Forum Discussion

DB's avatar
DB
Icon for Nimbostratus rankNimbostratus
Apr 25, 2018

DNS Wide-IP "Priority" Pools or members?

I'd like to have a Wide-IP that always resolves to a particular LTM service, but in the event that service goes down, have it fail over to a Disaster Recovery Site. No load balancing back and forth, a straight "if A is up, Give answer=a, if A is down, give answer=B". In the LTM world I would look to a priority group setup to accomplish this. For Wide-IP, I can use a Last Resort Pool, but is there another way to fail over to a DR site that should be considered?

 

1 Reply

  • You best option on gtm load balancing is Global Availability.

    Your understanding is correct, In LTM its Priority Groups, In GTM its GA.

    About the Global Availability load balancing method

    The Global Availability load balancing method distributes DNS name resolution requests based on the order of resources in a list. Using global availability, BIG-IP GTM sends a request to the first available resource in a list. Only when a resource becomes unavailable does BIG-IP GTM send requests to the next resource in the list.

    I'm hoping you already have your LTM's configured as servers in the GTM. So it would be something below to create the setup,

    tmsh modify gtm server dc1ltm virtual-servers add { 10_10_10_10_443 { destination 10.10.10.10:443 } }
    tmsh modify gtm server dc2ltm virtual-servers add { 20_20_20_20_443 { destination 20.20.20.20:443 } }
    
    tmsh create gtm pool appname_pool load-balancing-mode global-availability fallback-mode return-to-dns members add { dc1ltm:10_10_10_10_443  { order 0 } dc2ltm:20_20_20_20_443  { order 0 } } 
    
    tmsh create gtm wideip app.wideip.com ipv6-no-error-response enabled pools add { appname_pool }