TruCluster Server Handbook (HP Technologies)

Chapter 1: Introduction

Figure 1-1: ASE-style Cluster

Chapter 2: Tru64 UNIX & TruCluster Server Overview

Figure 2-1: Tru64 UNIX History

Figure 2-2: Computer Hardware Components

Figure 2-3: Dual-CPU Hardware Components

Figure 2-4: Asynchronous Multiprocessing

Figure 2-5: Symmetric Multiprocessing

Figure 2-6: Loosely-Coupled Computers

Figure 2-7: Two-Member Cluster

Figure 2-8: Cluster Interconnect

Figure 2-9: ICS to CI Subsystem Relationship

Figure 2-10: Connection Manager Subsystem Relationships

Figure 2-11: CAA Component Relationship

Figure 2-12: DLM Subsystem Relationships

Figure 2-13: CFS Subsystem Relationships

Figure 2-14: DRD Subsystem Relationships

Figure 2-15: CLSM Subsystem Relationships

Figure 2-16: CLUA Subsystem Relationships

Figure 2-17: Cluster Subsystem Components

Chapter 4: Cluster Configuration Planning

Figure 4-1: The Disk Configuration (diskconfig (8)) Program

Figure 4-2: RAID Controller Transparent Failover

Figure 4-3: RAID Controller Multi-bus Failover

Figure 4-4: Memory Channel 2 – Virtual Hub Configuration

Figure 4-5: Redundant MC Virtual Hub Configuration

Figure 4-7: Memory Channel 2 – Standard Hub Configuration

Figure 4-6: Redundant MC Standard Hub Configuration

Figure 4-8: LAN Cluster Interconnect

Chapter 5: Installation and Configuration of Tru64 UNIX

Figure 5-1: The Tru64 UNIX setup Program

Figure 5-2: The checklist Clipboard

Figure 5-3: SysMan Navigational Information

Figure 5-4: Quick Setup – Main Screen

Figure 5-5: Quick Setup – OSF-BASE License PAK

Figure 5-6: Quick Setup – NIC Setup

Figure 5-7: Quick Setup – Network Routing

Figure 5-8: Quick Setup – DNS/BIND

Figure 5-9: Quick Setup – NTP Server

Figure 5-10: Quick Setup – NIS

Figure 5-11: Quick Setup – NFS

Figure 5-12: Quick Setup – E-mail Server

Figure 5-13: Quick Setup – Default Printer & Print Server

Figure 5-14: Quick Setup – Summary Pages (1 & 2)

Figure 5-15: SysMan Menu

Chapter 6: Tru64 UNIX Cluster Hooks—File System Hierarchy, CDSL, & PID

Figure 6-1: CDSL Resolved

Figure 6-2: Runlevel Initialization

Figure 6-3: Runtime Configuration Execution Flow

Figure 6-4: File System Hierarchy (V5+)

Figure 6-5: The member0 directory hierarchy

Figure 6-6: PID Structure Illustrated

Chapter 7: Tru64 UNIX Cluster Hooks—Device Naming & Hardware Management

Figure 7-1: Device Directory Hierarchy

Figure 7-2: WWID for the HSZ explained

Chapter 8: Tru64 UNIX Cluster Hooks— Event Manager

Figure 8-1: EVM Event Path

Figure 8-2: EVM root directory hierarchy

Figure 8-3: EVM /usr directory hierarchy

Figure 8-4: EVM /var directory hierarchy

Chapter 9: Tru64 UNIX Cluster Hooks—NIFF, NetRAIN, & LAG

Figure 9-1: NIFF Flow

Figure 9-2: Traditional Network Interface Configuration

Figure 9-3: NetRAIN Interface Configuration

Chapter 11: Adding a Cluster Member/Deleting a Cluster Member

Figure 11-1: SCSI Device Location vs. Naming

Chapter 12: Cluster Hooks Revisited

Figure 12-1: File System Hierarchy in a Cluster

Figure 12-2: Hardware and Device Special Files Database Locations

Figure 12-3: EVM Additions in a Cluster

Chapter 13: The Cluster File System (CFS)

Figure 13-1: The CFS I/O Architecture

Figure 13-2: Direct I/O in V5.0A

Figure 13-3: Concurrent Direct I/O

Figure 13-4: Direct Access Cached Reads

Figure 13-5: Direct Access Cached Reads – No Physical Connection

Chapter 14: The Cluster Logical Storage Manager (CLSM)

Figure 14-1: What is RAID?

Figure 14-2: Cluster Device I/O Subsystem Relationships

Figure 14-3: Relationship of LSM Objects

Figure 14-4: LSM Disk Types

Figure 14-5: LSM Concatenated and Striped Plexes

Figure 14-6: CLSM Disk Group Connectivity

Figure 14-7: Top-Down vs. Bottom-Up

Chapter 15: The Device Request Dispatcher (DRD)

Figure 15-1: Cluster-wide Device Naming

Figure 15-2: TruCluster Server I/O Subsystems Architecture

Figure 15-3: Direct Access I/O

Figure 15-4: DAIO Device, Private Bus

Figure 15-5: Served Device, Shared Bus

Figure 15-6: DAIO Device on a Semi-Shared Bus in a Four-Member Cluster

Figure 15-7: DAIO Device on a Semi-Shared Bus – After Access Node Change

Figure 15-8: Load Balancing Multiple Devices on a Semi-Shared Bus

Figure 15-9: Served Device on a Shared bus in a Two-Member Cluster

Figure 15-10: Served Device on a Shared Bus in a Two-Member Cluster after Server Relocation

Chapter 16: The Cluster Alias Subsystem (CLUA)

Figure 16-1: Cluster Alias IP to MAC

Figure 16-2: Gratuitous ARP Ignored or Denied

Figure 16-3: Client ARP Cache Refreshed through Gratuitous ARP

Figure 16-4: vMAC in Action

Figure 16-5: Common Subnet: Physical vs. Logical View

Figure 16-6: Virtual Subnet: Physical vs. Logical View

Figure 16-7: Cluster Alias Connection Decision Tree

Chapter 17: The Connection Manager

Figure 17-1: CNX - ICS Communication

Figure 17-2: CNX Cluster Subsystem Communication

Figure 17-3: CNX Rebuild and Callouts

Figure 17-4: Cluster Partition

Figure 17-5: CNX Node Announcement

Figure 17-6: Describe Nodes

Figure 17-7: Connectivity Matrix (aka Topology Bitmap)

Figure 17-8: Cluster Formation Transaction

Figure 17-9: Send a Join Request

Figure 17-10: Describe Member & Cluster

Figure 17-11: Member Join Transaction

Figure 17-12: "clu_quorum -v" explained

Figure 17-13: Four-Member Cluster without a Quorum Disk

Figure 17-14: Four-Member Cluster without a Quorum Disk – One Member Down

Figure 17-15: Four-Member Cluster without a Quorum Disk – Two Members Down

Figure 17-16: Four-Member Cluster with a Quorum Disk

Figure 17-17: Four-Member Cluster with a Quorum Disk – Two Members Down

Chapter 18: Miscellaneous Subsystems

Figure 18-1: ICS Cluster Subsystem Communication

Figure 18-2: ICS Cluster Subsystem Communication – Detailed

Figure 18-3: DLM – Create a Lock

Figure 18-4: DLM – Directing Lock Request to the Lock Master

Figure 18-5: DLM Cluster Subsystem Communication

Figure 18-6: KGS Cluster Subsystem Communication

Figure 18-7: KGS – "I feel the need for a donut."

Figure 18-8: KGS – "Let's all go to the Donut Shop."

Figure 18-9: KGS – "How many of you want to go?"

Figure 18-10: KGS – "We're going for donuts."

Chapter 19: System Administration Tasks

Figure 19-1: SysMan Menu

Chapter 20: Network Administration Tasks

Figure 20-1: NFS Client – CFS Server

Chapter 21: Cluster Administration Tasks

Figure 21-1: SysMan Station – Main Screen

Figure 21-2: SysMan Station – Color/Icon Descriptions

Figure 21-3: SysMan Station – Hardware View

Figure 21-4: SysMan Menu Popups

Figure 21-5: SysMan Station -- Physical File System View

Chapter 23: Cluster Application Availability (CAA)

Figure 23-1: The CAA Architecture

Figure 23-2: CAA Directories and Files

Chapter 24: CAA by Example

Figure 24-1: The xhostname Tcl/Tk Script

Figure 24-2: CAA Main Window

Figure 24-3: CAA Setup Window

Figure 24-4: CAA Setup: Add Resource Popup and Resource Type Pulldown Menu

Figure 24-5: CAA Setup Resources

Figure 24-6: CAA Resource Options Pages

Figure 24-7: CAA Informational Popup

Figure 24-8: CAA Resource Registration Popup and Start Resource Informational Popup

Chapter 25: Performing a Rolling Upgrade

Figure 25-1: The Rolling Upgrade Flow

Chapter 26: Migrating to TruCluster Server

Figure 26-1: V1.[56] Cluster

Figure 26-2: Preparing for the New Cluster – Install Tru64 UNIX

Figure 26-3: Create a One-Member TruCluster Server Cluster

Figure 26-4: Create and Test Application Resources

Figure 26-5: Save the Current Cluster's Configuration Information

Figure 26-6: Apply the Configuration Data to the New Cluster

Figure 26-7: Connect the Second System to the New Cluster

Figure 26-8: Add the Second System to the New Cluster

Figure 26-9: Welcome to TruCluster Server!

Figure 26-10: Failed Migration Recovery

Appendix A: TruCluster Server Troubleshooting

Figure A-1: Memory Channel 2 Hub Configuration, Node ID 2 (Hub Slot 2)

Figure A-2: Memory Channel 2 Hub Configuration, Node ID 0 (Hub Slot 0)

Figure A-3: Memory Channel 2 Virtual Hub Configuration, VH0

Figure A-4: Memory Channel 2 Virtual Hub Configuration, VH1

Figure A-5: Memory Channel 2 Memory Window Size of 512MB