|New Reviews| |Software Methodologies| |Popular Science| |AI/Machine Learning| |Programming| |Java| |Linux/Open Source| |XML| |Software Tools| |Other| |Web| |Tutorials| |All By Date| |All By Title| |Resources| |About| |
Keywords: Linux, debugging, problem investigation, computer forensics Title: Self-Service Linux Authors: Mark Wilding and Den Behman Publisher: Prentice Hall PTR ISBN: 013147751X Media: Book Level: Advanced Verdict: An in-depth treatment for system administrators |
You have to look at the subtitle of this book - Mastering The Art of Problem Determination - to get an idea of what 'Self-Service Linux' is about. As the subtitle suggests, this is a book about advanced troubleshooting techniques for resolving a wide range of system or application problems on Linux. In part it's a description of a methodology and in part it's about learning the tools and techniques to track down and fix the causes of problems.
The opening chapter describes a phased approach to problem investigation. It describes how to come at problems using your own skills, then moves on to look at getting the most of other available resources (the internet, mailing lists, project developers etc). There's an emphasis on good discipline, showing how keeping detailed notes and grabbing the system status are essential skills.
The next few chapters dive deep into the technical core of the book: strace, the /proc file system, compiling with gcc, using the stack, debugging with gdb, handling system crashes and hangs, kernel debugging with kdb and finally the ELF file format. In each case the material is solidly handled, with good explanations and lots of console output for illustration. The writing is on the dry side, but then this kind of technical material doesn't really lend itself to jokey asides and inconsequential banter.
In addition to the guidance included in the core chapters, the book also includes two appendices. The first is a quick listing of the tools and utilities that comprise the recommended toolbox, and it covers everything from traceroute to ltrace to sar to vnc. This is followed by a comprehensive and extremely useful data collection script that can be used to collect a range of relevant data useful in diagnosing problem causes.
This really isn't the sort of book that's likely to be of use to the casual user of Linux. However, system administrators, computer forensics pesonnel and developers it's a guided tour of advanced techniques for tracking down the sort of intractable problems that no amount of Googling can solve.