#===============================================================================
# BROKER
#===============================================================================
# Description: The Broker is responsible for:
# - Exporting centralized logs of all Shinken daemon processes
# - Exporting status data
# - Exporting performance data
# - Exposing Shinken APIs:
# - Status data
# - Performance data
# - Command interface
#===============================================================================
define broker {
# 1 #======== Daemon name and address =========
# Daemon name. Must be unique
broker_name is a spare, 0 = is not a spare
spare broker-master0
# IP/fqdn spare_daemon: name of thisthe daemon (note: you MUST changethat will take this daemon job if it bydies
the real ip/fqdn of this server)# IMPORTANT:
address# * a spare_daemon can only be the spare of 1 (and only one) master localhostdaemon
# Port (HTTP/HTTPS) exposed* bya this spare_daemon
cannot have a portspare_daemon
# * the spare must have modules with the same module_type as the 7772master
# 0 = use HTTP, 1- =depending useof HTTPS
the value of use_ssl the broker__manage_spare__spare_must_have_the_same_list_of_module_type parameter
# Example: spare_daemon 0
#======== Master or spare selection =========broker-spare
spare_daemon
# 1 = is a spare, 0 = is not a spare (default) the spare defined with spare_daemon must have the same module_type as this master
spare# 0 = the spare module_type are not checked
# broker__manage_spare__spare_must_have_the_same_list_of_module_type 01
# spare_daemon: name of the daemon that will take this daemon job if it dies======== Daemon connection timeout and down state limit =========
# IMPORTANT:
# * a spare_daemon can only be the spare of 1 (and only one) master daemon
# * a spare_daemon cannot have a spare_daemon
# * the spare must have modules with the same module_type as the master
# - depending of the value of the broker__manage_spare__spare_must_have_the_same_list_of_module_type parameter
# Example: spare_daemontimeout: how many seconds to consider a node don't answer
timeout 3
# data_timeout: how many second to consider a configuration transfer to be failed
# because the network bandwidth is too small.
data_timeout broker-spare120
# spare_daemon
# 1 = (default) the spare defined with spare_daemon must have the same module_type as this mastermax_check_attempts: how many fail check to consider this daemon as DEAD
max_check_attempts 3
# 0 = the spare module_type are not checked
Check this daemon every X seconds
check_interval # broker__manage_spare__spare_must_have_the_same_list_of_module_type 160
#======== DaemonModules connectionto timeoutenable andfor downthis state limitdaemon =========
# timeout: how many seconds to consider a node don't answer
Available:
# - WebUI : Visualisation interface
# - Graphite-Perfdata timeout : Save all metrics into a graphite database
# - sla 3
# data_timeout: how many second to consider a configuration transfer to be failed
# because the network bandwidth is: tooSave small.
sla into a database
data_timeout # - Livestatus 120
# max_check_attempts: how many fail check to consider this: daemonTCP asAPI DEAD
to query element max_check_attempts 3
state, used by nagios external tools like NagVis or Thruk
# Check this daemon every X seconds
check_interval - broker-module-livedata : REST API to query all monitored element data (host, cluster or check)
# - event-manager-writer 60
#======== Modules to enable for this: daemon =========
# Available:
# - WebUI Save events for events manager (do not forget to activate the module in your webui to see data)
# - Simple-log : Visualisation interface
# -: Graphite-PerfdataSave all logs into a common file, Use this module only if you need :to Savehave all metricsthe check intoresults ain graphiteone databasefile.
# - broker--module-report-builder sla : External module used by webui--module-report-handler to generate reports
modules : Save sla into a databaseWebUI, Graphite-Perfdata, sla, event-manager-writer
#
- Livestatus #======== Realm and architecture settings =========
# Realm to set this daemon into
:realm TCP API to query element state, used by nagios external tools like NagVis or Thruk
# - broker-module-livedata All
# 1 = take data :from RESTthe APIdaemon torealm queryand allits monitoredsub elementrealms
data (host, cluster or# check)
0 = take data #only - event-manager-writer from the daemon realm
manage_sub_realms : Save events for1
events manager (do not# forgetIn toNATted activateenvironments, theyou moduledeclare ineach your webui to see data)satellite ip[:port] as seen by
# - Simple-log *this* daemon (if port not set, the port declared by satellite itself
# is used)
#satellitemap : Save all logs into a common file, Use this module only if you need to have all the check results in one file. scheduler-1=1.2.3.4:7768, scheduler-2=1.2.3.5:7771
# Exchange between brokers <- schedulers can be limited by packet size (in kB)
# - broker--module-report-builder Note: Externalas modulecompression usedis by webui--module-report-handler to generate reports
modules automatic, this is a higher limit, and in real case the
# packets will be WebUI, Graphite-Perfdata, sla, event-manager-writer
lower than this value
#======== Realm and architecture settings = broks_packet_size 1024
#========
# Realm to set this daemon into Memory protection =========
realm# Are the daemon module process and worker process are waiting for enough
# memory to be All
available before being launch. #Default: 1 =(enabled)
take data from the daemon realm and its sub realms broker__manage_brok__enable_sub_processes_memory_usage_protection 1
# 0The =sub takeprocess datamemory onlyusage fromprotection thecan daemonhave realm
a system reserved memory
manage_sub_realms # that won't be used 1
by theses sub process #when In NATted environments, you declare each satellite ip[:port] as seen bylaunched
# By default: 0 (no reserved memory)
# *this* Broker Example: 10 (ifmeans port not set,10% of the porttotal declaredmemory byis satellitereserved itself
for # is usedthe system)
#satellitemap scheduler-1=1.2.3.4:7768, poller-1=1.2.3.5:7771broker__manage_brok__sub_process_memory_usage_system_reserved_memory 0
# If Exchangea betweensub Brokersprocess <-cannot Schedulersbe canstarted bebecause limitedof bythe packetprotection, sizehow (inmany kB)seconds
# Note:it aswill compression is automatic, this is a higher limit, and in real case the
# packets will be lower than this value
# broks_packet_size 1024be retry and wait that the system memory is freed until it fail to start
# By default: 5 (seconds)
broker__manage_brok__sub_processes_memory_usage_protection_max_retry_time 5
#======== MemoryBrok pusher protectionworker =========
# Are the daemon moduleThe broker spawn broks pusher sub process andto workerpush processto areexternal waitingmodules for(like enoughWebUI)
# memorythe tobroker bewill availablelook beforeat beingthis launch.worker Default:execution 1 (enabled)
broker__manage_brok__enable_sub_processes_memory_usage_protection 1
time, and will kill if it timeout
# The subbroker processwill memorycompute usagethe protectionaverage canexecution havetime aof systemprevious reservedworkers memoryto
# decide thatabout won'thow bemany usedtime bythis thesesworker subwill processtake whenbased launchedon:
# By default: 0 (no reserved memory number of broks to send / past average send speed (broks/s)
# Example: 10 (means 10% of the total memory is reserved for the system)
broker__manage_brok__sub_process_memory_usage_system_reserved_memory 0
# If a sub process cannot be started because of the protection, how many seconds
# it will be retry and wait that the system memory is freed until it fail to startIf this time is reach, it means that the pusher process is killed
# For small amount of broks to send, it should lead to ridiculously small allowed execution time
# and the fac to spawn the sub process can be higher than this value, so we are using a minimal
# execution timeout
# By defaultDefault: 5 (secondssecond)
broker__manage_brok__sub_processesprocess_memorybroks_usagepusher_protectionmin_max_retry_timeexecution_timeout 5
#======== Brok pusher worker ========= In order to manage the fact that the server can slow down during this send, you can setup a
# Theratio Brokerthat spawnwill broksbe pusherused subto processincrease to pushthe toallowed externaltimeout modulesby (likemultiply WebUI)it
# theDefault: 5
Broker will look at this worker execution time, and will kill if it timeout
broker__manage_brok__sub_process_broks_pusher_security_ratio # The Broker will compute5
the average execution time# ofAt previousthe workersbroker to
start without stats, this #valid decidewill aboutbe howused manyfor timethe thistimeout
worker will take based# on:Default: 240 (seconds)
# number of broks to send / past average send speed (broks/s)broker__manage_brok__sub_process_broks_pusher_max_execution_timeout 240
# If this time is reacha sub process reach a timeout, it meanswill be thatkilled theand pusherrelaunched. processAfter ismax killedretry,
# Forthe smallattached amountmodule ofwill broksbe torestarted
send, it should lead# toDefault: ridiculously3
small allowed execution time
broker__manage_brok__sub_process_broks_pusher_max_retry # and the fac to spawn the sub process can be higher than this value, so we are using a minimal3
# execution timeoutbroker__manage_brok__sub_process_broks_pusher_queue_batch_size:
# Default: 5 (second)
broker__manage_brok__sub_process_broks_pusher_min_execution_timeout * defines the maximum number of broks the "queue brok pusher"
# process will handle per 5
send to external module #( Inlike orderWebUI to) manage.
the fact that the# server can slow* downRemaining duringbroks thiswill send,be youhandled canin setupnext asend.
# ratio that* willIMPORTANT: beincrease usedthis tovalue increasecan thelead allowedto timeouterror byon multiplythe itsocket
# Default: 5100000 (broks/batch)
# broker__manage_brok__sub_process_broks_pusher_securityqueue_batch_ratiosize 5100000
# AtBroks thewhose Brokerserialization start without stats,time exceeds this validthreshold will begenerate used for the timeouta warning
# Default: 240100 (secondsmilliseconds)
# broker__manage_brok__suboversized_processdata_brokswarning_pusherthreshold_max_executionserialization_timeouttime 240100
# IfBroks awhose subserialization processtime reachexceeds a timeout, itthis threshold will be killed and relaunched. After max retry,
# the attached module will be restartedgenerate an error
# Default: 500 3(milliseconds)
# broker__manage_brok__sub_process_broks_pusher_max_retry oversized_data_error_threshold__serialization_time 500
#======== VMWare / ESXi ==========
# 1 (default) = if vmware get the ESXi CPU stats value, 0 = do not get value
vmware__statistics_compute_enable 31
#======== Enable or not this daemon =========
# 1 = is enabled, 0 = is disabled
enabled 1
}
| 
Il est conseillé d'éditer les fichiers .cfg avec l'encodage utf-8