VAX/VMS Troubleshooting

This document provides a step-by-step approach to analyzing VAX/VMS crashdumps, processor halts, and process or system hangs. It is intended for software specialists, field service engineers, and others who need to troubleshoot VAX/VMS problems, assuming a level of familiarity with VMS internals and the System Dump Analyzer (SDA) utility.

The document is organized into sections, with the primary goal of guiding the user through the analysis of system failures. It covers various types of errors and system states, including:

• Messages: An index of messages encountered during troubleshooting.
• Starting Points: Guidance on where to begin the troubleshooting process for crashes, halts, and hangs.
• Accessing Process Context with SDA: Techniques for examining process-specific information using the System Dump Analyzer (SDA).
• Bugchecks: Information on analyzing various bugcheck types and the steps involved.
• Crashdump Requirements: Prerequisites for creating and analyzing crashdump files.
• Exceptions: Details on software and hardware exceptions and how VMS handles them.
• Forced Crashes: Procedures for analyzing system crashes that are intentionally initiated.
• Halts: Information on different types of processor halts and their indications, particularly for VAX-11/780 and VAX-11/785 systems.
• Hangs: Procedures for diagnosing and troubleshooting system and process hangs.
• Machine Checks: An overview of machine check errors, their causes, and how to analyze them.
• Resource Waits: Explanation of various resource wait states that processes may enter and how to investigate them.
• Stack Patterns: Guidance on deciphering stack contents to understand the flow of execution and identify the root cause of problems.
• System Service Vectors: Information on how system services are invoked and how to analyze their behavior.
• Virtual Addresses: A breakdown of virtual address spaces (PO, P1, System) and how to use SDA to examine them.

The document emphasizes the use of the System Dump Analyzer (SDA) for examining system states, memory, and process information. It also references various VMS manuals and documentation for more in-depth information. The content is organized to facilitate a systematic approach to problem diagnosis, starting with identifying the type of failure and then drilling down into specific areas of investigation.