Use the hacks in this chapter to explore the authentication options that are available to you in heterogeneous networked computing environments and simplify administering user accounts and passwords. This chapter also provides some quick and dirty tips for those unfortunate moments when, for one reason or another, you have to lock users out of specific systems quickly.

This chapter explores ways of connecting to remote systems. When you just can’t be everywhere at once, it’s incredibly useful to be able to access multiple consoles and graphical displays from the convenience of your office or machine room. You’ll find many of the hacks in this chapter to be handy tips that you may want to pass on to your users who also need to work on multiple systems, regardless of what operating systems they’re running.

Networks make it easy to set up servers on specific systems to address the needs of clients throughout your computing environment. The hacks in this chapter explain how to set up central servers that do things like synchronize the time on all the systems in your environment (via NTP), deliver IP addresses to newly connected hosts (using DHCP), and integrate these services with existing ones (with DHCP and name lookups done via DNS, for example). This chapter also discusses setting up centralized access to printers from both sides—how to set up your print servers, and how to access them from the various operating systems that your users may have running on their desktops.

This chapter presents a variety of cool sysadmin tips and techniques that we’ve accumulated over the years, including how to keep processes running without writing a daemon or staying logged in, how to use PXE to netboot Linux, how to share information with fellow sysadmins in a centralized fashion, how to get the most out of classic but incredibly useful terminal-oriented applications (such as minicom, screen, and vi), and so on. We also discuss how to quickly and easily create documentation for your sysadmin policies and procedures so that your successors can figure out how things work after Google hires you away from your current employer.

If everything just kept running forever, storage was infinite, and users never executed the rm command with the wrong arguments, this chapter would be unnecessary. Welcome to Earth! Things don’t actually work that way. However, the hacks in this chapter explore some cool ways of making it easier for you to manage storage, deploy new systems, do backups of today’s huge disks, and even reduce the need for some of the restore requests that occasionally clog every sysadmin’s inbox.

Networked computing environments make it easy to store data on different machines or on centralized servers. This chapter provides some tips and tricks for managing distributed storage and making sure the administrative environments on your servers are synchronized.

Security is not just a job; it’s an adventure with no end in sight. Crackers are always working on new ways to break into existing networks and machines, and you need to be able to either lock them out or at least understand what they’ve broken when they get in. The hacks in this chapter discuss a wide range of security tools and techniques that can help you sleep at night and protect your systems at the same time.

This chapter provides techniques for optimizing system performance, whether by figuring out who’s hogging the entire CPU and shooting down that user’s nethack sessions or by using cool knobs in the /proc filesystem to tweak system performance or using journaling filesystems to minimize system restart time. It also provides some useful X hacks, such as an easy way to use multiple monitors on a single system and a discussion of reducing desktop overhead by punting GNOME or KDE in favor of simpler X Window managers that eliminate CPU-intensive bells and whistles and actually just manage windows.

Logfiles aren’t just a diary for your system and its core applications; they are a useful record that you can use to spot emerging problems so you can correct them before they mature into catastrophes. This chapter includes hacks that enable you to centralize log information in a variety of ways, be warned when problems arise, and get the most out of system status information, whether it’s log information, internal disk controller status data, or remote hardware status information that you can collect via SNMP. It also discusses tools for monitoring your network and spotting the BitTorrent user who’s slowing down your CEO’s web browsing.

Sooner or later, some system that you’re responsible for will go down. If you can’t fix your problems by board-swapping, the hacks in this chapter will show you how to boot crippled systems so that you can diagnose problems, repair munged filesystems, and even (if you’re lucky) recover deleted files or data that was stored on disks that have gone belly up. Try the tips and tricks in this chapter if you’re having problems—there’s always plenty of time to panic later.

Used to indicate URLs, Unix utilities, filenames, filename extensions, and directory/folder names. For example, a path in the filesystem will appear as /Developer/Applications.

Used to show code examples, the contents of files, and console output, as well as the names of variables, commands, and other code excerpts in the text.

Used to indicate user input, such as commands to be entered by the user, and to highlight portions of code (typically new additions to old code).

Used in code examples to show sample text to be replaced with your own values.

Used to indicate a cross-reference within the text.