NimbostratusWe have two F5 in HA (Active/Standby), and each have two interfaces, one for external (use for VirtualServer only) and another for internal (for connection to back-end servers and VirtualServer)
Is it mandatory to have floatingIP on the external interface?
Thanks,
No it is not, you only need a floating IP if you use snat automap (on vlan between the BIG-IP and the pool members) or when using a IP forwarding virtual server. (the float should be on the vlan between the BIG-IP and the router pointing towards the BIG-IP (floating ip))
Cheers,
Kees
Altostratusnot mandatory, but n you will have to double your request for dmz or outside access for the device. firewall rules and such. the floater allows you to allow only 1 ip address for those boxes to communicate outside or thru the dmz.
CumulonimbusMonitors uses always self ip even if there is a floating ip!
So firewall rules may allow self ip and floating ips if snat is enabled!
Altostratuscorrect
NimbostratusIf your VS is not in the same network as your Self-IPs on the external vlan then you will need the floating IP on the external vlan as this will be the next hop route from your external FW/router to reach the Virtual Servers.
NimbostratusYes, In a High Availability configuration.
The Floating IP Address is used by the Active Device.
When Active it listens on this IP Address and then can utilize either the Floating IP Address or it's Self-IP Address in responding.
The Standby device uses its Self-IP Address to listen to what is going on and verify that requests are being responded to by the Active Device as well as determine its on ability to act as the Active Device (Example: See K13297: Overview of VLAN failsafe (10.x - 13.x)).
