Administration

Update

When upgrading from an earlier version than version 20.04, all data of the Host Discovery feature will be lost:

• Discovery tasks,
• Saved parameters/credentials.

This is due to the new hardened way credentials are stored in this version.

Discovered hosts through those tasks will remain.

Upgrading to 22.10 will keep all data stored since 20.04.

1. To update the module, run the following command:

   yum update -y centreon-auto-discovery-server

2. Connect to the Centreon web interface using an account allowed to install products and go to the Administration > Extensions > Manager page.

3. Make sure that License Manager and Plugin Packs Manager modules are up-to-date before updating the Auto Discovery module.

4. Click on the update icon corresponding to the Auto Discovery module:

The module is now updated:

Uninstallation

Uninstalling the module will also remove all the associated data. Data won't be restorable unless a database backup has been made.

1. Connect to the Centreon web interface using an account allowed to install products and go to the Administration > Extensions > Manager page.

2. Click on the delete icon corresponding to the Auto Discovery module:

3. A confirmation popup will appear. Confirm the action:

   

   The module is now uninstalled:

   

Gorgone module configuration

The Auto Discovery module brings a specific configuration for the Gorgone service on the Central server. The default configuration is /etc/centreon-gorgone/config.d/41-autodiscovery.yaml.

A maximum duration for hosts discovery jobs is set globally. If it is necessary to change it (large subnet SNMP discovery for example), edit the configuration and add the global_timeout directive.

If mail notifications are enabled in service discovery rules, mail parameters can be defined to choose the sender, subject or mail command.

Example of configuration:

gorgone:
  modules:
    - name: autodiscovery
      package: "gorgone::modules::centreon::autodiscovery::hooks"
      enable: true
      # Host Discovery
      check_interval: 15
      global_timeout: 300
      # Service Discovery
      mail_subject: Centreon Auto Discovery
      mail_from: centreon-autodisco
      mail_command: /bin/mail

Be sure to restart the Gorgone service after any configuration modification:

systemctl restart gorgoned

Distributed architecture

The hosts and services discoveries both rely on Gorgone to perform discoveries on both Central and Remote Servers or Pollers.

It is necessary to have a ZMQ communication between the Central server and a Remote Server to perform a discovery on a Poller attached to this Remote Server.

Look at the section presenting the different communication types to know more.

Service Discovery scheduled job

All the active discovery rules are periodically executed through a scheduled job managed by Gorgone's cron module. The Auto Discovery module brings a cron definition in the following file: /etc/centreon-gorgone/config.d/cron.d/41-service-discovery.yaml.

- id: service_discovery
  timespec: "30 22 * * *"
  action: LAUNCHSERVICEDISCOVERY

The default configuration runs the discovery every day at 10:30 PM.

If you had changed the legacy crond configuration file to adapt the schedule you must apply changes to the new configuration file.

It is also possible to run multiple service discoveries with different parameters:

- id: service_discovery_poller_1
  timespec: "15 9 * * *"
  action: LAUNCHSERVICEDISCOVERY
  parameters:
    filter_pollers:
      - Poller-1
- id: service_discovery_poller_2_linux
  timespec: "30 9 * * *"
  action: LAUNCHSERVICEDISCOVERY
  parameters:
    filter_pollers:
      - Poller-2
    filter_rules:
      - OS-Linux-SNMP-Disk-Name
      - OS-Linux-SNMP-Traffic-Name
- id: service_discovery_poller_2_windows
  timespec: "45 9 * * *"
  action: LAUNCHSERVICEDISCOVERY
  parameters:
    filter_pollers:
      - Poller-2
    filter_rules:
      - OS-Windows-SNMP-Disk-Name
      - OS-Windows-SNMP-Traffic-Name

Here is the list of all available parameters:

Key	Value
filter_rules	Array of rules to use for discovery (empty means all)
force_rule	Run disabled rules ('0': not forced, '1': forced)
filter_hosts	Array of hosts which will run the discovery (empty means all)
filter_pollers	Array of pollers for which linked hosts will undergo discovery (empty means all)
dry_run	Run discovery without configuration changes ('0': changes, '1': dry run)
no_generate_config	No configuration generation (even if there are some changes) ('0': generation, '1': no generation)

API accesses

When installing Gorgone, a default configuration to access the Centreon APIs is located at /etc/centreon-gorgone/config.d/31-centreon-api.yaml.

It defines accesses to both Centreon CLAPI and RestAPI to allow discovery to communicate with Centreon.

Example of configuration:

gorgone:
  tpapi:
    - name: centreonv2
      base_url: "http://127.0.0.1/centreon/api/latest/"
      username: api
      password: bpltc4aY
    - name: clapi
      username: cli
      password: PYNM5kcc

Access to RestAPI, represented by centreonv2, requires credentials of a user with Reach API Configuration access. It is used for Host Discovery.

Access to CLAPI requires credentials of an Admin user. It is used for Service Discovery.

One user can be used for both accesses. Furthermore, users don't need access to the Centreon UI.