force a refresh arp caches

taking a step back if you are not configured for HA then I am assuming you must have configured it for Active-ACtive.

 

 

***Excerpt from the LTM Network Guide***

 

For an active-active system, suppose that when you initially ran the Setup

 

utility on unit 1, you specified 11.12.11.3 as the internal floating IP address,

 

and when you ran Setup on unit 2, you specified 11.12.11.4 as its internal

 

floating IP address. When you synchronize the configurations later,

 

11.12.11.3 should appear on both units as the floating IP address belonging

 

to unit 1, and 11.12.11.4 should appear on both units as the floating IP

 

addresses belonging to unit 2.

 

Then, if unit 1 fails over:

 

• Unit 2 assumes the internal floating IP address of unit 1 (11.12.11.3).

 

• The back-end servers that normally send traffic to the internal address

 

11.12.11.3 on unit 1 continue to send their traffic to that same address,

 

even though this incoming traffic is now processed by unit 2.

 

• The back-end servers that normally send traffic to the internal address

 

11.12.11.4 on unit 2 continue to send their traffic to that same address.

 

Conversely, if unit 2 fails over:

 

• Unit 1 assumes the internal floating IP address of unit 2 (11.12.11.4).

 

• The back-end servers that normally send traffic to the internal address

 

11.12.11.4 on unit 2 continue to send their traffic to that same address,

 

even though this incoming traffic is now processed by unit 1.

 

• The back-end servers that normally send traffic to the internal address

 

11.12.11.3 on unit 1 continue to send their traffic to that same address.

 

 

If your LTM is not configured for Active-Active in this manner then yes it will never work until the unit is reloaded. Because one LTM thinks it's active over the other.

 

 

Under HA the unit ID's are not used because they know the state of each other since they share a single common floating address.

 

 

CB