PANIC! UNIX System Crash Dump Analysis Handbook (Bk/CD-ROM)
PANIC! UNIX System Crash Dump Analysis Handbook (Bk/CD-ROM)
ISBN: 0131493868
EAN: 2147483647
EAN: 2147483647
Year: 1994
Pages: 289
Pages: 289
Authors: Chris Drake
- Main Page
- Table of content
- Copyright
- Figures
- Tables
- Code Examples
- Acknowledgments
- Introduction
- What happened?
- What will Panic teach you?
- So many flavors of UNIX
- The audience
- Conventions used
- Contacting the authors
- Welcome to Panic
- Part 1: Getting Started
- Chapter 1. My System Has Crashed
- What is a system crash?
- What conditions cause panics?
- A word about bad traps
- The panic() routine
- How do you know if your system has panic ed?
- Capturing system crash information
- What is a program crash in comparison?
- Chapter 2. My System Is Hung
- What is a system hang?
- What conditions cause hangs?
- How do you know if your system is hung?
- What is a program hang in comparison?
- Capturing system hang information
- Chapter 3. The savecore Program
- What is savecore?
- How does savecore work?
- Disk space requirement locations
- Security issues
- Solaris 1: How to set up savecore
- Solaris 2: How to set up savecore
- Chapter 4. Hey We Got One
- What to do when your system has crashed
- Is the system still usable?
- Turn off savecore? (How many dumps will you need?)
- Saving the crash to tape for shipment or archives
- Chapter 5. Crashing Your Own System
- Crash your Solaris 2 system
- Crash your Solaris 1 system
- Chapter 6. Initial Analysis Without adb
- Identifying the UNIX release hardware architecture
- The message buffer, msgbuf
- Process status utilities: ps pstat
- Network status: netstat
- NFS status: nfsstat
- Address resolution protocol status: arp
- Interprocess communication status: ipcs
- The crash program
- Summary
- Chapter 7. Introduction to adb
- Other debuggers
- adb hardware software requirements
- The distribution of adb
- The different uses of adb kadb
- adb macros usrlibadb
- General startup syntax
- Security issues
- Other helpful files
- Chapter 8. adb: The Gory Details
- Basic commands
- Displaying data
- Locations sizes
- Miscellaneous commands
- Summary
- Chapter 9. Initial Analysis Using adb
- Starting an adb session
- System identification
- Boot time, crash time, and uptime
- Panic strings
- The message buffer, msgbuf
- Stack tracebacks
- Summary
- Chapter 10. The usrinclude Header Files
- What is a header file?
- The usrinclude directories
- usrkvmsys
- usrsharesrcuts
- usrucbinclude
- Summary
- Chapter 11. Symbol Tables
- Namelists the nm command
- Using adb to look at tiny s variables
- A tiny summary
- Chapter 12. adb Macros: Part One
- The macro library
- Reading and understanding macros
- The utsname macro
- The bootobj macro
- Summary
- Chapter 13. adb Macros: Par t Two
- The msgbuf and msgbuf.wrap macros
- The cpus, cpus.nxt, cpu macros
- Summary
- Chapter 14. adb Macros: Writing Your Own
- Exercise 1: Initial information
- Exercise 2: DNLC, the directory name lookup cache
- Exercise 3: Swap information
- Extra Credit Challenge: Which process on which CPU?
- Possible solutions
- Part 2: Advanced Studies
- Chapter 15. Introduction to Assembly
- High-level vs. low-level languages
- Assembly languages
- Basic CPU structure (all CPUs are similar)
- Instruction execution
- Instruction types
- Instruction formats and addressing modes
- Addressing and registers
- Data in memory
- On to SPARC
- Chapter 16. Introduction to SPARC
- Basic characteristics of SPARC assembly language
- SPARC instructions
- SPARC registers
- SPARC instruction types
- Delayed Control Transfer Instructions
- Looking at instructions in memory
- Want to learn more?
- Chapter 17. Stacks
- A generic stack
- The frame structure
- Instructions that affect windows frames
- What have we got so far?
- Chapter 18. Stack Tracebacks
- Compiling with optimization
- The trouble with more than six arguments
- Chapter 19. A Kernel Overview
- Major sections
- Entering the kernel
- Scheduling: processes and threads
- File systems
- Files, inodes, and processes
- Memory Management
- This was just a kernel overview
- Chapter 20. Virtual Memory
- The free list
- Swap space
- Page faults
- Working set, or resident set
- Keeping track of pages
- Keeping track of process space
- Anonymous memory
- Kernel functions
- Virtual memory routines
- Address spaces
- Segments
- Pages
- The hat layer
- Chapter 21. Scheduling
- SunOS 4.x
- Solaris 2
- How you change it
- Chapter 22. File Systems
- Basic disk structure
- The old original
- VFS functions
- UNIX File System (UFS)
- Other file systems
- Chapter 23. Hardware Devices and Drivers
- Drivers and device control
- Driver functions
- Real hardware
- Drivers and crashes
- Chapter 24. Interprocess Communication
- Semaphores
- Messages
- Shared memory
- Common functions
- Why all this?
- Chapter 25. STREAMS
- STREAMS structure
- Data structures
- Queues
- Message structures
- Data blocks
- Pipes
- Basic functions
- Support functions
- Digging around inside
- Chapter 26. Trap Handling
- Kinds of traps
- Trap sequence
- Trap frames
- Trap types
- Chapter 27. Watchdog Resets
- What is a watchdog?
- Can you get a core file?
- What do you do next?
- Watchdog analysis
- Summary
- For further reference
- Chapter 28. Interrupts
- SPARC systems
- Priority levels
- Serial devices
- Vectored interrupts
- Polled interrupts
- Interrupts in tracebacks
- Chapter 29. Multiprocessor Kernels
- Data protection
- SunOS 4.x multiprocessor systems
- Solaris 2
- Part 3: Case Histories
- Chapter 30. Network Troubles
- Initial analysis
- Check the instruction
- Bug check
- Resolution
- Chapter 31. A Stomped-on Module
- Strings output
- Analysis using adb
- Walking the stack by hand
- The ipcaccess() routine
- Loading the semsys module with modload
- Remember to use the same OS
- Is it a hardware failure?
- How about a software problem?
- Using nm to query symbol values
- adb s search command at work
- Conclusion
- Chapter 32. Hanging Instead of Swapping
- Initial information
- Process status
- Stack tracebacks for every process
- Resolution
- Chapter 33. Panic ed Pipes
- Always get initial information
- Walking the stack by hand
- Examining assembly code: fifo_rdwr()
- Calling parameters
- Examining assembly code: vno_rw()
- Don t work too hard Use SunSolve
- Resolution
- Chapter 34. A Sleeping Dragon
- Initial information
- Walking the stack by hand
- Using the threadlist macro
- Examining mutex locks
- Data address not found? Dig deeper
- Additional analysis tools
- Which processes were waiting for locks?
- Loadable kernel modules
- Conclusion
- Chapter 35. Once Is Not Enough
- The first captured crash
- The second captured crash
- The other crashes
- The solution
- Conclusion
- Chapter 36. Life Without A Root Directory
- Get initial information
- Invoking mutex_enter()
- What process was involved?
- Going back through time
- What s still useful in the stack traceback?
- Why was FrameMaker involved?
- Chapter 37. Disk Woes in the Wee Hours
- Get initial information
- Look for patterns
- Inodes vnodes
- Crash 11: A closer look
- Crash 12: A closer look
- Crashes 13 14: A closer look
- Resolution
- Appendix A. SPARC: The Gory Details
- All CPUs are similar
- The SPARC processor
- Integer Unit (IU)
- Floating-Point Unit (FPU)
- Coprocessor (CP)
- Windows use of SPARC registers
- In closing
- Appendix B. SPARC Instruction Set
- Instruction set summary
- Memory access instructions
- Arithmetic logical shift instructions
- Control transfer instructions
- State register instructions
- Floating-point unit instructions
- Coprocessor instructions
- Synthetic instructions
- Always keep an instruction set reference handy
PANIC! UNIX System Crash Dump Analysis Handbook (Bk/CD-ROM)
ISBN: 0131493868
EAN: 2147483647
EAN: 2147483647
Year: 1994
Pages: 289
Pages: 289
Authors: Chris Drake