First Day on a New ONTAP Cluster

1. Initial setup

cluster setup creates a cluster or joins a node. Establish the cluster-management LIF and admin password, then configure DNS and NTP so certificates, Kerberos, peers, and logs have reliable names and time. See time services.

cluster::> cluster setup
Welcome to the cluster setup wizard.
Do you want to create a new cluster or join? create
Cluster name: prod-cluster
Cluster management interface port: e0M
IP address: 192.0.2.20
Netmask: 255.255.255.0
Default gateway: 192.0.2.1

Prompts vary by release and platform. Keep console access until both HA partners and management access are proven.

2. Check the fresh cluster

Prove membership, node health, subsystem health, and storage before tenant objects.

prod-cluster::> cluster show
Node   Health Eligibility
node1  true   true
node2  true   true
prod-cluster::> system node show
Node   Health Eligibility Uptime
node1  true   true        00:42:18
node2  true   true        00:38:51

prod-cluster::> system health status show
Status
---------------
ok
prod-cluster::> storage aggregate show
Aggregate   Size  Available Used% State  Nodes
aggr0_node1 894GB 42.1GB    95%   online node1
aggr0_node2 894GB 42.0GB    95%   online node2

Root aggregates may be the only storage shown; factory/provisioning state varies. Stop on unhealthy, ineligible, or offline results.

3. Create a data aggregate if none exists

An aggregate is the capacity pool for FlexVol volumes. The reference accepts a disk count or disk list; automatic selection must fit the RAID/spare plan.

prod-cluster::> storage aggregate create -aggregate aggr1 -node node1 -diskcount 12
Info: aggregate "aggr1" will use 12 disks.
Do you want to continue? {y|n}: y
Aggregate creation succeeded.
prod-cluster::> storage aggregate show -aggregate aggr1
Aggregate Size  Available Used% State
aggr1    21.8TB 21.8TB    0%    online

Do not assume 12 disks is right. Partitioned disks, RAID layout, ASA r2, and platform provisioning differ; check your version and hardware procedure.

4. Create the SVM

The SVM is the data-serving and security boundary. Enable only needed services.

prod-cluster::> vserver create -vserver svm1 -rootvolume root_vol \
  -aggregate aggr1 -rootvolume-security-style unix -data-services nfs,cifs
[Job 41] Job succeeded: Vserver creation completed.
prod-cluster::> vserver show -vserver svm1
Vserver Type  Subtype State
svm1    data  default running

Valid -data-services values/defaults are release-dependent. Configure NFS, or DNS/time/Active Directory and the CIFS server, before client access. See SVM architecture.

5. Create a data LIF

A LIF gives the SVM a protocol address and a home port. Confirm VLAN, broadcast domain, IPspace, MTU, route, and failover targets.

prod-cluster::> network interface create -vserver svm1 -lif data1 \
  -role data -data-protocol nfs,cifs -home-node node1 -home-port e0d \
  -address 192.0.2.10 -netmask 255.255.255.0
prod-cluster::> network interface show -vserver svm1
Vserver LIF   Status Address       Current Node Port
svm1    data1 up/up  192.0.2.10/24 node1        e0d

Current releases emphasize service policies; -role is retained in documented syntax but check your version. Continue with network architecture.

6. Create the first volume and junction

The junction mounts the volume in the NAS namespace.

prod-cluster::> volume create -vserver svm1 -volume data -aggregate aggr1 \
  -size 10GB -junction-path /data
[Job 48] Job succeeded: Successful
prod-cluster::> volume show -vserver svm1 -volume data -fields state,junction-path
vserver volume state  junction-path
svm1    data   online /data

7. Authorize exports or publish a share

Export rules authorize NFS clients. SMB shares publish a namespace path but do not replace CIFS server setup and filesystem/share permissions.

prod-cluster::> vserver export-policy rule create -vserver svm1 \
  -policyname default -ruleindex 1 -protocol nfs3 \
  -clientmatch 192.0.2.0/24 -rorule sys -rwrule sys -superuser none
prod-cluster::> vserver export-policy rule show -vserver svm1
Policy  Rule Protocol Client Match  RO  RW
default 1    nfs3     192.0.2.0/24 sys sys

prod-cluster::> vserver cifs share create -vserver svm1 -share-name data -path /data
prod-cluster::> vserver cifs share show -vserver svm1 -share-name data
Vserver Share Path
svm1    data  /data

8. Choose snapshot retention

Inspect built-ins first; schedules and defaults vary. Snapshots are local recovery points, not independent backup.

prod-cluster::> snapshot policy show
Vserver Policy  Enabled Schedules
Cluster default true    hourly,daily,weekly
Cluster none    true    -
prod-cluster::> snapshot policy create -vserver svm1 -policy app-basic \
  -enabled true -schedule1 daily -count1 7 -prefix1 daily
prod-cluster::> volume modify -vserver svm1 -volume data -snapshot-policy app-basic

9. Record a monitoring baseline

Confirm AutoSupport configuration and EMS destinations. Follow AutoSupport and EMS and the production checklist.

prod-cluster::> system node autosupport show -fields state,transport,to
node  state  transport to
node1 enable https     storage-ops@example.com
node2 enable https     storage-ops@example.com
prod-cluster::> event notification destination show
Name       Type   Destination
syslog-ops syslog 192.0.2.60

10. Make the first off-cluster copy

Prepare the destination, peer the clusters/SVMs, then create and initialize a supported SnapMirror relationship. Policy, sizing, and endpoints depend on topology and version; use the DR runbook.

dr-cluster::> volume create -vserver svm_dr -volume data_dr \
  -aggregate aggr_dr -type DP -size 10GB
prod-cluster::> snapmirror create -source-path svm1:data \
  -destination-path svm_dr:data_dr -type XDP -policy MirrorAndVault
prod-cluster::> snapmirror initialize -destination-path svm_dr:data_dr
Operation is queued.

Before handover

Official sources

• cluster setup
• aggregate create and vserver create
• network interface create and volume create
• export rule and CIFS share
• Create a replication relationship