Bonjour,

j'ai un problème avec Shinken, je viens de l'installer, sans aucun problème, le service est OK, seul hic : je ne peux pas trouver la web UI

 (je suis sous Debian )

Quand je vais sur : 127.0.0.1:7772 ===> la page web est blanche avec noté "OK" en haut à gauche

Quand je vais sur 127.0.0.1:7767 ===> C'est comme si le site n'existait pas.

Voici mes configs :

la config de shinken.cfg :

# Configuration files with common objects like commands, timeperiods,
# or templates that are used by the host/service/contacts
cfg_dir=commands
cfg_dir=timeperiods
cfg_dir=escalations
cfg_dir=dependencies

# Now templates of hosts, services and contacts
cfg_dir=templates

# notification things
cfg_dir=notificationways

# Now groups
cfg_dir=servicegroups
cfg_dir=hostgroups
cfg_dir=contactgroups

# And now real hosts, services, packs and discovered hosts
# They are directory, and we will load all .cfg file into them, and
# their sub-directory
cfg_dir=hosts
cfg_dir=services
cfg_dir=contacts
cfg_dir=packs
cfg_dir=modules

cfg_dir=arbiters
cfg_dir=schedulers
cfg_dir=pollers
cfg_dir=reactionners
cfg_dir=brokers
cfg_dir=receivers
cfg_dir=realms

# You will find global MACROS into this file
#resource_file=resource.cfg
cfg_dir=resource.d

# Number of minutes between 2 retention save, here 1hour
retention_update_interval=60

# Number of interval (5min by default) to spread the first checks
# for hosts and services
max_service_check_spread=5
max_host_check_spread=5

# after 10s, checks are killed and exit with CRITICAL state (RIP)
service_check_timeout=60
timeout_exit_status=2

# flap_history is the lengh of history states we keep to look for
# flapping.
# 20 by default, can be useful to increase it. Each flap_history
# increases cost:
#    flap_history cost = 4Bytes * flap_history * (nb hosts + nb services)
# Example: 4 * 20 * (1000+10000) ~ 900Ko for a quite big conf. So, go for it!
flap_history=20


# Max plugin output for the plugins launched by the pollers, in bytes
max_plugins_output_length=65536


# Enable or not the state change on impact detection (like
# an host going unreach if a parent is DOWN for example). It's for
# services and hosts.
# Remark: if this option is absent, the default is 0 (for Nagios
# old behavior compatibility)
enable_problem_impacts_states_change=1


# if 1, disable all notice and warning messages at
# configuration checking
disable_old_nagios_parameters_whining=0


# If you need to set a specific timezone to your deamons, uncomment it
#use_timezone=Europe/Paris

# Disabling env macros is good for performances. If you really need it, enable it.
enable_environment_macros=0

# If not need, don't dump initial states into logs
log_initial_states=0

# By default don't launch even handlers during downtime. Put 0 to
# get back the default nagios behavior
no_event_handlers_during_downtimes=1


# [Optionnal], a pack distribution file is a local file near the arbiter
# that will keep host pack id association, and so push same host on the same
# scheduler if possible between restarts.
pack_distribution_file=/var/lib/shinken/pack_distribution.dat



## Arbiter daemon part, similar to ini

#If not specified will use lockfile direname
workdir=/var/lib/shinken/

# Lock file (with pid) for Arbiterd
lock_file=/var/run/shinken/arbiterd.pid

# The arbiter can have it's own local log
local_log=/var/log/shinken/arbiterd.log

# Accepted log level values: DEBUG,INFO,WARNING,ERROR,CRITICAL
#log_level=WARNING

# User that will be used by the arbiter.
# If commented, run as current user (root?)
shinken_user=shinken
shinken_group=shinken

# The path to the modules directory
modules_dir=/var/lib/shinken/modules

# Set to 0 if you want to make this daemon (arbiter) NOT run
daemon_enabled=1

#-- Security using SSL --
use_ssl=0
# WARNING : Put full paths for certs
ca_cert=/etc/shinken/certs/ca.pem
server_cert=/etc/shinken/certs/server.cert
server_key=/etc/shinken/certs/server.key
hard_ssl_name_check=0

# If cherrypy3 is not available, it will fail back to swsgiref
http_backend=auto


# kernel.shinken.io communication channel. Create an account to http://shinken.io
# and look at your profile to fill this.
#api_key=
#secret=
# if you need an http proxy to exchange with kernel.shinken.io
#http_proxy=


# Export all shinken inner performances
# into a statsd server. By default at localhost:8125 (UDP)
# with the shinken prefix
statsd_host=127.0.0.1
statsd_port=7767
statsd_prefix=shinken
statsd_enabled=1

La config de webui.cfg

define module {
   
    module_name webui
    module_type webui
    host 127.0.0.1
    port 7767
    auth_secret CHANGE_ME
    allow_html_output 1
    max_output_length 1024
    manage_acl 1
    play_sond 5
    login_text YOP !

    modules
    http_backend auto
}

La config de broker-master.cfg :

#===============================================================================
# BROKER (S1_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
#   - Configuration data
#   - Command interface
# https://shinken.readthedocs.org/en/latest/08_configobjects/broker.html
#===============================================================================
define broker {
    broker_name     broker-master
    address         localhost
    port            7772
    spare           0

    ## Optional
    manage_arbiters     1   ; Take data from Arbiter. There should be only one
                            ; broker for the arbiter.
    manage_sub_realms   1   ; Does it take jobs from schedulers of sub-Realms?
    timeout             3   ; Ping timeout
    data_timeout        120 ; Data send timeout
    max_check_attempts  3   ; If ping fails N or more, then the node is dead
    check_interval      60  ; Ping node every N seconds

    ## Modules
    # Default: None
    # Interesting modules that can be used:
    # - simple-log              = just all logs into one file
    # - livestatus              = livestatus listener
    # - tondodb-mysql           = NDO DB support
    # - npcdmod                 = Use the PNP addon
    # - graphite                = Use a Graphite time series DB for perfdata
    # - webui                   = Shinken Web interface
    # - glpidb                  = Save data in GLPI MySQL database
    modules    

    # Enable https or not
    use_ssl              0
    # enable certificate/hostname check, will avoid man in the middle attacks
    hard_ssl_name_check   0
   
    ## Advanced
    realm   All
}

=======================

Pouvez-vous me dire comment je peux faire pour obtenir l'interface web de Shinken ? J'ai installé tout ce qu'il fallait normalement, j'ai suivi beaucoup de tuto.

merci d'avance

-
Edité par Illoxx 29 décembre 2014 à 21:08:23

Salut,

A priori tu as pas mal modifié de paramètres et comme je n'ai pas la même organisation des fichiers de configuration...

Illoxx a écrit:

La config de webui.cfg

define module {
   
    module_name webui
    module_type webui
    host 127.0.0.1
    port 7767
    auth_secret CHANGE_ME
    allow_html_output 1
    max_output_length 1024
    manage_acl 1
    play_sond 5
    login_text YOP !

    modules
    http_backend auto
}

Essaye de mettre les valeurs par défaut au pire mais je me demande comment est interprété ton point d'exclamation après YOP

Hormis ça, difficile de te venir en aide car comme je te l'ai dit, chez moi ce ne sont pas les mêmes fichiers :

Bye