Set up the failover system

Follow this procedure to set up your failover system:

1. Set up a passwordless communications.

   • For Linux, use the instructions in “Set up passwordless communication between Active and Standby hosts” on page 20.

     If you plan to run the Failover Manager on a separate dedicated host, you need to set up passwordless communications between the Failover Manager host and between Standby and Active locations as well.

2. Install the products on the assigned hosts for the Active and Standby locations.

   The Domain Managers Installation Guide provides detailed installation and migration instructions.

   The installation locations of both the Active and Standby Managers must be identical on different hosts. For example:

   • For Active SAM on Host A, use: /opt/InCharge/SAM

     • For Standby SAM on Host B, use: /opt/InCharge/SAM

       By default, Failover Manager assumes that software is installed in the same directory in the Active and Standby locations. If you decide not to use identical directories, you need to change the StrictSitemod parameter to False in the BASEDIR/conf/failover/failover.conf file. The Setting Section provides details on the StrictSitemod parameter.

       Note:

       All hosts taking part in failover must be SSH and SCP-enabled to avoid the requirement for any additional authentication.

3. Configure the following products in the Active and Standby locations.

Table 1. Product configuration list

Product	Go to
SAM and Adapter Platform	Configure SAM, Adapter Platform, and BIM for failover
BIM	Configure SAM, Adapter Platform, and BIM for failover
IP Manager	Configure IP Manager for failover
Server Manager	Configure Server Manager for failover
Trap Adapter	Configure Trap Adapter
Trap Exploder	Configure the Trap Exploder for failover
Broker	Configure the Broker for failover
Failover Manager	Configure the Failover Manager

1. Verify the roles assigned by the Failover Manager to the Active and Standby components using the ic-failover-server command as described in “Display failover status” on page 57.

2. Start a topology synchronization on the Active IP Availability Manager. Type:

                  ./dmctl.exe -s INCHARGE-AM invoke ICF_TopologyManager::ICF-TopologyManager updateDiscoveryTimers
               

3. Perform a manual backup after the first discovery cycle completes for Domain Managers like IP Manager, and Server Manager. Performing a manual backup at this point ensures that the Standby components have saved the latest repository (RPS) files.

   For Standby Service Assurance Managers, it means that all users, profiles, notification lists and escalation policies, and so on will be synchronized with the Active SAM. The source domains will also be synchronized but they will be added as disabled.

   Consult Initiate manual backup for instructions.

4. Verify failover status.

   Check the failover logs in the BASEDIR/smarts/local/logs directory for any startup errors and to verify Failover Manager actions. The status of Active components must be UP/ACTIVE and the status of the Standby components must be UP/STANDBY.

   • By default, the BASEDIR/smarts/local/logs/<failover_manager>_en_US_UTF-8.log file lists startup messages for the Failover Manager.

     • The BASEDIR/smarts/local/logs/<Failover_manager>.actions.log file lists Failover Manager actions. The action logging is optional. To enable it, set the ActionLogging parameter to True in the failover.conf file.

5. Test the failover system as described in Test and Verify the Failover System chapter.