Customer Reviews

Distributed UNIX System Administration; Team Procedures for the Enterprise

5 star:		(0)
4 star:		(0)
3 star:		(0)
2 star:		(1)
1 star:		(0)

 

 

There is a saying that some people are too heavy for light work and too light for heavy work. In that sense, Distributed Unix System Administration: Team Procedures for the Enterprise is a book that parallels that saying.

The book is fact, really two books. The first 9 of the books 11 chapters deal with Unix fundamentals. The basics of file and block structure, shells, scheduling, processes, security and more are discussed. While Wetsch does give a good overview of these topics, it is hard to justify spending $50.00 for those topics alone, when similar information is available free on the net. If you do not want to spend a lot of money, check out the Unix FAQ at www.cis.ohio-state.edu/hypertext/faq/usenet/unix-faq/faq/contents/faq.html for answers to nearly every Unix question. Also, subscribing to the Usenet comp.unix.questions newsgroup is another good way to inexpensively get well informed about Unix.

The final two chapters of the book detail a program Wetsch wrote called SAmatrix. SAmatrix is an expert analysis tool designed to manage systems with the goal of bringing order, reliability and exception reporting to distributed systems. The outgrowth is SAmatrix is an Enterprise System Assessment that accesses an organization's environment and determines if the organization is at risk, and if so, what action to take.

SAmatrix breaks down systems into 5 modules: software, hardware, network, security and operations. Within the program, there are about a hundred questions. After answering all of the questions, the program provides module risk assessments and levels of risk.

While SAmatrix has a lot of potential, the problem with it is that it in its current incarnation, is way too light of a program to do anything meaningful for complex environments. The fact that it can not really be customized or its entries weighted, make the program for the most part unusable. As an example, the fact that you can not weigh the SAmatrix entry of: are major system crashes less than 6 per year, is a problem since 6 major system crashes a year can be catastrophic for some environments (factory floor, hospitals, etc.) or a minor inconvenience for others (library, home use, etc.) The main issue it that systems management is a very complex science. Taming such a beast requires rather advanced tools, not the type that you will find in a $49.95 book limited to only two chapters.