You are previewing Implementing High Availability and Disaster Recovery in IBM PureApplication Systems V2.
O'Reilly logo
Implementing High Availability and Disaster Recovery in IBM PureApplication Systems V2

Book Description

This IBM Redbooks publication describes and demonstrates common, prescriptive scenarios for setting up disaster recovery for common workloads using IBM WebSphere Application Server, IBM DB2, and WebSphere MQ between two IBM PureApplication System racks using the features in PureApplication System V2.

The intended audience for this book is pattern developers and operations team members who are setting up production systems using software patterns from IBM that must be highly available or able to recover from a disaster (defined as the complete loss of a data center).

Table of Contents

  1. Front cover
  2. Notices
    1. Trademarks
  3. IBM Redbooks promotions
  4. Preface
    1. Authors
    2. Now you can become a published author, too!
    3. Comments welcome
    4. Stay connected to IBM Redbooks
  5. Chapter 1. Overview
    1. 1.1 Define high availability and disaster recovery
      1. 1.1.1 High availability
      2. 1.1.2 Disaster recovery
    2. 1.2 PureApplication System support for HA and DR
      1. 1.2.1 IBM General Parallel File System (GPFS)
      2. 1.2.2 Shared service for GPFS
      3. 1.2.3 Block storage
      4. 1.2.4 Block storage replication
      5. 1.2.5 External storage
      6. 1.2.6 Multisystem deployments
    3. 1.3 Backup and recovery
    4. 1.4 Always On (Continuous Availability)
      1. 1.4.1 Always On Principles (Continuous Availability)
      2. 1.4.2 Always On Patterns
    5. 1.5 Overview of HADR use case scenarios
  6. Chapter 2. High availability and disaster recovery capabilities of PureApplication System V2.0
    1. 2.1 Storage volumes
      1. 2.1.1 The new type of storage volumes in PureApplication System V2.0
    2. 2.2 Block storage overview
      1. 2.2.1 Block storage replication
    3. 2.3 Block storage replication
      1. 2.3.1 Planned failover
      2. 2.3.2 Unplanned failover
    4. 2.4 External storage
    5. 2.5 GPFS Overview
    6. 2.6 GPFS topologies
    7. 2.7 Active/Active GPFS deployment
    8. 2.8 Active/Passive GPFS deployment
    9. 2.9 Shared Service for GPFS
    10. 2.10 GPFS file systems and file sets
    11. 2.11 Load balancing
  7. Chapter 3. High availability and disaster recovery scenarios
    1. 3.1 Overview for the scenarios
      1. 3.1.1 Nomenclature
      2. 3.1.2 Patterns
      3. 3.1.3 Rack topology
      4. 3.1.4 PureApplication Platform for testing scenarios
      5. 3.1.5 Scenario basics
    2. 3.2 HADR scenarios for WebSphere Application Server
      1. 3.2.1 Scenario WAS_1: WebSphere cell in the same rack (PDC-1) with transactions in GPFS
      2. 3.2.2 Scenario WAS_2: WebSphere cell across two racks in same data center
      3. 3.2.3 Scenario WAS_3: WebSphere active-passive cells - identical setup in PDC and SDC, with WebSphere transactions stored in GPFS
    3. 3.3 HADR scenarios for DB2
      1. 3.3.1 Scenario DB2_1: DB2 HADR from the same pattern and deployed on a single rack (PDC-1)
      2. 3.3.2 Scenario DB2_2: DB2 HADR from the same pattern and deployed the parts on two racks (PDC-1 and PDC-2)
      3. 3.3.3 Scenario DB2_3: Identical DB2 HADR deployments across primary (PDC) and secondary DR (SDC) data centers
    4. 3.4 HADR scenarios for WebSphere Application Server and DB2
      1. 3.4.1 Scenario WDB_1: WebSphere Application Server cluster and DB2 HADR deployed on a single rack with transactions stored in database
      2. 3.4.2 Scenario WDB_2: WebSphere Application Server cluster split across two racks in the same data center with DB2 HADR also split across the racks, with WebSphere transactions stored in database
      3. 3.4.3 Scenario WDB_3: Identical WebSphere Application Server cell and DB2 HADR replicated across DR site, with WebSphere transactions stored in DB
    5. 3.5 HADR scenarios for WebSphere MQ
      1. 3.5.1 Scenario WMQ_1: WebSphere MQ primary and standby in the same pattern deployed on a single rack
      2. 3.5.2 Scenario WMQ_2: WebSphere MQ primary and standby in different pattern deployed on two different racks within the same data center
      3. 3.5.3 Scenario WMQ_3: WebSphere MQ primary and passive in the different patterns deployed on separate racks across the data center
  8. Chapter 4. Infrastructure setup
    1. 4.1 Block storage configuration
      1. 4.1.1 Block storage configuration
    2. 4.2 Block storage replication configuration
      1. 4.2.1 Block storage replication: Steps
    3. 4.3 Configuring an Active/Active (Mirrored) GPFS deployment
      1. 4.3.1 Active/Active GPFS deployment: Steps
    4. 4.4 Configuring an Active/Passive GPFS deployment
      1. 4.4.1 Active/Passive GPFS deployment: steps
      2. 4.4.2 Active/Passive setup and takeover
    5. 4.5 Deploy GPFS Shared Service
    6. 4.6 External storage configuration
    7. 4.7 Network configuration and cloud resources configuration
      1. 4.7.1 Network configuration
      2. 4.7.2 Cloud resources configuration
    8. 4.8 Multisystem environment deployment
      1. 4.8.1 Management domains
      2. 4.8.2 Deployment subdomains
      3. 4.8.3 Additional requirements
      4. 4.8.4 System configuration
      5. 4.8.5 Create one or more deployment subdomains
      6. 4.8.6 Cloud resources configuration
    9. 4.9 DNS setup for primary and secondary (cross) rack scenarios
      1. 4.9.1 Network setup
    10. 4.10 Network configuration for WebSphere Application Server and DB2 scenarios
      1. 4.10.1 PDC
      2. 4.10.2 SDC
  9. Chapter 5. High availability and disaster recovery scenarios for DB2
    1. 5.1 Introduction to DB2 HADR
      1. 5.1.1 Primary, standby, and log shipping
      2. 5.1.2 DB2 client and automatic client rerouting
    2. 5.2 DB2 Client Setup
      1. 5.2.1 Deploy DB2 client on PureApplication System
      2. 5.2.2 Configure DB2 client
    3. 5.3 Building a DB2 Virtual System Pattern
      1. 5.3.1 Cloning the Default DB2 OLTP Pattern with HADR for Linux
      2. 5.3.2 Modifying the new pattern, ITSO DB2 OLTP HADR Pattern
      3. 5.3.3 Optional: Multiple HADR databases
    4. 5.4 Scenario DB2_1
      1. 5.4.1 Deployment
      2. 5.4.2 Validation
    5. 5.5 Scenario DB2_2: Two systems using a single pattern
      1. 5.5.1 Deployment
      2. 5.5.2 Validation
    6. 5.6 Scenario DB2_3: Two systems using block storage replication
      1. 5.6.1 Adding block storage to Virtual System Pattern
      2. 5.6.2 Block Storage configuration
      3. 5.6.3 Deploy Virtual System Pattern on both systems
      4. 5.6.4 Enable Block Storage replication
      5. 5.6.5 Validate the Virtual System Instance on the active system
      6. 5.6.6 Planned Failover to SDC
      7. 5.6.7 Unplanned Failover to SDC
    7. 5.7 Validation
      1. 5.7.1 Configure the DB2 client
      2. 5.7.2 Connect to the DB2 database and perform a simple query
      3. 5.7.3 Confirm DB2 HADR roles of primary and standby DB2 servers
    8. 5.8 Testing for outages
      1. 5.8.1 Planned outage: DB2 takeover
      2. 5.8.2 Unplanned outage - shutdown of primary database OS
  10. Chapter 6. High availability and disaster recovery scenarios for WebSphere Application Server
    1. 6.1 Scenario WAS_1: WebSphere cell in the same rack, transactions in GPFS
      1. 6.1.1 Configure Primary GPFS Server
      2. 6.1.2 Build the WebSphere Application Server cluster pattern
      3. 6.1.3 Add GPFS Client Policy
      4. 6.1.4 Deploy WebSphere Application Server cluster pattern
      5. 6.1.5 Create a WebSphere Application Server cluster
      6. 6.1.6 Configure transaction services
      7. 6.1.7 Test Scenario WAS_1 HA
    2. 6.2 Scenario WAS_2: Single WebSphere Cell Across two racks in PDC
      1. 6.2.1 Configure and Deploy GPFS Mirror Server at PDC-2
      2. 6.2.2 Configure and Deploy Tiebreaker Server at PDC-2
      3. 6.2.3 Configure and deploy GPFS Primary server on PDC-1
      4. 6.2.4 Deploy GPFS Shared Service at PDC-1
      5. 6.2.5 Configure WebSphere Application Server with GPFS client policy
      6. 6.2.6 Deploy WebSphere pattern to multiple domains
      7. 6.2.7 Configure WebSphere Application Server to write transaction log to GPFS storage volume
      8. 6.2.8 Test the Multi-domain WebSphere Split Cell HA using GPFS scenario
    3. 6.3 Scenario WAS_3: Active/Passive, identical setups in PDC and SDC, with transactions stored in GPFS
      1. 6.3.1 Network configuration
      2. 6.3.2 Cloud Group and Environment Profile
      3. 6.3.3 Storage Volume
      4. 6.3.4 GPFS
      5. 6.3.5 WebSphere Application Server
      6. 6.3.6 Planned outage at PDC
      7. 6.3.7 Recovery
  11. Chapter 7. High availability and disaster recovery scenarios for WebSphere Application Server and DB2
    1. 7.1 Common assets used in scenarios
    2. 7.2 Scenario WDB_1: WebSphere Application Server cluster and DB2 HADR deployed on a single rack with transactions stored in database
      1. 7.2.1 Build the WebSphere Application Server cluster pattern
      2. 7.2.2 Deploy WebSphere Application Server cluster pattern
      3. 7.2.3 Create a WebSphere Application Server cluster
      4. 7.2.4 Install DB2 JDBC driver
      5. 7.2.5 Create a JDBC Provider
      6. 7.2.6 Create a J2C alias
      7. 7.2.7 Create data sources
      8. 7.2.8 Install BankTransaction application
      9. 7.2.9 Validate the functionality of the application
    3. 7.3 Scenario WDB_2: WebSphere Application Server cluster with DB2 HADR, split across two racks with WebSphere transactions stored in database
    4. 7.4 Scenario WDB_3: Identical WebSphere Application Server cell and DB2 HADR replicated across DR site, with WebSphere transactions stored in database
  12. Chapter 8. High availability and disaster recovery scenarios for WebSphere MQ
    1. 8.1 Common assets used in WebSphere MQ scenarios
      1. 8.1.1 Image Parts
      2. 8.1.2 Policies
      3. 8.1.3 Script packages and parameters used in the pattern
      4. 8.1.4 Prerequisites
    2. 8.2 Scenario WMQ_1: WebSphere MQ Primary and standby in the same pattern deployed on a single rack
    3. 8.3 Scenario WMQ_2: WebSphere MQ primary and standby in different patterns deployed on two different racks in the same data center
    4. 8.4 Scenario WMQ_3: WebSphere MQ primary and standby in the different patterns deployed on two different racks across different data centers
      1. 8.4.1 Steps for creating a WebSphere MQ active/passive scenario
  13. Appendix A. Sample Application
    1. Database Setup
    2. Application Coding Explanation
    3. Testing
  14. Appendix B. Common WebSphere Application Server configuration tasks
    1. Build the WebSphere Application Server cluster pattern
    2. Deploy WebSphere Application Server cluster pattern: Single rack
    3. Deploy WebSphere Application Server cluster pattern: Multiple rack
    4. Create WebSphere Application Server cluster
    5. Configure database connectivity for BankTransaction application
    6. Install BankTransaction application
    7. Validate BankTransaction application
  15. Appendix C. Additional material
    1. Locating the web material
    2. Using the web material
  16. Related publications
    1. IBM Redbooks
    2. Online resources
    3. Help from IBM
  17. Back cover
  18. IBM System x Reference Architecture for Hadoop: IBM InfoSphere BigInsights Reference Architecture
    1. Introduction
    2. Business problem and business value
    3. Reference architecture use
    4. Requirements
    5. InfoSphere BigInsights predefined configuration
    6. InfoSphere BigInsights HBase predefined configuration
    7. Deployment considerations
    8. Customizing the predefined configurations
    9. Predefined configuration bill of materials
    10. References
    11. The team who wrote this paper
    12. Now you can become a published author, too!
    13. Stay connected to IBM Redbooks
  19. Notices
    1. Trademarks