Title and Copyright Information
 
About This Manual
Audience
New and Changed Features
Organization
Related Documents
Reader's Comments
Conventions
 
1    Managing Clusters Overview
1.1    Commands and Utilities for Clusters
1.2    Commands and Features That Are Different in a Cluster
 
2    Tools for Managing Clusters
2.1    Overview of Available Management Tools and Interfaces
2.1.1    Cluster Tools Quick Start
2.1.2    HP Insight Manager Integration
2.1.3    HP Insight Manager XE Integration
2.1.4    Available Management Tools and Interfaces
2.2    Cluster Configuration Tools and Related User Interfaces
2.3    Overview of SysMan Components
2.3.1    Introduction to SysMan Menu
2.3.2    Introduction to SysMan Station
2.3.3    Introduction to the SysMan Command Line
2.4    Using SysMan Menu in a Cluster
2.4.1    Getting in Focus
2.4.2    Specifying a Focus on the Command Line
2.4.3    Invoking SysMan Menu
2.5    Using SysMan Station in a Cluster
2.5.1    Invoking SysMan Station
2.6    Using the SysMan Java Applets in a Cluster
2.6.1    Invoking the SysMan Java Applets
2.7    Using SysMan Java PC Applications in a Cluster
2.7.1    Invoking SysMan Java Applications on a PC
2.8    Using SysMan Command-Line Interface in a Cluster
2.9    Using HP Insight Manager in a Cluster
2.9.1    Invoking HP Insight Manager
2.10    Using Tru64 UNIX Configuration Report
 
3    Managing the Cluster Alias Subsystem
3.1    Summary of Cluster Alias Features
3.2    Configuration Files
3.3    General Planning Considerations
3.4    Things to Do Before You Create an Alias
3.5    Specifying and Joining a Cluster Alias
3.6    Modifying Cluster Alias and Service Port Attributes
3.7    Monitoring Cluster Aliases
3.8    Leaving a Cluster Alias
3.9    Deleting a Cluster Alias
3.10    Application Load Balancing
3.11    Extending Clusterwide Port Space
3.12    Enabling Cluster Alias vMAC Support
3.13    Routing Configuration Guidelines
3.14    Routing Daemon and Static Routing Options
3.15    Cluster Alias and NFS
3.16    Cluster Alias and Cluster Application Availability
 
4    Managing Cluster Membership
4.1    Overview of Connection Manager
4.2    Overview of Quorum and Votes
4.2.1    How a System Becomes a Cluster Member
4.2.2    Expected Votes
4.2.3    Current Votes
4.2.4    Node Votes
4.2.5    Quorum Disk Votes
4.3    Overview of Calculating Cluster Quorum
4.4    Connection Manager Example
4.5    Using a Quorum Disk
4.5.1    Replacing a Failed Quorum Disk
4.6    Using the clu_quorum Command to Display Cluster Vote Information
4.7    Cluster Vote Assignment Examples
4.8    Monitoring the Connection Manager
4.9    Connection Manager Panics
4.10    Troubleshooting Unfortunate Expected Vote and Node Vote Settings
4.10.1    Joining a Cluster After a Cluster Member or Quorum Disk Fails and Cluster Loses Quorum
4.10.2    Forming a Cluster When Members Do Not Have Enough Votes to Boot and Form a Cluster
 
5    Managing Cluster Members
5.1    Managing Configuration Variables
5.2    Managing Kernel Attributes
5.3    Managing Remote Access Within and From the Cluster
5.4    Shutting Down the Cluster
5.5    Shutting Down and Starting One Cluster Member
5.5.1    Identifying a Critical Voting Member
5.5.2    Preparing to Halt or Delete a Critical Voting Member
5.5.3    Halting a Noncritical Member
5.5.4    Shutting Down a Hosting Member
5.6    Shutting Down a Cluster Member to Single-User Mode
5.7    Rebooting Cluster Members
5.8    Deleting a Cluster Member
5.9    Removing a Cluster Member and Restoring It as a Standalone System
5.10    Moving a Cluster to Another IP Subnet
5.10.1    File Edit Tables
5.10.1.1    Changing External IP Addresses Only
5.10.1.2    Changing External IP Addresses and Host Names
5.10.1.3    Changing External and Internal IP Addresses and Host Names
5.10.2    Attribute and Checklist Tables
5.10.2.1    External Host Names and IP Addresses
5.10.2.2    Cluster Name and Cluster Aliases
5.10.2.3    Interface IP Aliases
5.10.2.4    External Servers
5.10.2.5    Checklist
5.10.3    Verifying Success
5.10.4    Troubleshooting
5.11    Changing the Cluster Name or IP Address
5.11.1    Changing the Cluster Name
5.11.2    Changing the Cluster IP Address
5.12    Changing the Member Name, IP Address, or Cluster Interconnect Address
5.12.1    Check Quorum Before Removing Member
5.12.2    Check CAA Restricted Placement Policies Before Removing Member
5.12.3    Remove and Add the Member
5.13    Managing Software Licenses
5.14    Installing and Deleting Layered Applications
5.15    Managing Accounting Services
 
6    Managing Networks in a Cluster
6.1    Updating ifaccess.conf When Changing Network Interfaces
6.2    Providing Failover for Network Interfaces
6.3    Running IP Routers
6.4    Configuring the Network
 
7    Managing Network Services
7.1    Configuring DHCP
7.2    Configuring NIS
7.2.1    Configuring an NIS Master in a Cluster with Enhanced Security
7.3    Configuring Printing
7.4    Configuring DNS/BIND
7.5    Managing Time Synchronization
7.5.1    Configuring NTP
7.5.2    Using the Same External NTP Servers for All Members
7.5.2.1    Time Drift
7.6    Managing NFS
7.6.1    Configuring NFS
7.6.2    Considerations for Using NFS in a Cluster
7.6.2.1    CFS Support of NFS File Systems
7.6.2.2    Clients Must Use a Cluster Alias
7.6.2.3    Using CDSLs to Mount NFS File Systems
7.6.2.4    Loopback Mounts Not Supported
7.6.2.5    Do Not Mount Non-NFS File Systems on NFS-Mounted Paths
7.6.3    Using AutoFS in a Cluster
7.6.3.1    Forcibly Unmounting File Systems
7.6.3.1.1    Determining Whether a Forced Unmount Is Required
7.6.3.1.2    Correcting the Problem
7.6.4    Migrating from Automount to AutoFS
7.6.4.1    Migrating Without a Reboot
7.6.4.2    Migrating When Rebooting a Cluster Member
7.6.4.3    Migrating When Rebooting the Cluster
7.7    Managing inetd Configuration
7.8    Managing Mail
7.8.1    Configuring Mail
7.8.1.1    Mail Files
7.8.1.2    The Cw Macro (System Nicknames List)
7.8.1.3    Configuring Mail at Cluster Creation
7.8.1.4    Configuring Mail After the Cluster Is Running
7.8.2    Distributing Mail Load Among Cluster Members
7.9    Configuring a Cluster for RIS
7.10    Displaying X Window Applications Remotely
 
8    Managing Highly Available Applications
8.1    Learning the Status of a Resource
8.1.1    Learning the State of a Resource
8.1.2    Learning Status of All Resources on One Cluster Member
8.1.3    Learning Status of All Resources on All Cluster Members
8.1.4    Getting Number of Failures and Restarts and Target States
8.2    Reporting Resource Availability Measurements
8.3    Relocating Applications
8.3.1    Manual Relocation of All Applications on a Cluster Member
8.3.2    Manual Relocation of a Single Application
8.3.3    Manual Relocation of Dependent Applications
8.4    Starting and Stopping Application Resources
8.4.1    Starting Application Resources
8.4.2    Stopping Application Resources
8.4.3    No Multiple Instances of an Application Resource
8.4.4    Using caa_stop to Reset UNKNOWN State
8.5    Balancing Application Resources
8.5.1    Balancing All Applications on a Cluster
8.5.2    Balancing All Applications on a Member
8.5.3    Balancing Specific Applications
8.5.4    Balancing an Application at Specific Time
8.6    Registering and Unregistering Resources
8.6.1    Registering Resources
8.6.2    Unregistering Resources
8.6.3    Updating Registration
8.7    Network, Tape, and Media Changer Resources
8.8    Using SysMan to Manage CAA
8.8.1    Managing CAA with SysMan Menu
8.8.1.1    CAA Management Dialog Box
8.8.1.2    Start Dialog Box
8.8.1.3    Setup Dialog Box
8.8.2    Managing CAA with SysMan Station
8.8.2.1    Starting an Application with SysMan Station
8.8.2.2    Resource Setup with SysMan Station
8.9    CAA Considerations for Startup and Shutdown
8.10    Managing caad
8.10.1    Determining Status of the Local CAA Daemon
8.10.2    Restarting the CAA Daemon
8.10.3    Monitoring CAA Daemon Messages
8.11    Using EVM to View CAA Events
8.11.1    Viewing CAA Events
8.11.2    Monitoring CAA Events
8.12    Troubleshooting with Events
8.13    Troubleshooting a Command-Line Message
 
9    Managing File Systems and Devices
9.1    Working with CDSLs
9.1.1    Making CDSLs
9.1.2    Maintaining CDSLs
9.1.3    Kernel Builds and CDSLs
9.1.4    Exporting and Mounting CDSLs
9.2    Managing Devices
9.2.1    Managing the Device Special File
9.2.2    Determining Device Locations
9.2.3    Adding a Disk to the Cluster
9.2.4    Managing Third-party Storage
9.2.5    Tape Devices
9.2.6    Formatting Diskettes in a Cluster
9.2.7    CD-ROM and DVD-ROM
9.3    Managing the Cluster File System
9.3.1    When File Systems Cannot Fail Over
9.3.2    Direct Access Cached Reads
9.3.3    CFS Load Balancing
9.3.3.1    Starting and Stopping cfsd
9.3.3.2    EVM Events
9.3.3.3    Modifying the /etc/cfsd.conf Configuration File
9.3.3.4    Understanding cfsd Analysis and Implementing Recommendations
9.3.3.5    Automatic Relocations
9.3.3.6    Relationship to CAA Resources
9.3.3.7    Balancing CFS Load Without cfsd
9.3.3.8    Distributing CFS Server Load via cfsmgr
9.3.4    Distributing File Systems Via the mount -o Command
9.3.5    Freezing a Domain Before Cloning
9.3.5.1    Determining Whether a Domain Is Frozen
9.3.6    Optimizing CFS Performance
9.3.6.1    Changing the Number of Read-Ahead and Write-Behind Threads
9.3.6.2    Taking Advantage of Direct I/O
9.3.6.2.1    Differences Between Cluster and Standalone AdvFS Direct I/O
9.3.6.2.2    Cloning a Fileset with Files Open in Direct I/O Mode
9.3.6.2.3    Gathering Statistics on Direct I/O
9.3.6.3    Adjusting CFS Memory Usage
9.3.6.4    Using Memory Mapped Files
9.3.6.5    Avoiding Full File Systems
9.3.6.6    Other Strategies
9.3.7    MFS and UFS File Systems Supported
9.3.8    Partitioning File Systems
9.3.9    Block Devices and Cache Coherency
9.3.10    CFS Restrictions
9.4    Managing the Device Request Dispatcher
9.4.1    Direct-Access I/O and Single-Server Devices
9.4.1.1    Devices Supporting Direct-Access I/O
9.4.1.2    Replacing RZ26, RZ28, RZ29, or RZ1CB-CA as Direct-Access I/O Disks
9.4.1.3    HSZ Hardware Supported on Shared Buses
9.5    Managing AdvFS in a Cluster
9.5.1    Integrating AdvFS Files from a Newly Added Member
9.5.2    Create Only One Fileset in Cluster Root Domain
9.5.3    Adding Filesets to a Member's Boot Partition Not Recommended
9.5.4    Do Not Add a Volume to a Member's Root Domain
9.5.5    Using the addvol and rmvol Commands in a Cluster
9.5.6    User and Group File System Quotas Are Supported
9.5.6.1    Quota Hard Limits
9.5.6.2    Setting the quota_excess_blocks Value
9.5.7    Storage Connectivity and AdvFS Volumes
9.6    Considerations When Creating New File Systems
9.6.1    Verifying Disk Connectivity
9.6.2    Looking for Available Disks
9.6.2.1    Looking for the Location of the Quorum Disk
9.6.2.2    Looking for the Location of Member Boot Disks and Clusterwide AdvFS File Systems
9.6.2.3    Looking for Member Swap Areas
9.6.3    Editing /etc/fstab
9.7    Managing CDFS File Systems
9.8    Backing Up and Restoring Files
9.8.1    Suggestions for Files to Back Up
9.9    Managing Swap Space
9.9.1    Locating Swap Device for Improved Performance
9.10    Fixing Problems with Boot Parameters
9.11    Using the verify Utility in a Cluster
9.11.1    Using the verify Utility on Cluster Root
 
10    Using Logical Storage Manager in a Cluster
10.1    Differences Between Managing LSM in Clusters and in Standalone Systems
 
11    Troubleshooting Clusters
11.1    Resolving Problems
11.1.1    Booting Systems Without a License
11.1.2    Shutdown Leaves Members Running
11.1.3    Environmental Monitoring and Cluster Shutdowns
11.1.4    Dealing with CFS Errors at Boot
11.1.5    Backing Up and Repairing a Member's Boot Disk
11.1.5.1    Example of Recovering a Member's Boot Disk
11.1.6    Specifying cluster_root at Boot Time
11.1.7    Recovering the Cluster Root File System to a Disk Known to the Cluster
11.1.8    Recovering the Cluster Root File System to a New Disk
11.1.9    Dealing with AdvFS Problems
11.1.9.1    Responding to Warning Messages from addvol or rmvol
11.1.9.2    Resolving AdvFS Domain Panics Due to Loss of Device Connectivity
11.1.9.3    Forcibly Unmounting an AdvFS File System or Domain
11.1.9.4    Avoiding Domain Panics
11.1.10    Accessing Boot Partitions on Down Systems
11.1.11    Booting a Member While Its Boot Disk Is Already Mounted
11.1.12    Generating Crash Dumps
11.1.12.1    Generating a Crash Dump When a Member Is Hung
11.1.13    Fixing Network Problems
11.1.14    Running routed in a Cluster
11.2    Hints for Managing Clusters
11.2.1    Moving /tmp
11.2.2    Running the MC_CABLE Console Command
11.2.3    Korn Shell Does Not Record True Path to Member-Specific Directories
 
A    Cluster Events
 
B    Configuration Variables
 
C    clu_delete_member Log
 
D    LAN Cluster Administration
D.1    Configuring a NetRAIN Virtual Interface for a Cluster LAN Interconnect
D.2    Tuning the LAN Interconnect
D.2.1    Improving Cluster Interconnect Performance by Setting Its ipmtu Value
D.2.2    Set Ethernet Switch Address Aging to 15 Seconds
D.3    Obtaining Network Adapter Configuration Information
D.4    Monitoring LAN Interconnect Activity
D.5    Migrating from Memory Channel to LAN
D.6    Migrating from LAN to Memory Channel
D.7    Migrating from Fast Ethernet LAN to Gigabit Ethernet LAN
D.8    Troubleshooting
D.8.1    Many Broadcast Errors on Booting or Booting New Member Panics Existing Member
D.8.2    Cannot Manually Fail Over Devices in a NetRAIN Virtual Interface
D.8.3    Applications Unable to Map to Port
 
Examples
2-1    Example sysman Output
9-1    Sample /etc/cfsd.conf File
 
Figures
2-1    SysMan Menu Hierarchy
2-2    SysMan Menu Interfaces
2-3    SysMan Station Graphical Interface
2-4    SysMan Station Initial Cluster View
2-5    Sample SysMan Station Cluster Hardware View
2-6    Displaying Available Actions in SysMan Station
2-7    HP Insight Manager Display
2-8    Sample Configuration Report Display
4-1    Three-Member deli Cluster
4-2    Three-Member deli Cluster Loses a Member
4-3    Two-Member deli Cluster Without a Quorum Disk
4-4    Two-Member deli Cluster with Quorum Disk Survives Member Loss
8-1    CAA Branch of SysMan Menu
8-2    CAA Management Dialog Box
8-3    Start Dialog Box
8-4    Setup Dialog Box
8-5    SysMan Station CAA_Applications_(active) View
8-6    SysMan Station CAA_Applications_(all) View
8-7    SysMan Station CAA Setup Screen
9-1    SysMan Station Display of Hardware Configuration
9-2    Cluster with Semi-private Storage
9-3    Four Node Cluster
 
Tables
1-1    Cluster Commands
1-2    File Systems and Storage Differences
1-3    Networking Differences
1-4    Printing Differences
1-5    Security Differences
1-6    General System Management Differences
1-7    Features Not Supported
2-1    Cluster Tools Quick Start
2-2    Available Management Tools and Interfaces
2-3    Cluster Management Tools
2-4    Invoking SysMan Menu
2-5    Invoking SysMan Station
4-1    Effects of Various Member cluster_expected_votes Settings and Vote Assignments in a Two- to Four-Member Cluster
4-2    Examples of Resolving Quorum Loss in a Cluster with Failed Members or Quorum Disk
4-3    Examples of Repairing a Quorum Deficient Cluster by Booting a Member with Sufficient Votes to Form the Cluster
5-1    /etc/rc.config* Files
5-2    Kernel Attributes Not to Decrease
5-3    Configurable TruCluster Server Subsystems
8-1    Target and State Combinations for Application Resources
8-2    Target and State Combinations for Network Resources
8-3    Target and State Combinations for Tape and Media Changer Resources
9-1    Sources of Information of Storage Device Management
11-1    Member Boot Disk Partitions
B-1    Cluster Configuration Variables
 
Index