Preface

About this guide

This guide describes the following tasks and topics:

Preparing to install an F800 series cluster, a FAS900 series cluster, or a FAS3000 series cluster
Cabling a pair of F800 series, FAS900 series, or FAS3000 series storage systems into a cluster
Configuring the cluster when you first bring it up
Managing a cluster
Performing disaster recovery with a MetroCluster
Performing basic troubleshooting on the cluster

This guide does not cover basic system or network administration, which is covered in the Data ONTAP^TM System Administration Guide and Data ONTAP Network Administration Guide.

Audience

This guide is for qualified system administrators and service personnel who are familiar with Network Appliance^TM equipment and who need to

Cable and configure two systems into a standard or mirrored cluster system
Convert stand-alone systems into a standard or mirrored cluster system
Cable and configure a new Fabric Attached or nonswitched MetroCluster
Manage a cluster
Troubleshoot a cluster

Terminology

This guide uses the following terms:

General storage system terminology

NetApp® storage products (filers, FAS storage systems, and NearStore® systems) are all storage systems. All these products are also sometimes called filers, appliances, or storage appliances.
FAS3000 series refers to NetApp storage systems belonging to that series, such as the FAS3050c.
FAS900 series refers to NetApp storage systems belonging to that series, such as the FAS980c.
F800 series refers to NetApp storage systems belonging to that series, such as the F880c.
Storage system refers to any F800 series, FAS900 series, or FAS3000 series storage system.

Cluster terminology

Cluster refers to two storage systems, or nodes, that are connected in such a way that they can take over operations for each other, either automatically or with operator intervention. One node is called the local node, and the other node is called the partner node.
Active-active controller configuration is another term for cluster.
Standard cluster, or failover cluster, refers to a cluster configured so that one node automatically takes over for its partner when the partner node becomes impaired.
Mirrored cluster refers to a standard cluster where there are two copies, or plexes, of the data.
Fabric Attached MetroCluster refers to a cluster running syncmirror_local and cluster_remote, where the nodes (storage systems) and disk shelves are attached to two pairs of Fibre Channel switches, where the nodes and disk shelves are not cabled together in standard cluster fashion, and where the nodes are separated by more than 500 meters and up to 30 kilometers.
Nonswitched MetroCluster refers to a cluster running syncmirror_local and cluster_remote, where the nodes are separated without a switch by up to 500 meters.
Switched cluster refers to a cluster that has the local and partner nodes connected to each other, then connected to clients through a switch.
Cluster failover, also referred to as controller failover or CFO, refers to the technology that enables two storage systems to take over each other's data, thus improving data availability.
Node refers to either the local or partner storage system.
Local cluster node refers to the cluster node in the same area as the system administrator.
Remote cluster node refers to the cluster node that is not in the same area as the system administrator.
Remote storage refers to the storage that is accessible to the local cluster node, but is located at the location of the remote cluster node.

Clustering and storage hardware terminology

FC HBA for Disk or FC HBA refers to the Fibre Channel host bus adapter that connects the node to the switch or to the disks.
Disk shelf refers to a Fibre Channel StorageShelf FC7, Fibre Channel StorageShelf FC8, Fibre Channel StorageShelf FC9, DiskShelf14, DS14mk2 FC, or DS14mk2 AT.
The LRC (Loop Resiliency Circuit) module keeps the FC-AL loop intact during the addition and removal of disks within a disk shelf. It also contains the enclosure services processor, which communicates the environmental data of the disk shelf. The LRC reconditions the signal so that there is no accumulated error in the data signals. As the signal moves, waveforms might distort slightly. Over many hops these distortions would otherwise accumulate, causing high error rates. The LRC prevents this.
The ESH (Embedded Switching Hub) module provides a means of managing an FC-AL loop in an intelligent manner, such that a single drive failure does not take down the loop. It also contains the enclosure services processor, which communicates the environmental data of the disk shelf.
The ESH2 module is a second-generation ESH module.
AT-FCX is an enhanced FC-AL to Serial ATA bridge used in the DS14Mk2 AT disk shelf.

Command conventions

You can enter commands on the system console or from any client that can obtain access to the storage system using a Telnet session. In examples that illustrate commands executed on a UNIX® workstation, the command syntax and output might differ, depending on your version of UNIX.

Keyboard conventions

This guide uses capitalization and some abbreviations to refer to the keys on the keyboard. The keys on your keyboard might not be labeled exactly as they are in this guide.

What is in this guide...	What it means...
hyphen (-)	Used to separate individual keys. Example Ctrl-D means holding down the Ctrl key while pressing the D key.
Enter	Used to refer to the key that generates a carriage return, although the key is named Return on some keyboards.
type	Used to mean pressing one or more keys on the keyboard.
enter	Used to mean pressing one or more keys and then pressing the Enter key.

Formatting conventions

The following table lists the kinds of formatting this guide uses to identify special information.

Formatting convention	Type of information
Italic type	Words or characters that require special attention. Placeholders for information you must supply. For example, if the guide requires you to enter the `arp -d` hostname command, you enter the characters "arp -d" followed by the actual name of the host. Book titles in cross-references.
`Monospaced font`	Command and daemon names. Information displayed on the system console or other computer monitors. Contents of files.
`Bold monospaced` font	Words or characters you type. What you type is always shown in lowercase letters, unless you must type it in uppercase letters for it to work properly.

Special messages

This guide contains special messages that are described as follows:

Note
A note contains important information that helps you install or operate the system efficiently.

Attention
An attention contains instructions that you must follow to avoid damage to the equipment, a system crash, or loss of data.

WARNING
A warning contains instructions that you must follow to avoid personal injury.