Home‎ > ‎Barracuda‎ > ‎

Putting LB in HA mode


Overview: 

The Barracuda LBs can be configured so that there is a backup unit that monitors the health of the primary, and takes over if the primary goes down.  

The health checking, and syncing, is done from the backup unit.  The backup checks the configs of the primary every few minutes, so the two might not have synced up configs right after a change is made to the primary.  

Creating a Cluster:

The assumption is that you already have a working primary unit.  The following process would allow you to pair another unit to it and configure them as an HA pair. 

Get the backup box online: 

On the the secondary LB, and install the initial configuration including the WAN, DNS, and Activation.  DO NOT CONFIGURE THE LAN INTERFACE. 

Modify the Primary Box: 

From the Basic Tab, under the Status sub-tab, confirm that the current setting is that the device is in Stand-alone mode


From the Basic tab and the High Availability sub-tab, Enter in the following, and then select Save Changes. 
  • Cluster Shared Secret: Pick a random code that would be hard to guess.  Save this code, because you will need it for the other unit. 
  • Cluster Group ID: the default "200" is fine unless you have more then one pair of load balancers on the same broadcast.  
  • Failback Mode: set this to "Manual".  When the primary unit fails, the secondary will assume the role of the primary, and stay that way until it fails.  (Whatever happened to the primary might happen again, so best to keep the traffic off of it until you have figured out what the issue was.)
  • Failover if LAN link is down: set this to "YES" to ensure that the system fails if the inside link on the primary fails. 


Enable the Backup System: 

Like in the primary unit, on the backup unit goto the Basic tab and the High Availability sub-tab, Enter in the following, and then select Save Changes. 
  • Cluster Shared Secret: use the random code that that was used for the primary unit
  • Cluster Group ID: the default "200" or whatever was chosen for the primary unit.  
  • Failback Mode: set this to "Manual".  When the primary unit fails, the secondary will assume the role of the primary, and stay that way until it fails.  (Whatever happened to the primary might happen again, so best to keep the traffic off of it until you have figured out what the issue was.)
  • Failover if LAN link is down: set this to "YES" to ensure that the system fails if the inside link on the primary fails. 


Under the Basic tab, and IP Configuration sub-tab, make sure that the LAN IP Configuration is blank.  If not, delete the setting and select Save Changes.


Under the Basic tab, and the High Availability sub-tab, enter the IP address of the Primary unit in the Clustered Systems section, and select the Join Cluster button. 

Back in the Basic Tab, under the Status sub-tab, you will see the clustering proces in the Running Tasks section.  While this is happening, your backup unit will show as the primary, but after a couple of seconds or minutes (depending on the configs on the primary), this will change. 

Once it is complete, the High Availability Status will change to BACKUP, confirming that the backup unit is properly paired to the primary. 

Confirm the Primary Unit.

Just to make sure, you can check the primary unit's Basic tab and Status sub-tab to confirm that the High Availability Status is set to Primary. 



Comments