After initializing a Data ONTAP 10.0 system, perform the following steps to create a basic configuration. This example is for a mixed NFS and CIFS environment.
For each node in the cluster, perform the following steps:
Connect to the node by using Secure Shell (SSH) or Telnet.
The node displays a login prompt. Enter the user name admin and an empty password (that is, press Enter at the Password: prompt). The system then displays the following prompt:
node::>
where node is the hostname of the node.
Enter the storage aggregate show command to verify that the initialization process created a storage aggregate:
node::>storage aggregate showAggregate Size(MB) Used% State #Vols Node(s) RAID Type RAID Status aggr0 116262 97% online 1 node1a raid_dp raid_dp
The CLI provides tab completion for commands and parameters. For example, instead of typing storage aggregate show in full, you can type stoTab aggTab shTab. See “Using the Command Line Interface” for details.
Enter the storage aggregate rename to rename the storage aggregate and prevent naming conflicts with aggregates on other nodes in the cluster. An aggregate name can consist of alphanumeric characters and the underscore (_). After you rename the aggregate, run the storage aggregate show again to verify the name change.
node::>storage aggregate rename -aggregate aggr0 -newname node1a_aggr0node::>storage aggregate showAggregate Size(MB) Used% State #Vols Node RAID Type RAID Status node1a_aggr0 116262 97% online 1 node1a raid_dp raid_dp
Set the date, time, and time zone on the node by using the node date modify command. The following example sets the time zone to Etc/UTC, the date to 1 May 2006, and the time to 9:00 a.m. on a node named node1a:
node::>node date modify -node node1a -timezone Etc/UTC -date "05/01/2006 09:00:00"
For more information on the node date commands, see “Setting the Time Zone, Date, and Time on a Node”.
Add the node to a cluster by one of the following methods:
If this is the first node in the cluster, enter the cluster create command to create a cluster. The following example creates a cluster. The cluster's base license is ABCDEFG (obtain licenses for your cluster from your sales or support representative). The cluster is named clus1. The node's first cluster port has the IP address 10.29.127.0, the netmask 255.255.240.0, and the MTU setting 9000. The node's second cluster port has the IP address 10.29.127.1 and the same netmask and MTU setting as the first cluster port.
node::>cluster create -license ABCDEFG -clustername clus1 -ipaddr1 10.29.127.0 -netmask1 255.255.240.0 -mtu1 9000 -ipaddr2 10.29.127.1 -netmask2 255.255.240.0 -mtu2 9000
If this is not the first node in the cluster, enter the cluster join command to join the node to the existing cluster. In the following example, another node is added to the cluster created in the previous example; the -clusteripaddr parameter specifies the IP address of a cluster port of a node that is already a member of the cluster; in this case, the IP address 10.29.127.0 is used. The IP address of node1b's first cluster port is 10.29.127.2, its netmask is 255.255.240.0, and its MTU setting is 9000. The IP address of its second cluster port is 10.29.127.3, with the same netmask and MTU setting as the first cluster port.
node::>cluster join -clusteripaddr 10.29.127.0 -ipaddr1 10.29.127.2 -netmask1 255.255.240.0 -mtu1 9000 -ipaddr2 10.29.127.3 -netmask2 255.255.240.0 -mtu2 9000
After you have added the nodes into the cluster, enter the cluster show command to verify the cluster information.
node::>cluster showNode Health Eligibility node1a true true
For additional information about these commands, see “Clustering Tasks”.
Enter the storage disk show command to determine which disks are connected to the node.
node::>storage disk showDisk UsedSize(MB) Shelf Bay State RAID Type Aggr Owner node1a:0c.16 135979 1 0 present parity node1a_ node1a aggr0 node1a:0c.17 135979 1 1 present data node1a_ node1a aggr0 node1a:0c.18 - - - - - - - node1a:0c.19 - - - - - - - node1a:0c.20 - - - - - - - node1a:0c.21 - - - - - - - node1a:0c.22 135979 1 6 present data node1a_ node1a aggr0 node1a:0c.23 - - - - - - - node1a:0c.24 - - - - - - - node1a:0c.25 - - - - - - - node1a:0c.26 - - - - - - - node1a:0c.27 - - - - - - - node1a:0c.28 135979 1 12 spare pending - node1a node1a:0c.29 - - - - - - -
Typically, only three to four disks are owned and in a valid state; all other disks are unowned.
Assign ownership of unowned disks by using the storage disk modify command. The following example assigns a disk named node1a:0c.18 to a node named node1a:
node::>storage disk modify -disk node1a:0c18 -owner node1a
NetApp recommends that you assign half of the unassigned disks to one node in the failover pair and the other half to the other node in the failover pair. Assigning all disks to a single node prevents its failover partner from accessing the disks, rendering the failover partner unusable.
See “Disk Naming Conventions” for information about disk names and “Modifying a Disk's Attributes” for information about the storage disk modify command.
For each set of failover pairs, perform the following steps:
On one node of each failover pair in the cluster, enable storage failover by using the storage failover modify command:
node::>storage failover modify -nodenode_name-enabled true
Enabling storage failover on one node of a failover pair automatically enables it on the partner node. See “Managing Storage Failover” for more information about storage failover.
If your cluster consists of only two nodes, enable two-node high availability (HA) by using the cluster ha modify command:
node::>cluster ha modify -configured true
See “Configuring High Availability in a Two-Node Cluster” for more information on HA mode.
The following steps can be performed from any single node in the cluster.
Enter any licenses for optional software packages, such as asynchronous mirroring:
node::>license add -license-codelicense_code
where license_code is the code for the purchased license. Run the command for each license you have purchased.
If desired, enable and configure NTP for the cluster by using the system service ntp commands. The following examples enable NTP in the cluster and configure the node named node1a to use an NTP server named ntp1.example.com that uses NTP version 4:
node::>system service ntp config modify -enabled truenode::>system service ntp server new -node node1a -server ntp1.example.com -version 4
Create a storage aggregate to hold the virtual server's root volume by using the storage aggregate new command. Storage aggregate names must be unique in the cluster, and the number of disks specified cannot exceed the number of available spare disks. The following example creates a storage aggregate named node1a_aggr1 with 10 disks:
node::>storage aggregate new -aggregate node1a_aggr1 -diskcount 10
See “Aggregate Concepts” and “Creating a Storage Aggregate” for more information about storage aggregates.
Create a virtual server by performing the following steps:
Use the vserver new command to create a virtual server. The command creates both a virtual server and its root volume. The following example creates a virtual server named vs1 with a root volume named vs1_root on the storage aggregate named node1a_aggr1. NIS is used for network name inquiries, the default language used on the volume is U.S. English, and the volumes' security style is mixed (that is, it supports both NFS and CIFS).
node::>vserver new -vserver vs1 -rootvolume vs1_root -aggregate node1a_aggr1 -ns-switch nis -language en_US -rootvolume-security-style mixed
Verify the creation of the virtual server and its root volume by using the vserver show and volume show commands, respectively:
node::>vserver showVirtual Root Aggregate Name NIS Server Name Volume Name Service Domain vs1 vs1_root node1a_ nis - aggr1node::>volume showVserver Volume Aggregate State Type Size Available Used% ------------------------------------------------------------------------------ vs1 vs1_root node1a_aggr1 online RW 100GB 79.99GB 0%
See Chapter 7, Managing Virtual Servers for more information about virtual servers.
Create an NIS-domain configuration by using the services nis-domain new command. The following example configures a virtual server named vs1 to use an NIS domain named example.com, which uses NIS servers that have the IP addresses 10.98.10.0 and 10.98.12.0.
node::>services nis-domain new -vserver vs1 -domain example.com -active true -servers 10.98.l0.0,10.98.12.0
Verify the NIS-domain configuration by using the services nis-domain show command:
node::>services nis-domain showVirtual Domain Active NIS Server Name Server vs1 example.com true 10.98.10.0,10.98.12.0
See “Configuring an NIS-Domain Configuration” for more information about NIS-domain configurations.
Verify that the virtual server is using the specified NIS domain by using the vserver show command:
node::>vserver showVirtual Root Aggregate Name NIS Server Name Volume Name Service Domain vs1 vs1_root node1a_ nis example.com aggr1
Create a volume to hold user data by performing the following steps:
Use the volume new command to create a volume. The following example creates a volume named src on a virtual server named vs1 and a storage aggregate named node1a_aggr1. The volume is placed online upon its creation. The volume's export rule set is default_ruleset. The volume is owned by a user named admin and a group named eng. The security style on the volume is mixed (that is, it supports both NFS and CIFS). The volume's junction path is named src and is active upon the volume's creation. The volume's parent volume is vs1_root. The volume is 100 GB in size and uses the volume style of space guarantee. The volume uses the default snapshot policy.
node::>volume new -vserver vs1 -volume src -state online -ruleset default_ruleset -user admin -group eng -security-style mixed -junction-path src -junction-active true -parent vs1_root -size 100g -space-guarantee volume -snapshot-policy default
See Chapter 11, Managing Volumes for more information about volumes.
Verify the creation of the volume by using the volume show command. The first example displays information about all volumes; the second example displays detailed information about the new volume.
node::>volume showFiles MaxDir Vserver Volume State Type Size Used% Total Used Size ------------------------------------------------------------------------------ vs1 vs1_root online RW 20MB 0% 629 97 10MB vs1 src online RW 100GB 0% 629 97 10MBnode::>volume show -vserver vs1 -volume srcVserver Name: vs1 Volume Name: src Volume Data Set ID: 1129 Volume Master Data Set ID: 2147484777 Aggregate Name: node1a_aggr1 Volume State: online Volume Type: RW Export Rule Set Name: default_ruleset User ID: admin Group ID: eng Security Style: mixed Mount Path: src Is Mount Path Active: true Parent Volume: vs1_root Comment: Volume Size: 100GB Available Size: 100GB Used Size: 0 Total Files: 629 Files Used: 97 Space Guarantee Style: volume Space Guarantee in Effect: true Snapshot Policy: default Creation Time: Fri Aug 26 18:26:13 2005 Force option: -
Create a virtual interface for data and assign it to the virtual server by performing the following steps. For more information about network ports and virtual interfaces, see Chapter 4, Managing Network Ports and Chapter 8, Managing Virtual Interfaces.
Verify that all network ports are properly configured by using the network port show command:
node::>network port showNode Port Role Link MTU Auto Neg Duplex Speed (Mbps) Aggregate Admin/Oper Admin/Oper Admin/Oper Owner ------------------------------------------------------------------------------- node1a e0a cluster up 9000 true/true full/full 1000/1000 - e0b cluster up 9000 true/true full/full 1000/1000 - e0c data up 9000 true/true full/full 1000/1000 - e0d data up 9000 true/true full/full 1000/1000 - e1a mgmt up 1500 true/true full/full 100/100 - e1b mgmt down 1500 true/true half/half 10/10 - node1b e0a cluster up 9000 true/true full/full 1000/1000 - e0b cluster up 9000 true/true full/full 1000/1000 - e0c data up 9000 true/true full/full 1000/1000 - e0d data up 9000 true/true full/full 1000/1000 - e1a mgmt up 1500 true/true full/full 100/100 - e1b mgmt down 1500 true/true half/half 10/10 -
Create a virtual interface by using the network interface new command. The following example creates a virtual interface named vs1_data1 on a virtual server named vs1. The virtual interface serves data (that is, client access). The virtual interface's home node is named node1a and its home port is e0c. Its initial administrative status is set to up (that is, online). Its failover policy is nextavail (that is, if the physical port fails, the virtual interface migrates to the next available port). The virtual interface has the IP address 10.29.204.99 and the netmask 255.255.240.0.
node::>network interface new -server vs1 -vif vs1_data1 -role data -home-node node1a -home-port e0c -status-admin up -failover-policy nextavail -address 10.29.204.99 -netmask 255.255.240.0
Specify a static route for the virtual interface by using the network interface route new command. The following example creates a static route for the virtual interface named vs1_data1 on a virtual server named vs1. The static route uses the destination IP address 0.0.0.0, the netmask 0.0.0.0, and the gateway IP address 10.29.192.1.
node::>network interface route new -server vs1 -vif vs1_data1 -destination 0.0.0.0 -netmask 0.0.0.0 -gateway 10.29.192.1
Verify the creation of the virtual interface and static route by using the network interface show and network interface route show commands, respectively.
node::>network interface showServer Virtual Status Network Netmask Is Interface Admin/Oper Address Home node1a clus1 up/up 10.29.127.0 255.255.240.0 true clus2 up/up 10.29.127.1 255.255.240.0 true mgmt1 up/up 10.29.205.131 255.255.240.0 true node1b clus1 up/up 10.29.127.2 255.255.240.0 true clus2 up/up 10.29.127.3 255.255.240.0 true mgmt1 up/up 10.29.205.129 255.255.240.0 true vs1 vs1_data1 up/up 10.29.204.99 255.255.240.0 truenode::>network interface route showServer Virtual Destination Netmask Gateway Interface node1a mgmt1 0.0.0.0 0.0.0.0 10.29.192.1 node1b mgmt1 0.0.0.0 0.0.0.0 10.29.192.1 vs1 vs1_data1 0.0.0.0 0.0.0.0 10.29.192.1
Perform the following steps to add an export rule to the default export rule set. For more information on export rules, see “Export-Rule Concepts”, “Creating an Export-Rule Set”, and “Adding a Rule to an Export Rule Set”.
Use the protocol exportruleset rule new command to create an export rule to enable client access to data on the virtual server. The following example provides full access to all clients.
node::>protocol exportruleset rule new -vserver vs1 -rulesetname default_ruleset -clientmatch 0.0.0.0/0 -rorule any -rwrule any -superuser any -anon 0 -allow-suid true -allow-dev true
Verify the creation of the export rule by using the protocol exportruleset rule show command:
node::>protocol exportruleset rule showVirtual Ruleset Rule Access Client RO RW Server Name Index Protocol Match Rule Rule vs1 default_ruleset 1 any 0.0.0.0/0 any any
Enable and configure NFS service on the virtual server by performing the following steps. See “NFS Concepts” and “Creating an NFS Configuration” for more information about NFS.
Use the protocol nfs server new command to enable and configure NFS. The following example enables NFS access on a virtual server named vs1. NFS v2 and NFS v3 are both enabled over both TCP and UDP. SpinAuth authentication is enabled. A default Windows user named Administrator is specified.
node::>protocol nfs server new -vserver vs1 -access true -v2-enable true -v3-enable true -udp-enable true -tcp-enable true -spinauth true -default-win-user Administrator
Verify NFS service by using the protocol nfs server show command:
node::>protocol nfs server showVirtual Default General Enable Enable Enable Enable Server Windows User Access v2 v3 udp tcp vs1 Administrator true true true true true
Enable and configure CIFS service by performing the following steps. For more information about CIFS, see “CIFS Concepts”, “Creating a CIFS Domain”, “Enabling CIFS on a Virtual Server”, “Creating a CIFS Share”, and “Creating a CIFS Superuser”.
Create a valid computer account for the node on your CIFS domain controller. Include all trusted domains in the visible domain list. Use one of the following methods to create the node's account:
For Windows NT 4.0, use Server Manager.
For Windows 2000 and later, open the Active Directory, select New > Computer, add the node, and select the "Allow pre-Windows 2000 computers to use this account" check box.
Create a CIFS domain by using the protocol cifs domain new command. The following example creates a CIFS domain with the NetBIOS name ENG that uses a WINS server with the IP address 10.98.16.1, a primary domain controller named RALPH with the IP address 10.98.16.1, and a backup domain controller named NELSON with the IP address 10.98.17.1.
node::>protocol cifs domain new -domain ENG -wins-ip 10.98.16.1 -dc1-name RALPH -dc1-ip 10.98.16.1 -dc2-name NELSON -dc2-ip 10.98.17.1
Verify the creation of the CIFS domain by using the protocol cifs domain show command:
node::>protocol cifs domain showDomain WINS Kerberos DC DC Name IP Config Name Name IP ENG 10.98.16.1 - RALPH 10.98.16.1
Enable and configure CIFS on the virtual server by using the protocol cifs server new command. The following example enables CIFS on a virtual server named vs1. The virtual server uses the NetBIOS name DATA to identify itself to CIFS clients and uses the CIFS domain named ENG. The virtual server uses the peer NetBIOS type to register its NetBIOS name, does not use Kerberos authentication for CIFS clients, and automatically updates its Machine Trust Account password.
node::>protocol cifs server new -vserver vs1 -cifs-server DATA -domain ENG -broadcast disable -kerberos disable -password-update enable
Verify the creation of the CIFS server by using the protocol cifs server show command:
node::>protocol cifs server showVirtual Domain NetBIOS Kerberos Broadcast Server Name Name Enabled Enabled vs1 ENG DATA disable disable
Define a CIFS share that exports the root volume of the virtual server by using the protocol cifs share new command. The following example creates a share named share1 on a virtual server named vs1. The share's path is the root of the virtual server. The share is browsable.
node::>protocol cifs share new -vserver vs1 -share-name share1 -path / -share-properties browsable
Verify the creation of the CIFS share by using the protocol cifs share show command:
node::>protocol cifs share showVirtual Share Path Share Description Server Name Properties vs0 ADMIN$ / browsable - vs0 IPC$ / browsable - vs0 ROOTSHARE / browsable - vs1 ADMIN$ / browsable - vs1 IPC$ / browsable - vs1 ROOTSHARE / oplocks - browsable
Define the Administrator account as a superuser for the CIFS configuration by using the protocol cifs superuser new:
node::>protocol cifs superuser new -vserver vs1 -domain ENG -accountname Administrator
Verify the creation of the superuser account by using the protocol cifs superuser show command:
node::>protocol cifs superuser showVirtual CIFS Domain Account Server Server Name Name vs1 DATA ENG Administrator
Create name mappings between UNIX and Windows accounts by performing the following steps. See “Name-Mapping Concepts” and “Creating a Name Mapping” for more information about these steps.
Use the protocol name-mapping new command to create a name mapping that maps Windows account names in the form ENG\user_name to the UNIX form user_name:
node::>protocol name-mapping new -vserver vs1 -type user -direction win-unix -position 1 -pattern ENG\\(.+) -replacement \1
Create a name mapping that maps UNIX account names in the form user_name to the Windows form ENG\user_name:
node::>protocol name-mapping new -vserver vs1 -type user -direction unix-win -position 1 -pattern (.+) -replacement ENG\\\1
Verify the creation of the name mappings by using the protocol name-mapping show command:
node::>protocol name-mapping showVirtual Mapping Direction Pos Pattern Replacement Server Type vs1 user win-unix 1 ENG\\(.+) \1 user unix-win 1 (.+) ENG\\\1