Standalone Device 1 NIC - fadlytabrani/f5-azure-ha-fo GitHub Wiki

Topology

Topology

Deployment

Choose your flavor of BIG-IP from the Azure marketplace and deploy it normally using the deployment wizard.

Create an Azure load balancer with the following settings:

  • A public frontend IP configuration for the web service (1.1.1.1 for example),
  • a backend pool with the BIG-IP selfip of 10.4.2.10 as a member,
  • a TCP health probe at port 694,
  • a load balancing rule with the above, at port 80 with floating ip enabled.

The next steps are all done on the BIG-IP device.

Create a standard virtual server on the selfip to indicate to the the Azure load balancer that the device is active.

# bash
tmsh create ltm virtual _cloud_lb_probe_listener_ destination 10.4.2.10:694 source 168.63.129.16/32 ip-protocol tcp

Create a virtual server with the same address as the public frontend ip(1.1.1.1) of the Azure load balancer. Assign the web servers as a pool and the BIG-IP should respond accordingly.

How Does it Work?

  1. The Azure load balancer is a transparent device.
  2. The BIG-IP receives the packet on a VLAN where a selfip has been configured with a virtual server of the same public destination address, thus it responds.

Where Do You Go From Here?

This design supports as many public IPs as the Azure load balancer supports, 600 at the time of writing. Use the same steps above to configure them and virtual servers on the BIG-IP.

If you need to provide internal services, deploy an internal Azure load balancer instead. It works the same, the only difference is a private frontend ip.

Extend the deployment, Active Standby HA FO 1 NIC