12.1 Introduction

When something goes wrong and a Tru64 UNIX system crashes, the first person to be called is the system administrator. Unfortunately, these events seem to happen at the most inopportune time. Whether it is in the middle of an important demonstration, crunch time at month's end, or simply in the middle of the night, it is up to you to quickly diagnose the problem, identify a solution, and get the system back up. This skill is probably best defined as problem management. Problem management is a broad topic that includes dealing with hardware and software errors, handling system crashes, and administering the system logging facility. Tru64 UNIX is a complex system, and often a system administrator's only recourse when confronted with a system problem is to provide as much diagnostic information as possible to your support organization, whether that be Compaq's Customer Support Center or another organization. This information can include system crash dumps and hardware, software, and system error log entries. This chapter will provide some insight into how to interpret the output of these problem-reporting facilities as well as how to correlate system problems with their resulting crash dump output or error-log events. Then, as your knowledge and experience increase, these same Tru64 UNIX problem management facilities can allow for a more proactive problem resolution.

At your disposal in this effort are several Tru64 UNIX facilities to assist you in this problem management effort. These are:

• The system logging facility

• The binary error logging facility

• The...