Products/HA/DellRedHatHALinuxCluster/Overview

From DellLinuxWiki

(Difference between revisions)
Jump to: navigation, search
Line 1: Line 1:
-
<font size=1>[[Halinux|Dell|Red Hat HA Linux]] > Overview</font>
 
-
 
=Overview=
=Overview=
-
This document provides instructions to use Red Hat® Enterprise Linux® (RHEL) 5.1 Advanced Platform (AP), Red Hat Cluster Suite (RHCS), and Red Hat Global File System (GFS) to configure a Dell|Red Hat HA Linux Cluster with PowerEdge servers connected to Dell storage arrays.
+
This document provides instructions to use Red Hat® Enterprise Linux® (RHEL) 5.3 Advanced Platform (AP), Red Hat Clustering (RHCS), and Red Hat Global File System (GFS) to configure a Dell|Red Hat HA Linux Cluster with PowerEdge servers connected to Dell storage arrays.
==Types of Clustering==
==Types of Clustering==
-
Clustering is the process of configuring two or more systems to provide
+
Clustering is the process of configuring two or more systems to provide services as if they were on a single system. These systems are known as cluster nodes. You can use clustering to enable load balancing between nodes, parallel processing, high availability, and storage clustering.
-
services as if they were a single system. These systems are known as cluster
+
 
-
nodes. You can use clustering to enable load balancing between nodes,
+
'''Load Balancing''' - Load Balancing clusters accept requests from clients and use a defined algorithm to forward the requests to a pool of nodes. Red Hat Clustering contains Linux Virtual Server service that performs this function. Implementation of the Linux Virtual Server function is beyond the scope of this document. For more information, see the Red Hat website at [http://www.redhat.com/cluster_suite/ www.redhat.com/cluster_suite/].
-
parallel processing, high availability, and storage clustering.
+
-
Load Balancing—Load Balancing clusters accept requests from clients and
+
'''High Performance Compute Clustering''' ('''HPCC''') or '''Parallel Processing''' Typical HPC clusters are controlled by a master or front-end server. The master server accepts job requests and divides them up among the compute nodes to take advantage of parallel processing. This document will not discuss HPCC. For more information, see [http://www.dell.com/hpcc dell.com/hpcc]
-
use a defined algorithm to forward the requests to a pool of nodes. Red Hat
+
-
Cluster Suite contains Linux Virtual Server service that performs this
+
-
function. Implementation of the Linux Virtual Server function is beyond the
+
-
scope of this document. For more information, see the Red Hat website at
+
-
[http://www.redhat.com/cluster_suite/ www.redhat.com/cluster_suite/].
+
-
Parallel Processing or High Performance Compute Clustering (HPCC)—
+
'''High Availability''' or '''Failover Clustering''' - Includes a set of nodes capable of taking over the workload from another node in the same cluster, if there is an unexpected failure. Red Hat cluster suite supports high availability cluster configuration. For more information, see the Red Hat cluster suite website at [http://www.redhat.com/cluster_suite/ www.redhat.com/cluster_suite/].
-
Typical HPC clusters are controlled by a master or front-end server. The
+
-
master server accepts job requests and divides them up among the compute
+
-
nodes to take advantage of parallel processing. Red Hat Cluster Suite does
+
-
not support HPCC architecture.
+
-
High availability or failover clustering— Includes a set of nodes capable of
+
-
taking over the workload from another node in the same highly available
+
-
cluster, if there is an unexpected failure. Red Hat cluster suite supports high
+
-
availability cluster configuration. For more information, see the Red Hat
+
-
cluster suite website at [http://www.redhat.com/cluster_suite/ www.redhat.com/cluster_suite/].
+
-
Storage Cluster - A cluster comprised of systems that are accessing the same
+
'''Storage Cluster''' - A cluster comprised of systems that are accessing the same shared storage such as Red Hat Global File System. For more information, see the Red Hat GFS website at: [http://www.redhat.com/gfs www.redhat.com/gfs].
-
shared storage such as Red Hat Global File System. For more information, see
+
-
the Red Hat GFS website at: [http://www.redhat.com/gfs www.redhat.com/gfs].
+
-
This document describes the procedure to configure a high availability and
+
This document describes a procedure to configure a cluster that provides both high availability and shared storage.
-
storage cluster.
+
==Supported Configurations and Requirements==
==Supported Configurations and Requirements==
The Dell|Red Hat High-availability cluster is comprised of PowerEdge systems
The Dell|Red Hat High-availability cluster is comprised of PowerEdge systems
-
running Red Hat Cluster Suite and a storage cluster running Red Hat
+
running Red Hat Clustering and a storage cluster running Red Hat
Global File System. You can configure the high-availability functionality in a
Global File System. You can configure the high-availability functionality in a
cluster without a storage cluster, but you cannot configure a storage cluster
cluster without a storage cluster, but you cannot configure a storage cluster
Line 45: Line 25:
===Operating System===
===Operating System===
-
This solution supports RHEL 5.1 AP.  The following list provides an overview of the supported configuration:
+
This solution supports RHEL 5.3 AP.  The following list provides an overview of the supported configuration:
* Minimum disk space 1 GB per node
* Minimum disk space 1 GB per node
Line 51: Line 31:
===Cluster Nodes===
===Cluster Nodes===
-
The Dell|Red Hat HA Linux Cluster solution uses PowerEdge systems accessing shared storage. Each PowerEdge system runs RHEL 5.1 AP. Table 1-1 provides the requirements for each PowerEdge system in the cluster.
+
The Dell|Red Hat HA Linux Cluster solution uses PowerEdge systems accessing shared storage. Each PowerEdge system runs RHEL 5.3 AP. Table 1-1 provides the requirements for each PowerEdge system in the cluster.
'''Table 1-1. Cluster Node Requirements'''
'''Table 1-1. Cluster Node Requirements'''
Line 59: Line 39:
|-
|-
|RAM
|RAM
-
|At least 512 MB RAM installed on each cluster node for RHEL 5.1 Advanced Platform.
+
|At least 512 MB RAM installed on each cluster node for RHEL 5.3 Advanced Platform.
|-
|-
|Network Interface Cards (NICs)
|Network Interface Cards (NICs)
Line 76: Line 56:
* Package management for nodes
* Package management for nodes
-
Additionally a dedicated management node can also provide the following functions, which are outside the scope of this document:
+
Additionally, a dedicated management node can provide the following functions, which are outside the scope of this document:
* Kickstart network installation server
* Kickstart network installation server
-
* Pre-execution environment (PXE) server
+
* Pre-boot eXecution Environment (PXE) server
-
* Dynamic host control protocol (DHCP) server
+
* Dynamic Host Configuration Protocol (DHCP) server
-
* Domain Naming Service (DNS)
+
* Domain Name Service (DNS) server
* Hardware instrumentation via OpenManage
* Hardware instrumentation via OpenManage
Line 90: Line 70:
Fencing is used to ensure data integrity on the shared storage. If a node does not respond to the cluster services in a timely manner, it is removed from the cluster. There are several possible methods to use for fencing. For the complete list of fencing devices, see the Red Hat website at [http://www.redhat.com/cluster_suite/hardware/ www.redhat.com/cluster_suite/hardware/].
Fencing is used to ensure data integrity on the shared storage. If a node does not respond to the cluster services in a timely manner, it is removed from the cluster. There are several possible methods to use for fencing. For the complete list of fencing devices, see the Red Hat website at [http://www.redhat.com/cluster_suite/hardware/ www.redhat.com/cluster_suite/hardware/].
-
The Dell|Red Hat HA Linux Cluster system is tested with the following devices:
+
The Dell|Red Hat HA Linux Cluster system has been tested with the following devices:
* Dell Remote Access Controller (DRAC) 5
* Dell Remote Access Controller (DRAC) 5
-
* PowerEdge Intelligent Platform Management Interface (IPMI)
+
* Chassis Management Controller (CMC)
 +
* Integrated Dell Remote Access Controller (iDRAC)
 +
* Intelligent Platform Management Interface (IPMI)
* APC® network power switch (AP7867A)
* APC® network power switch (AP7867A)
 +
* WTI® network power switch (IPS-800-D20)
'''NOTE''': It is recommended that you use two fencing methods, such as a
'''NOTE''': It is recommended that you use two fencing methods, such as a
-
network power switch and a remote access controller (DRAC or IPMI).
+
network power switch (APC or WTI) and a remote access controller (DRAC, iDRAC or IPMI).
==Other Documents You May Need==
==Other Documents You May Need==
Line 112: Line 95:
* For Red Hat documentation, see the following links:
* For Red Hat documentation, see the following links:
** Red Hat Enterprise Linux manuals: [http://www.redhat.com/docs/manuals/enterprise/ www.redhat.com/docs/manuals/enterprise/]
** Red Hat Enterprise Linux manuals: [http://www.redhat.com/docs/manuals/enterprise/ www.redhat.com/docs/manuals/enterprise/]
-
** Red Hat Hardware Certification: [http://www.redhat.com/rhel/compare/ www.redhat.com/rhel/compare/]
+
** Red Hat Hardware Certification: [http://hardware.redhat.com hardware.redhat.com]
-
** Red Hat Cluster Suite: [http://www.redhat.com/cluster_suite/ http://www.redhat.com/cluster_suite/]
+
** Red Hat Clustering: [http://www.redhat.com/cluster_suite/ http://www.redhat.com/cluster_suite/]
** Red Hat Global File System: [http://www.redhat.com/gfs/ www.redhat.com/gfs/]
** Red Hat Global File System: [http://www.redhat.com/gfs/ www.redhat.com/gfs/]
** Red Hat Network [http://www.redhat.com/docs/manuals/RHNetwork/ www.redhat.com/docs/manuals/RHNetwork/]
** Red Hat Network [http://www.redhat.com/docs/manuals/RHNetwork/ www.redhat.com/docs/manuals/RHNetwork/]
-
** Additional information is also available from the Red Hat trainingclass RH436—Red Hat Enterprise Clustering and Storage Management  
+
** Additional information is also available from the Red Hat training class RH436 - Red Hat Enterprise Clustering and Storage Management  
Release notes or readme files may be included to provide last-minute updates to the system documentation or advanced technical reference material intended for experienced users or technicians.
Release notes or readme files may be included to provide last-minute updates to the system documentation or advanced technical reference material intended for experienced users or technicians.
 +
 +
'''Proceed to the next section [[../System|System]]'''

Revision as of 20:43, 21 April 2009

Contents

Overview

This document provides instructions to use Red Hat® Enterprise Linux® (RHEL) 5.3 Advanced Platform (AP), Red Hat Clustering (RHCS), and Red Hat Global File System (GFS) to configure a Dell|Red Hat HA Linux Cluster with PowerEdge servers connected to Dell storage arrays.

Types of Clustering

Clustering is the process of configuring two or more systems to provide services as if they were on a single system. These systems are known as cluster nodes. You can use clustering to enable load balancing between nodes, parallel processing, high availability, and storage clustering.

Load Balancing - Load Balancing clusters accept requests from clients and use a defined algorithm to forward the requests to a pool of nodes. Red Hat Clustering contains Linux Virtual Server service that performs this function. Implementation of the Linux Virtual Server function is beyond the scope of this document. For more information, see the Red Hat website at www.redhat.com/cluster_suite/.

High Performance Compute Clustering (HPCC) or Parallel Processing Typical HPC clusters are controlled by a master or front-end server. The master server accepts job requests and divides them up among the compute nodes to take advantage of parallel processing. This document will not discuss HPCC. For more information, see dell.com/hpcc

High Availability or Failover Clustering - Includes a set of nodes capable of taking over the workload from another node in the same cluster, if there is an unexpected failure. Red Hat cluster suite supports high availability cluster configuration. For more information, see the Red Hat cluster suite website at www.redhat.com/cluster_suite/.

Storage Cluster - A cluster comprised of systems that are accessing the same shared storage such as Red Hat Global File System. For more information, see the Red Hat GFS website at: www.redhat.com/gfs.

This document describes a procedure to configure a cluster that provides both high availability and shared storage.

Supported Configurations and Requirements

The Dell|Red Hat High-availability cluster is comprised of PowerEdge systems running Red Hat Clustering and a storage cluster running Red Hat Global File System. You can configure the high-availability functionality in a cluster without a storage cluster, but you cannot configure a storage cluster without the high-availability functionality.

For the cluster configuration support matrix, see: http://www.dell.com/ha

Operating System

This solution supports RHEL 5.3 AP. The following list provides an overview of the supported configuration:

  • Minimum disk space 1 GB per node
  • Minimum required RAM is 512 MB per node

Cluster Nodes

The Dell|Red Hat HA Linux Cluster solution uses PowerEdge systems accessing shared storage. Each PowerEdge system runs RHEL 5.3 AP. Table 1-1 provides the requirements for each PowerEdge system in the cluster.

Table 1-1. Cluster Node Requirements

Component Minimum Requirement
RAM At least 512 MB RAM installed on each cluster node for RHEL 5.3 Advanced Platform.
Network Interface Cards (NICs) At least two NICs per cluster node - one NIC for the public network and another NIC for the private network.
Power Supplies At least two power supplies per node supplied by different power sources.

Management Node

It is recommended that you configure a dedicated management node running RHEL, but not required. The management node performs the following functions in your Dell|Red Hat HA Linux Cluster system configuration:

  • Manage and Monitor the cluster
  • Manage and Monitor the shared storage
  • Package management for nodes

Additionally, a dedicated management node can provide the following functions, which are outside the scope of this document:

  • Kickstart network installation server
  • Pre-boot eXecution Environment (PXE) server
  • Dynamic Host Configuration Protocol (DHCP) server
  • Domain Name Service (DNS) server
  • Hardware instrumentation via OpenManage

Cluster Storage

The Dell|Red Hat HA Linux Cluster system requires shared storage accessible by all cluster nodes.

Fencing Mechanisms

Fencing is used to ensure data integrity on the shared storage. If a node does not respond to the cluster services in a timely manner, it is removed from the cluster. There are several possible methods to use for fencing. For the complete list of fencing devices, see the Red Hat website at www.redhat.com/cluster_suite/hardware/.

The Dell|Red Hat HA Linux Cluster system has been tested with the following devices:

  • Dell Remote Access Controller (DRAC) 5
  • Chassis Management Controller (CMC)
  • Integrated Dell Remote Access Controller (iDRAC)
  • Intelligent Platform Management Interface (IPMI)
  • APC® network power switch (AP7867A)
  • WTI® network power switch (IPS-800-D20)

NOTE: It is recommended that you use two fencing methods, such as a network power switch (APC or WTI) and a remote access controller (DRAC, iDRAC or IPMI).

Other Documents You May Need

CAUTION: Read the Product Information Guide for important safety and regulatory information. Warranty information may be included within this document or as a separate document.

  • The Rack Installation Guide included with your rack solution describes how to install your system into a rack.
  • The Setting Up Your System document provides an overview of initially setting up your system.
  • The System Administrator's Guide provides system operation and management information.
  • Documentation for any components you purchased separately provides information to configure and install these options.
  • Updates are sometimes included with the system to describe changes to the system, software, and/or documentation.

NOTE: Always read the updates first because they often supersede information in other documents.

Release notes or readme files may be included to provide last-minute updates to the system documentation or advanced technical reference material intended for experienced users or technicians.

Proceed to the next section System

Personal tools
Distributions