service firewall check

[m0le121]

i had a problem with the check of the code below.
The code has the name firewall_check.sh on the client.
When i run firewall_check.sh on the client it works correctly (it says running or not running).

Code: Select all

#!/bin/bash
SERVICE=firewall;
status="/etc/init.d/$SERVICE status"

if $status | grep "firewall is running"; then
   echo "Status 0 - OK : $service is running"
elif $status | grep "firewall is stopped"; then
   echo "Status 1 - Critical : $service is not running"
else
   echo "Status 2 - Information : $service is unrecognized service"
fi

At the nagios server i run

Code: Select all

/usr/lib64/nagios/plugins/check_nrpe -H externalip.com -c firewall_check

Code: Select all

define command{
        command_name    firewall_check
        command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -c firewall_check
        }

Code: Select all

define service {
 use                             generic-service
 host_name                      externalip.com
 service_description             10telecom firewall check
 display_name                    10telecom_firewall_check
 check_command                   10telecom_firewall_check
 contact_groups                  admins
}

The output is always

firewall is stopped
Status 1 - Critical : is not running

What goes wrong?

[lmiltchev]

When i run firewall_check.sh on the client it works correctly (it says running or not running).

Do you run the check as root or as nagios user? Can you show us the actual check run as nagios from the command line, along with the output of it? Also, show us how the check is defined in the nrpe.cfg. I suspect you will need to add the command to the sudoers.

[jdalrymple]

Most likely the user running the nrpe daemon doesn't have access to read the service status.

Confirm by launching the script as the user whom is running nrpe.

Repair with a sudoers rule and the appropriate adjustment of your script.

sudoers rule:

Code: Select all

nrpeuser ALL = NOPASSWD:/etc/init.d/firewall status

Adjustment to your script:

Code: Select all

status="sudo /etc/init.d/$SERVICE status"

[m0le121]

i added the code below on the nagios client.
vi /etc/sudoers

Code: Select all

nagios  ALL=NOPASSWD:/etc/init.d/firewall status

i edit my script with the following line:

Code: Select all

status="sudo /etc/init.d/$SERVICE status"

i added the code below on the nagios server.
vi /etc/nagios/cgi.cfg

Code: Select all

authorized_for_read_only=nagios

but if i do "su nagios" on the server...

This account is currently not available.

i have checked the commands as root.

Code: Select all

/usr/lib64/nagios/plugins/check_nrpe -H externalip.com -c 1firewall_check

Status 2 - Information : is unrecognized service

[jdalrymple]

but if i do "su nagios" on the server...

This account is currently not available.

Running check_nrpe on the server isn't necessary, although this is because the nagios user doesn't have a valid login shell

i have checked the commands as root.

Code: Select all

/usr/lib64/nagios/plugins/check_nrpe -H externalip.com -c 1firewall_check

Status 2 - Information : is unrecognized service

Why is it "1firewall_check"?

[m0le121]

sorry i mean:

Code: Select all

 /usr/lib64/nagios/plugins/check_nrpe -H externalip.com -c firewall_check

[jdalrymple]

Please change your script so that you get some usable output:

Code: Select all

#!/bin/bash
SERVICE=firewall;
status="/etc/init.d/$SERVICE status"

if $status | grep "firewall is running"; then
   echo "Status 0 - OK : $SERVICE is running"
elif $status | grep "firewall is stopped"; then
   echo "Status 1 - Critical : $SERVICE is not running"
else
   echo "Status 2 - Information : $SERVICE is unrecognized service"
fi

[m0le121]

edit on the nagios client
I changed the code to:

Code: Select all

#!/bin/bash
SERVICE=firewall;
T1=$(sudo /etc/init.d/$SERVICE status)
RUNNING="$SERVICE is running"
STOPPED="$SERVICE is stopped"
UNKNOWN="$SERVICE is unknown"
if [ "$T1" = "$RUNNING" ]; then
   echo "Status 0 - OK : $RUNNING"
   exit 0
elif [ "$T1" = "$STOPPED" ]; then
   echo "Status 1 - Critical : $STOPPED"
   exit 2
else
  echo "Status 2 - Information : $UNKNOWN"
  exit 3
fi

if i run it at the client it works perfectly.

i edited /etc/sudoers

Code: Select all

root    ALL=(ALL)       ALL
nagios ALL=(ALL)        ALL
nagiosadmin ALL=(ALL)   ALL

i edited /etc/nagios/nrpe.cfg

Code: Select all

command[firewall_check]=/usr/lib/nagios/plugins/firewall_check.sh

edit on the nagios server
edit: /etc/nagios/objects/commands.cfg

Code: Select all

define command{
        command_name   firewall_check
        command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -c firewall_check
        }

edit: /etc/nagios/conf.d/SERVICE-sip.cfg

Code: Select all

define service {
 use                             generic-service
 host_name                       externalip.com
 service_description             firewall check
 display_name                    firewall_check
 check_command                firewall_check
 contact_groups                  admins
}

When i run the command below from the nagios server as root,nagios,nagiosadmin:

Code: Select all

/usr/lib64/nagios/plugins/check_nrpe -H externalip.com -c firewall_check

the output is always:

Status 2 - Information : firewall is unknown

What did i missed?

[abrist]

Add some debug to output the contents of the $T1 var in case of unknown:

Code: Select all

#!/bin/bash
SERVICE=firewall;
T1=$(sudo /etc/init.d/$SERVICE status)
RUNNING="$SERVICE is running"
STOPPED="$SERVICE is stopped"
UNKNOWN="$SERVICE is unknown"
if [ "$T1" = "$RUNNING" ]; then
   echo "Status 0 - OK : $RUNNING"
   exit 0
elif [ "$T1" = "$STOPPED" ]; then
   echo "Status 1 - Critical : $STOPPED"
   exit 2
else
  echo "Status 2 - Information : $UNKNOWN - $T1"
  exit 3
fi

[lmiltchev]

What is the EXACT output of "/etc/init.d/firewall status" when the service is/is not running?