Dell CX4 User Manual

Download

Dell/EMC CX4-series

Fibre Channel Storage Arrays

With Microsoft

Server

Failover Clusters

Windows

Hardware Installation

and Troubleshooting

Guide

Notes, Cautions, and Warnings

NOTE: A NOTE indicates important information that helps you make better use

of your computer.

CAUTION: A CAUTION indicates either potential damage to hardware or

loss of data and tells you how to avoid the problem.

WARNING: A WARNING indicates a potential for property damage, personal

injury, or death.

___________________

Reproduction of these materials in any manner whatsoever without the written permission of Dell Inc. is strictly forbidden.

Trademarks used in this text: Dell, the DELL logo, PowerEdge, and PowerVault are trademarks of Dell Inc.; Active Directory , Microsoft, W indows, W indows Server , Windows XP and Windows NT are either trademarks or registered trademarks of Microsoft Corporation in the United States and/ or other countries.; EMC, Navisphere, and PowerPath are re gistered trademarks and MirrorView, SAN Copy, and SnapView are trademarks of EMC Corporation.

Other trademarks and trade names may be used in this document to refer to either the entities claiming the marks and names or their products. Dell Inc. disclaims any proprietary interest in trademarks and trade names other than its own.

October 2008 Rev A00

Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . 7

Cluster Solution . . . . . . . . . . . . . . . . . . . . . . 8

Cluster Hardware Requirements

Cluster Nodes

Cluster Storage

. . . . . . . . . . . . . . . . . . . . . 9

. . . . . . . . . . . . . . . . . . . 10

Supported Cluster Configurations

Direct-Attached Cluster

SAN-Attached Cluster

Introduction

A Dell™ Failover Cluster combines specific hardware and software components to provide enhanced availability for applications and services that are run on the cluster. A Failover Cluster is designed to reduce the possibility of any single point of failure within the system that can cause the clustered applications or services to become unavailable. It is recommended that you use redundant components like server and storage power supplies, connections between the nodes and the storage array(s), and connections to client systems or other servers in a multi-tier enterprise application architecture in your cluster.

This document provides information to configure your Dell/EMC CX4-series fibre channel storage arrays with one or more Failover Clusters. It provides specific configuration tasks that enable you to deploy the shared storage for your cluster.

For more information on deploying your cluster with Microsoft

Server Windows Server 2003 Installation and Troubleshooting Guide located on the Dell Support website at support.dell.com. For more information on deploying your cluster with Windows Server 2008 operating systems, see the

Dell Failover Clusters with Microsoft Windows Server 2008 Installation and Troubleshooting Guide located on the Dell Support website at

support.dell.com.

For a list of recommended operating systems, hardware components, and driver or firmware versions for your Dell Failover Cluster, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

2003 operating systems, see the Dell Failover Clusters with Microsoft

Windows

Introduction 7

Cluster Solution

Your cluster implements a minimum of two nodes to a maximum of either eight nodes (for Windows Server 2003) or sixteen nodes (for Windows Server

2008) and provides the following features:

• 8-Gbps and 4-Gbps Fibre Channel technology

• High availability of resources to network clients

• Redundant paths to the shared storage

• Failure recovery for applications and services

• Flexible maintenance capabilities, allowing you to repair, maintain, or upgrade a node or storage system without taking the entire cluster offline

Implementing Fibre Channel technology in a cluster provides the following advantages:

•

Flexibility

switches without degrading the signal.

•

Availability

providing multiple data paths and greater availability for clients.

•

Connectivity

Small Computer System Interface (SCSI). Because Fibre Channel devices are hot-pluggable, you can add or remove devices from the nodes without taking the entire cluster offline.

— Fibre Channel allows a distance of up to 10 km between

— Fibre Channel components use redundant connections

— Fibre Channel allows more device connections than

Cluster Hardware Requirements

Your cluster requires the following hardware components:

• Cluster nodes

• Cluster storage

8 Introduction

Cluster Nodes

Table 1-1 lists the hardware requirements for the cluster nodes.

Table 1-1. Cluster Node Requirements

Component Minimum Requirement

Cluster nodes A minimum of two identical PowerEdge servers are required.

The maximum number of nodes that are supported depend on the variant of the Windows Server operating system used in your cluster, and on the physical topology in which the storage system and nodes are interconnected.

RAM The variant of the Windows Server operating system that is

installed on your cluster nodes determines the minimum RAM required.

Host Bus Adapter (HBA) ports

NICs At least two NICs: one NIC for the public network and

Internal disk controller

Two Fibre Channel HBAs per node, unless the server employs an integrated or supported dual-port Fibre Channel HBA.

Where possible, place the HBAs on separate PCI buses to improve availability and performance.

another NIC for the private network.

NOTE: It is recommended that the NICs on each public network

are identical, and that the NICs on each private network are identical.

One controller connected to at least two internal hard drives for each node. Use any supported RAID controller or disk controller.

Two hard drives are required for mirroring (RAID 1) and at least three are required for disk striping with parity (RAID 5).

NOTE: It is strongly recommended that you use

hardware-based RAID or software-based disk-fault tolerance for the internal drives.

NOTE: For more information about supported systems, HBAs and operating system

variants, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

Introduction 9

Cluster Storage

Table 1-2 lists supported storage systems and the configuration requirements for the cluster nodes and stand-alone systems connected to the storage systems.

Table 1-2. Cluster Storage Requirements

Hardware Components Requirement

Supported storage systems

Cluster nodes All nodes must be directly attached to a single storage system

Multiple clusters and stand-alone systems

Table 1-3 lists hardware requirements for the storage processor enclosures (SPE), disk array enclosures (DAE), and standby power supplies (SPS).

Table 1-3. Dell/EMC Storage System Requirements

One to four supported Dell/EMC storage systems. See Table 1-3 for specific storage system requirements.

or attached to one or more storage systems through a SAN.

Can share one or more supported storage systems. See "Installing and Configuring the Shared Storage System" on page 45.

Processor Enclosure

CX4-120 One DAE-OS with at

CX4-240 One DAE-OS with at

CX4-480 One DAE-OS with at

CX4-960 One DAE-OS with at

NOTE: The DAE-OS is the first DAE enclosure that is connected to the CX4-series

(including all of the storage systems listed above). Core software is preinstalled on the first five hard drives of the DAE-OS.

Minimum Storage Possible Storage

Expansion

Up to seven DAE’s with least five and up to 15 hard drives

least five and up to 15 hard drives

a maximum of 15 hard

drives each

Up to fifteen DAE’s with

a maximum of 15 hard

drives each

Up to thirty one DAE’s

with a maximum of 15

hard drives each

Up to sixty three DAE’s

with a maximum of 15

hard drives each

SPS

Two for SPE and DAE-OS

10 Introduction

Each storage system in the cluster is centrally managed by one host system (also called a management station) running EMC Navisphere

Manager—a centralized storage management application used to configure Dell/EMC storage systems. Using a graphical user interface (GUI), you can select a specific view of your storage arrays, as shown in Table 1-4.

Table 1-4. Navisphere Manager Storage Views

View Description

Storage Shows the logical storage components and their relationships to each

other and identifies hardware faults.

Hosts Shows the host system's storage group and attached logical unit

numbers (LUNs).

Monitors Shows all Event Monitor configurations, including centralized and

distributed monitoring configurations.

You can use Navisphere Manager to perform tasks such as creating RAID arrays, binding LUNs, and downloading firmware. Optional software for the shared storage systems include:

• EMC MirrorView™ — Provides synchronous or asynchronous mirroring between two storage systems.

• EMC SnapView™

— Captures point-in-time images of a LUN for backups

or testing without affecting the contents of the source LUN.

• EMC SAN Copy™ — Moves data between Dell/EMC storage systems without using host CPU cycles or local area network (LAN) bandwidth.

For more information about Navisphere Manager, MirrorView, SnapView, and SAN Copy, see "Installing and Configuring the Shared Storage System" on page 45.

Introduction 11

Supported Cluster Configurations

The following sections describe the supported cluster configurations.

Direct-Attached Cluster

In a direct-attached cluster, all the nodes of the cluster are directly attached to a single storage system. In this configuration, the RAID controllers (or storage processors) on the storage system are connected by cables directly to the Fibre Channel HBA ports in the nodes.

Figure 1-1 shows a basic direct-attached, single-cluster configuration.

Figure 1-1. Direct-Attached, Single-Cluster Configuration

public network

cluster node

Fibre Channel connections

EMC PowerPath Limitations in a Direct-Attached Cluster

private network

storage system

cluster node

Fibre Channel connections

EMC PowerPath® provides failover capabilities, multiple path detection, and dynamic load balancing between multiple ports on the same storage processor. However, the direct-attached clusters supported by Dell connect to a single port on each storage processor in the storage system. Because of the single port limitation, PowerPath can provide only failover protection, not load balancing, in a direct-attached configuration.

12 Introduction

SAN-Attached Cluster

In a SAN-attached cluster, all nodes are attached to a single storage system or to multiple storage systems through a SAN using redundant switch fabrics. SAN-attached clusters are superior to direct-attached clusters in configuration flexibility, expandability, and performance.

Figure 1-2 shows a SAN-attached cluster.

Figure 1-2. SAN-Attached Cluster

public network

cluster node

private network

Fibre Channel connections

Fibre Channel switch

storage system

cluster node

Fibre Channel connections

Fibre Channel switch

Cabling Your Cluster Hardware

NOTE: To configure Dell blade server modules in a Dell PowerEdge cluster, see the

Using Dell Blade Servers in a Dell PowerEdge High Availability Cluster document located on the Dell Support website at support.dell.com.

Cabling the Mouse, Keyboard, and Monitor

When installing a cluster configuration in a rack, you must include a switch box to connect the mouse, keyboard, and monitor to the nodes. See the documentation included with your rack for instructions on cabling connections of each node to the switch box.

Cabling the Power Supplies

See the documentation for each component in your cluster solution and ensure that the specific power requirements are satisfied.

The following guidelines are recommended to protect your cluster solution from power-related failures:

• For nodes with multiple power supplies, plug each power supply into a separate AC circuit.

• Use uninterruptible power supplies (UPS).

• For some environments, consider having backup generators and power from separate electrical substations.

Figure 2-1 and Figure 2-2 illustrate recommended methods for power cabling for a cluster solution consisting of two PowerEdge systems and two storage systems. To ensure redundancy, the primary power supplies of all the components are grouped into one or two circuits and the redundant power supplies are grouped into a different circuit.

Cabling Your Cluster Hardware 15

Figure 2-1. Power Cabling Example With One Power Supply in the PowerEdge Systems

primary power supplies on one AC power strip (or on one AC Power Distribution Unit [not

0123

redundant power supplies on one AC power strip (or on one AC PDU [not shown])

shown])

NOTE: This illustration is intended only to demonstrate the power

distribution of the components.

16 Cabling Your Cluster Hardware

Figure 2-2. Power Cabling Example With Two Power Supplies in the PowerEdge Systems

primary power supplies on one AC power strip (or on one AC PDU [not shown])

0123

redundant power supplies on one AC power strip (or on one AC PDU [not shown])

NOTE: This illustration is intended only to demonstrate the power

distribution of the components.

Cabling Your Cluster for Public and Private Networks

The network adapters in the cluster nodes provide at least two network connections for each node, as described in Table 2-1.

NOTE: To configure Dell blade server modules in a Dell PowerEdge cluster, see the

Using Dell Blade Servers in a Dell PowerEdge High Availability Cluster document located on the Dell Support website at support.dell.com.

Cabling Your Cluster Hardware 17

Table 2-1. Network Connections

Network Connection Description

Public network All connections to the client LAN.

At least one public network must be configured for Mixed mode for private network failover.

Private network A dedicated connection for sharing cluster health and

status information only.

Figure 2-3 shows an example of cabling in which dedicated network adapters in each node are connected to each other (for the private network) and the remaining network adapters are connected to the public network.

Figure 2-3. Example of Network Cabling Connection

public network

public network adapter

cluster node 1

private network adapter

private network

cluster node 2

Cabling the Public Network

Any network adapter supported by a system running TCP/IP may be used to connect to the public network segments. You can install additional network adapters to support additional public network segments or to provide redundancy in the event of a faulty primary network adapter or switch port.

18 Cabling Your Cluster Hardware

Cabling the Private Network

The private network connection to the nodes is provided by a different network adapter in each node. This network is used for intra-cluster communications. Table 2-2 describes three possible private network configurations.

Table 2-2. Private Network Hardware Components and Connections

Method Hardware Components Connection

Network switch

Point-to-Point Gigabit Ethernet (two-node clusters only)

NOTE: Throughout this document, Gigabit Ethernet is used to refer to either Gigabit

Ethernet or 10 Gigabit Ethernet.

Using Dual-Port Network Adapters

You can configure your cluster to use the public network as a failover for private network communications. If you are using dual-port network adapters, do not configure both ports simultaneously to support both public and private networks.

Gigabit Ethernet network adapters and switches

Copper Gigabit Ethernet network adapters

Connect standard Ethernet cables from the network adapters in the nodes to a Gigabit Ethernet switch.

Connect a standard Ethernet cable between the Gigabit Ethernet network adapters in both nodes.

NIC Teaming

NIC teaming combines two or more NICs to provide load balancing and fault tolerance. Your cluster supports NIC teaming, only in a public network. NIC teaming is not supported in a private network.

Use the same brand of NICs in a team. Do not mix brands in NIC teaming.

Cabling the Storage Systems

This section provides information on cabling your cluster to a storage system in a direct-attached configuration or to one or more storage systems in a SANattached configuration.

Cabling Your Cluster Hardware 19

Cabling Storage for Your Direct-Attached Cluster

A direct-attached cluster configuration consists of redundant Fibre Channel host bus adapter (HBA) ports cabled directly to a Dell/EMC storage system.

Figure 2-4 shows an example of a direct-attached, single cluster configuration with redundant HBA ports installed in each cluster node.

Figure 2-4. Direct-Attached Cluster Configuration

public network

cluster node

Fibre Channel connections

cluster node

private network

Fibre Channel connections

storage system

20 Cabling Your Cluster Hardware

Cabling a Cluster to a Dell/EMC Storage System

Each cluster node attaches to the storage system using two Fibre optic cables with duplex local connector (LC) multimode connectors that attach to the HBA ports in the cluster nodes and the storage processor (SP) ports in the Dell/EMC storage system. These connectors consist of two individual Fibre optic connectors with indexed tabs that must be aligned properly into the HBA ports and SP ports.

CAUTION: Do not remove the connector covers until you are ready to insert the

connectors into the HBA port, SP port, or tape library port.

Cabling a Two-Node Cluster to a Dell/EMC Storage System

NOTE: The Dell/EMC storage system requires at least 2 front-end fibre channel

ports available on each storage processor.

Connect cluster node 1 to the storage system:

Install a cable from cluster node 1 HBA port 0 to the first front-end fibre channel port on SP-A.

Install a cable from cluster node 1 HBA port 1 to the first front-end fibre channel port on SP-B.

Connect cluster node 2 to the storage system:

Install a cable from cluster node 2 HBA port 0 to the second front-end fibre channel port on SP-A.

Install a cable from cluster node 2 HBA port 1 to the second front-end fibre channel port on SP-B.

Cabling Your Cluster Hardware 21

Figure 2-5. Cabling a Two-Node Cluster to a CX4-120 or CX4-240 Storage System

cluster node 1

HBA ports (2)

0123

SP-B

0123

SP-A

CX4-120 or CX4-240 storage system

Figure 2-6. Cabling a Two-Node Cluster to a CX4-480 Storage System

cluster node 1

HBA ports (2)

cluster node 2

SP-B

0123

CX4-480 storage system

22 Cabling Your Cluster Hardware

0123

SP-A

Figure 2-7. Cabling a Two-Node Cluster to a CX4-960 Storage System

cluster node 1

HBA ports (2)

SP-B

0123 0123

0123001123

cluster node 2

HBA ports (2)

SP-A

CX4-960 storage system

Cabling a Multi-Node Cluster to a Dell/EMC Storage System

You can configure a cluster with more than two nodes in a direct-attached configuration using a Dell/EMC storage system, depending on the availability of front-end fibre channel ports. The CX4-120 and CX4-240 storage systems can support up to 6-node cluster, the CX4-480 storage system can support up to 8-node cluster, and the CX4-960 can support up to 12-node cluster.

The following example explains how to cable a four-node cluster:

NOTE: The Dell/EMC storage system requires at least 4 front-end fibre channel

ports available on each storage processor.

NOTE: The following steps can be modified to add more clusters or standalone

systems.

Connect cluster node 1 to the storage system:

Install a cable from cluster node 1 HBA port 0 to the first front-end fibre channel port on SP-A.

Install a cable from cluster node 1 HBA port 1 to the first front-end fibre channel port on SP-B.

Cabling Your Cluster Hardware 23

Connect cluster node 2 to the storage system:

Install a cable from cluster node 2 HBA port 0 to the second front-end fibre channel port on SP-A.

Install a cable from cluster node 2 HBA port 1 to the second front-end fibre channel port on SP-B.

Connect cluster node 3 to the storage system:

Install a cable from cluster node 3 HBA port 0 to the third front-end fibre channel port on SP-A.

Install a cable from cluster node 3 HBA port 1 to the third front-end fibre channel port on SP-B.

Connect cluster node 4 to the storage system:

Install a cable from cluster node 4 HBA port 0 to the fourth front-end fibre channel port on SP-A.

Install a cable from cluster node 4 HBA port 1 to the fourth front-end fibre channel port on SP-B.

Cabling Multiple Clusters to a Dell/EMC Storage System

The high number of available front-end fibre channel ports on the CX4-series storage system also allows you to configure multiple clusters or a mix of cluster(s) and non-clustered server(s) in a direct-attached configuration.

For example, the 6 front-end fibre channel ports per storage processor on Dell/EMC CX4-120 and CX4-240 storage systems allows you to connect 3 two-node clusters, or 2 two-node clusters and 2 non-clustered systems in a direct-attached configuration. Similarly, the 8 front-end fibre-channel ports per storage processor on Dell/EMC CX4-480 storage system allows you to connect 4 two-node clusters, or 2 two-node clusters and 4 non-clustered systems in a direct-attached environment.

Similarly, the 12 front-end fibre channel ports per storage processor on Dell/EMC CX4-960 storage system allows you to connect 6 two-node clusters or 5 two-node clusters and 2 non-clustered servers in a direct-attached environment.

NOTE: Enable EMC® Access Control if the CX4-series storage system is connected

to more than one cluster in a direct-attached configuration.

24 Cabling Your Cluster Hardware

Cabling Two Two-Node Clusters to a Dell/EMC Storage System

The following steps are an example of how to cable a two two-node cluster. The Dell/EMC storage system needs to have at least 4 front-end fibre channel ports available on each storage processor.

In the first cluster, connect cluster node 1 to the storage system:

Install a cable from cluster node 1 HBA port 0 to the first front-end fibre channel port on SP-A.

Install a cable from cluster node 1 HBA port 1 to the first front-end fibre channel port on SP-B.

In the first cluster, connect cluster node 2 to the storage system:

Install a cable from cluster node 2 HBA port 0 to the second front-end fibre channel port on SP-A.

Install a cable from cluster node 2 HBA port 1 to the second front-end fibre channel port on SP-B.

In the second cluster, connect cluster node 1 to the storage system:

Install a cable from cluster node 1 HBA port 0 to the third front-end fibre channel port on SP-A.

Install a cable from cluster node 1 HBA port 1 to the third front-end fibre channel port on SP-B.

In the second cluster, connect cluster node 2 to the storage system:

Install a cable from cluster node 2 HBA port 0 to the fourth front-end fibre channel port on SP-A.

Install a cable from cluster node 2 HBA port 1 to the fourth front-end fibre channel port on SP-B.

Cabling Storage for Your SAN-Attached Cluster

A SAN-attached cluster is a cluster configuration where all cluster nodes that are attached to a single storage system or to multiple storage systems through SAN use a redundant switch fabric.

SAN-attached cluster configurations provide more flexibility, expandability, and performance than direct-attached configurations.

See "Implementing Zoning on a Fibre Channel Switched Fabric" on page 42 for more information on Fibre Channel switch fabrics.

Cabling Your Cluster Hardware 25

Figure 2-8 shows an example of a two node SAN-attached cluster.

Figure 2-9 shows an example of an eight-node SAN-attached cluster.

Similar cabling concepts can be applied to clusters that contain a different number of nodes.

NOTE: The connections listed in this section are representative of one proven

method of ensuring redundancy in the connections between the cluster nodes and the storage system. Other methods that achieve the same type of redundant connectivity may be acceptable.

Figure 2-8. Two-Node SAN-Attached Cluster

public network

cluster node

private network

Fibre Channel connections

Fibre Channel switch

storage system

26 Cabling Your Cluster Hardware

cluster node

Fibre Channel connections

Fibre Channel switch

Figure 2-9. Eight-Node SAN-Attached Cluster

public network

private network

cluster nodes (2-8)

Fibre Channel switch

storage system

Cabling Your Cluster Hardware 27

Cabling a SAN-Attached Cluster to a Dell/EMC Storage System

The cluster nodes attach to the storage system using a redundant switch fabric and Fibre optic cables with duplex LC multimode connectors.

The switches, the HBA ports in the cluster nodes, and the SP ports in the storage system use duplex LC multimode connectors. The connectors consist of two individual fibre optic connectors with indexed tabs that must be inserted and aligned properly in the small form-factor pluggable (SFP) module connectors on the Fibre Channel switches and the connectors on the cluster nodes and storage systems.

Each HBA port is cabled to a port on a Fibre Channel switch. One or more cables connect from the outgoing ports on a switch to a storage processor on a Dell/EMC storage system.

Table 2-3 provides information for cabling your storage system to the Fibre Channel switches.

Table 2-3. Storage System Cabling Description

Storage System Front-end Fibre

Channel ports per SP

CX4-120, CX4-240 Two to six ports Four to twelve Attach one cable

CX4-480 Four to eight ports Eight to sixteen

CX4-960 Four to twelve

ports

NOTE: Adding more cables from the storage system to the switches can increase

the I/O bandwidth and high availability of data. Although the CX4-960 has a maximum of 12 front-end fibre channel ports per SP, only 8 of them can be connected to fibre channel switches.

Fibre Optic Cables Required

Eight to sixteen

Cabling Description

from each storage processor port to the Fibre Channel switch.

Figure 2-10 illustrates the method for cabling a SAN-attached cluster to the CX4-120 and CX4-240 storage systems.

Figure 2-11 illustrates a method for cabling a SAN-attached cluster to a CX4480 storage system.

Figure 2-12 illustrates a method for cabling a SAN-attached cluster to a CX4-960 storage system.

28 Cabling Your Cluster Hardware

Cabling a SAN-Attached Cluster to a Dell/EMC CX4-120 or CX4-240 Storage System

Connect cluster node 1 to the SAN:

Connect a cable from HBA port 0 to Fibre Channel switch 0 (sw0).

Connect a cable from HBA port 1 to Fibre Channel switch 1 (sw1).

Repeat step 1 for each additional cluster node.

Connect the storage system to the SAN:

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 1 (sw1) to the second front-end fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 1 (sw1) to the second front-end fibre channel port on SP-B.

NOTE: Additional cables can be connected from the fibre channel switches to the

storage system if there are available front-end fibre channel ports on the storage processors.

Cabling Your Cluster Hardware 29

Figure 2-10. Cabling a SAN-Attached Cluster to the Dell/EMC CX4-120 or CX4-240

cluster node 1

sw0

HBA ports (2)

SP-B

0123

HBA ports (2)

SP-A

0123

cluster node 2

sw1

CX4-120 or CX4-240 storage system

Cabling a SAN-Attached Cluster to the Dell/EMC CX4-480 or CX4-960 Storage System

Connect cluster node 1 to the SAN:

Connect a cable from HBA port 0 to Fibre Channel switch 0 (sw0).

Connect a cable from HBA port 1 to Fibre Channel switch 1 (sw1).

Repeat step 1 for each additional cluster node.

Connect the storage system to the SAN:

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 0 (sw0) to the second front-end fibre channel port on SP-A.

30 Cabling Your Cluster Hardware

Connect a cable from Fibre Channel switch 0 (sw0) to the second front-end fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 1 (sw1) to the third frontend fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 1 (sw1) to the third frontend fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 1 (sw1) to the fourth front-end fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 1 (sw1) to the fourth front-end fibre channel port on SP-B.

NOTE: Additional cables can be connected from the fibre channel switches to the

storage system if there are available front-end fibre channel ports on the storage processors.

Figure 2-11. Cabling a SAN-Attached Cluster to the Dell/EMC CX4-480

cluster node 1

sw0

HBA ports (2)

0123

SP-B

CX4-480 storage system

Cabling Your Cluster Hardware 31

HBA ports (2)

0123

SP-A

cluster node 2

sw1

Figure 2-12. Cabling a SAN-Attached Cluster to the Dell\EMC CX4-960

cluster node 2cluster node 1

HBA ports (2)

sw0 sw1

Fibre Channel switch

HBA ports (2)

Fibre Channel switch

SP-B

0 1 23 0 1 23

0123001123

CX4-960 storage system

SP-A

Cabling Multiple SAN-Attached Clusters to a Dell/EMC Storage System

To cable multiple clusters to the storage system, connect the cluster nodes to the appropriate Fibre Channel switches and then connect the Fibre Channel switches to the appropriate storage processors on the processor enclosure.

For rules and guidelines for SAN-attached clusters, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

NOTE: The following procedures use Figure 2-10, Figure 2-11, and Figure 2-12 as

examples for cabling additional clusters.

32 Cabling Your Cluster Hardware

Cabling Multiple SAN-Attached Clusters to the CX4-120 or CX4-240 Storage System

In the first cluster, connect cluster node 1 to the SAN:

Connect a cable from HBA port 0 to Fibre Channel switch 0 (sw0).

Connect a cable from HBA port 1 to Fibre Channel switch 1 (sw1).

In the first cluster, repeat step 1 for each additional cluster node.

For each additional cluster, repeat step 1 and step 2.

Connect the storage system to the SAN:

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 1 (sw1) to the second front-end fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 1 (sw1) to the second front-end fibre channel port on SP-B.

NOTE: Additional cables can be connected from the fibre channel switches to the

storage system if there are available front-end fibre channel ports on the storage processors.

Cabling Multiple SAN-Attached Clusters to the CX4-480 or CX4-960Storage System

In the first cluster, connect cluster node 1 to the SAN:

Connect a cable from HBA port 0 to Fibre Channel switch 0 (sw0).

Connect a cable from HBA port 1 to Fibre Channel switch 1 (sw1).

In the first cluster, repeat step 1 for each additional cluster node.

For each additional cluster, repeat step 1 and step 2.

Connect the storage system to the SAN:

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 0 (sw0) to the first frontend fibre channel port on SP-B.

Cabling Your Cluster Hardware 33

Connect a cable from Fibre Channel switch 0 (sw0) to the second front-end fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 0 (sw0) to the second front-end fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 1 (sw1) to the third frontend fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 1 (sw1) to the third frontend fibre channel port on SP-B.

Connect a cable from Fibre Channel switch 1 (sw1) to the fourth front-end fibre channel port on SP-A.

Connect a cable from Fibre Channel switch 1 (sw1) to the fourth front-end fibre channel port on SP-B.

NOTE: Additional cables can be connected from the fibre channel switches to the

storage system if there are available front-end fibre channel ports on the storage

processors.

Zoning Your Dell/EMC Storage System in a Switched Environment

Dell only supports single-initiator zoning for connecting clusters to a Dell/EMC storage system in a switched environment. When using EMC PowerPath, a separate zone is created from each HBA port to the SPE.

Connecting a PowerEdge Cluster to Multiple Storage Systems

You can increase your cluster storage capacity by attaching multiple storage systems to your cluster using a redundant switch fabric. Failover Clusters can support configurations with multiple storage units attached to clustered nodes. In this scenario, the Microsoft Cluster Service (MSCS) software can fail over disk drives in any cluster-attached shared storage array between the cluster nodes.

NOTE: Throughout this document, MSCS is used to refer to either the Microsoft

Windows Server 2003 Cluster Service or the Microsoft Windows Server 2008 Failover Cluster Service.

When attaching multiple storage systems with your cluster, the following rules apply:

• There is a maximum of four storage systems per cluster.

• The shared storage systems and firmware must be identical. Using dissimilar storage systems and firmware for your shared storage is not supported.

34 Cabling Your Cluster Hardware

• MSCS is limited to 22 drive letters. Because drive letters A through D are reserved for local disks, a maximum of 22 drive letters (E to Z) can be used for your storage system disks.

• Windows Server 2003 and 2008 support mount points, allowing greater than 22 drives per cluster.

Figure 2-13 provides an example of cabling the cluster nodes to four Dell/EMC storage systems. See "Implementing Zoning on a Fibre Channel Switched Fabric" on page 42 for more information.

Figure 2-13. PowerEdge Cluster Nodes Cabled to Four Storage Systems

cluster node

private network

Fibre Channel switch

storage systems (4)

Connecting a PowerEdge Cluster to a Tape Library

cluster node

Fibre Channel switch

To provide additional backup for your cluster, you can add tape backup devices to your cluster configuration. The Dell PowerVault™ tape libraries may contain an integrated Fibre Channel bridge or Storage Network Controller (SNC) that connects directly to your Fibre Channel switch.

Figure 2-14 shows a supported Failover Cluster configuration using redundant Fibre Channel switches and a tape library. In this configuration, each of the cluster nodes can access the tape library to provide backup for your local disk resources, as well as your cluster disk resources. Using this configuration allows you to add more servers and storage systems in the future, if needed.

Cabling Your Cluster Hardware 35

NOTE: While tape libraries can be connected to multiple fabrics, they do not

provide path failover.

Figure 2-14. Cabling a Storage System and a Tape Library

Fibre Channel switch

tape library

cluster node

private network

storage system

cluster node

Fibre Channel switch

Obtaining More Information

See the storage and tape backup documentation for more information on configuring these components.

Configuring Your Cluster With SAN Backup

You can provide centralized backup for your clusters by sharing your SAN with multiple clusters, storage systems, and a tape library.

Figure 2-15 provides an example of cabling the cluster nodes to your storage systems and SAN backup with a tape library.

36 Cabling Your Cluster Hardware

Figure 2-15. Cluster Configuration Using SAN-Based Backup

cluster 2cluster 1

Fibre Channel switch

tape library

Fibre Channel switch

storage systems

Cabling Your Cluster Hardware 37

38 Cabling Your Cluster Hardware

Preparing Your Systems for Clustering

WARNING: Only trained service technicians are authorized to remove and

access any of the components inside the system. See your safety information for complete information about safety precautions, working inside the computer, and protecting against electrostatic discharge.

Cluster Configuration Overview

Ensure that your site can handle the cluster’s power requirements. Contact your sales representative for information about your region's power requirements.

Install the systems, the shared storage array(s), and the interconnect switches (for example, in an equipment rack), and ensure that all the components are turned on.

NOTE: For more information on step 3 to step 7 and step 10 to step 13, see the

"Preparing your systems for clustering" section of Dell Failover Clusters with

Microsoft Windows Server 2003 Installation and Troubleshooting Guide or Dell Failover Clusters with Microsoft Windows Server 2008 Installation and Troubleshooting Guide located on the Dell Support website at support.dell.com.

Deploy the operating system (including any relevant service packs and hotfixes), network adapter drivers, and storage adapter drivers (including Multipath I/O (MPIO) drivers) on each cluster node. Depending on the deployment method that is used, it may be necessary to provide a network connection to successfully complete this step.

NOTE: To help in planning and deployment of your cluster, record the relevant

cluster configuration information in the Cluster Data Form located at "Cluster Data Form" on page 63 and the Zoning configuration information in the Zoning Configuration form located and "Zoning Configuration Form" on page 61.

Establish the physical network topology and the TCP/IP settings for network adapters on each cluster node to provide access to the cluster public and private networks.

Preparing Your Systems for Clustering 39

Configure each cluster node as a member in the same Windows Active Directory Domain.

NOTE: You can configure the cluster nodes as Domain Controllers. For more

information, see the “Selecting a Domain Model” section of Dell Failover

Clusters with Microsoft Windows Server 2003 Installation and Troubleshooting Guide or Dell Failover Clusters with Microsoft Windows Server 2008 Installation and Troubleshooting Guide located on the Dell

Support website at support.dell.com.

Establish the physical storage topology and any required storage network settings to provide connectivity between the storage array and the systems that you are configuring as cluster nodes. Configure the storage system(s) as described in your storage system documentation.

Use storage array management tools to create at least one logical unit number (LUN). The LUN is used as a cluster Quorum disk for Windows Server 2003 Failover cluster and as a Witness disk for Windows Server 2008 Failover cluster. Ensure that this LUN is presented to the systems that you are configuring as cluster nodes.

NOTE: For security reasons, it is recommended that you configure the LUN on

a single node as mentioned in step 8 when you are setting up the cluster. Later, you can configure the LUN as mentioned in step 9 so that other nodes in the cluster can access it.

Select one of the systems and form a new failover cluster by configuring the cluster name, cluster management IP, and quorum resource. For more information, see "Preparing Your Systems for Clustering" on page 39.

NOTE: For Failover Clusters configured with Windows Server 2008, run the

Cluster Validation Wizard to ensure that your system is ready to form the cluster.

Join the remaining node(s) to the failover cluster. For more information, see "Preparing Your Systems for Clustering" on page 39.

Configure roles for cluster networks.

Test the failover capabilities of your new cluster.

NOTE: For Failover Clusters configured with Windows Server 2008, you can

also use the Cluster Validation Wizard.

40 Preparing Your Systems for Clustering

Configure highly-available applications and services on your Failover Cluster. Depending on your configuration, this may also require providing additional LUNs to the cluster or creating new cluster resource groups. Test the failover capabilities of the new resources.

Configure client systems to access the highly-available applications and services that are hosted on your failover cluster.

Installation Overview

Each node in your Dell Failover Cluster must be installed with the same release, edition, service pack, and processor architecture of the Windows Server operating system. For example, all nodes in your cluster may be configured with Windows Server 2003 R2, Enterprise x64 Edition. If the operating system varies among nodes, it is not possible to configure a Failover Cluster successfully. It is recommended to establish server roles prior to configuring a Failover Cluster, depending on the operating system configured on your cluster.

For a list of Dell PowerEdge Servers, Fibre Channel HBAs and switches, and recommended list of operating system variants, specific driver and firmware revisions, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

For a general overview of cluster configuration tasks and more detailed information about deploying your cluster with Windows Server 2003 operating system, see the Dell Failover Clusters with Microsoft Windows Server 2003 Installation and Troubleshooting Guide located on the Dell Support website at support.dell.com.

For more information on deploying your cluster with Windows Server 2008 operating systems, see the Dell Failover Clusters with Microsoft Windows Server 2008 Installation and Troubleshooting Guide located on the Dell Support website at support.dell.com.

The following sub-sections describe steps that must be taken to enable communication between the cluster nodes and your shared Dell/EMC CX4-series Fibre Channel storage array, and to present disks from the storage array to the cluster.

Preparing Your Systems for Clustering 41

Installing the Fibre Channel HBAs

For dual -HBA configurations, it is recommended that you install the Fibre Channel HBAs on separate peripheral component interconnect (PCI) buses. Placing the adapters on separate buses improves availability and performance.

For more information about your system's PCI bus configuration and supported HBAs, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

Installing the Fibre Channel HBA Drivers

For more information, see the EMC documentation that is included with your HBA kit.

For more information about installing and configuring Emulex HBAs and EMC-approved drivers, see the Emulex support website located at www.emulex.com or the Dell Support website at support.dell.com.

For more information about installing and configuring QLogic HBAs and EMC-approved drivers, see the QLogic support website at www.qlogic.com or the Dell Support website at support.dell.com.

For more information about supported HBA controllers and drivers, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

Implementing Zoning on a Fibre Channel Switched Fabric

A Fibre Channel switched fabric consists of one or more Fibre Channel switches that provide high-speed connections between servers and storage devices. The switches in a Fibre Channel fabric provide a connection through inbound and outbound points from one device (sender) to another device or switch (receiver) on the network. If the data is sent to another switch, the process repeats itself until a connection is established between the sender and the receiver.

Fibre Channel switches provide you with the ability to set up barriers between different devices and operating environments. These barriers create logical fabric subsets with minimal software and hardware intervention. Similar to subnets in the client/server network, logical fabric subsets divide a fabric into similar groups of components, regardless of their proximity to one another. The logical subsets that form these barriers are called zones.

42 Preparing Your Systems for Clustering

Zoning automatically and transparently enforces access of information to the zone devices. More than one PowerEdge cluster configuration can share Dell/EMC storage system(s) in a switched fabric using Fibre Channel switch zoning and with Access Control enabled. By using Fibre Channel switches to implement zoning, you can segment the SANs to isolate heterogeneous servers and storage systems from each other.

Using Zoning in SAN Configurations Containing Multiple Hosts

Using the combination of zoning and Access Control in SAN configurations containing multiple hosts, you can restrict server access to specific volumes on a shared storage system by preventing the hosts from discovering a storage volume that belongs to another host. This configuration allows multiple clustered or nonclustered hosts to share a storage system.

Using Worldwide Port Name Zoning

PowerEdge cluster configurations support worldwide port name zoning.

A worldwide name (WWN) is a unique numeric identifier assigned to Fibre Channel interfaces, such as HBA ports, storage processor (SP) ports, and Fibre Channel to SCSI bridges or storage network controllers (SNCs).

A WWN consists of an 8-byte hexadecimal number with each byte separated by a colon. For example, 10:00:00:60:69:00:00:8a is a valid WWN. Using WWN port name zoning allows you to move cables between switch ports within the fabric without having to update the zones.

Table 3-1 provides a list of WWN identifiers that you can find in the Dell/EMC cluster environment.

Table 3-1. Port Worldwide Names in a SAN Environment

Identifier Description

xx:xx:

xx:xx:xx:

50:06:01:6

xx:xx:00:

xx:xx:

xx:xx

00:60:69:

00:88:

x:xx:xx:xx:xx

00:C9:

00:E0:8B:

:00:0F:1F:

xx:xx:xx

Dell/EMC or Brocade switch

McData switch

Dell/EMC storage processor

Emulex HBA ports

QLogic HBA ports (non-embedded)

Dell 2362M HBA port

Preparing Your Systems for Clustering 43

Table 3-1. Port Worldwide Names in a SAN Environment (continued)

Identifier Description

xx:xx:xx:

xx:xx:xx

xx:xx:xx:C0:97:xx:xx:xx PowerVault ML6000 Fibre Channel tape drives

CAUTION: When you replace a Fibre Channel HBA in a PowerEdge server,

60:45:

E0:02:

:C0:01:

reconfigure your zones to provide continuous client data access. Additionally, when you replace a switch module, reconfigure your zones to prevent data loss or corruption.

xx:xx:xx

PowerVault 132T and 136T tape libraries

PowerVault 128T tape autoloader

PowerVault 160T tape library and Fibre Channel tape drives

CAUTION: You must configure your zones before you configure the logical unit

numbers (LUNs) and storage groups. Failure to do so may cause data loss, data corruption, or data unavailability.

Single Initiator Zoning

Each host HBA port in a SAN must be configured in a separate zone on the switch with the appropriate storage ports. This zoning configuration, known as single initiator zoning, prevents different hosts from communicating with each other, thereby ensuring that Fibre Channel communications between the HBAs and their target storage systems do not affect each other.

When you create your single-initiator zones, follow these guidelines:

• Create a zone for each HBA port and its target storage devices.

• Each CX4-series storage processor port can be connected to a maximum of 64 HBA ports in a SAN-attached environment.

• Each host can be connected to a maximum of four storage systems.

•The

integrated bridge/SNC or fibre-channel interface

on a tape library

can be added to any zone.

NOTE: If you are sharing a storage system with multiple clusters or a combination

of clustered and nonclustered systems (hosts), you must enable Access Control. Otherwise, you can only have one nonclustered system or one PowerEdge cluster attached to the Dell/EMC storage system.

44 Preparing Your Systems for Clustering

Installing and Configuring the Shared Storage System

See "Cluster Hardware Requirements" on page 8 for a list of supported Dell/EMC storage systems.

To install and configure the Dell/EMC storage system in your cluster:

Update the core software on your storage system and enable Access Control (optional) and install any additional software options, including EMC SnapView™, EMC MirrorView™, and SAN Copy™. See your EMC Navisphere

Install the EMC Navisphere Agent and EMC PowerPath software on each cluster node.

See your Navisphere documentation for more information.

Update the storage system configuration settings using Navisphere Manager.

See "Enabling Access Control and Creating Storage Groups Using Navisphere" on page 49 for more information.

The following subsections provide an overview of the storage management software and procedures for connecting the host systems to the storage systems.

Access Control

Fibre Channel topologies allow multiple clusters and stand-alone systems to share a single storage system. However, if you cannot control access to the shared storage system, you can corrupt your data. To share your Dell/EMC storage system with multiple heterogeneous host systems and restrict access to the shared storage system, you need to enable Access Control.

Enabling Access Control prevents all host systems from accessing any data on the storage system until they are given explicit access to a LUN through a storage group. By enabling Access Control, you can prevent the host systems from taking ownership of all LUNs on the storage system and prevent unauthorized access to sensitive information.

documentation for more information.

Preparing Your Systems for Clustering 45

Access Control is enabled using Navisphere Manager. After you enable Access Control and connect to the storage system from a management station, Access Control appears in the Storage System Properties window of

Navisphere Manager.

After you enable Access Control, the host system can only read from and write to specific LUNs on the storage system. This organized group of LUNs and hosts is called a storage group.

Storage Groups

A storage group is a collection of one or more LUNs that are assigned to one or more host systems. Managed by Navisphere Manager, storage groups provide an organized method of assigning multiple LUNs to a host system. After you create LUNs on your storage system, you can assign the LUNs to a storage group in Navisphere Manager and then assign the storage group to a specific host. Because the host can only access its assigned storage group, it cannot access any LUNs assigned to other host systems, thereby protecting your data from unauthorized access.

To create the storage groups for your host systems, you must use Navisphere Manager and enable Access Control in the storage system.

NOTE: A host system can access only one storage group per storage system.

Table 3-2 describes the properties in the storage group.

46 Preparing Your Systems for Clustering

Table 3-2. Storage Group Properties

Property Description

Unique ID A unique identifier that is automatically assigned to the

storage group that cannot be changed.

Storage group name The name of the storage group. The default storage

group name is formatted as Storage Group n, where n equals the existing number of storage groups plus one.

Connected hosts Lists the host systems connected to the storage group.

Each host entry contains the following fields:

•

Name

— Name of the host system

IP address

•

system

— IP address of the host system

— Operating system that is running on the host

NOTE: In a clustered environment, all nodes of a cluster

must be connected to the same storage group.

Used host connection paths

LUNs in storage group Lists the LUNs in the storage group.

An additional storage group feature that performs the following tasks:

• Lists all of the paths from the host server to the storage group

• Displays whether the path is enabled or disabled

Each path contains the following fields:

HBA

— Device name of the HBA in the host system

–

HBA Port

–

to the storage system

SP Port

–

connected to the HBA port

SP ID

–

Each LUN entry contains the following fields:

•

Identifier

Name

•

Capacity

• LUN

— Unique ID for the HBA port connected

— Unique ID for the storage processor port

— ID of the storage processor

— LUN icon representing the LUN

— Name of the LUN

— Amount of allocated storage space on the

Preparing Your Systems for Clustering 47

Navisphere Manager

Navisphere Manager provides centralized storage management and configuration from a single management console. Using a graphical user interface (GUI), Navisphere Manager allows you to configure and manage the disks and components in one or more shared storage systems.

You can access Navisphere Manager through a web browser. Using Navisphere Manager, you can manage a Dell/EMC storage system either locally on the same LAN or through an Internet connection. Navisphere components (Navisphere Manager user interface (UI) and Storage Management Server) are installed on a Dell/EMC storage system. You can access Navisphere Manager by opening a browser and entering the IP address of the storage system’s SP. Navisphere Manager downloads components to your system and runs in the web browser.

Optionally, you can run Navisphere Management Server for Windows. This software component installs on a host system connected to a Dell/EMC storage system, allowing you to run Navisphere Storage Management Server on the host system.

Using Navisphere Manager, you can:

• Create storage groups for your host systems

• Create, bind, and unbind LUNs

• Change configuration settings

• Monitor storage systems

Navisphere Agent

Navisphere Agent is installed on the host system and performs the following tasks:

• Registers each host with the storage system

• Communicates configuration information from the host to the storage system

In a cluster environment, a NIC may have multiple IP addresses bound to it. To ensure that the agent registers the host to the storage system with the correct IP address, perform the following steps:

Using a text editor to create a file named

Files\EMC\Navisphere (x86)\EMC\Navisphere

48 Preparing Your Systems for Clustering

directory (or

for Windows Server 2008 and 2003 (64-bit) systems).

agentID.txt

C:\Program Files

in the

C:\Program

Add the following two separate lines to the

agentID.txt

file, with no

special formatting:

• First line: Fully qualified hostname. For example, enter

node1.domain1.com

domain1

, if the host name is

node1

and the domain name

• Second line: IP address that you want the agent to register and use to communicate with the storage system.

EMC PowerPath

EMC PowerPath® automatically reroutes Fibre Channel I/O traffic from the host system and a Dell/EMC CX4-series storage system to any available path if a primary path fails for any reason. Additionally, PowerPath provides multiple path load balancing, allowing you to balance the I/O traffic across multiple SP ports.

Enabling Access Control and Creating Storage Groups Using Navisphere

The following subsection provides the required procedures for creating storage groups and connecting your storage systems to the host systems.

CAUTION: Before enabling Access Control, ensure that no hosts are attempting to

access the storage system. Enabling Access Control prevents all hosts from accessing any data until they are given explicit access to a LUN in the appropriate storage group. You must stop all I/O before enabling Access Control. It is recommended to turn off all hosts connected to the storage system during this procedure or data loss may occur. After you enable the Access Control software, it cannot be disabled.

Ensure that Navisphere Agent is started on all host systems.

Click the and then select

In the

•In the

Open a Web browser.

Start

Services

Name

Status

Startup Type

button and select

Services

Programs→ Administrative Tools

window, verify the following:

column,

Navisphere Agent

column,

Navisphere Agent

appears.

is set to

Started

Automatic

Preparing Your Systems for Clustering 49

Enter the IP address of the storage management server on your storage system and then press <Enter>.

NOTE: The storage management server is usually one of the SPs on your

storage system.

In the

Enterprise Storage

Right-click the icon of your storage system.

In the drop-down menu, click

Storage Systems Properties

The

window, click the

Properties

window appears.

Storage

tab.

7Click the

Select the

Storage Access

tab.

Access Control Enabled

check box.

A dialog box appears, prompting you to enable

Click

Yes

to enable

Click OK.

Right-click the icon of your storage system and select

Create Storage Group

The

In the

Storage Group Name

Click

Apply

Add new LUNs to the storage group.

Right-click the icon of your storage group and select

Click the

In the

Available LUNs

Click the right-arrow button to move the selected LUN to the

Selected LUNs

Click

Apply

Add new hosts to the

In the

Storage Group Properties

In the

Available Hosts

Access Control

LUNs

tab.

pane.

Sharable

dialog box appears.

field, enter a name for the storage group.

window, click an available LUN.

storage group.

dialog box, click the

window pane, click the host system that you

want to add to the storage group.

Using the right-arrow button, move the selected host to the

be Connected

window pane.

Access Control

Create Storage Group

Properties

Hosts

tab.

Hosts to

50 Preparing Your Systems for Clustering

Repeat step b and step c to add additional hosts.

Click

Apply

Click OK to exit the

Storage Group Properties

dialog box.

Configuring the Hard Drives on the Shared Storage System(s)

This section provides information for configuring the hard drives on the shared storage systems. The shared storage system hard drives must be configured before use. The following sections provide information on these configurations.

Configuring and Managing LUNs

Configuring and managing LUNs is accomplished using the Navisphere Manager utility. Before using Navisphere Manager, ensure that the Navisphere Agent service is started on your cluster nodes.

In some cases, the LUNs may have been bound when the system was shipped. It is still important, however, to install the management software and to verify that the desired LUN configuration exists.

You can manage your LUNs remotely using Navisphere Manager. A minimum of one LUN (RAID drive) is required for an active/passive configuration; at least two drives are required for an active/active configuration.

It is recommended that you create at least one LUN or virtual disk for each application. If multiple NTFS partitions are created on a single LUN or virtual disk, these partitions will not be able to fail over individually from node-to-node.

Configuring the RAID Level for the Shared Storage Subsystem

The hard drives in your shared storage subsystem must be configured into LUNs or virtual disks using Navisphere Manager. All LUNs or virtual disks, especially if they are used for the quorum resource, should be bound and incorporate the appropriate RAID level to ensure high availability.

NOTE: It is recommended that you use a RAID level other than RAID 0 (which is

commonly called striping). RAID 0 configurations provide very high performance, but do not provide the level of availability required for the quorum resource. See the documentation for your storage system for more information about setting up RAID levels for the system.

Preparing Your Systems for Clustering 51

Assigning LUNs to Hosts

If you have Access Control enabled in Navisphere Manager, you must create storage groups and assign LUNs to the proper host systems.

Optional Storage Features

Your Dell/EMC CX4-series storage array may be configured to provide optional features that can be used in conjunction with your cluster. These features include MirrorView, SnapView, and SANCopy.

MirrorView

MirrorView automatically duplicates primary storage system data from a cluster or stand-alone system to a secondary storage system. It can be used in conjunction with SnapView and is managed from within Navisphere Manager.

SnapView

SnapView captures images of a LUN and retains the images independently of subsequent changes to the files. The images can be used to share LUNs with another system without affecting the contents of the source LUN.

SnapView creates copies of LUNs using either snapshots or clones. Snapshots are virtual copies that create an image of the source LUN at the time the snapshot was created. This snapshot is retained independently of subsequent changes to the source LUN. Clones are duplicate copies of a source LUN. You can use snapshots and clones to facilitate backups or to allow multiple hosts to access data without affecting the contents of the source LUN.

The source LUN and each snapshot or clone must be accessed from a different host or a different cluster.

SnapView, which is installed on the storage processors as a non-disruptive upgrade, can be used in conjunction with MirrorView and is managed from within Navisphere Manager.

SAN Copy

SAN Copy allows you to move data between storage systems without using host processor cycles or LAN bandwidth. It can be used in conjunction with SnapView or MirrorView and is managed from within Navisphere Manager.

52 Preparing Your Systems for Clustering

Updating a Dell/EMC Storage System for Clustering

If you are updating an existing Dell/EMC storage system to meet the cluster requirements for the shared storage subsystem, you may need to install additional Fibre Channel disk drives in the shared storage system. The size and number of drives you add depend on the RAID level you want to use and the number of Fibre Channel disk drives currently in your system.

See your storage system's documentation for information on installing Fibre Channel disk drives in your storage system.

Upgrade the core software version that is running on the storage system or enable Access Control. For specific version requirements, see the Dell Cluster Configuration Support Matrix on the Dell High Availability website at www.dell.com/ha.

Installing and Configuring a Failover Cluster

After you have established the private and public networks and have assigned the shared disks from the storage array to the cluster nodes, you can configure the operating system services on your Dell Failover Cluster. The procedure to configure the Failover Cluster depends on the version of the Windows Server operating system that is running on the system.

For more information on deploying your cluster with Windows Server 2003 operating systems, see the

Server 2003 Installation and Troubleshooting Guide

website at

For more information on deploying your cluster with Windows Server 2008 operating systems, see the

Server 2008 Installation and Troubleshooting Guide

website at

support.dell.com

Dell Failover Clusters with Microsoft Windows

located on the Dell Support

Dell Failover Clusters with Microsoft Windows

located on the Dell Support

Preparing Your Systems for Clustering 53

54 Preparing Your Systems for Clustering

Troubleshooting

This appendix provides troubleshooting information for your cluster configuration. Table A-1 describes general cluster problems you may encounter and the probable causes and solutions for each problem.

Table A-1. General Cluster Troubleshooting

Problem Probable Cause Corrective Action

The nodes cannot access the storage system, or the cluster software is not functioning with the storage system.

The storage system is not cabled properly to the nodes or the cabling between the storage components is incorrect.

The length of the interface cables exceeds the maximum allowable length.

One of the cables is faulty.

Access Control is not enabled correctly.

The cluster is in a SAN, and one or more zones are not configured correctly.

Ensure that the cables are connected properly from the node to the storage system. See "Cabling Your Cluster for Public and Private Networks" on page 17 for more information.

Ensure that the fibre optic cables do not exceed 300 m (multimode) or 10 km (single mode switch-to-switch connections only).

Replace the faulty cable.

Verify the following:

• All switched zones are configured correctly.

• The Access Control software is enabled on the storage system.

• All LUNs and hosts are assigned to the proper storage groups.

Verify the following:

• Each zone contains only one initiator (Fibre Channel daughter card).

• Each zone contains the correct initiator and the correct storage port(s).

Troubleshooting 55

Table A-1. General Cluster Troubleshooting (continued)

Problem Probable Cause Corrective Action

One of the nodes takes a long time to join the cluster.

One of the nodes fail to join the cluster.

The node-to-node network has failed due to a cabling or hardware failure.

One or more nodes may have the Internet Connection Firewall enabled, blocking Remote Procedure Call (RPC) communications between the nodes.

Check the network cabling. Ensure that the node-to-node interconnection and the public network are connected to the correct NICs.

Configure the Internet Connection Firewall to allow communications that are required by the Microsoft Service (MSCS) and the clustered applications or services.

See Microsoft Knowledge Base article KB883398 at the Microsoft Support website at support.microsoft.com for more information.

Long delays in node-to-node communications may be normal.

Verify that the nodes can communicate with each other by running the ping command from each node to the other node. Try both the host name and IP address when using the ping command.

Cluster

56 Troubleshooting

Table A-1. General Cluster Troubleshooting (continued)

Problem Probable Cause Corrective Action

Attempts to connect to a cluster using Cluster Administrator fail.

The Cluster Service has not been started.

A cluster has not been formed on the system.

The system has just been booted and services are still starting.

The cluster network name is not responding on the network because the Internet Connection Firewall is enabled on one or more nodes.

Verify that the Cluster Service is running and that a cluster has been formed. Use the Event Viewer and look for the following events logged by the Cluster Service:

Microsoft Cluster Service successfully formed a cluster on this node.

Microsoft Cluster Service successfully joined the cluster.

If these events do not appear in Event Viewer, see the Microsoft Cluster Service Administrator’s Guide for instructions on setting up the cluster on your system and starting the Cluster Service.

Configure the Internet Connection Firewall to allow communications that are required by MSCS and the clustered applications or services.

See Microsoft Knowledge Base article KB883398 at the Microsoft Support website at support.microsoft.com for more information.

Troubleshooting 57

Table A-1. General Cluster Troubleshooting (continued)

Problem Probable Cause Corrective Action

You are prompted to configure one network instead of two during MSCS installation.

The TCP/IP configuration is incorrect.

The node-to-node network and public network must be assigned static IP addresses on different subnets. For more information about assigning the network IPs, see "Assigning Static IP Addresses to Cluster Resources and Components" of Dell Failover Clusters

with Microsoft Windows Server 2003 Installation and Troubleshooting Guide or Dell Failover Clusters with Microsoft Windows Server 2008 Installation and Troubleshooting Guide.

The private (point-to-point) network is

Ensure that all systems are powered on so that the NICs in the private network are available.

disconnected.

Using Microsoft Windows NT

to remotely administer a Windows

Some resources in

4.0

Windows Server 2003 are not supported in Windows NT 4.0.

It is strongly recommended that you use Microsoft Windows XP Professional or Windows Server 2003 for remote administration of a cluster

running Windows Server 2003. Server 2003 cluster generates error messages.

58 Troubleshooting

Table A-1. General Cluster Troubleshooting (continued)

Problem Probable Cause Corrective Action

Unable to add a node to the cluster.

The disks on the shared cluster storage appear unreadable or uninitialized in Windows Disk Administration

The new node cannot access the shared disks.

The shared disks are enumerated by the operating system differently on the cluster nodes.

One or more nodes may have the Internet Connection Firewall enabled, blocking RPC communications between the nodes.

This situation is normal if you stopped the Cluster Service. If you are running Windows Server 2003, this situation is normal if the cluster node does not own the cluster disk.

Ensure that the new cluster node can enumerate the cluster disks using Windows Disk Administration. If the disks do not appear in Disk Administration, check the following:

• Check all cable connections

• Check all zone configurations

• Check the Access Control settings on the attached storage systems

• Use the

Configure the Internet Connection Firewall to allow communications that are required by the MSCS and the clustered applications or services.

See Microsoft Knowledge Base article KB883398 at the Microsoft Support website at support.microsoft.com for more information.

No action required.

Advanced

Minimum

option

with

Troubleshooting 59

Table A-1. General Cluster Troubleshooting (continued)

Problem Probable Cause Corrective Action

Cluster Services does not operate correctly on a cluster running Windows Server 2003 and the Internet Firewall enabled.

The Windows Internet Connection Firewall is enabled, which may conflict with Cluster Services.

Perform the following steps:

On the Windows desktop, right-click

My Computer

In the

Computer Management

window, double-click

In the

Services

Cluster Services

In the

Cluster Services

Recovery

the

Click the

First Failure

arrow and select

Click the

Second Failure

arrow and select

Click OK.

For information on how to configure your cluster with the Windows Internet Connection Firewall enabled, see Microsoft Base (KB) articles 258469 and 883398 at the Microsoft Support website at support.microsoft.com and the Microsoft Windows Server 2003 Technet website at www.microsoft.com/technet.

Public network clients cannot access the applications or services that are provided by the cluster.

One or more nodes may have the Internet Connection Firewall enabled, blocking RPC communications between the nodes.

Configure the Internet Connection Firewall to allow communications that are required by the MSCS and the clustered applications or services.

See Microsoft Knowledge Base article KB883398 at the Microsoft Support website at support.microsoft.com for more information.

and click

Manage

Services

window, double-click

window, click

tab.

drop-down

Restart the Service

drop-down

Restart

the service.

60 Troubleshooting

Zoning Configuration Form

Node HBA WWPNs

or Alias Names

Storage WWPNs or Alias Names

Zone Name Zone Set for

Configuration Name

Zoning Configuration Form 61

62 Zoning Configuration Form

Cluster Data Form

You can attach the following form in a convenient location near each cluster node or rack to record information about the cluster. Use the form when you call for technical support.

Table C-1. Cluster Information

Cluster Information Cluster Solution

Cluster name and IP address

Server type

Installer

Date installed

Applications

Location

Notes

Table C-2. Cluster Node Information

Node Name Service Tag

Number

Public IP Address Private IP Address

Cluster Data Form 63

Additional Networks

Table C-3. Storage Array Information

Array Array xPE Type Array Service Tag

Number or World Wide Name Seed

Number of Attached DAEs

64 Cluster Data Form

Index

Access Control

about, 45

cable configurations

cluster interconnect, 19 for client networks, 18 for mouse, keyboard, and

monitor, 15

for power supplies, 15

cluster

optional configurations, 12

cluster configurations

connecting to multiple shared

storage systems, 34

connecting to one shared storage

system, 12 direct-attached, 12, 20 SAN-attached, 13

cluster storage

requirements, 10

clustering

overview, 7

Dell/EMC CX4-series, 7

Cabling a Multi-Node Cluster, 23 Cabling a Two-Node Cluster, 21 Cabling Description, 28 Cabling Multiple Clusters, 24 Cabling Multiple SAN-Attached

Clusters, 32 Cabling Storage, 25 CX4-120, 10 CX4-240, 10 CX4-480, 10 CX4-960, 10 Zoning, 34

direct-attached cluster

about, 20

drivers

installing and configuring

Emulex, 42

Emulex HBAs

installing and configuring, 42 installing and configuring

drivers, 42

Index 65

HBA drivers

installing and configuring, 42

host bus adapter

configuring the Fibre Channel

HBA, 42

keyboard

cabling, 15

LUNs

assigning to hosts, 52 configuring and managing, 51

MirrorView

about, 11

monitor

cabling, 15

mouse

cabling, 15

MSCS

installing and configuring, 53

Navisphere Manager

about, 11, 48 hardware view, 11 storage view, 11

network adapters

cabling the private

network, 18-19

cabling the public network, 18

operating system

Windows Server 2003, Enterprise

Edition

installing,41

power supplies

cabling, 15

PowerPath

about, 49

private network

cabling, 17, 19 hardware components, 19 hardware components and

connections, 19

public network

cabling, 17

Navisphere Agent

about, 48

66 Index

RAID

configuring the RAID level, 51

SAN

configuring SAN backup in your

cluster, 36

troubleshooting

connecting to a cluster, 57 shared storage subsystem, 55

SAN-Attached Cluster, 13

SAN-attached cluster

about, 25 configurations, 12

shared storage

assigning LUNs to hosts, 52

single initiator zoning

about, 44

SnapView

about, 11

storage groups

about, 46

storage management software

Access Control, 45 Navisphere Agent, 48 Navisphere Manager, 48 PowerPath, 49

storage system

configuring and managing

LUNs, 51

configuring the hard drives, 51

warranty, 13

worldwide port name zoning, 43

zones

implementing on a Fibre Channel

switched fabric, 42 in SAN configurations, 43 using worldwide port names, 43

tape library

connecting to a PowerEdge

cluster, 35

Index 67

68 Index

Dell CX4 User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Notes, Cautions, and Warnings

Introduction

Cluster Solution

Cluster Hardware Requirements

Cluster Nodes

Cluster Storage

Supported Cluster Configurations

Direct-Attached Cluster

SAN-Attached Cluster

Other Documents You May Need

Cabling Your Cluster Hardware

Cabling the Mouse, Keyboard, and Monitor

Cabling the Power Supplies

Cabling Your Cluster for Public and Private Networks

Cabling the Public Network

Cabling the Private Network

NIC Teaming

Cabling the Storage Systems

Cabling Storage for Your Direct-Attached Cluster

Cabling Storage for Your SAN-Attached Cluster

Preparing Your Systems for Clustering

Cluster Configuration Overview

Installation Overview

Installing the Fibre Channel HBAs

Installing the Fibre Channel HBA Drivers

Implementing Zoning on a Fibre Channel Switched Fabric

Using Zoning in SAN Configurations Containing Multiple Hosts

Using Worldwide Port Name Zoning

Installing and Configuring the Shared Storage System

Access Control

Storage Groups

Navisphere Manager

Navisphere Agent

EMC PowerPath

Enabling Access Control and Creating Storage Groups Using Navisphere

Configuring the Hard Drives on the Shared Storage System(s)

Optional Storage Features

Updating a Dell/EMC Storage System for Clustering

Installing and Configuring a Failover Cluster

Troubleshooting

Zoning Configuration Form

Cluster Data Form

Index