ONTAP LIF failover troubleshooting

A data LIF can be administratively up yet unreachable, operationally down, hosted away from home, or left with no useful failover target. This runbook separates those states before anyone changes a failover group or forces a revert.

Know which behavior you are testing

NAS and SAN behave differently. This guide’s failover-group workflow is for IP LIFs, especially NAS data LIFs. SAN data LIFs normally rely on host multipathing and are shown with failover disabled; do not “fix” an iSCSI or FC LIF by applying a NAS failover policy.

A failover group is the candidate port set. The failover policy filters that set to produce the LIF’s actual targets. A broadcast domain describes layer-2 reachability, while the LIF’s home node and home port define where a manual or automatic revert tries to return it. Treat those as four separate facts.

Preserve the working path. Before modifying a group, policy, home port, VLAN, ifgrp, or broadcast domain, record the current and home locations and test another management path. Removing the last valid target can turn a configuration correction into an outage.

Five-minute read-only triage

cluster::> network interface show
cluster::> network interface show -vserver svm1 -lif data1 -instance
cluster::> network interface show -vserver svm1 -lif data1 \
  -fields status-admin,status-oper,home-node,home-port,curr-node,curr-port,is-home,auto-revert,failover-policy,failover-group
cluster::> network interface show -vserver svm1 -lif data1 -failover
cluster::> network port show
cluster::> network port broadcast-domain show
cluster::> network interface failover-groups show

Interpret the result in order: down/down may be intentional; up/down means ONTAP wants the LIF up but cannot operate it; up/up with is-home false is not automatically a fault. A healthy failover leaves a NAS LIF available on another valid port. Confirm client reachability and application health instead of using “not home” alone as an incident trigger.

For a client outage, also verify the current port’s link, VLAN/ifgrp state, broadcast-domain membership, IPspace, routing, upstream VLAN carriage, DNS, and client network path. LIF mobility cannot repair a switch-side or routing design that makes every candidate port unreachable from the clients.

Validate the target set before testing failover

cluster::> network interface show -vserver svm1 -lif data1 -failover
cluster::> network interface show -vserver svm1 -lif data1 \
  -fields failover-group,failover-policy
cluster::> network interface failover-groups show \
  -failover-group Default
cluster::> network port broadcast-domain show

The first command is decisive: it shows the targets produced by the group and policy together. Check that each target is a real data port, VLAN, or ifgrp in the intended layer-2 network. A data LIF should not land on the out-of-band management port e0M. NetApp recommends retaining the default policy for the LIF type unless a documented topology requires a change.

LIF type	Expected model	Do not assume
NAS data	Uses failover targets; commonly the system-defined policy	Every cluster port is client-reachable
Cluster / node management	Policy is deliberately constrained by role	A NAS data policy is suitable
Intercluster	Normally stays local to its node	NAS-style node mobility is required
SAN data	Host multipathing supplies path resilience	LIF failover should be enabled

Revert only after the home path is healthy

A revert is movement back to the configured home node and port; it is not a generic repair command. Confirm that the home port is operational and reachable on the correct layer-2 network, then use the narrowest command:

cluster::> network interface show -vserver svm1 -lif data1 \
  -fields home-node,home-port,curr-node,curr-port,is-home,status-oper
cluster::> network port show -node node1 -port a0a-120

# Run after validating client impact and the home path
cluster::> network interface revert -vserver svm1 -lif data1
cluster::> network interface show -vserver svm1 -lif data1

If the home port is unavailable, ONTAP leaves the LIF where it is. If you want automatic return after recovery, inspect the existing setting and change it only as an operational-policy decision:

cluster::> network interface show -vserver svm1 -lif data1 \
  -fields auto-revert
cluster::> network interface modify -vserver svm1 -lif data1 \
  -auto-revert true

Automatic revert can move an address when a port or node returns. Decide whether that behavior matches the protocol, maintenance process, and change window; do not enable it merely to clear a dashboard warning.

Repair the configuration deliberately

Prefer repairing broadcast-domain membership and the system-created default group when those describe the intended network. Use a custom failover group only when the topology truly requires a restricted target set. Inventory first, make one change, then re-run network interface show -failover.

# Inspect supported values on this exact ONTAP release
cluster::> network interface modify -vserver svm1 -lif data1 -failover-policy ?

# Example only: use names validated in your environment
cluster::> network interface modify -vserver svm1 -lif data1 \
  -failover-policy broadcast-domain-wide -failover-group data-bd
cluster::> network interface show -vserver svm1 -lif data1 -failover

Example, not a universal prescription: broadcast-domain-wide is valid for appropriate IP LIFs, but the default depends on LIF type and release. The group name data-bd must already exist and contain only verified targets. Avoid rewriting a whole target list with failover-groups modify during incident triage; that command replaces the list.

Symptom-to-check map

Symptom	First evidence	Direction
LIF is up/up but clients fail	Current node/port, route, VLAN, DNS, client path	Treat reachability separately from LIF status
LIF is up/down	Port state, broadcast domain, VLAN/ifgrp, EMS	Restore a valid hosting path before reverting
No useful target	`network interface show -failover`	Correct group/domain membership or policy
LIF stays non-home	Home port health and `auto-revert`	Leave stable or perform an approved manual revert
SAN LIF says disabled	LIF service/data protocols and host paths	Validate ALUA/MPIO; disabled failover is expected
Failure repeats after port recovery	Physical errors, switch logs, EMS, port health	Fix the underlying link or network fault

Primary sources and currency

Reviewed 12 September 2026 against NetApp’s current ONTAP networking and CLI documentation. Verify syntax and supported policies with man network interface on the exact ONTAP release and platform in service.