Datasheet

Red Hat calls their unstable Fedora release rawhide. It’s available via the Fedora website, but you’ll usu-
ally perform a “yum update” (after uncommenting the development YUM repository entries in
/etc/yum.repos.d) from the most recent stable release in order to switch over, rather than trying to
perform an install of rawhide directly. Rawhide is a hive of activity from which you can often garner
some insight into what might make it into the next Fedora release—ultimately, that may even affect
what goes into the Red Hat Enterprise product at some point in the future.
Novell calls its unstable OpenSuSE release Factory. It’s available from the OpenSuSE website and can be
installed using network bootable images that are available for download. You will need to follow the
installation instructions carefully if you perform a network install as to do so necessitates changing vari-
ous options early on in the boot process—before the YaST installer has even started. You can also
upgrade using YUM (documented online), but that process is much newer as of this writing. Factory is
updated on a semi-regular basis, ultimately feeding technology into SuSE Linux Enterprise Server.
Ubuntu, like Debian, is available in an unstable release. This release is updated frequently, whenever the
packages within it are modified. Thus, it’s sometimes the case that a given system is unable to perform
an update to the latest unstable release at a particular moment in time. Unlike the other distributions
mentioned here, Ubuntu and Debian provide an interim testing version of their distribution, which
always contains packages that have been known to be usable after being released into unstable release
of the distribution. You will usually “apt-get upgrade” your system to unstable after modifying
/etc/apt/sources.list).
Scratch Boxes and Virtualization Technologies
As you become happier with Linux development and become more adventurous, you’ll want to have a
scratch box that you can just test out ideas on (or completely trash) without risking breaking anything it
had installed. This is especially true if you later decide to write your own Linux kernel device drivers or
otherwise modify critical system components — you don’t want to be doing that on a machine you need
to remain stable during that process. Many hackers resort to using old PCs for this purpose.
Virtualization technologies, such as VWware, qemu and Xen, can be useful ways to gain access to a large
number of test virtual machines that you can happily trash all day long, all without actually having to
buy any additional hardware. Not only is virtualization is good cost-saving idea, but it’s also very prac-
tical when it comes to setting up standard test environments and sharing ideas with your colleagues.
Most virtualization technologies will allow you to set up snapshots of systems configured in a particular
way that you can then store or send to your coworkers via a network storage area of some kind.
VWware
VWware allows you to manage large collections of virtual machines using their proprietary graphical
software. It’s trivial to configure a new virtual machine and to then install a Linux distribution within it.
Using VWware, you can easily install a range of different PC-based Linux distributions, all without actu-
ally changing the software on your machine. Great when you want to store preconfigured test machines
or try out some experimental features that don’t rely on having some specific custom hardware device,
not so great for testing out your custom Linux kernel device drivers!
You can find out more information about VWware at
www.vmware.com.
14
Chapter 1: Working with Linux
04_776130 ch01.qxp 2/2/07 10:11 PM Page 14