Availability Guide for Application Design

Instrumenting an Application for Availability
Availability Guide for Application Design—525637-004
8-15
Automating Object Management
inform the user that there is a temporary delay in processing. In addition, any error that
requires the user to reenter data must be reported to the user.
Application problems reported to the user must, of course, be worded in a way
appropriate to the expected knowledge level of the user.
Message Conventions
Messages sent to the operator’s console or to the user must be meaningful and easily
referenced if they are to achieve their purpose. To this end, your messages should
conform to specified guidelines that make it easy to find explanatory information. HP
messages follow one such set of guidelines; refer to the Operator Messages Manual
for examples. The format of your messages should be appropriate to your application.
Automating Object Management
Application availability can be increased by partially or completely automating the
operations and management of both the application and the subsystems upon which it
depends. Automation covers the spectrum from one application monitoring and
consolidating status information about another through having the monitoring
application interpret events and dispose of them appropriately in place of the operator.
How much automation you should do depends on such diverse considerations as:
•
The size and level of technical sophistication of your operations staff
What do they need to know? What do they want to know? What activities do they
want automated? What activities do application users want automated?
•
Whether the chosen automation interface allows disposition of an event
•
Whether the response time requirements of the application permit the application
to be slowed by the resource requirements and intervention of a higher priority
automation program
Automation can be performed through either of the following:
•
Generic, industry standard interfaces such as the Simple Network Management
Protocol (SNMP). SNMP provides a basic, cross-platform interface suitable for
monitoring and logging application events. The HP implementation of SNMP
provides access to the Event Management Service (EMS) messages of the HP
Distributed Systems Management software.
•
The HP Distributed Systems Management (DSM) architecture and its related
software. HP’s DSM architecture provides a framework for managing user
applications and HP subsystems. DSM provides management services that
capture events and filter and distribute those events for processing by
management applications. It also provides the underlying protocols for control of
applications and subsystems by management applications.