Argus - The All Seeing, System and Network Monitoring Software

Home
Features
Testimonials
Screen Shots
Demo
Download
Docs
History
Future
Links
Contributing
Mailing List
Contacting

Example Config


# RPC tests

Host "server.example.com" {

	Service UDP/RPC/100005		# default version
	Service UDP/RPC/100005/3	# specify version
	Service TCP/RPC/100005		# UDP or TCP
	Service UDP/RPC/ypbind		# by name
	Service UDP/RPC/mountd/3
}

################
# asterisk server (voip)

Host "asterisk.example.com" {
	# set in /etc/asterisk/manager.conf
	user:	argus
	pass:	secret

	Service TCP/Asterisk {
		# make sure PRI is up
		cmd:		pri show span 1
		expect:		Provisioned, Up, Active
	}

	Service TCP/Asterisk {
		# graph number of channels in use
		cmd:		show channels
		pluck:		(\d+) active channel
		graph:		yes
		title:		Channels In Use
	}

	# test iax + sip
	Service UDP/IAX2
	Service UDP/SIP
}


################
# DNS over TCP is the same as UDP, except for the spelling

Service TCP/DNS {
	zone:	example.com
	class:	IN
	query:	MX
	test:	nanswers
	expect: 2
}

################

# enable the async resolver

Resolv {
        frequency:      15
        timeout:        20
        max_queries:    1000
}

# or just

Resolv

################
# remote argus test port

# server 1
test_port:	3074

# server 2
Service TCP/Argus {
	hostname:	server1.example.com
	port:		3074
}

################

# snmp v3
Service UDP/SNMPv3 {
	hostname:	router.example.com
	oid:		ifInOctets.1
	calc:		ave-rate-bits
	maxvalue:	20000000
	snmpuser:	UserName
	snmppass:	Password
	snmpprivpass:	Secret
	# use default MD5+DES, auto-discover engine-id
}

################

# cron jobs
Service TCP/HTTP {
	hostname:	www.example.com

	# every saturday at 00:30
	cron "30 0 * * * 6" {
		func:		override
		text:		weekly maintenance downtime
		expires:	3600		# 1 hour
		mode:		manual
		quiet:		yes		# don't log
	}

	# lower the severity outside of busn hours
	cron "0 17 * * *" {
		func:		setparam
		param:		severity
		value:		warning
	}
	cron "0 9 * * *" {
		func:		setparam
		param:		severity
		value:		critical
	}

}



# database testing
Group "Databases" {

	Service DB {
		# dsn  - the db to connect to, using DBI syntax
		# user - the username
		# pass - the password
		# sql  - a select statement
		dsn:		dbi:Pg:dbname=mydatabase;host=dbhost
		user:		argus
		pass:		password
		sql:		select COUNT(*) from mytable
		minvalue:	10
	}
}

# graphing demo
Group "Bandwidth" {
	# verify usage is <20Mbps, and graph usage
	graph: yes
	ylabel: bits per second
	title:  Bandwidth
	Group "Sprint T3" {
		service UDP/SNMP {
			label:		Out
			calc:		ave-rate-bits
			hostname:       gw-amb3.example.com
			oid:		.1.3.6.1.2.1.2.2.1.16.1
			maxvalue:	20000000
		}
		service UDP/SNMP {
			label:		In
			calc:		ave-rate-bits
			hostname:       gw-amb3.example.com
			oid:		.1.3.6.1.2.1.2.2.1.10.1
			maxvalue:	20000000
		}
	}
}

# extended DNS testing
Group "DNS" {
	Host "ns1.example.com" {
		# is DNS running?
		Service DNS			# same as UDP/DNS in 3.1

		# are we authoritative?
		Service Domain/example.com	# same as UDP/Domain in 3.1

		Service DNS/Serial/example.com {
			# make sure serial number is at least this tall
			minvalue:	2001110100
		}

		Service DNS/A/www.example.com {
			# make sure the web server is where is should be
			expect:		10.1.2.3
		}

		# fully specified in "long form"
		# make sure we have 2 MX servers
		Service DNS {
			zone:	example.com
			class:	IN
			query:	MX
			test:	nanswers
			expect: 2
		}
	}
}

# advanced generic test functions
Group "NTP" {
	Host "ntp.example.com" {

		# pull out the stratum, and make sure is at most '6'
		Service UDP/NTP {
			label:    Stratum
			uname:    NTP-Stratum
			unpack:	  xC
			maxvalue: 6
		}

		# pull out the root-dispersion
		# verify that our clock has not drifted too far
		# and graph the result
		Service UDP/NTP {
			graph:	  yes
			title:	  Root Dispersion
			ylabel:	  seconds
			label:	  Dispersion
			uname:	  NTP-Disp
			unpack:	  x8N
			scale:	  65536
			maxvalue: 2
		}
	}
}


# my own stats
Group "My Stats" {
	graph:	yes

	# number of notifications
	Service Self/Notifs

	Service Self/idle {
		title:	Percent Idle
		calc:	ave-rate
		scale:	0.01
		# let someone know when it is time to upgrade h/w
		minvalue:  20
		messagedn: time to buy faster server
	}

	Service Self/tested {
		title:	Monitor Rate
		ylabel: tests per second
		calc:	ave-rate
	}

	Service Prog {
		# your ps may be different
		command:	ps -p $ARGUS_PID -o vsz | tail -1
		uname:		VSZ
		title:		Memory Use
		ylabel:		kBytes
	}
}


# create an audit log of everything that happens
Method "auditlog" {
	command:	echo "%T{%F %T} - %Y %P %O - %A" >> /tmp/audit.log
	qtime:		0
	nolotsmsgs:	yes
}

notifyaudit:	auditlog: