Skip to main content
Version: 24.04

Centreon HA for small infrastuctures

If your platform only handles small volumes of data, where databases are hosted on central servers, Centreon can set up a "simplified" HA. In the event of a problem, the central server and database switch over simultaneously, thanks to a colocation constraint.

The threshold below which the databases can be integrated into the central servers is the same as for a standard Centreon platform (see the decision tree on the Architectures page).

Diagram of a simplified HA architecture​

image

Elements of a simplified HA architecture​

In its simplified version, the HA cluster is a 2-node cluster. It includes:

  • 2 central servers, each with its own integrated database
  • a quorum device, that decides which central server is the active node.

Differences with a 4-node HA architecture​

All the procedures described in the HA section of this documentation are also correct for a 2-node HA, except for the following points:

  • The pcs status command will return a message like this one:
Cluster Summary:
  * Stack: corosync
  * Current DC: @CENTRAL_ACTIVE_NAME@ (version 2.0.5-9.0.1.el8_4.1-ba59be7122) - partition with quorum
  * Last updated: Wed Sep 15 16:35:47 2021
  * Last change:  Wed Sep 15 10:41:50 2021 by root via crm_attribute on @CENTRAL_ACTIVE_NAME@
  * 2 nodes configured
  * 14 resource instances configured
Node List:
  * Online: [ @CENTRAL_ACTIVE_NAME@ @CENTRAL_PASSIVE_NAME@ ]
Full List of Resources:
  * Clone Set: ms_mysql-clone [ms_mysql] (promotable):
    * Masters: [ @CENTRAL_ACTIVE_NAME@ ]
    * Slaves: [ @CENTRAL_PASSIVE_NAME@ ]
  * Clone Set: php-clone [php]:
    * Started: [ @CENTRAL_ACTIVE_NAME@ @CENTRAL_PASSIVE_NAME@ ]
  * Clone Set: cbd_rrd-clone [cbd_rrd]:
    * Started: [ @CENTRAL_ACTIVE_NAME@ @CENTRAL_PASSIVE_NAME@ ]
  * Resource Group: centreon:
    * vip       (ocf::heartbeat:IPaddr2):        Started @CENTRAL_ACTIVE_NAME@
    * http      (systemd:httpd):         Started @CENTRAL_ACTIVE_NAME@
    * gorgone   (systemd:gorgoned):      Started @CENTRAL_ACTIVE_NAME@
    * centreon_central_sync     (systemd:centreon-central-sync):         Started @CENTRAL_ACTIVE_NAME@
    * cbd_central_broker        (systemd:cbd-sql):       Started @CENTRAL_ACTIVE_NAME@
    * centengine        (systemd:centengine):    Started @CENTRAL_ACTIVE_NAME@
    * centreontrapd     (systemd:centreontrapd):         Started @CENTRAL_ACTIVE_NAME@
    * snmptrapd (systemd:snmptrapd):     Started @CENTRAL_ACTIVE_NAME@
  • The only constraints to be defined are those that link each central server to its database, so that failovers are simultaneous. The output of the pcs constraint command should look like this:
Location Constraints:
Ordering Constraints:
Colocation Constraints:
  centreon with ms_mysql-clone (score:INFINITY) (rsc-role:Started) (with-rsc-role:Master)
  ms_mysql-clone with centreon (score:INFINITY) (rsc-role:Master) (with-rsc-role:Started)
Ticket Constraints:
  • There are no VIPs for databases, as these are integrated into the central nodes.