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 1.74 2003/03/09 21:49:12 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.


############################################################
#### 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=300
# 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

# 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 configuration
############################################################

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


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

# The client interface listens for connections from clients
# on a TCP port. It is specified here.
ClientInterface.listenPort=4510

# The client interface needs to connect to another one to
# receive data. There is a special client interface in the
# root filter which is the top of the tree.
ClientInterface.parent=root

# 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


############################################################
#### LocalClient Configuration
############################################################

# The parent for the Local Client
Client.parent=clientinterface1


############################################################
#### 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=60

# How long a hosts config is valid for (a TTL effectively)
# after this time, in seconds, has passed a host should
# consider it's configuration invalid and check for a new
# configuration as soon as it can.
Host.ConfigTTL=3600

# 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;DiskIO;\
                NetIO;Paging;

# 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

# Paging monitor threshold values
Monitor.Paging.threshold.LOWER=1000
Monitor.Paging.threshold.UPPER=5000

# 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
Monitor.DiskInode.threshold.LOWER=90
Monitor.DiskInode.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
Monitor.DiskInode.thresholdMeasure=PERCENTAGE

# DiskIO monitor threshold values (in bytes)
Monitor.DiskIO.rbytes.threshold.LOWER=10000000
Monitor.DiskIO.rbytes.threshold.UPPER=50000000
Monitor.DiskIO.wbytes.threshold.LOWER=10000000
Monitor.DiskIO.wbytes.threshold.UPPER=50000000

# NetIO monitor threshold values (in bytes)
Monitor.NetIO.rxbytes.threshold.LOWER=10000000
Monitor.NetIO.rxbytes.threshold.UPPER=50000000
Monitor.NetIO.txbytes.threshold.LOWER=10000000
Monitor.NetIO.txbytes.threshold.UPPER=50000000

# 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:	1.75
Committed:	Mon May 5 22:04:59 2003 UTC (21 years, 7 months ago) by tdb
Branch:	MAIN
Changes since 1.74:	+13 -26 lines
Log Message:	Tidy up of the client interface code - more commonly known as the "right hand side of the server". Right since the start the filter side of the server has been nice and tree like - every Filter sent data to another Filter. At the top of the tree there was a "special" Filter known as the RootFilter, which to the other Filters just looked like a normal Filter. This was nice, and simple, and expandable. The Client Interface on the other hand was messy. The root filter had some hacky wrapper threads which pulled from a queue and pushed to the relevant client interfaces (one for real time stuff, and the other for databases). There was no simple room for expandability - it was all hardwired to do just what was needed at the time. This commit changes that. A Client Interface now connects to another Client Interface, with a special one being found in the RootFilter (yes, maybe that needs a name change now :-). So we can chain client interfaces, and move other bits and bobs around in the server - for example, alerting no longer needs to be connected to the Client Interface, it can connect straight to the RootFilter (or, wherever the config tells it ;). Hopefully this sanitizes the underlying layout of the server a bit. As a final note, I dropped the DBInterface. This used to insert data in to a MySQL database. We've long since stopped using that, and it's fallen behind and is way out of date. For now, it's gone in to the attic.
#	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	1.75	# $Id: system.conf,v 1.74 2003/03/09 21:49:12 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	ajm	1.14
57	ajm	1.16	############################################################
58			#### Misc system-wide configuration
59			############################################################
60
61	tdb	1.70	# The ConfigurationProxy caches configuration for various
62			# parts of the server. This setting defines how often, in
63			# seconds, the proxy will check if the data it contains
64			# is still valid.
65	ajm	1.16	ConfigurationProxy.updateTime=60
66
67	tdb	1.70	# The internal Queue's can be monitored. This specifies
68			# the interval, in seconds, at which their state should
69			# be monitored.
70	tdb	1.73	Queue.MonitorInterval=300
71	tdb	1.70	# The maximum size a Queue can be before items are
72			# dropped when new ones arrive.
73	tdb	1.40	Queue.SizeLimit=1000
74	tdb	1.70	# And which items we should drop when the queue is full
75			# and new items arrive. Choices are:
76			# RANDOM - drop a random item from the queue
77			# FIRST - drop the first item from the queue
78			# LAST - drop the last item from the queue
79			# DROP - drop the new item from the queue
80	tdb	1.40	Queue.RemoveAlgorithm=FIRST
81	tdb	1.2
82	tdb	1.70	# The XMLCache caches data passing through the system.
83			# This setting specifies the time, in seconds, at which
84			# the caches contents will be rotated. In real terms
85			# this means the minimum age at which an XMLPacket can
86			# be before it is up for deletion from the cache.
87	tdb	1.68	XMLCache.cleanupPeriod=30
88	ajm	1.14
89	tdb	1.70
90	ajm	1.14	############################################################
91			#### FilterManager configuration
92			############################################################
93	tdb	1.2
94	tdb	1.70	# The FilterManager accepts connections from hosts
95			# upon startup. It provides them with configuration
96			# and details of which filter to talk to. This setting
97			# specifies which port (TCP) it listens on.
98	tdb	1.2	FilterManager.listenPort=4567
99
100	tdb	1.70	# An Access Control List for the FilterManager. One
101			# might consider something like this:
102			#
103			# FilterManager.ACL=DEFAULT:DENY;*.example.com:ALLOW
104			FilterManager.ACL=DEFAULT:ALLOW
105	ajm	1.14
106
107			############################################################
108			#### Filter configuration
109			############################################################
110
111	tdb	1.71	# A Filter accepts data from hosts on a UDP port. It is
112			# specified here.
113	tdb	1.2	Filter.UDPListenPort=4589
114
115	tdb	1.70	# Filters contain plugins for checking the data passing
116			# through them. This plugins do the job of "filtering".
117			# This setting specifies the java package they're
118			# contained in.
119	tdb	1.55	Filter.PluginsPackage=uk.org.iscream.cms.server.filter.plugins
120	tdb	1.70	# And the names of the plugins to be loaded.
121	tdb	1.71	Filter.Plugins=TypeChecker;EnforceEssentialData;SourceChecker
122	ajm	1.14
123	tdb	1.70	# Every Filter, except the root, needs a parent to
124			# send their data to. This is usually the root filter,
125			# although in more complex setups it might be another
126			# filter - giving a tree structure.
127	ajm	1.14	Filter.parentFilter=root
128
129	tdb	1.71	# A Filter can accept data through UDP, or from
130	tdb	1.70	# an upstream Filter (CORBA). These methods can be
131			# individually turned of on a filter. The default
132			# is to have them all on.
133	tdb	1.61	# (comment or set to 0 to disable)
134			Filter.ActivateUDPReader=1
135			Filter.ActivateCORBAReader=1
136	tdb	1.62
137	tdb	1.70	# We can also specify Access Control Lists for the
138	tdb	1.71	# UDP input of the Filters. One might want something
139			# restrictive like this:
140	tdb	1.70	#
141			# Filter.UDPACL=DEFAULT:DENY;*.example.com:ALLOW
142			Filter.UDPACL=DEFAULT:ALLOW
143
144			# A Filter plugin called the SourceChecker verifies
145			# that packets come from permitted hosts. It is
146			# given an ACL to check the hosts against. A possible
147			# ACL might be:
148			#
149			# Filter.SourceCheckerPluginACL=DEFAULT:DENY;*.example.com:ALLOW
150			Filter.SourceCheckerPluginACL=DEFAULT:ALLOW
151	tdb	1.65
152	ajm	1.14
153			############################################################
154	tdb	1.75	#### Root filter configuration
155	ajm	1.14	############################################################
156
157	tdb	1.75	# The special filter known as the "Root Filter" needs
158			# a name within the system.
159			RootFilter.name=root
160	ajm	1.14
161
162			############################################################
163			#### Client interface configuration
164			############################################################
165
166	tdb	1.70	# The client interface listens for connections from clients
167			# on a TCP port. It is specified here.
168	tdb	1.2	ClientInterface.listenPort=4510
169	tdb	1.75
170			# The client interface needs to connect to another one to
171			# receive data. There is a special client interface in the
172			# root filter which is the top of the tree.
173			ClientInterface.parent=root
174	tdb	1.63
175	tdb	1.70	# The Client Interface has two channels, control and data.
176			# Each of these can have an ACL for allowing connections.
177			# As per previous examples in this configuration file they
178			# could be set restrictively, but we'll leave them open
179			# by default
180			ClientInterface.TCPControlChannelACL=DEFAULT:ALLOW
181			ClientInterface.TCPDataChannelACL=DEFAULT:ALLOW
182	tdb	1.52
183
184			############################################################
185	tdb	1.75	#### LocalClient Configuration
186	tdb	1.52	############################################################
187
188	tdb	1.75	# The parent for the Local Client
189			Client.parent=clientinterface1
190	tdb	1.2
191	ajm	1.14
192			############################################################
193			#### Host Configuration
194			############################################################
195	tdb	1.2
196	tdb	1.70	# A host can use any filter. We'll set the default filter
197			# for all hosts to be 'filter1'. This can be a list.
198			Host.filter=filter1;
199	tdb	1.13
200	tdb	1.71	# Hosts send data every UDPUpdateTime, in seconds.
201	tdb	1.72	Host.UDPUpdateTime=60
202
203			# How long a hosts config is valid for (a TTL effectively)
204			# after this time, in seconds, has passed a host should
205			# consider it's configuration invalid and check for a new
206			# configuration as soon as it can.
207			Host.ConfigTTL=3600
208	tdb	1.2
209	tdb	1.70	# We can run service checks on a host. The java package
210			# which contains them is specified here.
211	tdb	1.55	Host.serviceChecksPackage=uk.org.iscream.cms.server.filter.plugins
212	tdb	1.2
213	ajm	1.14
214			############################################################
215			#### Monitor Configuration
216			############################################################
217
218	tdb	1.70	# Monitors allow data to be watched for possible problems.
219			# This setting specifies which java package they're in.
220	tdb	1.55	Monitor.PluginsPackage=uk.org.iscream.cms.server.client.monitors
221	tdb	1.70	# And which ones should be loaded.
222	tdb	1.51	Monitor.Plugins=CPU;Load;Process;Disk;Memory;Swap;Services;\
223	tdb	1.74	Heartbeat;Queue;UserCount;WebFeeder;DiskIO;\
224			NetIO;Paging;
225	tdb	1.24
226	tdb	1.70	# Alerts timeout after a given time in seconds and go up
227			# to the next level. These can be overridden for monitors
228			# on an individual basis.
229	tdb	1.47	Monitor.alertTimeout.NOTICE=60
230			Monitor.alertTimeout.WARNING=900
231			Monitor.alertTimeout.CAUTION=1800
232			Monitor.alertTimeout.CRITICAL=3600
233	ajm	1.14
234	tdb	1.47	## Threshold values
235			#
236	tdb	1.70	# Here we specify the threshold for each monitor. When
237			# a threshold is broken an alert is raised. If the
238			# UPPER threshold is broken the alertTimeout values
239			# given above are halved to make the alert escalate
240			# quicker.
241	tdb	1.47
242			# CPU monitor threshold values
243	ajm	1.16	Monitor.CPU.threshold.LOWER=90
244	tdb	1.70	Monitor.CPU.threshold.UPPER=95
245	tdb	1.44
246	tdb	1.47	# Load monitor threshold values
247	tdb	1.44	Monitor.Load.threshold.LOWER=10.0
248			Monitor.Load.threshold.UPPER=20.0
249
250	tdb	1.74	# Paging monitor threshold values
251			Monitor.Paging.threshold.LOWER=1000
252			Monitor.Paging.threshold.UPPER=5000
253
254	tdb	1.47	# Process Count threshold values
255	tdb	1.70	Monitor.Process.threshold.LOWER=500
256			Monitor.Process.threshold.UPPER=1000
257	tdb	1.24
258	tdb	1.47	# Memory monitor threshold values
259	tdb	1.24	Monitor.Memory.threshold.LOWER=80
260			Monitor.Memory.threshold.UPPER=90
261	tdb	1.67	# Whether to include 'cache' in the 'free' value
262			# (1 is true, anything else is false)
263			Monitor.Memory.useCacheAsFree=1
264	tdb	1.24
265	tdb	1.47	# Swap monitor threshold values
266	tdb	1.24	Monitor.Swap.threshold.LOWER=80
267			Monitor.Swap.threshold.UPPER=90
268	tdb	1.50
269			# UserCount monitor threshold values
270			Monitor.UserCount.threshold.LOWER=100
271			Monitor.UserCount.threshold.UPPER=200
272	tdb	1.24
273	tdb	1.47	# Disk monitor threshold values
274	tdb	1.42	Monitor.Disk.threshold.LOWER=90
275			Monitor.Disk.threshold.UPPER=95
276	tdb	1.74	Monitor.DiskInode.threshold.LOWER=90
277			Monitor.DiskInode.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	tdb	1.74	Monitor.DiskInode.thresholdMeasure=PERCENTAGE
283
284			# DiskIO monitor threshold values (in bytes)
285			Monitor.DiskIO.rbytes.threshold.LOWER=10000000
286			Monitor.DiskIO.rbytes.threshold.UPPER=50000000
287			Monitor.DiskIO.wbytes.threshold.LOWER=10000000
288			Monitor.DiskIO.wbytes.threshold.UPPER=50000000
289
290			# NetIO monitor threshold values (in bytes)
291			Monitor.NetIO.rxbytes.threshold.LOWER=10000000
292			Monitor.NetIO.rxbytes.threshold.UPPER=50000000
293			Monitor.NetIO.txbytes.threshold.LOWER=10000000
294			Monitor.NetIO.txbytes.threshold.UPPER=50000000
295	ajm	1.33
296	tdb	1.56	# Queue monitor threshold values
297	tdb	1.49	Monitor.Queue.threshold.LOWER=50
298			Monitor.Queue.threshold.UPPER=80
299	tdb	1.47	# Thresholds given are PERCENTAGE or VALUE
300	tdb	1.70	# where PERCENTAGE is a percentage of the queue in use
301			# and VALUE is an absolute value of free space
302	tdb	1.46	Monitor.Queue.thresholdMeasure=PERCENTAGE
303
304	tdb	1.47	# Services monitor threshold values
305	ajm	1.34	#
306			# THE SERVICES MONITOR USES A HACK
307			# WHEREBY IT USES THE SERVICE STATUS
308	tdb	1.70	# AS THE THRESHOLD LEVEL. THIS SHOULD
309	ajm	1.34	# BE LOOKED AT!
310			#
311	tdb	1.47	# how many times we should be at the highest level before
312			# we go to a FINAL alert
313	tdb	1.38	Monitor.Services.reachFINALcount=5
314
315	tdb	1.47	# Heartbeat threshold values
316	tdb	1.25	# these are values in seconds from when the heartbeat
317			# was expected to arrive (ie. last + hearbeat period)
318	tdb	1.48	Monitor.Heartbeat.threshold.LOWER=180
319	tdb	1.38	Monitor.Heartbeat.threshold.UPPER=300
320	tdb	1.47	# this is how often we will run a "check" of heartbeats
321	tdb	1.48	Monitor.Heartbeat.checkPeriod=50
322	tdb	1.47	# how many times we should be at the highest level before
323			# we go to a FINAL alert
324			Monitor.Heartbeat.reachFINALcount=5
325	tdb	1.58	# list of hosts we expect to have at startup
326			# this ensures we notice hosts that aren't running, not
327			# just those that have gone down recently.
328	tdb	1.70	#Monitor.Heartbeat.initialHosts=importantserver.example.com;myserver.example.com
329	ajm	1.14
330	ajm	1.23
331	ajm	1.14	############################################################
332			#### Alerter Configuration
333			############################################################
334	tdb	1.70	# The following variables are understood by ALL alerters
335			# for messages. They will be substituted for their
336			# appropriate value.
337	ajm	1.23	#
338			# %level% - the alert level (eg, WARNING)
339			# %threshold% - the threshold broken (eg, LOWER)
340			# %source% - the source of the alert (eg, raptor.ukc.ac.uk)
341			# %value% - the value reached (eg, 95)
342			# %thresholdValue% - the value of the threshold broken (eg, 90)
343			# %attributeName% - the attribute that has caused the alert (eg, CPU User)
344			# %timeTillNextAlert% - the time the next alert will be sent out
345			# %timeSinceFirstAlert% - the time elapsed since the first alert for this problem
346			# %timeOfFirstAlert% - the time the first alert was sent
347			#
348	ajm	1.14
349	tdb	1.70	# The java package containing the alerters
350	tdb	1.55	Alerter.PluginsPackage=uk.org.iscream.cms.server.client.alerters
351	tdb	1.70	# And the names of the ones to activate
352			# You might want to turn them all on, but we only activate
353			# two basic ones by default.
354			# Alerter.Plugins=EMail;IRC;WebFeeder;Logging;
355			Alerter.Plugins=WebFeeder;Logging;
356	tdb	1.47
357	tdb	1.70	# Configuration for the EMail alerter
358			# The level alerts have to bypass to be e-mailed out
359	tdb	1.38	Alerter.EMail.level = WARNING
360	tdb	1.70	# A list of people to send alerts to
361			Alerter.EMail.destList = me@example.com
362			# The sender of the e-mail
363			Alerter.EMail.sender = me@example.com
364			# The SMTP server to send alerts to
365			Alerter.EMail.smtpServer = smtp.example.com
366			# The subject line of the e-mail
367	ajm	1.16	Alerter.EMail.subject = i-scream alert: %level% alert on %source% for %attributeName%
368	tdb	1.70	# The message in the e-mail
369	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%.
370
371	tdb	1.70	# Configuration for the IRC alerter
372			# The level alerts have to bypass to be sent
373	ajm	1.16	Alerter.IRC.level = OK
374	tdb	1.70	# The IRC server name
375			Alerter.IRC.IRCServer = irc.example.com
376			# and it's port
377	ajm	1.14	Alerter.IRC.IRCPort = 6667
378	tdb	1.70	# A list of nicks to try and be
379	tdb	1.19	Alerter.IRC.nickList = iscreamBot;_iscreamBot;i-screamBot
380	tdb	1.70	# Username to be
381	tdb	1.17	Alerter.IRC.user = i-scream
382	tdb	1.70	# Responses to CTCP prods
383	tdb	1.59	Alerter.IRC.comment = i-scream alerting bot (based on PircBot)
384			Alerter.IRC.finger = i-scream alerting bot (based on PircBot)
385	tdb	1.70	# Channel to join
386	ajm	1.14	Alerter.IRC.channel = #i-scream
387	tdb	1.70	# Message to use when sending an alert
388	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%
389	tdb	1.70	# How long to wait before reconnecting
390	tdb	1.17	Alerter.IRC.reconnectDelay = 30
391	tdb	1.70	# Notice to send when starting up
392	tdb	1.17	Alerter.IRC.startupNotice = i-scream alerting bot activated
393	tdb	1.70	# Commands the bot will recognise
394	tdb	1.17	Alerter.IRC.stopCommand = stop alerts
395			Alerter.IRC.startCommand = start alerts
396			Alerter.IRC.lastAlertCommand = last alert
397			Alerter.IRC.joinCommand = join
398	tdb	1.18	Alerter.IRC.nickChangeCommand = nick
399			Alerter.IRC.statCommand = statistics
400			Alerter.IRC.uptimeCommand = uptime
401			Alerter.IRC.timeSinceLastAlertCommand = time since last alert
402	tdb	1.19	Alerter.IRC.versionCommand = version
403	tdb	1.17	Alerter.IRC.helpCommand = help
404	tdb	1.70	# Message to send when the bot is addressed and
405			# doesn't understand the command
406	tdb	1.17	Alerter.IRC.rejectMessage = sorry, I don't understand your request
407	tdb	1.54
408	tdb	1.70	# Configuration for the Logging Alerter
409			# This alerter writes alerts to the i-scream logfile.
410			# The level alerts have to bypass to be logged
411	tdb	1.54	Alerter.Logging.level = WARNING
412	tdb	1.70	# The message to use when logging the alert
413	tdb	1.57	Alerter.Logging.message = %level%: %attributeName% on %source% has passed %threshold%(%thresholdValue%) threshold with %value%
414	tdb	1.29
415	tdb	1.47
416	tdb	1.29	############################################################
417			#### WebFeeder Configuration
418			############################################################
419
420	tdb	1.70	# The WebFeeder drops alerts and latest data in to a
421			# directory for the web interface to use. It is essentially
422			# the link between the web interface and the server.
423
424			# Set these to something to tell the WebFeeder to drop
425			# latest data, alerts, or both. Comment themt o disable.
426	tdb	1.32	WebFeeder.latestActive = true
427			WebFeeder.alertActive = true
428	tdb	1.43
429	tdb	1.70	# We can make the WebFeeder only drop alerts that go
430			# past a certain level. The default is "OK", which makes
431			# all alerts get passed on.
432	tdb	1.53	Alerter.WebFeeder.level = OK
433	tdb	1.37
434	tdb	1.70	# The WebFeeder cleans out old and stale alerts when it
435			# performs a check. This setting is the period, in seconds,
436			# at which a check happens.
437	tdb	1.42	WebFeeder.checkPeriod = 120
438	tdb	1.37
439			# Delete alerts older than this, in seconds.
440	tdb	1.42	WebFeeder.alertDeleteOlderThan = 300
441	tdb	1.32
442	tdb	1.70	# Data will be dropped to a subdirectory (specified
443			# below) of this directory.
444			WebFeeder.rootPath = /web/i-scream
445
446			# SubDir is the directory below rootPath which latest
447			# data will be placed in. Below that a directory will
448			# be created with the machine name, and a file named
449			# FileName (specified here) will be created containg
450			# the data.
451	tdb	1.29	WebFeeder.latestSubDir = latest
452	tdb	1.31	WebFeeder.latestFileName = latest_data
453	tdb	1.32
454	tdb	1.70	# As above, but for the alerts.
455	tdb	1.32	WebFeeder.alertSubDir = alert
456			WebFeeder.alertFileName = alert_data
457	ajm	1.16
458	tdb	1.47
459	tdb	1.17	############################################################