Duration: 5 Days
In this course, you will learn how to effectively use Serviceguard to protect mission critical applications from a wide variety of hardware and software failures. You will spend 60 percent of class using HP servers in hands-on labs. This course has been updated to support HP-UX 11.31 and Serviceguard version A.11.20 but is still applicable to earlier versions of the product.
What You Will Learn
- Configure a volume group or disk group that can be used on multiple systems
and a basic package to run in a Serviceguard environment
- Configure and
maintain a Serviceguard cluster
- Configure and implement an application
monitor
- Replace a failed LVM lock disk
- Change the cluster configuration
and add/delete a package to a running cluster
- Set up a Serviceguard package
using the NFS toolkit and Oracle toolkit
- Configure a highly available network
using redundant hubs, routers, and networks
- Perform a rolling upgrade
- Use
Serviceguard Manager to manage a Serviceguard cluster
- Perform troubleshooting
activities to resolve Serviceguard configuration problems
- Protect your
mission critical applications against a wide variety of hardware and software
failures through effective use of Serviceguard
- Deliver highly available
application services to your LAN-attached clients by configuring up to 16-nodes
in an enterprise cluster
- Reduce your application downtime to near zero by
learning how to configure your Serviceguard cluster and using Serviceguard's
rolling upgrade facility
- Minimize, and in some instances eliminate, your
application downtime by automating the detection of failures and restoration of
application service
Audience
HP-UX system administrators who currently, or soon will, develop, design, implement, and monitor Serviceguard clusters
Prerequistes
- HP-UX System and Network Administration I
- HP-UX System and Network Administration II
or
- HP-UX for Experienced UNIX System Administrators
- HP-UX Logical Volume Manager (LVM) or HP-UX VERITAS Volume Manager
(VxVM)
- POSIX Shell Programming
Course Outline
1. High Availability
- What Are the Risks?
- Reducing the Risk
- Redundant Data
- Asymmetric Clusters
- Minimizing Downtime
- Network Redundancy
- Disk Configurations
- Multi-Network Environment
- Redundant
- LAN Cards
- Hubs
- Routers
- Client Networks
2.
High Availability with Serviceguard
- Serviceguard
- High Availability with
Serviceguard
- Features and Benefits of Serviceguard
- Serviceguard Packages
- Redistributing Application Packages
- Minimizing Planned Downtime
- Serviceguard Bundle/Products
- HP Serviceguard Solutions
- Serviceguard and
Integrity VM
- Cluster File System
- Multi-Node Packages and Package
Dependencies
- Oracle RAC and CFS
- Storage Choices Based on Your Priorities
- SGeRAC and LVM MORE
- System Management Homepage
- Serviceguard Manager
- Serviceguard Storage Management Suite
- Metrocluster
- Continentalclusters
3. Storage for Servicegard
- Serviceguard Disk
Space Management
- Boot and Root Disk Concepts
- General Disk Concepts
- Volume Management in Serviceguard for HP-UX
- LUN, Disk, and DVD DSF Names for
11.31
- Review of LVM Concepts
- Configure a Shareable LVM Volume Group
- LVM
Issues with Serviceguard
- Common LVM Commands
- LVM and VxVM Command
Comparison
- Configure a Shareable VxVM Disk Group
- CFS Filesystem
4.
Cluster Concepts and Configuration
- Cluster
- Major Components of a Cluster
- Network Interface Configuration
- Cluster Lock Configuration Using LVM Disks
- Cluster Lock Configuration Using a Quorum Server
- Quorum Server Redundancy
- Cluster Lock Configuration Using Lock LUN
- Lock LUN Support for HP-UX
- Lock
LUN Rules
- Comparison
- Heartbeat Configuration
- cmcld Process
- Cluster
Formation Requirements
- Steps to Configure a Cluster
- Cluster Configuration
Procedure
- Viewing the Cluster: cmviewcl Command
- Checking the Cluster Log
5. Additional Cluster Features
- Serviceguard
Volume Groups
- Marking Volume Groups for Use in Serviceguard
- Exclusive Mode
Volume Group Activation
- Cluster Formation and Reformations
- Ways to
Initially Form the Cluster
- Node Failures and Node Joins
- Cluster
Reformation ExamplevLocal LAN Card Failover
- Local LAN Card Failover
- Network Flow to
Standby LAN Card
6. Packages and Services
- Packaging Concepts
- Sample Package Configuration
- Sample Configuration after
Node Failure
- Package Switching
- Viewing Package Status
- Package Owner and
State
- Modifying Package Status
- Review Commands for Controlling a Cluster
- Configuration of Packages
- Package Modules Types
- Modularized Package
Control Script
- Modularized Package Control File
- Package Configuration
Procedure
- Create the Package Configuration File
- Edit the Package
Configuration File
- Verify and Distribute the Binary File
- Enhancements to
cmapplyconf/cmcheckconf
- Package Script Log File
7.
Package Policies
- Package Policies
- Package Type
- Failover Policies
- Failback Policies
- Automatic Failback
- Access Control Policies
- Node Fail
Fast and Service Fail Fast
- Package Dependencies
- Dependency Rules
- Complex Dependencies
- Cross Node Dependencies
- Exclusionary Dependency
- cmapplyconf Warning
- Complex Dependency Challenge
- Package Priorities
- Setting Package Priority
- Resource Dependency
- cmrunpkg
- Package
Weight/Node capacity
- Node Capacity/Package Weight Challenge
- Configuration
File
- Node Capacity Configuration
- Package Weight Configuration Methods
- Package Weight/Node Capacity Rules
- Capacity "ackage_limit"
- Package
Priority and Package Weight Rules
- Priority and Weight Challenge
8.
Application Monitoring Scripts and ECMT Toolkits
- Package Startup and Shutdown
- Rules for Service Processes
- Application Monitoring Script
- HA Toolkit
- ECMT Version B.06.00
- Modular Toolkit Architecture
- Modular Oracle DB
Toolkit Deployment
- Configuration Directory Operations for Modular Package
9. Cluster Troubleshooting
- Troubleshooting in
Serviceguard
- Approaches to Identifying Problems
- Double-Checking Supported
Configurations
- Log Files
- Monitoring the syslog File
- Monitoring the
Package Log File
- Useful Troubleshooting Commands
- Using cmviewconf and
cmgetconf
- Using cmquerycl and cmcheckconf
- Using cmviewcl
- Common Problem
Categories
- System Administration Errors
- Package Control Script Hangs or
Failures
- Package Log File Enhancements
- Causes of Cluster Reformations
- Approaches to Fixing Problems
- Resolving Serviceguard Command Hangs
- Modifying Debugging Options
- Modifying Startup Debugging Options
- Starting
Applications Outside of a Serviceguard Package
- cmgetconf
- Built-In Safety
Net
- Patch Recommendations
- Common Cluster Configuration Issues
- Testing
- Cluster Operations
- Package
Manager
- Cluster Manager
- Network
Manager
- Notification for Package Failure
10. Cluster and
Package Online Reconfiguration
- Serviceguard Online Reconfiguration
- Online
Cluster Reconfiguration
- Storage Reconfiguration
- Add a Node while a Cluster
is Running
- Remove a Node while a Cluster is Running
- Add a Package while a
Cluster is Running
- Remove a Package while a Cluster is Running
- Modify a
Package while the Cluster and Package are Running
- Modify a Package while the
Cluster is Running, but the Package is Down
- New Preview Functionality
- SG
commands - t Option
11. Highly Available NFS
- Highly Available
NFS Server Package
- Using the modular NFS Server Toolkit
- Highly Available
NFS Client Package
12. Highly Available Oracle Database
- Highly
Available Oracle Package
- Serviceguard Toolkits
- ECMT Modular Toolkit
Contents
- Create Oracle Package
13. WBEM, EMS Resources, and
Serviceguard Packages
- EMS
- Configuration of EMS Requests
- Select EMS
Resources
- Select EMS Configuration
- Resource Dependency
- Setting up a
Package to Use an EMS Resource
- WBEM
- WBEM Services Value Proposition
- WBEM relative to SNMP, DMI, and EMS
- HP WBEM Based Enterprise Management
- HP-UX 11i WBEM Providers
14. High Availability Networking
- Network Redundancy
- Multi-Network Environment
- Redundant LAN Ca
Course Labs
