Storage Admin HubConfiguration · Protection · Recovery
All guides/Interview Q&A
Interview Q&A · ONTAP 9

NetApp interview: intermediate operations

Ten operational questions on access, paths, snapshots, QoS, capacity and change validation.

Scenario

An intermediate answer should give the order of checks and a validation step. The examples use ONTAP 9 concepts but exact options and defaults depend on release and platform.

Before you change production

Commands and screens can differ by release and platform. Replace example names and documentation IP addresses. Check prerequisites, impact, current health and rollback with your change owner.

1. An NFS client can mount but cannot write. Where do you look?

  1. First establish the effective client IP, user UID/GID, protocol version, mount options and exact file path. A successful mount proves some connectivity and namespace traversal, but not necessarily read-write permission. Inspect the volume export policy and ordered rule: client match, protocol, read/write security flavor, root treatment and parent traversal.
  2. Then inspect file mode, ownership, ACL and identity mapping. Test a new file in a disposable directory as a named non-root user. If the client receives “read-only filesystem,” also check the volume or mount state rather than only the export rule.
Read-only checks 
vserver export-policy rule show -vserver svm_prod
volume show -vserver svm_prod -fields policy,junction-path,state
Verify

The correct user writes one test file while unauthorized clients remain denied.

2. Why might SMB authentication work but a folder fail?

  1. A successful SMB session confirms authentication; authorization still passes through share ACL and file/directory ACL, possibly an identity mapping when the object uses UNIX security. Ask for exact user, group, workstation, path, time and operation. Compare with a working user under the same share.
  2. Use ONTAP effective-permission inspection or Windows Security tab with the correct administrative context. Fix the narrow missing group or file ACL. Do not grant Everyone full access as a troubleshooting shortcut.
Read-only checks · adapt identity 
vserver cifs show -vserver svm_prod
vserver cifs share show -vserver svm_prod
vserver security file-directory show-effective-permissions -vserver svm_prod -win-user-name EXAMPLE\testuser -path /projects -share-name projects
Verify

Describe which permission layer blocked the user and prove the corrected operation.

3. What happens when a NAS data LIF’s home port fails?

  1. ONTAP may move an eligible NAS data LIF to a failover target according to its policy and broadcast domain. The alternate port must actually reach the client VLAN and routes. Existing connections may pause or reconnect depending on protocol and client behavior; a LIF marked up on an unsuitable port does not mean users can reach it.
  2. Inspect current versus home node/port, failover group, broadcast-domain membership and switch tagging. Test from the affected client subnet and restore the home path when safe. Avoid assuming identical behavior for SAN LIFs.
Read-only checks 
network interface show
network interface show -fields home-node,home-port,curr-node,curr-port,failover-policy
network port show
Verify

The LIF is reachable on the intended network from representative clients.

4. A LUN is mapped but the host sees no device. What is the sequence?

  1. Check the host initiator identity, target connectivity or fabric login, SVM protocol service and online SAN LIF, igroup membership, LUN online state, mapping and reporting nodes. Then rescan the host and inspect multipath. FC zoning and iSCSI session/CHAP failures occur before a valid mapping can be used.
  2. Do not recreate the LUN or format an unknown device. Compare its ONTAP serial to the host device identity and check the host utility/driver version against the support matrix.
Read-only checks 
network interface show
igroup show
lun show
lun mapping show
Verify

The host discovers exactly the expected serial through supported paths.

5. Why can a thin volume or LUN run out of physical space?

  1. Thin provisioning allocates logical capacity without reserving all backing blocks. Several volumes can commit more logical capacity than the aggregate can physically hold. Snapshots, growth and clones may retain blocks; host free space does not guarantee aggregate free space.
  2. Check volume active/snapshot usage, aggregate physical headroom, any space-reservation guarantees, autosize thresholds and LUN space reporting. Add real capacity or resize within known headroom after approving application impact; do not delete required SnapMirror snapshots to silence an alert.
Read-only checks 
volume show-space
storage aggregate show
volume snapshot show
lun show -fields space-reserve
Verify

Explain the exhausted layer and the corrective capacity action.

6. What is the difference between volume move and SnapMirror migration?

  1. A volume move relocates a volume within the same ONTAP cluster to a different aggregate or node, with a managed cutover. SnapMirror transfers data to another volume, often another cluster, with a replication relationship and an explicit destination activation step. Their network, client cutover and rollback plans differ.
  2. For a SAN volume moving between HA pairs, update SLM reporting nodes as required before the move so hosts retain optimized paths; after the move, verify host path states. For cross-cluster SnapMirror, plan DNS, SVM protocols and only one writable copy.
Verify

Choose the correct method for intra-cluster placement versus site migration.

7. How do you select a Snapshot schedule and retention?

  1. Start with application RPO, restore use cases, rate of change, space, backup/replication dependencies and compliance. A schedule defines when points are taken; retention controls how many remain. A storage snapshot alone may be crash consistent for an active database, so involve the application backup method.
  2. Example: hourly local snapshots for rapid mistaken-file recovery plus daily off-cluster copies for a week. Test a file restore and a full application restore; record actual recovery time. Do not claim a seven-day recovery point unless the policy and transfers demonstrably retain it.
Verify

Name one local and one off-cluster recovery test tied to an RPO/RTO.

8. What is QoS and when would you apply it?

  1. Quality of Service policies can cap or guarantee performance within supported platform and release capabilities. A cap can stop a noisy workload from monopolizing resources, but a low cap can also cause application latency. A minimum or adaptive policy has different placement and capacity requirements.
  2. Example: a backup scan is saturating a shared aggregate during business hours. Measure per-workload IOPS and latency, identify the competing application, then model a targeted QoS policy and verify both workloads after applying it. Do not use QoS as a substitute for diagnosing a bad host path.
Verify

Show baseline, policy objective and post-change latency/throughput evidence.

9. What are deduplication, compression and FlexClone?

  1. Deduplication and compression reduce physical storage use by sharing repeated data or encoding it more compactly. FlexClone creates a writable point-in-time copy that initially shares blocks with its parent; differences consume space as writes occur. Reported savings depend on which efficiency and snapshot effects the metric includes.
  2. Example: a test environment can clone a production volume for a safe restore drill. The clone is not a separate full backup; deleting or changing recovery objects still requires retention planning. Validate physical aggregate capacity and clone lifecycle before creating many copies.
Verify

Distinguish logical provisioned size from actual used blocks and the clone’s future growth.

10. How do you validate an ONTAP upgrade?

  1. Before an upgrade, check hardware and host compatibility, supported version path, cluster and HA health, active jobs, replication, capacity, firmware requirements and current alerts. Run NetApp automated prechecks and address blocking warnings. Schedule a maintenance window and application owner monitoring.
  2. After the upgrade, compare node versions, HA takeover readiness, LIF status, SMB/NFS/SAN client I/O, volume capacity and SnapMirror lag to the baseline. A successful upgrade job does not guarantee a specific application or host stack is healthy.
Read-only postchecks 
cluster show
storage failover show
network interface show
snapmirror show
system health alert show
Verify

A named client and application transaction still work, and protection resumes.

NetApp references