IBM’s Websphere MQ Version 5.3
Implementation and Certification
with Hewlett-Packard’s MC/ServiceGuard
High Availability Software Version 11.14
The HP Partner Technology Access Center (PTAC) has successfully integrated IBM’s Websphere MQ
v.5.3 with Hewlett-Packard MC/ServiceGuard A.11.14 High Availability product on HP-UX 11i. This
document describes the setup and the specific steps of the certification.
The implementation and certification of Websphere MQ v.5.3 followed the PTAC’s High Availability
Implementation and Certification Process. The general process followed for the certification of the ISV
application in a MC/ServiceGuard environment is detailed in:
HP’s Partner Technology Access Center High Availability Implementation and Certification Services
Process and Methodology
Implementation and Certification Process Participants:
Hewlett-Packard: Guy Patrick (Partner Technology Access Center)
IBM Software: Andy Ford (IBM WMQI Development),
November 18th, 2002 – Paramus, New Jersey
PAR T NE R
TECH NOLOGY
ACCESS
CENTER
IBM’s Websphere MQ and MC/ServiceGuard
03/06/03
Page 1
Table of Contents
High Availability with HP MC/ServiceGuard
IBM’s Websphere MQ in a Highly Available Environment
Websphere MQ Implementation and Certification in an MC/ServiceGuard Cluster
Test Equipment Architecture
Installation Outline for Websphere MQ and MC/ServiceGuard
MC/ServiceGuard Configuration
Test Suite
Appendix 01: Websphere MQ installation addendum for a MC/ServiceGuard environment
Appendix 02: “Cluster1” Cluster configuration file (cluster1.ascii)
Appendix 03: “mqm” package configuration file (mqm.ascii)
Appendix 04: “mqm” package control script (control.sh)
Appendix 05: “mqm” start, stop, and monitor script (mqm.sh)
Appendix 06: Output of bdf on rp5470a
Appendix 07: Contents of /etc/fstab on rp5470a
Appendix 08: Output of ioscan –f on rp5470a
Appendix 09: Output of mount on rp5470a
Appendix 10: Contents of /etc/rc.config.d/netconf on rp5470a
Appendix 11: Output of <netstat –r> on rp5470a
Appendix 12: Contents of /etc/passwd on rp5470a
Appendix 13:
Contents of /etc/services on rp5470a
Appendix 14: Contents of /stand/system on rp5470a
Appendix 15: Output of swlist on rp5470a
Appendix 16:
Output of <vgdisplay -v> on rp5470a
Appendix 17: HP’s PTAC High Availability Implementation and Certification Services Data Sheet
Appendix 18: HP’s PTAC High Availability Implementation and Certification Services Process and
Methodology
IBM’s Websphere MQ and MC/ServiceGuard
03/06/03
Page 2
High Availability with HP MC/ServiceGuard
HP Multi-Computer/ServiceGuard (MC/ServiceGuard) is a specialized facility for protecting mission-
critical applications from a wide variety of hardware and software failures. With MC/ServiceGuard,
multiple— up to 16—nodes (systems) are organized into an enterprise cluster that delivers highly available
application services to LAN-attached clients.
HP MC/ServiceGuard monitors the health of each node and quickly responds to failures in a way that
minimizes or eliminates application downtime. MC/ServiceGuard is able to automatically detect and
respond to failures in the following components:
• System processors
• System memory
• LAN media and adapters
• System processes
• Application processes
Application Packages
With HP MC/ServiceGuard, application services and all the resources needed to support the application are
bundled into special entities called application packages. These application packages are the basic units that
are managed and moved within an enterprise cluster. Packages simplify the creation and management of
highly available services and provide outstanding levels of flexibility for workload balancing.
Fast Detection of Failure, Fast Restoration of Applications
Within an enterprise cluster, HP MC/ServiceGuard monitors hardware and software components, detects
failures, and responds by promptly allocating new resources to support mission-critical applications. The
process of detecting the failure and restoring the application service is completely automated—no operator
intervention is needed.
Recovery times provided by HP MC/ServiceGuard for LAN adapter failures are extremely fast, typically
within a few seconds. Recovery times for failures requiring the switch of an application to an alternate node
will vary, depending on the software services being used by the application. For example, a database
application that is using a logging facility would need to perform transaction rollbacks as part of the
recovery process. The time needed to perform this transaction rollback would be part of the total time to
recover the application. MC/ServiceGuard will detect the node failure, reconfigure the cluster, and begin
executing the startup script for the application package on an alternate node in less than 30 seconds.
High Availability for Mission-Critical Applications
99.95% Uptime Commitment—The Mission-Critical Server Suites
(MCSS) is HP’s platform solution,
offering an unprecedented 99.95% uptime commitment. Preconfigured and tested, MCSS is based on
MC/ServiceGuard-enabled HP 9000 Enterprise Servers, packaged with the consulting and support services
necessary to ensure success in your mission-critical environment.
MCSS solutions are available on HP 9000 servers; and high availability storage is offered with HP Virtual
Arrays, HP SureStore E XP arrays, or the EMC Symmetrix Enterprise Storage System, providing the
ultimate in data availability, reliability, and high performance.
Other Benefits of MC/ServiceGuard:
-Availability during Hardware and Software Maintenance
-Online Reconfiguration Reduces Planned Downtime
-Workload Balancing
-Protecting Data Integrity
For more information, contact any of our worldwide sales offices or HP Channel Partners (in the U.S. call
1-800-637-7740).
IBM’s Websphere MQ and MC/ServiceGuard
03/06/03
Page 3
IBM’s Websphere MQ v5.3 in a Highly Available Environment
IBM’s Websphere MQ v5.3 is validated and certified with Hewlett Packard's MC/ServiceGuard high
availability software, and will be referred to as Websphere MQ from here on. Websphere MQ can be
monitored and controlled by MC/ServiceGuard to provide for a highly available product for use in a
mission critical environment. The level of monitoring is highly configurable and subject to the
implementation plans of the customer.
The MC/ServiceGuard solution for Websphere MQ consists of one package, mqm. The mqm package
contains two services, mqm and mqm_lsr. The mqm package can be configured to provide flexibility in the
configuration and characteristics of the high availability solution.
In a representative Websphere MQ architecture, there is a single instance of Websphere MQ running on a
single server. For this reason, the MC/ServiceGuard restart value for the Websphere MQ package is 0.
This means Websphere MQ will be moved to the adoptive host and restarted if any of the resources
configured by MC/ServiceGuard fail; a power failure, a NIC failure, or a network failure. For more
information on responses to failures reference "HP Managing MC/ServiceGuard Manual".
In the certification of the Websphere MQ with MC/ServiceGuard solution, the “client” used for testing was
anWindows system. Since the MC/ServiceGuard solution utilizes a relocatable IP address for the mqm
package, the client does not need to re-configure it’s session connectivity settings if a fail-over scenario
occurs. The client simply continues to re-attempt communications if the host goes down. Once the package
is running on the fail-over node, the Websphere MQ session is re-established. Depending on the type of
failure, a client in a real-world scenario may be unable to log in for the duration of the fail-over, or the
session may time-out and may need to be restarted. This behavior is completely dependent on the client
application(s).
This document details a set of tests that show that Websphere MQ is compatible with MC/ServiceGuard.
The tests show that if a node running the Websphere MQ server fails, Websphere MQ can be up and
running on an adoptive node in as little as 30 seconds, without human intervention. This time will vary
depending on the amount of activity occurring on the Websphere MQ server, the size of log files, the
number of messages in the queues, the client application using Websphere MQ, and other application
specific conditions.
A MC/ServiceGuard package is dependent on having automated application startup and shutdown scripts.
HP Partner Technology Access Center engineers, teaming with IBM engineers, have developed and tested
package configuration and control scripts which are suitable for use (after system-specific modification) in
an Websphere MQ production environment.
IBM’s Websphere MQ and MC/ServiceGuard
03/06/03
Page 4
Websphere MQ Implementation and Certification in an MC/ServiceGuard Cluster
Test Equipment Architecture
IBM’s Websphere MQ and MC/ServiceGuard
03/06/03
Page 5