Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

This might be a pedestrian question but what is the difference between a "Floating IP" address and a "Virtual IP" address? Are they synonyms?

share|improve this question

2 Answers 2

up vote 2 down vote accepted

To me, the terms mean different things.

A floating IP address is used to support failover in a high-availability cluster. The cluster is configured such that only the active member of the cluster "owns" or responds to that IP address at any given time. Should the active member fail, then "ownership" of the floating IP address would be transferred to a standby member to promote it as the new active member. Specifically, the member to be promoted issues a gratuitous ARP, announcing the new MAC address–to–IP address association.

A virtual IP address refers to the IP address of a virtual server, and is a more nebulous term. With F5 load balancers, for example, the virtual servers are the services (websites, etc.) you want to host.

More concretely, suppose you have a pair of load balancers in an active-standby cluster. For each interface or VLAN, the load balancers would each have a self IP address, as well as a floating IP address that is shared between both members. When the load balancer relays incoming requests to the back-end nodes, it uses the floating IP address as the source address, so if the load balancer dies, its partner will be able to take over and receive the response. Each website or other service being hosted on the load balancers would have its own IP address, which you could call a "virtual" IP address. (You could say that these virtual IPs "float" as well, since control of them would transfer to the standby node in the event of a failover.)

share|improve this answer
    
And doing a active standby cluster is faster than simply restarting a load balancer? –  CMCDragonkai Nov 10 at 15:38
    
IF two load balancers have the same floating IP address, then when a request is made to that floating IP address, who decides which Load balancer must be executed. –  JavaTechnical Dec 10 at 16:23
    
@JavaTechnical IP-to-Ethernet mapping is done through ARP. The members of the cluster coordinate amongst themselves such that at any time, only the active unit will respond to ARP requests for the floating IP. During failover, the new active unit sends a gratuitous ARP announcement informing all other devices on the subnet (especially the router) that the new active unit's MAC address should be associated with the floating IP. –  200_success Dec 10 at 17:25
    
@200_success Then who first receives the request to floating IP? Can it be any member (no matter whether that member is busy or not). Also if there is a fail over of a member, who takes care of removing that load balancer from the list of available balancers? –  JavaTechnical Dec 11 at 6:40
    
@JavaTechnical The cluster members negotiate amongst themselves which one is active. For BIG-IP load balancers, the election is influenced by the Redundancy State Preference setting. –  200_success Dec 11 at 6:43

Yes, they're same. These are terms usually used in load balancing configuration etc

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.