NRDS Passive Check - Error After Upgrade to 2012

[reclinerhead]

Hi,

I'm having a problem when trying to run/install the passive NRDS configuration on one of my hosts.

I get the error: ERROR: CONFIG_OS is not a valid cfg field.

It seems that this happens only to configurations that I create or update after the upgrade to 2012 a few weeks ago. My other passive clients don't have "CONFIG_OS" in their configuration files. But I figure as soon as I update them from within XI, they will start having errors too.

Here's the error I get... i tried reinstalling the client completely:

Code: Select all

# sh installnrds fms1.nps.edu 5
nagios group already exists.
nagios user already exists.
Installing NRDS Client

Adding cron jobs for fms1.nps.edu at a 5 minute interval

Crontabs installed OK
Updating config and plugins
ERROR: CONFIG_OS is not a valid cfg field.
Installation complete

Here's the top part of a config that works OK (I removed the token):

Code: Select all

CONFIG_VERSION=1.3
CONFIG_NAME=FMS_WEB
URL=https://nagios.nps.edu/nrdp/
TOKEN=xxxxxxxxxxxxxxxxxxxxxxx
PLUGIN_DIR=/usr/local/nagios/libexec
SEND_NRDP=/usr/local/nrdp/clients/send_nrdp.sh
TMPDIR=/usr/local/nrdp/clients/tmp
COMMAND_PREFIX=
UPDATE_CONFIG=1
UPDATE_PLUGINS=1

Here's one that is reporting the error... it seems the format has been changed in the new XI version.

Code: Select all

CONFIG_VERSION="0.2"
CONFIG_OS="Linux"
CONFIG_NAME="fms1.nps.edu"
URL="https://nagios.nps.edu/nrdp/"
TOKEN="xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
HOSTNAME="fms1.nps.edu"
PLUGIN_DIR="/usr/local/nagios/libexec"
SEND_NRDP="/usr/local/nrdp/clients/send_nrdp.sh"
TMPDIR="/usr/local/nrdp/clients/tmp"
COMMAND_PREFIX=""
LOG_FILE=""
UPDATE_CONFIG="1"
UPDATE_PLUGINS="1"

This is my first support request - so I apologize if I'm not including some important info from our instance. Thanks for any help!!

[reclinerhead]

Our XI server is running Redhat 6, and we did the manual installation of XI (i.e. we did not use the vm image).

[scottwilkerson]

There was a transition that happened a while back (which included a switch to perl as well as the CONFIG_OS addition) and I think your version of the NRDS Config Manager is somehow different than what is deployed to your machines.

Can you re-deploy the agent to the remote machine?

[reclinerhead]

Hi thanks for the reply. So I did a brand new client install using NRDP. On the client, I nuked the /usr/local/nagios and /usr/local/nrdp folders. On the server, I created a brand new NRDS config package using the default checks, and selected Linux as the O/S.

But I am getting the same error on the client when I download the new client and try to install. It just doesn't like the CONFIG_OS parameter in the config file. Maybe our XI server is delivering an incorrect set of scripts that haven't been updated yet somehow?

Code: Select all

[root@fms1 clients]# sh installnrds fms1.nps.edu 5
nagios group already exists.
nagios user already exists.
Installing NRDS Client

Adding cron jobs for fms1.nps.edu at a 5 minute interval

Crontabs installed OK
Updating config and plugins
ERROR: CONFIG_OS is not a valid cfg field.
Installation complete

[reclinerhead]

I checked the contents of the newly installed script at /usr/local/nrdp/clients/nrds/nrds.sh .... trying to find some version information. Here's what I found from the top of the file.

Code: Select all

#!/bin/bash
#
# Written by: Scott Wilkerson ([email protected])
# Copyright (c) 2010-2011 Nagios Enterprises, LLC.
#
#
###########################

# Setup Defaults
CONFIG=/usr/local/nrdp/clients/nrds/nrds.cfg
SEND_NRDP=/usr/local/nrdp/clients/send_nrdp.sh
COMMAND_PREFIX=""

PROGNAME=$(basename $0)
RELEASE="Revision 0.1"

[scottwilkerson]

Did you re-download the client from the XI server?

Lets make sure it didn't rename the tar.gz.1

Code: Select all

ls -l /tmp/|grep tar.gz

[reclinerhead]

Hi - yes I did it twice even to be sure. Each time I cleared out any related files/folders from /tmp, and I redownloaded the install package using a fresh link from the NRDS config manager page.

[scottwilkerson]

Can you please run the following on the client and report the results

Code: Select all

ls -l /usr/local/nrdp/clients/nrds/

Thanks

[reclinerhead]

Code: Select all

[wyatt@fms1 ~]$ ls -l /usr/local/nrdp/clients/nrds/
total 28
-rwxrwxr-- 1 nagios nagios  813 Oct 22 16:02 nrds.cfg
-rwxrwx--- 1 nagios nagios 3173 Oct 22 16:02 nrds.sh
-rwxrwxr-- 1 nagios nagios 6081 Oct 22 16:02 nrds_updater.sh
[wyatt@fms1 ~]$

[scottwilkerson]

I think I see the problem, the nrds config manager must not have upgraded on the server

Can you run the following on the XI server

Code: Select all

cd /usr/local/nagiosxi/html/includes/components/nrdsconfigmanager/
rm -f installed.nrds
./installnrds

Then lets re-download the client tarball and install