It seems like your browser didn't download the required fonts. Please revise your security settings and try again.
Barracuda Load Balancer ADC

How do I configure High Availability on my two Barracuda Load Balancers?

  • Type: Knowledgebase
  • Date changed: 2 years ago
Solution #00001316

All Barracuda Load Balancers models 340 and above, all firmware versions.

The Barracuda Load Balancer works in an Active/Passive High Availability configuration. High availability between Barracuda Load Balancers is easily configured from the web interface.

  • Two (and only two) Barracuda Load Balancers of the same model.
  • Each box must be on the same version of firmware (1.5.015 or later recommended).
  • Both Barracuda Load Balancers must be configured with a unique WAN IP address on the network.
  • Each Barracuda Load Balancer (primary and secondary) should be able to see each other on the WAN network. Specifically, they need port 8002 open between them.
  • Both Barracuda Load Balancers need to be able to access all Real Servers (the backup Barracuda Load Balancer must be connected to the network and real servers in the same way as the primary, in parallel).
Once the prerequisites have been met, the procedure for configuring High Availability is as follows:
  1. Configure the primary Barracuda Load Balancer as needed (only after proper configuration and operation of the Primary Load Balancer should High Availability be configured), and generate a backup of the configuration on the Advanced > Backup page (in case the Barracuda Load Balancers are not clustered together properly).
  2. On the primary Barracuda Load Balancer, go to the Advanced > High Availability page. In the Cluster Shared Secret field, type a passphrase to be used to authenticate communication between the Barracuda Load Balancers and click Save Changes. It is recommended to not use any special characters (%@#$!^&*).
  3. On the Barracuda Load Balancer intended to be the backup, go to the Advanced > High Availability page. Enter the same Cluster Shared Secret here, and click Save Changes.
  4. If the Cluster Host Name field is present, leave it blank on both Barracuda Load Balancers.
  5. Regardless of the deployment mode, make sure the LAN IP Address and LAN Netmask fields are blank on the Basic > IP Configuration page of the intended backup Barracuda Load Balancer.
  6. On the Barracuda Load Balancer intended to be the backup, go back to the Advanced > High Availability page. Enter the WAN IP Address of the primary Barracuda Load Balancer under the Clustered Systems heading and click the Join Cluster button. this will begin the clustering process. Please note that this will wipe out all settings (beyond the administrator password, timezone, and basic IP configuration) on the Barracuda Load Balancer joining the other.
  7. The clustering process will complete after a few minutes.
  8. Once the clustering process is completed, reboot the backup unit to complete the configuration and the High Availability setup will be complete.
If everything went well, you should be able to review the cluster members by going to the Advanced > High Availability page on either box. You should see both the primary and secondary Barracuda Load Balancers listed under the Clustered Systems heading, each with a green indicator. To test whether the cluster is set up properly, go to the Basic > Services of the backup Barracuda Load Balancer and see whether the services already configured on the primary have appeared on the backup. If they appear with red status indicators, it is because these services are not active on the backup while the primary is up and functioning. You can also test the cluster by making a change on either Barracuda Load Balancer's configuration on the Basic > Services page and watching that change replicate to the other Barracuda Load Balancer in just a few seconds. If (after a few minutes) either of the High Availability indicators is red or the Service configurations are not syncing between them, please contact Barracuda Networks Technical Support.

Once clustered, the Baracuda Load Balancers maintain a heartbeat signal between them to determine the health of each other. In the event that the primary Barracuda Load Balancer fails, the backup should activate and continue normal operations within 1 to 10 seconds. The configuration will automatically return to normal if or when the primary Barracuda Load Balancer comes online again. At that point, the switch will also occur within 1 to 10 seconds.

Additional Notes:
The Barracuda Load Balancer model 240 does not support High Availability.

If you need to change the WAN IP address of one or both Load Balancers configured in High Availability, you will need to first break the HA cluster. To break a High Availability cluster see solution #00004238.

Link to This Page: