nagios-nrpe service don't start automatically after reboot

[mroter]

After reboot the nagios-nrpe service on a linux server don't start automatically.
We have to run the following to get it up again:

Code: Select all

 /etc/init.d/nagios-nrpe-server restart

I found some info on the net regarding this issue. The service is started before the network is ready.
How can should we resolve this?

[slansing]

What document did you use to install NRPE? Did you follow this one?:

http://assets.nagios.com/downloads/nagi ... _Agent.pdf

If you used an off-site document could you please link it?

[mroter]

We did not used the suggested document above
We where installing a zen-lb plugin and followed the instruction in it:

3. Install nagios NRPE
apt-get install nagios-nrpe-server
(accepted WORKGROUP)
4. Edit /etc/nagios/nrpe_local.cfg
allowed_hosts=<ip>
command[<command name>]=<full path to plugin and any arguments>

// E.g.
allowed_hosts=<nagios_server_ip>
server_address=<local_bind_ip>
server_port=<local_bind_port>
command_timeout=60
connection_timeout=300
command[check_status_carp]=/etc/nagios/EKHScripts/check_zenlb_carp.php
command[check_status_penfarms]=/etc/nagios/EKHScripts/check_zenlb_penfarms.php

5. Restart NRPE daemon
/etc/init.d/nagios-nrpe-server restart

[scottwilkerson]

Run

Code: Select all

chkconf nagios-nrpe-server on

[mroter]

I assume you were referring to chkconfig command, right?
Anyway the box I'm monitoring is running Debian and don't have this command:

cat /proc/version
Linux version 2.6.32-5-686 (Debian 2.6.32-41squeeze2) ([email protected]) (gcc version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Mon Mar 26 05:20:33 UTC 2012

Can you advise on the equivalent? Manual fix...

[GldRush98]

I assume you were referring to chkconfig command, right?
Anyway the box I'm monitoring is running Debian and don't have this command:

cat /proc/version
Linux version 2.6.32-5-686 (Debian 2.6.32-41squeeze2) ([email protected]) (gcc version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Mon Mar 26 05:20:33 UTC 2012

Can you advise on the equivalent? Manual fix...

Debian uses update-rc.d instead of chkconfig. Check the syntax on it as I know it is a bit different than the chkconfig syntax.

[mroter]

I ran the following and got this reply. Is it OK?

root@ZLB-104-QA:~# update-rc.d nagios-nrpe-server defaults
update-rc.d: using dependency based boot sequencing

[mroter]

After few reboots I got a networking error (see below). After restart its OK => Problem not solved

Dec 25 15:19:59 ZLB-104-QA nrpe[948]: Starting up daemon
Dec 25 15:19:59 ZLB-104-QA nrpe[948]: Network server bind failure (99: Cannot assign requested address)
Dec 25 15:23:23 ZLB-104-QA nrpe[1229]: Starting up daemon
Dec 25 15:23:23 ZLB-104-QA nrpe[1229]: Warning: Daemon is configured to accept command arguments from clients!
Dec 25 15:23:23 ZLB-104-QA nrpe[1229]: Listening for connections on port 5666
Dec 25 15:23:23 ZLB-104-QA nrpe[1229]: Allowing connections from: 172.20.1.106

[slansing]

This is generally caused by having the wrong IP defined in nrpe.cfg, make sure the IP set in the server_address= field, are you running NRPE under a daemon, or under xinetd/inetd? If you are forcing it to bind to an address it has to be one available to that server. Once you have made the changes restart NRPE and then:

Code: Select all

tail -f /var/log/messages