build/etc/system.conf

# This is the default system configuration file
# used by all components of the i-scream system
#
# ALL configuration that uses the centralised
# configuration system has this file as its 
# root of configuration
#
# This is a working sample configuration, which
# aims to provide a working "out of the box"
# setup, aswell as a some example configuration.
#
# $Author: tdb $
# $Id: system.conf,v 1.70 2003/02/23 12:45:21 tdb Exp $
#
# http://www.i-scream.org.uk/
#

############################################################
#### Configuration Files & Groups
############################################################

# All components, such as hosts, will take their configuration
# as given in this file. Unless, that is, they have an extra
# configuration file specified here, or they're in a group
# that has a configuration file. If this is the case the extra
# configuration will override what is given in this file.

# We can define an extra configuration file for any component
# of the system. This is done by specifying the following:
#
# config.component=someconfigfile.conf
#
# Component names will be defined in the documentation for
# the component. An example is that a host component is
# known by the name "Host.hostname" or "Host.ip".
# It is also possible to use the wildcard "*" in a
# hostname or IP address.

# This defines that server.example.com has some extra
# configuration in the file 'server.conf'.
config.Host.server.example.com=exserver.conf

# All webservers are placed in a group named 'www'.
group.www=Host.www1.example.com;Host.www2.example.com
# Then we define some extra config for them.
config.www=webservers.conf

# Our workstations are in their own subnet.
group.workstations=Host.192.168.10.*;
# They can have their own config too.
config.workstations=workstations.conf

# Any hosts not specified by now have the default
# config as given in this file.

# Another component of the system is the mySQL
# database part of the server call the DBInterface.
# This has it's own config so the passwords for the
# mySQL database can be kept seperate.
#config.mySQL=mySQL.conf


############################################################
#### Misc system-wide configuration
############################################################

# The ConfigurationProxy caches configuration for various
# parts of the server. This setting defines how often, in
# seconds, the proxy will check if the data it contains
# is still valid.
ConfigurationProxy.updateTime=60

# The internal Queue's can be monitored. This specifies
# the interval, in seconds, at which their state should
# be monitored.
Queue.MonitorInterval=15
# The maximum size a Queue can be before items are
# dropped when new ones arrive.
Queue.SizeLimit=1000
# And which items we should drop when the queue is full
# and new items arrive. Choices are:
# RANDOM - drop a random item from the queue
# FIRST  - drop the first item from the queue
# LAST   - drop the last item from the queue
# DROP   - drop the new item from the queue
Queue.RemoveAlgorithm=FIRST

# The XMLCache caches data passing through the system.
# This setting specifies the time, in seconds, at which
# the caches contents will be rotated. In real terms
# this means the minimum age at which an XMLPacket can
# be before it is up for deletion from the cache.
XMLCache.cleanupPeriod=30


############################################################
#### FilterManager configuration
############################################################

# The FilterManager accepts connections from hosts
# upon startup. It provides them with configuration
# and details of which filter to talk to. This setting
# specifies which port (TCP) it listens on.
FilterManager.listenPort=4567

# An Access Control List for the FilterManager. One
# might consider something like this:
#
# FilterManager.ACL=DEFAULT:DENY;*.example.com:ALLOW
FilterManager.ACL=DEFAULT:ALLOW


############################################################
#### Filter configuration
############################################################

# A Filter accepts data from hosts on a UDP port. It is
# specified here.
Filter.UDPListenPort=4589

# Filters contain plugins for checking the data passing
# through them. This plugins do the job of "filtering".
# This setting specifies the java package they're
# contained in.
Filter.PluginsPackage=uk.org.iscream.cms.server.filter.plugins
# And the names of the plugins to be loaded.
Filter.Plugins=TypeChecker;EnforceEssentialData;SourceChecker

# The special filter known as the "Root Filter" needs
# a name within the system.
RootFilter.name=root

# Every Filter, except the root, needs a parent to
# send their data to. This is usually the root filter,
# although in more complex setups it might be another
# filter - giving a tree structure.
Filter.parentFilter=root

# A Filter can accept data through UDP, or from
# an upstream Filter (CORBA). These methods can be
# individually turned of on a filter. The default
# is to have them all on.
# (comment or set to 0 to disable)
Filter.ActivateUDPReader=1
Filter.ActivateCORBAReader=1

# We can also specify Access Control Lists for the
# UDP input of the Filters. One might want something
# restrictive like this:
#
# Filter.UDPACL=DEFAULT:DENY;*.example.com:ALLOW
Filter.UDPACL=DEFAULT:ALLOW

# A Filter plugin called the SourceChecker verifies
# that packets come from permitted hosts. It is
# given an ACL to check the hosts against. A possible
# ACL might be:
#
# Filter.SourceCheckerPluginACL=DEFAULT:DENY;*.example.com:ALLOW
Filter.SourceCheckerPluginACL=DEFAULT:ALLOW


############################################################
#### Root filter interfaces configuration
############################################################

# The Root FIlter can send data to various upstream
# data processors. There are two of these at present
# in the server - the DBInterface and the ClientInterface.
# Their component names are given here as defined below.
# (the DBInterface is switched off - commented - by default)
#RootFilter.dbInterfaceName=database
RootFilter.realtimeInterfaceName=realtimeclients


############################################################
#### Client interface configuration
############################################################

# The client interface listens for connections from clients
# on a TCP port. It is specified here.
ClientInterface.listenPort=4510
# The component name for the Client Interface. This should
# match the name specified to the Root Filter above.
ClientInterface.name=realtimeclients

# The Client Interface has two channels, control and data.
# Each of these can have an ACL for allowing connections.
# As per previous examples in this configuration file they
# could be set restrictively, but we'll leave them open
# by default
ClientInterface.TCPControlChannelACL=DEFAULT:ALLOW
ClientInterface.TCPDataChannelACL=DEFAULT:ALLOW


############################################################
#### Database interface configuration
############################################################

# The name for the Database Interface. This should match
# the name given to the Root Filter.
DBInterface.name=database


############################################################
#### Host Configuration
############################################################

# A host can use any filter. We'll set the default filter
# for all hosts to be 'filter1'. This can be a list.
Host.filter=filter1;

# Hosts send data every UDPUpdateTime, in seconds.
Host.UDPUpdateTime=10

# We can run service checks on a host. The java package
# which contains them is specified here.
Host.serviceChecksPackage=uk.org.iscream.cms.server.filter.plugins


############################################################
#### Monitor Configuration
############################################################

# Monitors allow data to be watched for possible problems.
# This setting specifies which java package they're in.
Monitor.PluginsPackage=uk.org.iscream.cms.server.client.monitors
# And which ones should be loaded.
Monitor.Plugins=CPU;Load;Process;Disk;Memory;Swap;Services;\
                Heartbeat;Queue;UserCount;WebFeeder;

# Alerts timeout after a given time in seconds and go up
# to the next level. These can be overridden for monitors
# on an individual basis.
Monitor.alertTimeout.NOTICE=60
Monitor.alertTimeout.WARNING=900
Monitor.alertTimeout.CAUTION=1800
Monitor.alertTimeout.CRITICAL=3600

## Threshold values
#
# Here we specify the threshold for each monitor. When
# a threshold is broken an alert is raised. If the
# UPPER threshold is broken the alertTimeout values
# given above are halved to make the alert escalate
# quicker.

# CPU monitor threshold values
Monitor.CPU.threshold.LOWER=90
Monitor.CPU.threshold.UPPER=95

# Load monitor threshold values
Monitor.Load.threshold.LOWER=10.0
Monitor.Load.threshold.UPPER=20.0

# Process Count threshold values
Monitor.Process.threshold.LOWER=500
Monitor.Process.threshold.UPPER=1000

# Memory monitor threshold values
Monitor.Memory.threshold.LOWER=80
Monitor.Memory.threshold.UPPER=90
# Whether to include 'cache' in the 'free' value
# (1 is true, anything else is false)
Monitor.Memory.useCacheAsFree=1

# Swap monitor threshold values
Monitor.Swap.threshold.LOWER=80
Monitor.Swap.threshold.UPPER=90

# UserCount monitor threshold values
Monitor.UserCount.threshold.LOWER=100
Monitor.UserCount.threshold.UPPER=200

# Disk monitor threshold values
Monitor.Disk.threshold.LOWER=90
Monitor.Disk.threshold.UPPER=95
# Thresholds given are PERCENTAGE or VALUE
# where PERCENTAGE is a percentage of disk in use
# and VALUE is an absolute value of space free
Monitor.Disk.thresholdMeasure=PERCENTAGE

# Queue monitor threshold values
Monitor.Queue.threshold.LOWER=50
Monitor.Queue.threshold.UPPER=80
# Thresholds given are PERCENTAGE or VALUE
# where PERCENTAGE is a percentage of the queue in use
# and VALUE is an absolute value of free space
Monitor.Queue.thresholdMeasure=PERCENTAGE

# Services monitor threshold values
# 
# THE SERVICES MONITOR USES A HACK
# WHEREBY IT USES THE SERVICE STATUS
# AS THE THRESHOLD LEVEL. THIS SHOULD
# BE LOOKED AT!
#
# how many times we should be at the highest level before
# we go to a FINAL alert
Monitor.Services.reachFINALcount=5

# Heartbeat threshold values
# these are values in seconds from when the heartbeat
# was *expected* to arrive (ie. last + hearbeat period)
Monitor.Heartbeat.threshold.LOWER=180
Monitor.Heartbeat.threshold.UPPER=300
# this is how often we will run a "check" of heartbeats
Monitor.Heartbeat.checkPeriod=50
# how many times we should be at the highest level before
# we go to a FINAL alert
Monitor.Heartbeat.reachFINALcount=5
# list of hosts we expect to have at startup
# this ensures we notice hosts that aren't running, not
# just those that have gone down recently.
#Monitor.Heartbeat.initialHosts=importantserver.example.com;myserver.example.com


############################################################
#### Alerter Configuration
############################################################
# The following variables are understood by ALL alerters
# for messages. They will be substituted for their
# appropriate value.
#
# %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
#

# The java package containing the alerters
Alerter.PluginsPackage=uk.org.iscream.cms.server.client.alerters
# And the names of the ones to activate
# You might want to turn them all on, but we only activate
# two basic ones by default.
# Alerter.Plugins=EMail;IRC;WebFeeder;Logging;
Alerter.Plugins=WebFeeder;Logging;

# Configuration for the EMail alerter
# The level alerts have to bypass to be e-mailed out
Alerter.EMail.level = WARNING
# A list of people to send alerts to
Alerter.EMail.destList = me@example.com
# The sender of the e-mail
Alerter.EMail.sender = me@example.com
# The SMTP server to send alerts to
Alerter.EMail.smtpServer = smtp.example.com
# The subject line of the e-mail
Alerter.EMail.subject = i-scream alert: %level% alert on %source% for %attributeName%
# The message in the e-mail
Alerter.EMail.message = The i-scream distributed central monitoring system has\nraised a %level% alert for the host %source%.\n\nThe value for %attributeName% of %value% has exceeded the\n%threshold% threshold value of %thresholdValue%.\n\nThis alert was originally raised at %timeOfFirstAlert%,\nwhich was %timeSinceFirstAlert% ago.\n\nThe next alert (should one occur) will be sent in %timeTillNextAlert%.

# Configuration for the IRC alerter
# The level alerts have to bypass to be sent
Alerter.IRC.level = OK
# The IRC server name
Alerter.IRC.IRCServer = irc.example.com
# and it's port
Alerter.IRC.IRCPort = 6667
# A list of nicks to try and be
Alerter.IRC.nickList = iscreamBot;_iscreamBot;i-screamBot
# Username to be
Alerter.IRC.user = i-scream
# Responses to CTCP prods
Alerter.IRC.comment = i-scream alerting bot (based on PircBot)
Alerter.IRC.finger = i-scream alerting bot (based on PircBot)
# Channel to join
Alerter.IRC.channel = #i-scream
# Message to use when sending an alert
Alerter.IRC.message = %level%: %attributeName% on %source% has passed %threshold%(%thresholdValue%) threshold with %value% - time till next alert (should one occur), %timeTillNextAlert%
# How long to wait before reconnecting
Alerter.IRC.reconnectDelay = 30
# Notice to send when starting up
Alerter.IRC.startupNotice = i-scream alerting bot activated 
# Commands the bot will recognise
Alerter.IRC.stopCommand = stop alerts
Alerter.IRC.startCommand = start alerts
Alerter.IRC.lastAlertCommand = last alert
Alerter.IRC.joinCommand = join
Alerter.IRC.nickChangeCommand = nick
Alerter.IRC.statCommand = statistics
Alerter.IRC.uptimeCommand = uptime
Alerter.IRC.timeSinceLastAlertCommand = time since last alert
Alerter.IRC.versionCommand = version
Alerter.IRC.helpCommand = help
# Message to send when the bot is addressed and
# doesn't understand the command
Alerter.IRC.rejectMessage = sorry, I don't understand your request

# Configuration for the Logging Alerter
# This alerter writes alerts to the i-scream logfile.
# The level alerts have to bypass to be logged
Alerter.Logging.level = WARNING
# The message to use when logging the alert
Alerter.Logging.message = %level%: %attributeName% on %source% has passed %threshold%(%thresholdValue%) threshold with %value%


############################################################
#### WebFeeder Configuration
############################################################

# The WebFeeder drops alerts and latest data in to a
# directory for the web interface to use. It is essentially
# the link between the web interface and the server.

# Set these to something to tell the WebFeeder to drop
# latest data, alerts, or both. Comment themt o disable.
WebFeeder.latestActive = true
WebFeeder.alertActive = true

# We can make the WebFeeder only drop alerts that go
# past a certain level. The default is "OK", which makes
# all alerts get passed on.
Alerter.WebFeeder.level = OK

# The WebFeeder cleans out old and stale alerts when it
# performs a check. This setting is the period, in seconds,
# at which a check happens.
WebFeeder.checkPeriod = 120

# Delete alerts older than this, in seconds.
WebFeeder.alertDeleteOlderThan = 300

# Data will be dropped to a subdirectory (specified
# below) of this directory.
WebFeeder.rootPath = /web/i-scream

# SubDir is the directory below rootPath which latest
# data will be placed in. Below that a directory will
# be created with the machine name, and a file named
# FileName (specified here) will be created containg
# the data.
WebFeeder.latestSubDir = latest
WebFeeder.latestFileName = latest_data

# As above, but for the alerts.
WebFeeder.alertSubDir = alert
WebFeeder.alertFileName = alert_data


############################################################
Revision:	1.71
Committed:	Mon Feb 24 20:18:45 2003 UTC (21 years, 6 months ago) by tdb
Branch:	MAIN
Changes since 1.70:	+8 -21 lines
Log Message:	Fairly major commit. This will break the current version of ihost, but this had to be done really to give Pete something to test the new ihost against. The main change here is removal of the TCP Heartbeat functionality from the filter. This meant the following features stopped working :- - Heartbeat testing - Configuration checking - Service checks The heartbeat testing, specifically the monitor, now looks at the presence of UDP packets instead. Before it just looked for the presence of a TCP heartbeat packet, so the change their is fairly negligible. Of course this means heartbeat testing now relies on the UDP working... but I don't see this as a problem. Configuration checking has been repositioned in to the filtermanager. This is a backwards compatible change - the filtermanager should still perform as it should for older hosts. But now there's an extra command to check the configuration is up-to-date, with a similar format to the old TCP protocol in the filter. (although we may optimise this soon) The service checks are broken. This isn't a major issue for us as they were pretty useless in the first place. The concept is good, but the checks are just far too primitive. I expect at some point I'll work on a seperate component that just monitors services, which will replace this function. Further changes in the server include removal of the key checking code, as this relied on a bolt on to the TCP heartbeat protocol to ship the key. This got more akward than originally planned, so I'm happy to drop the idea. In the long term we hope to replace this with a public key systems for signing and even encryption. Finally, general tidy up to remove other bits of code that check for TCP heartbeat packets when they don't need to any more.
#	User	Rev	Content
1	tdb	1.2	# This is the default system configuration file
2			# used by all components of the i-scream system
3			#
4			# ALL configuration that uses the centralised
5			# configuration system has this file as its
6			# root of configuration
7			#
8	tdb	1.70	# This is a working sample configuration, which
9			# aims to provide a working "out of the box"
10			# setup, aswell as a some example configuration.
11			#
12	tdb	1.59	# $Author: tdb $
13	tdb	1.71	# $Id: system.conf,v 1.70 2003/02/23 12:45:21 tdb Exp $
14	tdb	1.70	#
15			# http://www.i-scream.org.uk/
16	ajm	1.14	#
17	tdb	1.70
18	ajm	1.14	############################################################
19			#### Configuration Files & Groups
20			############################################################
21
22	tdb	1.70	# All components, such as hosts, will take their configuration
23			# as given in this file. Unless, that is, they have an extra
24			# configuration file specified here, or they're in a group
25			# that has a configuration file. If this is the case the extra
26			# configuration will override what is given in this file.
27
28			# We can define an extra configuration file for any component
29			# of the system. This is done by specifying the following:
30			#
31			# config.component=someconfigfile.conf
32			#
33			# Component names will be defined in the documentation for
34			# the component. An example is that a host component is
35			# known by the name "Host.hostname" or "Host.ip".
36			# It is also possible to use the wildcard "*" in a
37			# hostname or IP address.
38
39			# This defines that server.example.com has some extra
40			# configuration in the file 'server.conf'.
41			config.Host.server.example.com=exserver.conf
42
43			# All webservers are placed in a group named 'www'.
44			group.www=Host.www1.example.com;Host.www2.example.com
45			# Then we define some extra config for them.
46			config.www=webservers.conf
47
48			# Our workstations are in their own subnet.
49			group.workstations=Host.192.168.10.*;
50			# They can have their own config too.
51			config.workstations=workstations.conf
52
53			# Any hosts not specified by now have the default
54			# config as given in this file.
55
56			# Another component of the system is the mySQL
57			# database part of the server call the DBInterface.
58			# This has it's own config so the passwords for the
59			# mySQL database can be kept seperate.
60			#config.mySQL=mySQL.conf
61	tdb	1.15
62	ajm	1.14
63	ajm	1.16	############################################################
64			#### Misc system-wide configuration
65			############################################################
66
67	tdb	1.70	# The ConfigurationProxy caches configuration for various
68			# parts of the server. This setting defines how often, in
69			# seconds, the proxy will check if the data it contains
70			# is still valid.
71	ajm	1.16	ConfigurationProxy.updateTime=60
72
73	tdb	1.70	# The internal Queue's can be monitored. This specifies
74			# the interval, in seconds, at which their state should
75			# be monitored.
76	ajm	1.16	Queue.MonitorInterval=15
77	tdb	1.70	# The maximum size a Queue can be before items are
78			# dropped when new ones arrive.
79	tdb	1.40	Queue.SizeLimit=1000
80	tdb	1.70	# And which items we should drop when the queue is full
81			# and new items arrive. Choices are:
82			# RANDOM - drop a random item from the queue
83			# FIRST - drop the first item from the queue
84			# LAST - drop the last item from the queue
85			# DROP - drop the new item from the queue
86	tdb	1.40	Queue.RemoveAlgorithm=FIRST
87	tdb	1.2
88	tdb	1.70	# The XMLCache caches data passing through the system.
89			# This setting specifies the time, in seconds, at which
90			# the caches contents will be rotated. In real terms
91			# this means the minimum age at which an XMLPacket can
92			# be before it is up for deletion from the cache.
93	tdb	1.68	XMLCache.cleanupPeriod=30
94	ajm	1.14
95	tdb	1.70
96	ajm	1.14	############################################################
97			#### FilterManager configuration
98			############################################################
99	tdb	1.2
100	tdb	1.70	# The FilterManager accepts connections from hosts
101			# upon startup. It provides them with configuration
102			# and details of which filter to talk to. This setting
103			# specifies which port (TCP) it listens on.
104	tdb	1.2	FilterManager.listenPort=4567
105
106	tdb	1.70	# An Access Control List for the FilterManager. One
107			# might consider something like this:
108			#
109			# FilterManager.ACL=DEFAULT:DENY;*.example.com:ALLOW
110			FilterManager.ACL=DEFAULT:ALLOW
111	ajm	1.14
112
113			############################################################
114			#### Filter configuration
115			############################################################
116
117	tdb	1.71	# A Filter accepts data from hosts on a UDP port. It is
118			# specified here.
119	tdb	1.2	Filter.UDPListenPort=4589
120
121	tdb	1.70	# Filters contain plugins for checking the data passing
122			# through them. This plugins do the job of "filtering".
123			# This setting specifies the java package they're
124			# contained in.
125	tdb	1.55	Filter.PluginsPackage=uk.org.iscream.cms.server.filter.plugins
126	tdb	1.70	# And the names of the plugins to be loaded.
127	tdb	1.71	Filter.Plugins=TypeChecker;EnforceEssentialData;SourceChecker
128	ajm	1.14
129	tdb	1.70	# The special filter known as the "Root Filter" needs
130			# a name within the system.
131	tdb	1.2	RootFilter.name=root
132	ajm	1.14
133	tdb	1.70	# Every Filter, except the root, needs a parent to
134			# send their data to. This is usually the root filter,
135			# although in more complex setups it might be another
136			# filter - giving a tree structure.
137	ajm	1.14	Filter.parentFilter=root
138
139	tdb	1.71	# A Filter can accept data through UDP, or from
140	tdb	1.70	# an upstream Filter (CORBA). These methods can be
141			# individually turned of on a filter. The default
142			# is to have them all on.
143	tdb	1.61	# (comment or set to 0 to disable)
144			Filter.ActivateUDPReader=1
145			Filter.ActivateCORBAReader=1
146	tdb	1.62
147	tdb	1.70	# We can also specify Access Control Lists for the
148	tdb	1.71	# UDP input of the Filters. One might want something
149			# restrictive like this:
150	tdb	1.70	#
151			# Filter.UDPACL=DEFAULT:DENY;*.example.com:ALLOW
152			Filter.UDPACL=DEFAULT:ALLOW
153
154			# A Filter plugin called the SourceChecker verifies
155			# that packets come from permitted hosts. It is
156			# given an ACL to check the hosts against. A possible
157			# ACL might be:
158			#
159			# Filter.SourceCheckerPluginACL=DEFAULT:DENY;*.example.com:ALLOW
160			Filter.SourceCheckerPluginACL=DEFAULT:ALLOW
161	tdb	1.65
162	ajm	1.14
163			############################################################
164			#### Root filter interfaces configuration
165			############################################################
166
167	tdb	1.70	# The Root FIlter can send data to various upstream
168			# data processors. There are two of these at present
169			# in the server - the DBInterface and the ClientInterface.
170			# Their component names are given here as defined below.
171			# (the DBInterface is switched off - commented - by default)
172			#RootFilter.dbInterfaceName=database
173	tdb	1.2	RootFilter.realtimeInterfaceName=realtimeclients
174	ajm	1.14
175
176			############################################################
177			#### Client interface configuration
178			############################################################
179
180	tdb	1.70	# The client interface listens for connections from clients
181			# on a TCP port. It is specified here.
182	tdb	1.2	ClientInterface.listenPort=4510
183	tdb	1.70	# The component name for the Client Interface. This should
184			# match the name specified to the Root Filter above.
185	tdb	1.52	ClientInterface.name=realtimeclients
186	tdb	1.63
187	tdb	1.70	# The Client Interface has two channels, control and data.
188			# Each of these can have an ACL for allowing connections.
189			# As per previous examples in this configuration file they
190			# could be set restrictively, but we'll leave them open
191			# by default
192			ClientInterface.TCPControlChannelACL=DEFAULT:ALLOW
193			ClientInterface.TCPDataChannelACL=DEFAULT:ALLOW
194	tdb	1.52
195
196			############################################################
197			#### Database interface configuration
198			############################################################
199
200	tdb	1.70	# The name for the Database Interface. This should match
201			# the name given to the Root Filter.
202	tdb	1.52	DBInterface.name=database
203	tdb	1.2
204	ajm	1.14
205			############################################################
206			#### Host Configuration
207			############################################################
208	tdb	1.2
209	tdb	1.70	# A host can use any filter. We'll set the default filter
210			# for all hosts to be 'filter1'. This can be a list.
211			Host.filter=filter1;
212	tdb	1.13
213	tdb	1.71	# Hosts send data every UDPUpdateTime, in seconds.
214	tdb	1.7	Host.UDPUpdateTime=10
215	tdb	1.2
216	tdb	1.70	# We can run service checks on a host. The java package
217			# which contains them is specified here.
218	tdb	1.55	Host.serviceChecksPackage=uk.org.iscream.cms.server.filter.plugins
219	tdb	1.2
220	ajm	1.14
221			############################################################
222			#### Monitor Configuration
223			############################################################
224
225	tdb	1.70	# Monitors allow data to be watched for possible problems.
226			# This setting specifies which java package they're in.
227	tdb	1.55	Monitor.PluginsPackage=uk.org.iscream.cms.server.client.monitors
228	tdb	1.70	# And which ones should be loaded.
229	tdb	1.51	Monitor.Plugins=CPU;Load;Process;Disk;Memory;Swap;Services;\
230			Heartbeat;Queue;UserCount;WebFeeder;
231	tdb	1.24
232	tdb	1.70	# Alerts timeout after a given time in seconds and go up
233			# to the next level. These can be overridden for monitors
234			# on an individual basis.
235	tdb	1.47	Monitor.alertTimeout.NOTICE=60
236			Monitor.alertTimeout.WARNING=900
237			Monitor.alertTimeout.CAUTION=1800
238			Monitor.alertTimeout.CRITICAL=3600
239	ajm	1.14
240	tdb	1.47	## Threshold values
241			#
242	tdb	1.70	# Here we specify the threshold for each monitor. When
243			# a threshold is broken an alert is raised. If the
244			# UPPER threshold is broken the alertTimeout values
245			# given above are halved to make the alert escalate
246			# quicker.
247	tdb	1.47
248			# CPU monitor threshold values
249	ajm	1.16	Monitor.CPU.threshold.LOWER=90
250	tdb	1.70	Monitor.CPU.threshold.UPPER=95
251	tdb	1.44
252	tdb	1.47	# Load monitor threshold values
253	tdb	1.44	Monitor.Load.threshold.LOWER=10.0
254			Monitor.Load.threshold.UPPER=20.0
255
256	tdb	1.47	# Process Count threshold values
257	tdb	1.70	Monitor.Process.threshold.LOWER=500
258			Monitor.Process.threshold.UPPER=1000
259	tdb	1.24
260	tdb	1.47	# Memory monitor threshold values
261	tdb	1.24	Monitor.Memory.threshold.LOWER=80
262			Monitor.Memory.threshold.UPPER=90
263	tdb	1.67	# Whether to include 'cache' in the 'free' value
264			# (1 is true, anything else is false)
265			Monitor.Memory.useCacheAsFree=1
266	tdb	1.24
267	tdb	1.47	# Swap monitor threshold values
268	tdb	1.24	Monitor.Swap.threshold.LOWER=80
269			Monitor.Swap.threshold.UPPER=90
270	tdb	1.50
271			# UserCount monitor threshold values
272			Monitor.UserCount.threshold.LOWER=100
273			Monitor.UserCount.threshold.UPPER=200
274	tdb	1.24
275	tdb	1.47	# Disk monitor threshold values
276	tdb	1.42	Monitor.Disk.threshold.LOWER=90
277			Monitor.Disk.threshold.UPPER=95
278	tdb	1.47	# Thresholds given are PERCENTAGE or VALUE
279			# where PERCENTAGE is a percentage of disk in use
280			# and VALUE is an absolute value of space free
281	tdb	1.46	Monitor.Disk.thresholdMeasure=PERCENTAGE
282	ajm	1.33
283	tdb	1.56	# Queue monitor threshold values
284	tdb	1.49	Monitor.Queue.threshold.LOWER=50
285			Monitor.Queue.threshold.UPPER=80
286	tdb	1.47	# Thresholds given are PERCENTAGE or VALUE
287	tdb	1.70	# where PERCENTAGE is a percentage of the queue in use
288			# and VALUE is an absolute value of free space
289	tdb	1.46	Monitor.Queue.thresholdMeasure=PERCENTAGE
290
291	tdb	1.47	# Services monitor threshold values
292	ajm	1.34	#
293			# THE SERVICES MONITOR USES A HACK
294			# WHEREBY IT USES THE SERVICE STATUS
295	tdb	1.70	# AS THE THRESHOLD LEVEL. THIS SHOULD
296	ajm	1.34	# BE LOOKED AT!
297			#
298	tdb	1.47	# how many times we should be at the highest level before
299			# we go to a FINAL alert
300	tdb	1.38	Monitor.Services.reachFINALcount=5
301
302	tdb	1.47	# Heartbeat threshold values
303	tdb	1.25	# these are values in seconds from when the heartbeat
304			# was expected to arrive (ie. last + hearbeat period)
305	tdb	1.48	Monitor.Heartbeat.threshold.LOWER=180
306	tdb	1.38	Monitor.Heartbeat.threshold.UPPER=300
307	tdb	1.47	# this is how often we will run a "check" of heartbeats
308	tdb	1.48	Monitor.Heartbeat.checkPeriod=50
309	tdb	1.47	# how many times we should be at the highest level before
310			# we go to a FINAL alert
311			Monitor.Heartbeat.reachFINALcount=5
312	tdb	1.58	# list of hosts we expect to have at startup
313			# this ensures we notice hosts that aren't running, not
314			# just those that have gone down recently.
315	tdb	1.70	#Monitor.Heartbeat.initialHosts=importantserver.example.com;myserver.example.com
316	ajm	1.14
317	ajm	1.23
318	ajm	1.14	############################################################
319			#### Alerter Configuration
320			############################################################
321	tdb	1.70	# The following variables are understood by ALL alerters
322			# for messages. They will be substituted for their
323			# appropriate value.
324	ajm	1.23	#
325			# %level% - the alert level (eg, WARNING)
326			# %threshold% - the threshold broken (eg, LOWER)
327			# %source% - the source of the alert (eg, raptor.ukc.ac.uk)
328			# %value% - the value reached (eg, 95)
329			# %thresholdValue% - the value of the threshold broken (eg, 90)
330			# %attributeName% - the attribute that has caused the alert (eg, CPU User)
331			# %timeTillNextAlert% - the time the next alert will be sent out
332			# %timeSinceFirstAlert% - the time elapsed since the first alert for this problem
333			# %timeOfFirstAlert% - the time the first alert was sent
334			#
335	ajm	1.14
336	tdb	1.70	# The java package containing the alerters
337	tdb	1.55	Alerter.PluginsPackage=uk.org.iscream.cms.server.client.alerters
338	tdb	1.70	# And the names of the ones to activate
339			# You might want to turn them all on, but we only activate
340			# two basic ones by default.
341			# Alerter.Plugins=EMail;IRC;WebFeeder;Logging;
342			Alerter.Plugins=WebFeeder;Logging;
343	tdb	1.47
344	tdb	1.70	# Configuration for the EMail alerter
345			# The level alerts have to bypass to be e-mailed out
346	tdb	1.38	Alerter.EMail.level = WARNING
347	tdb	1.70	# A list of people to send alerts to
348			Alerter.EMail.destList = me@example.com
349			# The sender of the e-mail
350			Alerter.EMail.sender = me@example.com
351			# The SMTP server to send alerts to
352			Alerter.EMail.smtpServer = smtp.example.com
353			# The subject line of the e-mail
354	ajm	1.16	Alerter.EMail.subject = i-scream alert: %level% alert on %source% for %attributeName%
355	tdb	1.70	# The message in the e-mail
356	tdb	1.38	Alerter.EMail.message = The i-scream distributed central monitoring system has\nraised a %level% alert for the host %source%.\n\nThe value for %attributeName% of %value% has exceeded the\n%threshold% threshold value of %thresholdValue%.\n\nThis alert was originally raised at %timeOfFirstAlert%,\nwhich was %timeSinceFirstAlert% ago.\n\nThe next alert (should one occur) will be sent in %timeTillNextAlert%.
357
358	tdb	1.70	# Configuration for the IRC alerter
359			# The level alerts have to bypass to be sent
360	ajm	1.16	Alerter.IRC.level = OK
361	tdb	1.70	# The IRC server name
362			Alerter.IRC.IRCServer = irc.example.com
363			# and it's port
364	ajm	1.14	Alerter.IRC.IRCPort = 6667
365	tdb	1.70	# A list of nicks to try and be
366	tdb	1.19	Alerter.IRC.nickList = iscreamBot;_iscreamBot;i-screamBot
367	tdb	1.70	# Username to be
368	tdb	1.17	Alerter.IRC.user = i-scream
369	tdb	1.70	# Responses to CTCP prods
370	tdb	1.59	Alerter.IRC.comment = i-scream alerting bot (based on PircBot)
371			Alerter.IRC.finger = i-scream alerting bot (based on PircBot)
372	tdb	1.70	# Channel to join
373	ajm	1.14	Alerter.IRC.channel = #i-scream
374	tdb	1.70	# Message to use when sending an alert
375	ajm	1.16	Alerter.IRC.message = %level%: %attributeName% on %source% has passed %threshold%(%thresholdValue%) threshold with %value% - time till next alert (should one occur), %timeTillNextAlert%
376	tdb	1.70	# How long to wait before reconnecting
377	tdb	1.17	Alerter.IRC.reconnectDelay = 30
378	tdb	1.70	# Notice to send when starting up
379	tdb	1.17	Alerter.IRC.startupNotice = i-scream alerting bot activated
380	tdb	1.70	# Commands the bot will recognise
381	tdb	1.17	Alerter.IRC.stopCommand = stop alerts
382			Alerter.IRC.startCommand = start alerts
383			Alerter.IRC.lastAlertCommand = last alert
384			Alerter.IRC.joinCommand = join
385	tdb	1.18	Alerter.IRC.nickChangeCommand = nick
386			Alerter.IRC.statCommand = statistics
387			Alerter.IRC.uptimeCommand = uptime
388			Alerter.IRC.timeSinceLastAlertCommand = time since last alert
389	tdb	1.19	Alerter.IRC.versionCommand = version
390	tdb	1.17	Alerter.IRC.helpCommand = help
391	tdb	1.70	# Message to send when the bot is addressed and
392			# doesn't understand the command
393	tdb	1.17	Alerter.IRC.rejectMessage = sorry, I don't understand your request
394	tdb	1.54
395	tdb	1.70	# Configuration for the Logging Alerter
396			# This alerter writes alerts to the i-scream logfile.
397			# The level alerts have to bypass to be logged
398	tdb	1.54	Alerter.Logging.level = WARNING
399	tdb	1.70	# The message to use when logging the alert
400	tdb	1.57	Alerter.Logging.message = %level%: %attributeName% on %source% has passed %threshold%(%thresholdValue%) threshold with %value%
401	tdb	1.29
402	tdb	1.47
403	tdb	1.29	############################################################
404			#### WebFeeder Configuration
405			############################################################
406
407	tdb	1.70	# The WebFeeder drops alerts and latest data in to a
408			# directory for the web interface to use. It is essentially
409			# the link between the web interface and the server.
410
411			# Set these to something to tell the WebFeeder to drop
412			# latest data, alerts, or both. Comment themt o disable.
413	tdb	1.32	WebFeeder.latestActive = true
414			WebFeeder.alertActive = true
415	tdb	1.43
416	tdb	1.70	# We can make the WebFeeder only drop alerts that go
417			# past a certain level. The default is "OK", which makes
418			# all alerts get passed on.
419	tdb	1.53	Alerter.WebFeeder.level = OK
420	tdb	1.37
421	tdb	1.70	# The WebFeeder cleans out old and stale alerts when it
422			# performs a check. This setting is the period, in seconds,
423			# at which a check happens.
424	tdb	1.42	WebFeeder.checkPeriod = 120
425	tdb	1.37
426			# Delete alerts older than this, in seconds.
427	tdb	1.42	WebFeeder.alertDeleteOlderThan = 300
428	tdb	1.32
429	tdb	1.70	# Data will be dropped to a subdirectory (specified
430			# below) of this directory.
431			WebFeeder.rootPath = /web/i-scream
432
433			# SubDir is the directory below rootPath which latest
434			# data will be placed in. Below that a directory will
435			# be created with the machine name, and a file named
436			# FileName (specified here) will be created containg
437			# the data.
438	tdb	1.29	WebFeeder.latestSubDir = latest
439	tdb	1.31	WebFeeder.latestFileName = latest_data
440	tdb	1.32
441	tdb	1.70	# As above, but for the alerts.
442	tdb	1.32	WebFeeder.alertSubDir = alert
443			WebFeeder.alertFileName = alert_data
444	ajm	1.16
445	tdb	1.47
446	tdb	1.17	############################################################