Forum Discussion

Victor_Barros_1's avatar
Victor_Barros_1
Icon for Nimbostratus rankNimbostratus
Mar 01, 2015

Configure LTM for HA and not Load Balance

Hi, I have a simple setup i need to configure. I have one pool with two nodes: A and B. All traffic goes to Node A until Node A become unavailable at which point all traffic should automatically go to Node B. Traffic should remain on Node B even after Node A comes back on line.

 

Can some please point me to what feature of the LTM or documentation I need to look at in order to achieve this functionality?

 

Regards.

 

application delivery
LTM
management

2 Replies

  • amolari's avatar
    amolari
    Icon for Cirrostratus rankCirrostratus
    Mar 01, 2015

    search for "priority group" in the documentation. With that, when higher prio node comes back online, it will send the traffic to it. You can either use the "manual resume" monitor functionality to avoid that, or use the "Single Node Persistence" iRule (search on DevCentral).

     

  • StephanManthey's avatar
    StephanManthey
    Icon for Nacreous rankNacreous
    Mar 01, 2015

    Hi Victor,

     

    I´m using the "destination address affinity" on BIG-IP successfully exactly for this purpose. It designed for something else but will stick all incoming traffic to the same pool member until it becomes unavailable. Will flip back only, if this one goes down.

     

    Thanks, Stephan