Free diskless software for linux

MJS is a console mp3-player with some special features not found in most console based mp3-players.
About oktober 2000, at my studentsunion you could only play tapes or compact discs. The cds had often meny scratches on them and listening to them had become extermely hard. This is when I thought that maybe it was time to do something about it.
I had this old pentium-60 lying somewhere and I had some experience with diskless linux-clients. So I decided to give it a try. After equiping the Pentium with a cheapo soundcard, nic and mda and setting up linux, I came to the problem of choosing an mp3-player.
It had to be console based to keep it simple, but it being simple to use was a must. After searching a couple of days on the internet I came up with mms (Matts MP3 Selector). This software fulfilled many of our wishes but missed some. So I decided as it was GPL licensed to have a go at it.
At that time I didnt have any knowledge of C at all, so my efforts were done by trial and error. After about half a year I had a C++ programming class at the university and a few months later I had enough knowledge to begin seriously hacking the old program into something better fulfilling our wishes.
I added a search function, that uses slocate databases, playlist loading / saving, somewhat more studentproof interface, easier directory navigation, more configuration options and so on.
Enhancements:
- Added config option to remove a song from the playlist once it has been played

Diskless Remote Boot in Linux (DRBL) provides a diskless or systemless environment for client machines. DRBL Live Standard works on Debian, Ubuntu, Mandriva, Red Hat, Fedora, CentOS and SuSE. DRBL uses distributed hardware resources and makes it possible for clients to fully access local hardware. It also includes Clonezilla, a partitioning and disk cloning utility similar to Symantec Ghost.

The purpose of DRBL live is to let you run a machine as DRBL server without installation. Its based on Debian Live. DRBL live includes all the DRBL functions, plus Clonezilla. Therefore it can provide PXEBoot Clonezilla, which can be used to do massively clone in a computer classrom or similar environment. Multicast clone is also supported.

Generally speaking, DRBL live is server edition, while Clonezilla live is personal edition. The differece between DRBL live and Clonezilla live is: DRBL live provides DRBL functions (DHCP, TFTP, NFS, NIS services), so client can boot via PXE and be cloned. Since DRBL live includes all the Clonezilla programs, it can be used as an alternative of Clonezilla live.

GeDI, the GEneric Diskless Installer, is a set of bash scripts charged with the simple purpose of making it easier to install and manage a set of diskless clients.
Enhancements:
- Removed the Make-rpm.mk and fixed the Makefile to create rpms.
- TODO: Fix the makefile to pay attention to actual tags (rather than arbitrarily grabbing what is in the META file.

Plume is a complete set of packages to create and administrate a network of diskless computers used as X terminals. Stations root filesystem is extensively based on Debian with as few new packages as possible. It is clean and lean, as well as easy to understand and adapt if needed.

It is all about setting up easily a server with thin clients, each based on Debian. It is meant and known to be particularly light on the clients (ex.: it requires as few as 4MB RAM on a typical client) and run on commodity hardware.

Thin client [1]: A thin client is a computer (client) in client-server architecture networks which has little or no application logic, so it has to depend primarily on the central server for processing activities. The word "thin" refers to the small boot image which such clients typically require - perhaps no more than required to connect to a network and start up a graphical interface.

Plume is working fine at several places for a long time now (early year 2000), without much maintenance and without specialists needed when something is to be done. It has successfully and painlessly passed a migration from Potato to Woody on at least one server and from Woody to Sarge on at least one other server. As it is available as Debian packages (only sources on this site, though you can find binary packages on apt.bxlug.be), it should be a snap to install on any Debian machine out there.

It is meant first for schools, and as such fits well on really old PC (as little as 486 with 8MB RAM). It has been thought as a clean replacement of LTSP.


The project is currently hosted at SourceForge.net. You can find it here.

You can download the latest source release here. There are currently packages for Debian Woody, but as of now, it still works fine on Sid also. For binary releases (Debian packages) I suggest you contact us on the mailing-list(archives and more available here), you dont even have to subscribe to do it. There is even a French-speaking mailing-list dedicated to current users of Plume installations here.

BCCD - Bootable Cluster CD was created to facilitate instruction of parallel computing aspects and paradigms. Part of the difficulty instructors face is lack of dedicated resources to explore distributed computing aspects lack of time to preconfigure and test the supporting environment.

The BCCD image addresses this problem by providing a non-destructive overlay way to run a full-fledged parallel computing environment on just about any workstation-class system...Were happy to say that this now includes the MAC too!

The BCCD does share similarities with a few diskless solutions for clustering, such as the Warewulf project, the thin-OSCAR approach, Cluster Knoppix (only an openMosix system, no MPI/LAM/PVM build tools, ...), and so on. This is definitely the trend in HPC. But the main differences are that the BCCD will always fit in your pocket, be highly customizable for specific institutions needs, and will always be geared toward education and not dedicated clusters.

The "gar" build system also sets the BCCD apart from other projects. "gar" is a mix between BSDs "ports" system, Linux from scratch, and gentoo Linux. With gar, you can build an entire BCCD image from net-fetched sources in about two hours (assuming you have a primed ccache!).

The BCCD is also distinctly different from NPACI-Rocks, OSCAR, Cluster in a box or other type of mass-imaging clustering project for two reasons:

1. Its a non-destructive overlay on top of the current hardware. Once a system is rebooted, it reverts back to its original state. It is intended to be booted "over top" of a currently-configured Windows/Linux/BSD/etc. system.
2. Its focus in on educational aspects of High-Performance Computing (HPC) instead of the HPC core. Students will have a much better appreciation and understanding of how to tweak an MTU setting or wire the topology across a cluster if they understand how a distributed computation is laid out! Emphasis is placed upon building, configuring, and running distributed applications.

Microdrive is basically a set of scripts which you can use to make your own live linux cd.
It depends on busybox, isolinux and linux of course.
I am planning to build a small bunch of diskless, monitorless, keyboardless nodes to do my rendering work [3D]. For them to run, I have various options but live cd seem to be most trouble free.
But then I couldnt find a live cd, that more or less faithfully follows a ordinary installed linux system. They seem to use squashfs or unionfs or some other form of pseudo writable file systems. I couldnt even get the slax to show me a screen. I am sure it is to do with my laptop graphics card.
But it sort of inspired me to make my own live cd, so I can match the components to the same library versions of my installed systems.
To this end instead of making a single cd, I broke the work needed in to a set of scripts so that the whole thing is manageable and reusable.
Hopefully it will be useful to someone else out there.
To make a live cd :
1. Get the Microdrive Linux Live script set from above.
2. Untar the files to a directory.
3. Make a kernel appropriate for live cd.
needs initrd, ramdisk and ext2
preferably also support basic devices as below. [ cd, devfs, framebuffer, vesa etc ]
No hard and fast rules. Afterall its your damn cd!
4. Copy the kernel to ./src/bzImage .
5. Download busybox source tarball and put it in ./src .
6. Download syslinux source tarball and put it in ./src .
7. Run the scripts listed in ./scripts directory. [ Please note. Dont cd in to scripts directory. Rather run them as ./scripts/SCRIPT_NAME . All the scripts assume that they are working from directory where ./src and ./scripts ./doc etc are the sub directories.]
8. The purpose of scripts [ At first run them in this order as well ] :
./scripts/make-prelim-dirs : This will make all the necessary directories such as itree - where the initrd image will be made. cdtree - whatever is here will endup on cd. cdtree/zz - whatever other software you want to install, install it here. Also /etc/rc.d/links on the live cd will link lib, usr, etc, etc/X11/* from here to the live system during runtime. cdtree/isolinux - where the initrd.img, isolinux end up.
./scripts/build-busybox : This will build and install busybox in to itree. Will also chmod u+s the busybox library.
./scripts/build-isolinux : Well, guess what.
./scripts/find-libs : After installing the busybox binaries, you will ofcourse need the libraries i:e; libc, libm, ld-so etc. Well, this does exactly that. It cannibalises them from your host system.
./scripts/install-base : This will install a basic /itree/etc to go in to initrd.img. You can customize the actions there to your hearts content.
./scripts/update-itree : As root, this will ldconfig the libs in itree. Then chroots to the itree. You should get a shell. If it doesnt work here, it aint gonna work on the cd.
./scripts/make-initrd : You need to run this as root. Please read the script before reaching out for the "su" . It will create a initrd.img from itree/ and puts it in cdtree/isolinux/ . This is what the kernel from live cd loads and feels content that it has a root file system.
./scripts/make-iso : Finally ! . Make a iso9660 bootable image to burn to cd or cdrw.
9. In other words :
./scripts/make-prelim-dirs
./scripts/build-busybox
./scripts/build-isolinux
./scripts/find-libs
./scripts/install-base
./scripts/update-itree [ as root ]
./scripts/make-initrd [ as root ]
./scripts/make-iso
Enhancements:
- Updated for the new kernel.
- No devfs now.

Cluster Live can do the following:
- Boots a cluster of diskless Thin Clients.
- Automatically loads a Web browser.
- Share workload between Thin Clients.
Why?
- No individual installations required.
- Fast deployment of centralised software over multiple computers.
- No need to access the hard disk of your existing system.
- Portable yet scalable infrastructure.
- Less spending on high end computer hardware such as hard drives, processors and memory.
Who can benefit from it?
- Public kiosk
- Internet cafe
- Home, Office and other intranet premises
- Training centre
- Government sectors, and other organisations requiring large deployment of dedicated applications.
- Ideal for academics in developing countries to make use of computers cost effectively.
- Open Source developers can freely create their own Cluster Live CDs.
How does it work?
- A server gets booted by a Cluster Live CD.
- The clients BIOS is configured to boot through the LAN by default.
- When the server has finished the bootup sequence, the client can be booted through the LAN.
What was it tested with?
- IBM Thinkpad T30 (CPU: Pentium 4 - 1GHz; RAM: 512MB)
- Virtual Machines (Software to allow running of multiple O.S. simultaneously)
- IDE CDRW (To rewrite the ISO image on to the CD for testing)
- A hub with several LAN cables connecting the infrastructure.
- Source built with Red Hat Fedora Core 3 Linux