October 1, 2004
A new release of Web-Kickstart is available. Its home page and lots of information about it can be found at
You can also see the main web page for the production Web-Kickstart service at NCSU here
What, you don’t want to click the links and read elsewhere about what Web-Kickstart is? You want me to tell you right here? Okay. So be it. Web-Kickstart is a tool to help manage a large base of Red Hat or Fedora Core workstations and servers. Its goal is to build upon the features of Red Hat’s Kickstart to completely automate installation of machines.
A lot of folks hear about Web-Kickstart and stick there nose up at it. I don’t blame them, but I think its worth a look. Its one of the most successful little projects that I have created. The folks here at NCSU are very familar with Jumpstart for Solaris. (Its like Kickstart for Red Hat.) That’s basically what the ‘config’ files look like. The idea is that you maintain a directory tree of these config files, one per machine you wish to install. The Web-Kickstart code (mod_python and apache) receive a request from Anaconda for a kickstart. Web-Kickstart resolves that IP into a FQDN and looks for that string as a filename in the directory tree. The config file is loaded and a normal kickstart is dynamically generated and sent off to the installing machine.
What are the benifits of such a system?
- One central place to manage the config files
- Kickstarts are long and complex for any configuration. A simple configuration with Web-Kickstart is one or two lines
- Installs can be logged into a database
- Seamlessly handles multiple versions of Red Hat based products and even multiple type via an object tree
- Kickstart code that must be run on every install machine can be garunteed to run
- Include files and symlinks used to simply configuration further
Anyway, I hope folks will get some use or at least an idea or two about managing Red Hat based products in an academic level. Suggestions, and patches welcome.