Control Center High Availability Failover¶

This guide contains instructions for activating the standby Lantronix Control Center in the event of a loss of the primary Lantronix Control Center, and instructions for reestablishing the High Availability backup after the failover.

Failover Procedure¶

1. Connect to the primary Control Center (if available) via ssh as emsadmin and become root. Use emsadmin's password when prompted.

# ssh emsadmin@172.30.161.26
emsadmin@172.30.161.26's password: 
[emsadmin@ems26 ~]$ sudo su -
[sudo] password for emsadmin: 
[root@ems26 ~]#

2. Run the following commands to deactivate the primary Control Center:

[root@ems26 ~]# ucc stop
[root@ems26 ~]# ucc chkconfigoff
[root@ems26 ~]# su – oracle
[oracle@ems26 ~]$ cd /home/oracle/ddl/ha_scripts
[oracle@ems26 ha_scripts]$ ./deactivate.sh

3. Connect to the secondary Control Center via ssh as emsadmin and become root. Use emsadmin's password when prompted.

# ssh emsadmin@172.30.161.52
emsadmin@172.30.161.52's password: 
[emsadmin@ems52 ~]$ sudo su -
[sudo] password for emsadmin: 
[root@ems52 ~]#

Switch users to oracle and then activate the standby Control Center as primary:

[root@ems52 ~]# su – oracle
[oracle@ems52 ~]$ cd /home/oracle/ddl/ha_scripts
[oracle@ems52 ha_scripts]$ ./become-primary

4. Reconnect to the secondary Control Center via ssh as emsadmin and become root. Use emsadmin's password when prompted.

Run the following commands:

[root@ems52 ~]# ucc start
[root@ems52 ~]# ucc chkconfigon

5. If using a DNS or a VIP, update the DNS record or VIP configuration to point the record or VIP for the Control Center at the secondary Control Center’s IP address.

6. If using automatic CC failover via multiple IPs, the LMs will automatically Heartbeat to the newly active Control Center.

7. Connect to the secondary Control Center's web interface. Verify that it is running and has up-to-date data. All appliances should be Heartbeating to the secondary Control Center.

Post-Failover Tasks¶

After failover has been completed, perform the following tasks to recover the unavailable Control Center and create a standby database on it.

Recover the failed Control Center, either by repairing the failed hardware or by remediating the conditions that caused the outage.
Create a new standby database on the recovered primary by running the “become-standby” script on the recovered primary by following the below steps. This should take 3 to 6 hours to complete.
Either continue to run the systems in this state, or alternatively failover back to the recovered primary using the above failover procedure.