exit $STATE_CRITICAL;
else
if [[ $nbUsers -gt $4 ]]
then
echo "$nbUsers utilisateur(s)."
exit $STATE_WARNING;
else
echo "$nbUsers utilisateur(s)."
exit $STATE_OK;
fi
fi
exit $STATE_UNKNOWN;


Someone have an idea about this difference ?

thanks a lot

[cdienger]

Is the snmpwalk causing a time out? Find out how long it is taking:

Code: Select all

time snmpwalk 172.16.0.12 -v2c -c NAGIOS -Oqv 1.3.6.1.2.1.25.4.2.1.2 |grep winlogon.exe |wc -l

The timeout can be adjusted on the snmpwalk command with the '-t' flag if needed.

The service check timeout is usually set to 60, but double check it in the nagios.cfg file. The directive would look like:

Code: Select all

service_check_timeout=60