The number of "Live" Linux distributions have been growing like weeds over the last two years. It's not surprising when you think about it. Live CD's give you the flexibility of running Linux on any system without the need to actually "install" it.
There's also another advantage to Live CD's, you can have custom functionality configurations designed for specific purposes. There are Live CD's for Desktops, Servers, Clusters, Gaming, Multi-Media, and, of course, Security. In fact the Live CD format lends itself to Security tasks extremely well. One of these Security focused distributions is Plan-B. Here is an interview with the projects creator, J. McDaniel, on his background, the history & future of Plan-B.
Joe Klemmer: Who is J. McDaniel? (What does the "J" stand for?)
J. McDaniel: The J stands for Jeremy. It never occurred to me, during the entire course of creating the CD and the website, that I hadn't used my first name. I have no reason why, I just didn't.
# whoami jmcdaniel # _Sorry couldn't resist=)
JK: What's your background?
JM: I'm from a small town in West Virginia. My freshman year in high school, I took what I thought was a keyboarding class. As it turned out, it was actually a programming class for an IBM with BASIC. (Or was it BASICA? I can't remember for sure.) It wasn't long before I had a Commodore 64 at home with a 13" TV and a tape drive (audio tape that is). A couple of years later, I got a PC with DOS 3.0. I still have the 386DX and BIOS chip as a souvenir. I joined the Army after high school and started out in Signal Corp. I was a Radio and Teletype (RATT) Operator, and then I luckily got switched to a computer tech. They were running SunOS then, now called Solaris. After my eight years were up, I jumped ship and it was almost the end of the computer scene.
JK: When/How did you get involved in Linux and Open Source?
JM: Still interested in computers outside the service, around late `96, I was informed of an OS Project that allowed you to connect a PC to an amateur ham radio rig - best of all, it was free. I immediately grabbed the first copy I could find, and Debian v1.0 became my best friend and enemy as we developed a love/hate relationship. Although I never did get the radio to work right with it, I did realize I had to get back into the IT field. In `98, when I got discharged, I was working any job I could get. None of my jobs were in IT, though. I quickly learned I needed a degree, and fast. I finally managed, in 1999, to start a program at a local school in Fairmont, WV called Computer Tech., working toward an Associates Degree. By the time I graduated, three years later, they had changed their name to IADT. I also quickly learned college costs money. Now things are better though. I'm working there as an Adjunct Instructor teaching Introduction to Unix and Unix Administration, and I'm working part time on the side for an accounting firm (T and T Inc.) as their Network Admin. Meanwhile, I'm attempting to complete a Bachelors then Masters in IT at AIU.
JK: What is Plan-B Linux?
JM: Finding it harder to locate a machine away from home with Linux installed, I resorted to a few "floppy based" distributions. I outgrew them in record time, though. Looking for a bit more, I started toying with several Live CDs. A Live CD is an OS that allows you to run it from the CD without having to install it first. After having to change all of them to suite me in one-way or another, I decided to create one made just for me. I know there are close to 150 or more of them now, yet they didn't address my personal needs (OK, wants.) I also decided I had to learn something from the experience. I found that most were based on Knoppix at their root, which is Debian Linux underneath. Knoppix is definitely the most popular and easy to use. At the same time I found it included tons of software I would never use, and very little of what I wanted. However, what I was trying to find was a CD using a modified Red Hat Install. It had to have as many, if not all, of the typical server daemons included in an installable distribution, root user authority by default, a small and easy to configure X Windows interface, hardware recognition and configuration, utilities for security scanning, auditing, and system recovery. It should also have, if necessary, forensic analysis and read/write access to as many file systems as possible, along with the ability to do everyday tasks i.e. email, browse the web, chat, write a report, shutdown and go home. I found a page that listed a CD created by H. Peter Anvin, the "SuperRescueCD". This was it, the perfect groundwork for I what I wanted. It was based in Red Hat 6/7 and was primarily built for recovery. I used a stripped installation of Red Hat 9 and reorganized the build structure Anvin used. After months of trial and error caused by the read/write permissions required on a lot of the software, along with countless coasters burned, I had a working model.
JK: What makes Plan-B unique?
JM: Tough question. It's unique to me, I suppose, because I've gotten to know it intimately over the last couple years while I molded it into something usable. I believe, out of the swarm of Live CDs available, there isn't too much unique about it on the surface. I would have to say it would be the closest to running an installed version of Red Hat. That is to say if you're currently a Technician or Administrator of Red Hat systems, you should feel right at home in Plan-B. It might disappoint you, though, if you are expecting to see KDE or GNOME. I chose a lightweight desktop (BlackBox) instead for the Window Manager. I didn't see an advantage of using anything more elaborate. It also uses a file compression I rarely see used on other CDs of its kind (not that I've researched the matter). I used zisofs, another project by Anvin, which lets me fill ~1.4GB of data on the CD.
JK: How did you go about deciding what to include or not include in Plan-B?
JM: I started with the basic necessities and a server class installation of Red Hat 9. After thinning the install of rarely used files, I started a log of the software I used most often. Then I began to stage a step-by-step scenario of routines I would use as a starting point for auditing systems and networks. Versions 0.1.0 through 0.8.0 were built and rebuilt again based on the ability to reenact each scenario. All software needed to work without failure and using the least amount of resources possible. The most precious resource on a project such as this, of course, is "Space". I scoured the Internet & reviewed hundreds of software projects looking for applications that provided the same capability of large "Feature Rich Applications," yet with smaller file sizes. Practically, as long as it works, it's great. Once I had reached what I felt was the space limit, I rebuilt the CD and used it as a desktop for a week or so to evaluate what I actually use and what was just wasting space. As for the Field Study Applications (Forensics, Security, and Auditing) I had a few that I use most often. However, knowing not everyone works under the same conditions or uses the same approach that I do, I requested information for resources from several authorities in each profession. After getting a list compiled of all the recommendations, I proceeded to add them a few at a time, rebuild the CD, test, and iterate ad nausea.
JK: What are your plans for the future Plan-B?
JM: Currently the Plan-B Project as a whole is being moved to the school here at the International Academy of Design and Technology (IADT) and will become a Student Project. Students will be offered the chance to become part of the development team. This would give them an opportunity to work in an Open Source Environment. They will be working at each phase of the development process for both the CD and the website. IADT has courses involving Network/System Administration, Programming, and Website Design, all of which fit in with the nature of the project perfectly. Research and Development, Software/Hardware Testing, Programming, System and Network Analysis, Project Management, etc. are only a few possibilities. Now, at a technical level, I will be compiling a list of known issues with version one and assessing all of the field requests for software additions. Those will be the primary changes to begin with. It is also time to make the switch to Fedora. You can also expect to see PB2 sporting a new kernel, actually one of each version - 2.4 and 2.6. I'd like to begin work on special software made specifically for Plan-B such as the ability to save a users or system configuration with ability to automatically load that configuration during the boot process. We will be researching the individual applications to see if it's possible to make them more intuitive. The less you have to setup, the faster you can get to work.
The biggest news at the moment, though, is that recently I exchanged a few emails with H. Peter Anvin, the creator of the SuperRescueCD and several well know Linux Utilities, about his plans for continuing with his project. After finding that he wished to continue and create a version 3 of SuperRescue, but doesn't have the time, I offered our project to jump in and begin work on it. He agreed happily. As a result, we now will also be building and maintaining the next SuperRescueCD. I'll be the first to admit I am very honored to do so. If it hadn't been for Anvin's project, ours may not exist. We only covered briefly changes in the current process used to create a custom version of either CD. We agree the current method is very difficult to manage or add personal files with and it is even harder to remove them. This is another area high on the to do list. We might, possibly, build from an rpm repository instead. The differences between the two will mostly deal with your need for use.
While PB2 will continue to move forward in Security, Forensics, and Auditing, SR3 will add a greater base of supported hardware and utilities for system diagnosis at a hardware and software level.
JK: Is there anything you'd like to see happen with Plan-B?
JM: For now, the projects will be worked on internally. However, I would like to see them grow out to the community here as well. I can see a benefit of having a LUG or Open Software Group locally to promote and aid in the use of Linux for home and small business use. I believe if we're going to see Linux in those areas it will be due to organizations such as those who apply the effort to make its existence known. I recall a conversation in which I was discussing some of the technology and offered, "Personally I run Linux." In reply, the gentleman said, "I've heard of that, Toyota or Honda make it, right?" It appears, then, that in the small business world one of the obstacles we face is just the awareness that there are alternatives in the market for Operating Systems and Software. Despite this, I still have high hopes for us yet.