--- projects/cms/source/server/build/etc/system.conf 2001/03/05 13:40:48 1.22 +++ projects/cms/source/server/build/etc/system.conf 2001/03/07 23:36:31 1.28 @@ -6,7 +6,7 @@ # root of configuration # # $Author: tdb $ -# $Id: system.conf,v 1.22 2001/03/05 13:40:48 tdb Exp $ +# $Id: system.conf,v 1.28 2001/03/07 23:36:31 tdb Exp $ # ############################################################ #### Configuration Files & Groups @@ -94,7 +94,7 @@ ClientInterface.listenPort=4510 ############################################################ # The default filter for a new host -Host.filter=computingFilter; +Host.filter=filter1; # The default update times for a new host Host.UDPUpdateTime=10 @@ -103,7 +103,7 @@ Host.AveragerUpdateTime=5 # The services checks to run on a host Host.serviceChecksPackage=uk.ac.ukc.iscream.filter.plugins -Host.serviceChecks=HTTP;SMTP +Host.serviceChecks=HTTP;SMTP;POP3;SSH;FTP;IMAP;Telnet @@ -113,8 +113,9 @@ Host.serviceChecks=HTTP;SMTP # the monitors to run Monitor.PluginsPackage=uk.ac.ukc.iscream.client.monitors -Monitor.Plugins=CPU;Heartbeat +Monitor.Plugins=CPU;Memory;Swap;Heartbeat; + ## CPU monitor configuration # threshold values # @@ -130,6 +131,39 @@ Monitor.CPU.alertTimeout.WARNING=900 Monitor.CPU.alertTimeout.CAUTION=1800 Monitor.CPU.alertTimeout.CRITICAL=3600 + +## Memory monitor configuration +# threshold values +# +# note: if the UPPER threshold is broken, +# the alertTimeout values are halved. +# This is to escalate the alert quicker. +Monitor.Memory.threshold.LOWER=80 +Monitor.Memory.threshold.UPPER=90 + +# timings of the alerts (in seconds) +Monitor.Memory.alertTimeout.NOTICE=60 +Monitor.Memory.alertTimeout.WARNING=900 +Monitor.Memory.alertTimeout.CAUTION=1800 +Monitor.Memory.alertTimeout.CRITICAL=3600 + + +## Swap monitor configuration +# threshold values +# +# note: if the UPPER threshold is broken, +# the alertTimeout values are halved. +# This is to escalate the alert quicker. +Monitor.Swap.threshold.LOWER=80 +Monitor.Swap.threshold.UPPER=90 + +# timings of the alerts (in seconds) +Monitor.Swap.alertTimeout.NOTICE=60 +Monitor.Swap.alertTimeout.WARNING=900 +Monitor.Swap.alertTimeout.CAUTION=1800 +Monitor.Swap.alertTimeout.CRITICAL=3600 + + ## Heartbeat monitor configuration # threshold values # @@ -137,6 +171,8 @@ Monitor.CPU.alertTimeout.CRITICAL=3600 # the alertTimeout values are halved. # This is to escalate the alert quicker. Monitor.Heartbeat.checkPeriod=50 +# these are values in seconds from when the heartbeat +# was *expected* to arrive (ie. last + hearbeat period) Monitor.Heartbeat.threshold.LOWER=120 Monitor.Heartbeat.threshold.LOWER=300 @@ -146,9 +182,32 @@ Monitor.Heartbeat.alertTimeout.WARNING=900 Monitor.Heartbeat.alertTimeout.CAUTION=1800 Monitor.Heartbeat.alertTimeout.CRITICAL=3600 +# this means it will reach a FINAL level, +# and thus the host will be removed from the monitor list, +# if the highest level alert is reached and occours this +# number of times. +# +# int this setup, that means 5 CRITICAL alerts would occour +# on a Heartbeat before reaching FINAL. +# If this value is wrong or not present, FINAL is never reached +# +Monitor.Heartbeat.reachFINALcount=5 + ############################################################ #### Alerter Configuration ############################################################ +# The following var's are understood by ALL alerters for messages +# +# %level% - the alert level (eg, WARNING) +# %threshold% - the threshold broken (eg, LOWER) +# %source% - the source of the alert (eg, raptor.ukc.ac.uk) +# %value% - the value reached (eg, 95) +# %thresholdValue% - the value of the threshold broken (eg, 90) +# %attributeName% - the attribute that has caused the alert (eg, CPU User) +# %timeTillNextAlert% - the time the next alert will be sent out +# %timeSinceFirstAlert% - the time elapsed since the first alert for this problem +# %timeOfFirstAlert% - the time the first alert was sent +# # alerters to run Alerter.PluginsPackage=uk.ac.ukc.iscream.client.alerters