Distributed Monitoring

[Dmitriy Kirhlarov]

On Mon, Jan 22, 2007 at 11:28:43AM +0200, Moayad Mohammad wrote:

>                 I have one nagios server working in my company, and I
> need to add another nagios server to monitor another servers in other
> subnets,
> 
> I don't know if there's any solution to have 2 nagios servers(1 central
> nagios) and 1 monitor screen... it's mean the second server will send
> all check results to the central nagios.

It's work in our installation.
But, I don't find no one web config tool support this. :(

It's not complicated.

Look schema:

Central Host (location1 + location2)
    ^            ^
    |	         |
    |	         |
location1    location2

You making service_template_locations and host_template_locations. On
central host and on locations it's different templates
(active_check_enable 0 and 1 correspondingly)
All checks for locations based on this templates.

Next, you must setup nsca on central host and send_nsca on
location{1,2},
configure on location{1,2}:
obsess_over_services=
ocsp_command=
obsess_over_hosts=
ochp_command=
and disable direct notification on locations -- only central host
sending notifications.

Also, good idea make checks on location{1,2} for nagios on central
host (over nrpe) and make eventhandler in this check for enabling
notification directly on location[1,2}, when nagios on central host
dead.

eventhadler example:
--------------------
#!/bin/sh

# $Id: service_notification,v 1.1 2006/01/18 14:40:02 dkirhlarov Exp $
# this handler switching notifications.
# for details contact dimma at oilspace.com

DATE="`which date 2>/dev/null`" || DATE="/bin/date"
CAT="`which cat 2>/dev/null`" || CAT="/bin/cat"
BASENAME="`which basename 2>/dev/null`" || BASENAME="/bin/basename"
ECHO="`which echo 2>/dev/null`" || ECHO="/bin/echo"
AWK="`which awk 2>/dev/null`" || AWK="/bin/awk"
PRINTF="`which printf 2>/dev/null`" || PRINTF="/usr/bin/printf"
PROGNAME=`${BASENAME} $0`
NAGIOSCMD="/usr/local/nagios/var/rw/nagios.cmd"

NOW=`${DATE} +%s`

print_help() {
	${CAT} <<EOF

Event Handler for switch on|off notifications
Usage: $PROGNAME -s $SERVICESTATE$ -st $SERVICESTATETYPE$ [-c /path/to/nagios/cmd/file]
Keys:
-s -- STATE of service
-st -- TYPE of service STATE
(for details read nagios documentation)
-c -- command nagios socket. By default /usr/local/nagios/var/rw/nagios.cmd
Usage: $PROGNAME --help
EOF
}

enable_event() {
${PRINTF} "[%lu] ENABLE_NOTIFICATIONS\n" ${NOW} > ${NAGIOSCMD}
}

disable_event() {
${PRINTF} "[%lu] DISABLE_NOTIFICATIONS\n" ${NOW} > ${NAGIOSCMD}
}

# Make sure the correct number of command line
# arguments have been supplied
if [ $# -lt 4 ]; then
	print_help
	exit -1
fi

$ECHO $* >> /usr/local/nagios/var/event.out
# Grab the command line arguments
while [ -n "$1" ]; do
	case "$1" in
		--help|-h)
			print_help
			exit -1
			;;
		-s)
			# state of service
			STATE=$2
			shift
			;;
		-st)
			# state type of service
			STATETYPE=$2
			shift
			;;

		-c)
			# command file
			NAGIOSCMD=$2
			shift
			;;
		*)
			${ECHO} "Wrong param '$1'"
			print_help
			exit -1
			;;
	esac
	shift
done

# Run check
case "$STATETYPE" in
	HARD)
		case "$STATE" in
			CRITICAL)
				# Time to enabling local notification feature
				enable_event
				;;
			WARNING|UNKNOWN)
				# Just smoke
				;;
			OK)
				# Disable local notification feature
				disable_event
				;;
			*)
				# What the shit we are get?
				${ECHO} "What?.. Can't understand '$STATE'"
				exit -1
				;;
		esac
		;;
esac
exit 0
--------------------

WBR.
Dmitriy

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null