Search within this manual
Search all Support content
Data ONTAP® 8.2
High Availability and MetroCluster Configuration Guide
For 7-Mode
Updated for 8.2.1
Understanding HA pairs
What an HA pair is
How HA pairs support nondisruptive operations and fault tolerance
How the HA pair improves fault tolerance
Single point of failure definition
Single point of failure analysis for HA pairs
Standard HA pair diagram
Comparison of HA pair types
Understanding mirrored HA pairs
Advantages of mirrored HA pairs
Asymmetrically mirrored HA pairs
Understanding takeover and giveback
When takeovers occur
Failover event cause-and-effect table
How hardware-assisted takeover speeds up takeover
What happens during takeover
What happens during giveback
Background disk firmware update and takeover and giveback
Planning your HA pair configuration
Best practices for HA pairs
Setup requirements and restrictions for standard HA pairs
Setup requirements and restrictions for mirrored HA pairs
Multipath HA requirement
What multipath HA for HA pairs is
How the connection types are used
Advantages of multipath HA
Requirements for hardware-assisted takeover
Possible storage configurations in the HA pairs
HA pairs and storage system model types
Single-chassis and dual-chassis HA pairs
Interconnect cabling for systems with variable HA configurations
HA configuration and the HA state PROM value
Table of storage system models and HA configuration differences
Installing and cabling an HA pair
System cabinet or equipment rack installation
HA pairs in an equipment rack
HA pairs in a system cabinet
Required documentation
Required tools
Required equipment
Preparing your equipment
Installing the nodes in equipment racks
Installing the nodes in a system cabinet
Cabling a standard HA pair
Determining which Fibre Channel ports to use for Fibre Channel disk shelf connections
Cabling Node A to DS14mk2 or DS14mk4 disk shelves
Cabling Node B to DS14mk2 or DS14mk4 disk shelves
Cabling the HA interconnect (all systems except 32xx)
Cabling the HA interconnect (32xx systems in separate chassis)
Cabling a mirrored HA pair
Determining which Fibre Channel ports to use for Fibre Channel disk shelf connections
Creating your port list for mirrored HA pairs
Cabling the Channel A DS14mk2 or DS14mk4 disk shelf loops
Cabling the Channel B DS14mk2 or DS14mk4 disk shelf loops
Cabling the redundant multipath HA connection for each loop
Cabling the HA interconnect (all systems except 32xx)
Cabling the HA interconnect (32xx systems in separate chassis)
Required connections for using uninterruptible power supplies with standard or mirrored HA pairs
Configuring an HA pair
Bringing up the HA pair
Considerations for HA pair setup
Configuring shared interfaces during setup
Configuring dedicated interfaces with setup
Configuring standby interfaces with setup
Verifying and setting the HA state on the controller modules and chassis
Enabling HA mode capability and controller failover
Setting options and parameters
Option types for HA pairs
Setting matching node options
Parameters that must be the same on each node
Best practices for cf options
Disabling the change_fsid option in MetroCluster configurations
When data loss can occur if the change_fsid option is enabled
Configuring hardware-assisted takeover
How hardware-assisted takeover speeds up takeover
Disabling and enabling the hardware-assisted takeover option
Setting the partner address for hardware-assisted takeover
Setting the partner port for hardware-assisted takeover
Configuring network interfaces for HA pairs
Understanding interfaces in an HA pair
What the networking interfaces do
Shared, dedicated, and standby interfaces
Interface roles in normal and takeover modes
Takeover configuration with shared interfaces
Takeover configuration with dedicated and standby interfaces
Interface types and configurations
IPv6 considerations in an HA pair
Making nondisruptive changes to the interface groups
Configuring network interfaces for the HA pair
Configuring a dedicated interface in an HA pair
Configuring a partner interface in an HA pair
Configuring partner addresses on different subnets (MetroCluster configurations only)
The /etc/mcrc file
Creating an /etc/mcrc file
Setting the system to use the partner's /etc/mcrc file at takeover
Verifying the HA pair configuration
Testing takeover and giveback
Managing takeover and giveback
Monitoring an HA pair in normal mode
Monitoring HA pair status
Description of HA pair status messages
Monitoring the hardware-assisted takeover feature
Checking the hardware-assisted takeover status of the local and partner node
Testing the hardware-assisted takeover configuration
Checking hardware-assisted takeover statistics
Displaying the partner's name
Displaying disk and array LUN information on an HA pair
What takeover and giveback are
When takeovers occur
What happens during takeover
What happens after takeover
What happens during giveback
Configuring automatic takeover
Reasons that automatic takeover occurs
Commands for performing a manual takeover
Halting a node without takeover
Rebooting a node without takeover
Enabling and disabling takeover
Enabling and disabling takeover on reboot
Enabling and disabling automatic takeover of a panicked partner
Specifying the time period before takeover
Enabling or disabling negotiated failover for a network interface
Takeover of vFiler units and the vFiler limit
Managing an HA pair in takeover mode
Determining why takeover occurred
Statistics in takeover mode
Managing emulated nodes
Management exceptions for emulated nodes
Accessing the emulated node from the takeover node
Using the partner command without arguments
Using the partner command with arguments
Using Remote Shell
Emulated node command exceptions
Performing dump and restore operations for a failed node
Giveback operations
About manual giveback
Option for shortening giveback time
Removing failed disks prior to attempting giveback
Initiating normal giveback
Forcing giveback
If giveback is interrupted
Configuring giveback
Configuring automatic giveback
Adjusting the giveback delay time for automatic giveback
Setting giveback delay time for CIFS clients
Terminating long-running processes to speed automatic giveback
Setting giveback to terminate long-running processes and override vetos
Enabling and disabling automatic giveback after takeover due to partner panicking
Reconfiguring an HA pair into two stand-alone systems
Ensuring uniform disk ownership within disk shelves and loops in the system
Configuring a node for non-HA (stand-alone) use
Reconfiguring nodes using disk shelves for stand-alone operation
Requirements when changing a node using array LUNs to stand-alone
Reconfiguring nodes using array LUNs for stand-alone operation
Managing DS14mk2 or DS14mk4 disk shelves in an HA pair
Adding DS14mk2 or DS14mk4 disk shelves to a multipath HA loop
Upgrading or replacing modules in an HA pair
About the disk shelf modules
Restrictions for changing module types
Best practices for changing module types
Testing the modules
Determining path status for your HA pair
Hot-swapping a module
Nondisruptive operations with HA pairs
Where to find procedures for nondisruptive operations with HA pairs
Hot-removing disk shelves in systems running Data ONTAP 8.2.1 or later
Overview of a MetroCluster configuration
Types of MetroCluster configurations
How mirroring works in a MetroCluster configuration
How Data ONTAP works with MetroCluster configurations
Advantages of stretch MetroCluster configurations
Configuration variations for stretch MetroCluster configurations
Advantages of fabric-attached MetroCluster configurations
Configuration variations for fabric-attached MetroCluster configurations
Implementing stretch MetroCluster configurations with disks
Planning a stretch MetroCluster configuration using disks
Setup requirements and restrictions for stretch MetroCluster configurations with disks
Hardware requirements for using disks in a MetroCluster configuration with V-Series systems
Required connections for using uninterruptible power supplies with MetroCluster configurations
Stretch MetroCluster configuration with disks
Required documentation, tools, and equipment
Required documentation for MetroCluster configurations
Required tools
Required equipment for MetroCluster configurations
Cabling a stretch MetroCluster configuration with disks
Stretch MetroCluster configuration on single-enclosure HA pairs
Assigning disk pools in a stretch MetroCluster configuration
Managing the default configuration speed of a stretch MetroCluster configuration
Changing the default configuration speed of a stretch MetroCluster configuration
Resetting a stretch MetroCluster configuration to the default speed
Implementing fabric-attached MetroCluster configurations with disks
Planning a fabric-attached MetroCluster configuration using disks
Setup requirements and restrictions for fabric-attached MetroCluster configurations with disks
Configuration limitations for fabric-attached MetroCluster configurations with disks
Hardware requirements for using disks in a MetroCluster configuration with V-Series systems
Required connections for using uninterruptible power supplies with MetroCluster configurations
Requirements for a shared-switches configuration
Fabric-attached MetroCluster configuration with disks
How fabric-attached MetroCluster configurations use Brocade and Cisco Fibre Channel switches
Planning the fabric-attached MetroCluster installation
Required documentation, tools, and equipment
Required documentation for MetroCluster configurations
Required tools
Required equipment for MetroCluster configurations
Converting an HA pair to a fabric-attached MetroCluster configuration
Cabling a fabric-attached MetroCluster configuration with disks
Configuring the switches
Cabling Node A
Cabling the Node A controller to the switches
Cabling the Node A disk shelves to the switches
Cabling the Node A FC-VI adapter and inter-switch link
Cabling Node B
Cabling the Node B controller to the switches
Cabling the Node B disk shelves to the switches
Cabling the Node B FC-VI adapter and inter-switch link
Assigning disk pools
Verifying disk paths
Setting up a shared-switches configuration
Cabling the FC-VI adapter and ISL in a shared-switches configuration
Fabric-attached MetroCluster configuration on single-enclosure HA pairs
Configuration differences for fabric-attached MetroCluster configurations on single-enclosure HA pairs
Setting a preferred primary port in a MetroCluster configuration
Removing the preferred primary port in a fabric-attached MetroCluster configuration
Enabling unidirectional port mirroring in a MetroCluster configuration
What unidirectional port mirroring is
Verifying the MetroCluster configuration
Gathering the information for verifying the MetroCluster configuration
Testing takeover and giveback
Testing MetroCluster configuration during a switch failure
Testing a MetroCluster configuration during FibreBridge 6500N bridge failure
Testing a MetroCluster configuration during cluster failover on disaster
Implementing a stretch or fabric-attached MetroCluster configuration with array LUNs
Planning for a MetroCluster configuration with array LUNs
Types of systems that can use array LUNs in a MetroCluster configuration
Implementation overview for a MetroCluster configuration with array LUNs
Requirements for a MetroCluster configuration with array LUNs
Requirements for a shared-switches MetroCluster configuration with array LUNs
Recommended fabric-attached MetroCluster configuration with array LUNs
Recommended stretch MetroCluster configuration with array LUNs
Cabling guidelines for a MetroCluster configuration with array LUNs
Planning zoning for a MetroCluster configuration with array LUNs
Connecting devices in a MetroCluster configuration with array LUNs
Connecting the local systems in a MetroCluster configuration with array LUNs
Connecting the remote systems in a MetroCluster configuration with array LUNs
Connecting the switch fabric in a MetroCluster configuration with array LUNs
Connecting the fabric and storage array in a MetroCluster configuration with array LUNs
Configuring zoning in a MetroCluster configuration with array LUNs
Setting up a shared-switches configuration
Cabling the FC-VI adapter and ISL in a shared-switches MetroCluster configuration with array LUNs
Setting a preferred primary port in a MetroCluster configuration
Removing the preferred primary port in a fabric-attached MetroCluster configuration
Enabling unidirectional port mirroring in a MetroCluster configuration
Changing the configuration speed of a stretch MetroCluster configuration
Setting up Data ONTAP after connecting devices in a MetroCluster configuration with array LUNs
Testing a MetroCluster configuration with array LUNs
Testing zoning of FC-VI ports in a MetroCluster configuration with array LUNs
Verifying proper setup at MetroCluster sites with storage arrays
Simulating a disaster recovery in a MetroCluster configuration with array LUNs
Implementing a MetroCluster configuration with both disks and array LUNs
Types of systems that can use array LUNs in a MetroCluster configuration
Planning a MetroCluster configuration with disks and array LUNs
Guidelines for implementing a MetroCluster configuration with disks and array LUNs
Supported fabric-attached MetroCluster configuration with disks and array LUNs
Supported stretch MetroCluster configuration with disks and array LUNs
Testing a MetroCluster configuration with disks and array LUNs
Recovering from a disaster by using MetroCluster configurations
Conditions that constitute a disaster
Ways to determine whether a disaster occurred
Failures that do not require disaster recovery
Recovering from a disaster
Restricting access to the disaster site node
Powering off the node
Fencing off the node
Forcing a node into takeover mode
Remounting volumes of the failed node
Recovering LUNs of the failed node
Fixing failures caused by the disaster
Reestablishing the MetroCluster configuration
Rejoining the mirrored aggregates
Re-creating the mirrored aggregates
Feature update record
Copyright information
Trademark information
How to send your comments
