Linux on Low-End Hardware
I work in a research lab with many Unix boxes. My problem is that these boxes are dedicated to research almost all the time. Our system administrator doesn't really have time to support people like me, who want more than the traditional dial-in e-mail reading. Sure, term is quite useful, but I'm a traditionalist.
You see, I run Linux at home. I have the full power of a Unix machine in my house, and I am not going to be satisfied by using it as a terminal to dial into a server with it. I want my own e-mail!
In the old days of Unix, before Internet, machines transferred e-mail, news, and files flawlessly [Well, pretty well, anyway—Ed] by calling each other up over phone lines. A machine could have dozens of modems and would accept calls and periodically call up other machines using the system called uucp—Unix-to-Unix copy. Instead of getting files through anonymous ftp, people used “public uucp”. Instead of writing some nightmarish scripts to call in, transfer files, hang up, and process them, I thought—why not use uucp?
So I began looking for a machine to use. About that time, a friend of mine was hired, and we found an old computer in his office. It was an old 16 MHz 386SX. In its previous life, it ran Procomm and was a full-time terminal. What a change we had in mind—we transformed it into a multi-tasking Unix server! It had 4MB of RAM, which was enough to run Linux. But the hard disk wasn't much to gawk at—40MB. We named it “trappen”.
The purpose of trappen was two-fold. First, we wanted uucp connectivity to our home Linux boxes. Second, we wanted to provide anonymous ftp and http for ourselves; not that we badly needed it (though it would be handy) but more because it's fun to have your own personal Unix server.
We decided to split the disk in half—20MB for the system and swap space, and 20MB for file storage. Now, it needed about 5MB of swap space, so how was I supposed to squeeze a Linux system into 15MB and still have plenty of room for news, e-mail, and ftp-able files?
I've installed many Linux systems over the years, but I always simple used Ethernet to copy my setup onto the target computer. (I have almost created my own distribution.) But my system would take me days to prune down to 15MB. I had to resort to using someone else's distribution, instead.
In the beginning, Linux distributions were not really very fancy. They gave you the files, and you were on your own. If you weren't a Unix expert, you could get stuck in a swamp of daemons, config files, and strange programs. But I knew installations were getting better, so I decided it was worth a try.
I grabbed the latest Slackware distribution and looked at the disk contents; I could do it. I didn't need any programming libraries, and I didn't need X-Windows or anything fancy. Just the basics. About 20 minutes later, I had the tiniest Unix system I've ever seen! I went ahead and installed everything I needed, and it took 14MB! I had all the binaries I needed: news-readers, mail-readers, and uucp. The next step was to configure everything.
Now I expected to have a lot of things to configure. We had smail, anonymous ftp, and the rest, but the Slackware system did such a good job of setting up default configuration files that most things worked out of the box. What I did spend time on was smail and uucp. I set up uucp to route things to our two Linux boxes at home, and configured smail to use it. Then we went through and stripped out any functionality we didn't need, to make the system even smaller.
To support us, our network administrator here at the lab agreed to put in DNS MX records for our home computers, pointing to trappen. This would allow e-mail addressed to us at our home machines to be sent to trappen, which would process and spool it and forward it to us when we were connected.
Through trappen, we now had e-mail at home—not just for us, but for anyone who had accounts on our machines. My whole family has access, and they're learning to use e-mail and news. (It takes quite a bit of disk space to store news at home, but I only have a few groups, and I don't keep articles for very long.)
Trappen has plenty of power for all its tasks, though it is pretty slow interactively. But we have its console, as well as a Wyse terminal in its office. We decided to make the ultimate use of the console and terminal—as telnet terminals to other, more powerful machines. We decided to write a program which would ask for a host to telnet to, which would allow trappen to function as a terminal server. It turned out to be easy, thanks to the getty_ps package. Getty_ps is so versatile that we were able to set up configuration files such that instead of spawning login, it would spawn telnet. We changed the login prompt accordingly, and voila! Trappen's VGA console is the fastest terminal we have in the building, especially because it runs over Ethernet. The serial terminal attached to it is quite handy to check e-mail with.
Everything worked perfectly! We didn't have to touch trappen for a week. In fact, when we had time we decided to add a few enhancements. We added a uucp-ftp gateway. Someone can anonymously ftp and place files in a directory named to computer and trappen will automatically copy the files to a directory on the machine with that name.
Then we really went all-out. We didn't want to put an entire X-Windows environment on trappen (it was quite under powered), but it did have a VGA card. We got the X-Windows server program and all the X-Windows fonts and copied them over. Then we ran X-Windows with the -query option, which causes it to be an X-terminal to another machine, in this case to one of our high-powered machines in the lab. It worked! Trappen didn't have nearly enough power to run X-Windows applications locally but it wasn't bad as a graphical terminal. We eventually decided it was too slow, but it did work, and did not require much disk space.
Trappen is still going strong, over a year after it was “born”. It now allows slip access for our home machines. The CPU is severely limited in power, and during interactive use, it is quite slow, but transferring files and telnet sessions through trappen to other machines is almost instantaneous. Trappen now supports several users who use it to access the Internet.
UUCP is almost completely error-proof, so when I want to transfer a file between home and work, I uucp it. Trappen doesn't call out right away, but the machines connect every few hours. It doesn't matter if it can't connect right away; uucp always gets through eventually, and I know that when I get home, the file will be waiting in my incoming directory.
Trent Tuggle is an Engineering student at the University of Central Florida. His other job is programming virtual reality simulations at the Institute for Simulation and Training. Between the two of them he doesn't have any time, so consequently he's into water sports, model airplanes, and music synthesizers. His e-mail address is tuggle@vsl.ist.ucf.edu.
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Sponsored by AMD
Built-in forensics, incident response, and security with Red Hat Enterprise Linux 6
Every security policy provides guidance and requirements for ensuring adequate protection of information and data, as well as high-level technical and administrative security requirements for a system in a given environment. Traditionally, providing security for a system focuses on the confidentiality of the information on it. However, protecting the data integrity and system and data availability is just as important. For example, when processing United States intelligence information, there are three attributes that require protection: confidentiality, integrity, and availability.
Learn more about catching the bad guy in this free white paper.
Sponsored by DLT Solutions
| Designing Electronics with Linux | May 22, 2013 |
| Dynamic DNS—an Object Lesson in Problem Solving | May 21, 2013 |
| Using Salt Stack and Vagrant for Drupal Development | May 20, 2013 |
| Making Linux and Android Get Along (It's Not as Hard as It Sounds) | May 16, 2013 |
| Drupal Is a Framework: Why Everyone Needs to Understand This | May 15, 2013 |
| Home, My Backup Data Center | May 13, 2013 |
- Designing Electronics with Linux
- New Products
- Making Linux and Android Get Along (It's Not as Hard as It Sounds)
- Dynamic DNS—an Object Lesson in Problem Solving
- Linux Systems Administrator
- Senior Perl Developer
- Technical Support Rep
- UX Designer
- Web & UI Developer (JavaScript & j Query)
- Using Salt Stack and Vagrant for Drupal Development
Enter to Win an Adafruit Pi Cobbler Breakout Kit for Raspberry Pi

It's Raspberry Pi month at Linux Journal. Each week in May, Adafruit will be giving away a Pi-related prize to a lucky, randomly drawn LJ reader. Winners will be announced weekly.
Fill out the fields below to enter to win this week's prize-- a Pi Cobbler Breakout Kit for Raspberry Pi.
Congratulations to our winners so far:
- 5-8-13, Pi Starter Pack: Jack Davis
- 5-15-13, Pi Model B 512MB RAM: Patrick Dunn
- 5-21-13, Prototyping Pi Plate Kit: Philip Kirby
- Next winner announced on 5-27-13!
Featured Jobs
| Linux Systems Administrator | Houston and Austin, Texas | Host Gator |
| Senior Perl Developer | Austin, Texas | Host Gator |
| Technical Support Rep | Houston and Austin, Texas | Host Gator |
| UX Designer | Austin, Texas | Host Gator |
| Web & UI Developer (JavaScript & j Query) | Austin, Texas | Host Gator |
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?




3 hours 44 min ago
4 hours 18 min ago
5 hours 17 min ago
6 hours 7 min ago
10 hours 9 min ago
13 hours 56 min ago
14 hours 4 min ago
16 hours 19 min ago
18 hours 49 min ago
1 day 4 hours ago