Why is my ONTAP volume full even after deleting files?

Snapshots may still reference deleted blocks. Check volume usage, snapshot sizes, and snapshot autodelete configuration before deleting any recovery point.

Why does an NFS mount to ONTAP say permission denied?

The export policy may not match the client, its protocol or security flavor may be wrong, or file permissions may deny access. Use export-policy check-access to test the client path.

What should I do when an ONTAP aggregate is offline after power loss?

Start with read-only health, disk, shelf, aggregate and event checks. Do not force an aggregate online or run destructive recovery commands until the RAID and disk state is understood.

Home / Troubleshooting / Common ONTAP Problems

Common ONTAP Problems — Symptoms, Causes & Fixes

Q: Why is SnapMirror lag high or still growing?

The source change rate may exceed transfer rate, the intercluster network may be constrained, or source production I/O may be competing with replication. Inspect relationship detail, lag time, LIFs, and port errors.

Start with the plain-language symptom. Each answer gives the likely causes, safe first commands, and the detailed runbook for complete diagnosis and recovery.

Why is my ONTAP volume full even after I deleted files?

Likely causes

Snapshots still reference deleted blocks, the FlexVol itself is full, or its aggregate has run out of headroom.

Quick checks

volume show -vserver vs1 -volume vol1 -fields size,used,percent-used,space-guarantee,space-slo
snapshot show -vserver vs1 -volume vol1
storage aggregate show -aggregate aggr1 -fields size,used,percent-used,percent-snapshot-reserve

Review recovery requirements before deleting a snapshot. See the volume space exhaustion runbook.

Why has the Snapshot reserve been consumed?

Likely causes

Snapshots grew beyond the reserve because they still hold changed or deleted blocks; autodelete may be disabled or use an unsuitable trigger.

Quick checks

snapshot autodelete show -vserver vs1 -volume vol1
volume show -vserver vs1 -volume vol1 -fields snapshot-reserve,percent-snapshot-space

Deleting snapshots permanently removes recovery points. See snapshots holding deleted data.

Why is SnapMirror lag high or still growing?

Likely causes

The source change rate exceeds the transfer rate, the intercluster network is constrained, or busy source I/O competes with replication.

Quick checks

snapmirror show -fields destination-path,state,status,lag-time,last-transfer-end,last-transfer-size
network interface show -role intercluster
network port show -fields link,error-*

See the SnapMirror lag runbook.

Why did my SnapMirror transfer fail?

Likely causes

Cluster or SVM peering is broken, an intercluster LIF is down, the destination is full, or the required common snapshot is missing.

Quick checks

snapmirror show -destination-path vsB:volB -detail
cluster peer show
vserver peer show

The detailed error determines the fix. See SnapMirror transfer failures.

What does “Failed to find a sync marker” mean in SnapMirror?

Likely causes

The expected baseline snapshot or transfer marker was deleted, the destination was reverted, or initialization never completed.

Quick checks

snapshot show -vserver vsA -volume volA
snapshot show -vserver vsB -volume volB
snapmirror resync -destination-path vsB:volB

Try resync before a full re-seed. See the sync-marker recovery procedure.

Why does my NFS mount fail with “Permission denied”?

Likely causes

The export policy does not match the client, its protocol or security flavor is wrong, root is being squashed, or file permissions deny access.

Quick checks

vserver export-policy check-access -vserver vs1 -client-ip 10.1.2.50 -volume vol1 -authentication-method sys
vserver export-policy rule show -vserver vs1 -policyname default

See the NFS mount failure runbook.

Why does an NFS mount time out or hang?

Likely causes

The data LIF is unreachable, routing is wrong, a firewall blocks NFS/RPC traffic, or the LIF’s physical port is down.

Quick checks

network interface show -vserver vs1 -fields address,status-admin,status-oper,home-node,home-port
network route show -vserver vs1
network port show -fields link,health-status

See NFS connectivity diagnosis and the network runbook.

Why do NFS clients report a stale file handle?

Likely causes

The client holds a handle for an object that was deleted, renamed, moved, or re-exported.

Quick fix

Remount the share on the client. If the error persists after a volume move, mount it fresh and verify the export and junction.

See the matching entry in the ONTAP error reference.

Why is ONTAP latency high or performance suddenly slow?

Likely causes

Node CPU, disk utilization, protocol workload, a hot volume, or a QoS limit may be the bottleneck.

Quick checks

system node run -node node1 -command sysstat -M 1 10
qos statistics volume show -interval 5
qos statistics volume show -vserver vs1 -volume vol1 -interval 5

See the performance analysis runbook.

What should I do when an aggregate is offline after a power loss?

Likely causes

A dirty shutdown may leave disks missing or broken, a shelf unreachable, RAID degraded, or the aggregate unable to come online safely.

Read-only triage

system health alert show
storage disk show -broken
storage shelf show
storage aggregate show-status -aggregate aggr1
event log show -severity ERROR

Do not force the aggregate online until its state is understood. See the post-power-loss runbook.

Why is a management or data LIF unreachable?

Likely causes

The LIF or physical port is down, it failed over to the wrong network, its route is missing, or VLAN, MTU, or broadcast-domain configuration is inconsistent.

Quick checks

network interface show -fields vserver,lif,role,home-node,home-port,curr-node,curr-port,address,state,failover-policy
network port show -fields link,health
network route show

See the network and connectivity runbook.

Why does SMB authentication fail or return access denied?

Likely causes

A bad or locked account can fail session authentication. If login succeeds, both the share ACL and NTFS permissions must allow access; AD, DNS, Kerberos, and clock health also matter.

Quick checks

vserver cifs show -vserver vs1
vserver cifs domain discovered-servers show -vserver vs1
vserver cifs share access-control show -vserver vs1 -share data
vserver security file-directory show -vserver vs1 -volume vol1 -path /

See the SMB/CIFS troubleshooting runbook.

Why am I out of inodes when the volume still has space?

Likely cause

The FlexVol exhausted its available filesystem entries, so clients can see 100% inode use even when byte capacity remains.

Quick checks and fix

volume show -vserver vs1 -volume vol1 -fields files,files-used,percent-inode-used
volume modify -vserver vs1 -volume vol1 -files 2000000

Adding inodes requires space and is limited by the volume maximum. See inode exhaustion.

Why are dedupe or compression savings not appearing?

Likely causes

Storage efficiency may not be enabled, or its scan may be scheduled, incomplete, or not yet reflected in the logical-versus-physical view.

Quick checks

volume efficiency show -volume vol1
volume show -fields logical-used,logical-space-used

Enable or schedule efficiency only after checking its current state. See the storage efficiency command reference.

Why is takeover or giveback stuck or vetoed?

Likely causes

The partner is in an incompatible state, an unhealthy or offline aggregate blocks takeover, or giveback has unresolved vetoes.

Quick checks

storage failover show
storage failover show-takeover
storage failover show-giveback

Resolve every veto before retrying; do not force giveback without understanding the impact. See the relevant entries in the ONTAP error reference.

Use placeholders carefully Replace example SVMs, volumes, nodes, IPs, and aggregates with values from your cluster. Commands target ONTAP 9.x; use man <command> for the flags available on your release.

← All troubleshooting guides