Skip to main content

Deciding which LIF failover policy to use

Decide whether to use the recommended, default failover policy or whether to change it based on your LIF type and environment.

Failover policy decision tree​

  1. Is this a NAS data LIF?
    1. Yes: go to step 2.
    2. No: use the default failover policy for your LIF type.
  2. Do you have a four-node or larger cluster?
    1. Yes: go to step 3.
    2. No: use the default failover policy.
  3. Are all nodes connected to the same data subnets?
    1. Yes: go to step 4.
    2. No: use the broadcast-domain-wide failover policy.
  4. Consider the following to choose the failover policy:
    • Choose the default system-defined failover policy to optimize node and cluster survivability, but risk not being able to fail over to more than one other node.
    • Choose the broadcast-domain-wide failover policy to optimize LIF survivability beyond two failed nodes, but risk cascading node failures.

Default failover policies by LIF type​

LIF typeDefault failover policyDescription
BGP LIFsdisabledLIF doesnot fail over to another port.
Cluster LIFslocal-onlyLIF fails over to ports on the same node only.
Cluster-mgmt LIFbroadcast-domain-wideLIF fails over to ports in the same broadcast domain, on any and every node in the cluster.
Intercluster LIFslocal-onlyLIF fails over to ports on the same node only.
NAS data LIFssystem-definedLIF fails over to one other node that is not the HA partner.
Node management LIFslocal-onlyLIF fails over to ports on the same node only.
SAN data LIFsdisabledLIF does not fail over to another port.
Note
The sfo-partner-only failover policy is deprecated. Do not use the sfo-partner-only failover policy.