Lab 5: Create & Enable HA Groups

In Lab 4, we observed that basic HA configuration does not take a physical interface link failure into account.

In Lab 5, to address this limitation, we will use the HA Group Failover object.

Lab Tasks:

• Task 1: Create a Gateway Pool

• Task 2: Create HA Groups

• Task 3: Enable HA Group as the Preferred Failover Method

Task 1: Create a Gateway Pool

In Task 1, we will create a pool that contains the default gateway as the pool member. The function of this pool is to help the BIG-IP identify network related problems upstream.

We will use this pool object as a failover event trigger in Task 2 of our HA Group configuration.

1. Navigate to: Local Traffic > Pools > Pool List > click the "+" sign to create a new pool:

   

2. Create a Gateway Pool with the following settings:

   • Name: ext_gw_pool

   • Health Monitors: gateway_icmp

   • Within the Resources Section:

     • New Node Address: 10.1.20.1

     • Service Port: * All Services

   • Click the Add button

     

   Click the Finished Button:

   

Task 2: Create HA Groups

In Task 2, we will create HA Group configurations on each BIG-IP.

NOTES:

• HA group configuration is device specific and is not synced between members of a DSC group.

  • You MUST create a separate HA group on every device in the device group for this traffic group.

1. On each BIG-IP, create an HA Group:

2. Navigate to: System > High Availability > HA Group List > click the "+" button:

   

3. Add an HA Group Name; example: bigip-ha-group

   

4. Next, we will add our External Gateway Pool we created earlier to the Pools HA Group Configuration:

   • Click the "Add" button

     

5. In the Pool drop-down, select "ext_gw_pool," and click the "Add" button:

   

6. Next, we will add our Trunk links to our HA Group Configuration:

   • Under the Trunks section, click the "Add" button:

     

7. From the drop-down, select the the External Trunk object (ext_trunk), and click the "Add" button:

   

8. Repeat this step for the Internal Trunk object (int_trunk):

   

9. Verify that your HA Group Trunk configuration reflects the following, and Click the "Create HA Group" button:

   

10. After HA Group Creation, you should be presented with the following screen, showing your HA Group configuration object:

    

About HA Group Scoring

The BIG-IP with the highest HA Group scoring value will become the active device.

The HA Group score is being calculated by the sum of all active trunks and pool member objects.

In addition, there is an "Active Bonus" value that is added to the Active device, preventing a race-condition of both devices having the same score when all HA objects are "up."

The Active Bonus is added for two reasons:

• To prevent an active-active split-brain scenario

• To allow the active device to stay active in specific use cases.

One use-case is:

• If there are multiple links in the HA trunk and one interface goes down, but the remaining links can still handle the traffic, then the Admin might decide that this is not a reason to failover because of service stability.

For detailed information and HA Best Practices, see following K-article: K16947: F5 recommended practices for the HA group feature

Task 3: Enable HA Group as the Preferred Failover Method

1. On the Active BIG-IP, Navigate to: Device Management > Traffic Groups > traffic-group-1 hyperlink:

   

2. From the "Health Monitor" Section, select the drop-down for "HA Group:"

   

3. Select our HA Group we created previously:

   

4. Under the Failover Configuration section, select the radio button for "Failover to Device With Best HA Score," then click the "Save" button:

   

5. Review the Traffic Groups screen, and verify the Failover Method is set to HA Score:

   

6. Perform this same procedure to setup the HA Group configuration on bigipB.f5demo.com

Lab Summary

In this lab, you enhanced your HA configuration to leverage HA Groups. HA Groups provide the ability to monitor an "object" and take action upon an event. After completion of these lab tasks, your BIG-IP should be leveraging HA Groups for Failover. In the following lab, we will test & validate how this configuration enhances the failover timing.

This completes Lab 5.