Learning The Cause Of A Problem: A Systematic Approach; Tools For Identifying Problems - HP Guardian User Manual

A
Problem Solving Techniques
This appendix presents an approach you can use in your operations environment to
determine the possible causes of problems, to systematically fix or escalate such
problems, and to develop ways of preventing the same problems from recurring.
Continuous availability of your Compaq NonStop™ Kernel system is important to
system users, and your problem-solving processes can help make such availability a
reality.
Learning the Cause of a Problem: A Systematic
Approach
Determining the cause of a problem on your system is similar to learning about a
problem you have with your car. If your car fails to start, what is the first thing you do?
You would not replace the engine as your first step. You would start with the easiest,
least expensive possibilities, try them, and move to more complex, expensive
possibilities only if the easier solutions fail.
With this in mind, the four basic steps in systematic problem solving are:
Task
Task 1: Get the Facts and Log the Problem
Task 2: Find and Eliminate the Cause of the Problem
Task 3: Escalate the Problem
Task 4: Focus on Prevention

Tools for Identifying Problems

Several tools might be available on your system to assist you or your operations
management in problem identification and tracking, including:
The Event Management Service Analyzer (EMSA) is used to extract specific types
of event messages from EMS log files and to create an ENSCRIBE database that
you can query to analyze problem trends. For more information, see the EMS
Analyzer User's Guide and Reference Manual.
The Measure program is used to collect and display system performance statistics,
performance of processors, processes, communication and network lines, files,
disks, and terminals. Operations management personnel often use Measure to help
fine-tune and balance a system. For more information, see the Measure User's
Guide and the Measure Reference Manual.
The PEEK program is used to gather statistical information about processor activity,
system storage pools, paging activity, message information, send instructions, and
interrupt conditions. For more information, see the PEEK Reference Manual.
Guardian User's Guide —425266-001
A- 1
Page
A-3
A-4
A-5
A-6