Groklaw’s PJ has the goods on SCO’s loss:

Court Rules: Novell owns the UNIX and UnixWare copyrights! Novell has right to waive!

Friday, August 10 2007 @ 04:52 PM EDT

Hot off the presses: Judge Dale Kimball has issued a 102-page ruling [PDF] on the numerous summary judgment motions in SCO v. Novell. Here it is as text. Here is what matters most:

[T]he court concludes that Novell is the owner of the UNIX and UnixWare Copyrights.

That’s Aaaaall, Folks! The court also ruled that “SCO is obligated to recognize Novell’s waiver of SCO’s claims against IBM and Sequent”. That’s the ball game. There are a couple of loose ends, but the big picture is, SCO lost. Oh, and it owes Novell a lot of money from the Microsoft and Sun licenses.

One of the witty commentators at slashdot pithily summarized the whole sorry ordeal:

There once was a CEO called McBride
Who thought he could take on Linux in stride.
But the creep from Santa Cruz
Was destined to lose
And get fscked in the ass by New York Gay Pride.

lnx4n6.be (get it?) is a Belgian site that has a Knoppix based forensics CD with a fascinating variety of tools. They can be used on both Micro$oft and Unix boxes — extremely useful for data recovery.

In following some of the trackbacks to the ZDNet article, Five crucial things the Linux community doesn’t understand about the average computer user, I stumbled across an excellent blog, Diary of a Mad IT Manager. As I was reading through the posts, I began to come across tales of the all too typical kind of disastrous, Microsoft-induced IT expenditures that I too had witnessed in my days as a tech support person, and later as a network administrator. For example:

A thought occurred to me today about the state of our computers here. We’re so used to rebuilding them, we’ve gotten better and more efficient at it. I mean, we’ve got all manner of support software designed to do nothing more than recreating and restoring an operating system and associated loaded applications on a computer in as efficient and timely manner as possible. Why? Because Windows sucks so much that we have to reload it on a regular basis.

And so I decided to put the question directly (though a bit profanely): why do corporations persist in using Microsoft products, when they so clearly cost so much more money than the alternatives? His reply, I thought, clearly articulated those reasons.

I replied in turn, and I think the exchange describes fairly concisely, at least in the corporate shop, the reasons for resistance to change: the cost, due to the weight of history, is too high. You may read the full exchange on Mad IT Manager’s blog here, or archived on my site here.

A couple of points (not in the order of your arguments) [with edits to my original reply]:

1. *Desktop* linux is brand new — it, and the dozens of application software projects necessary to make a desktop linux go has only begun to come together over the past two or three years. To wit: Ubuntu, now the most popular distro, is only four years old. Keep in mind this principle: once you written the third or fourth version of something like .pdf/web viewers, or word processors, little further innovation is needed. How much different is MS Word 95 from 98, 2000, XP, or 2007/8, or Abiword and OpenOffice for that matter? Not much. MS got the word processor gui right when it still was on the Mac, where it ironically began. So, there will be an open source equivalent for every type of software, and many are just now getting to an 85% to 95% equivalency to their proprietary cousins.

2. If you have any doubt as to the profound power of a monopolist to retain control of markets without juridical oversight, or of the mechanics thereof, it might be time to take a cursory review of Microsoft’s misdeeds, or of Standard Oil’s.

The latest catastrophe about to be visited on world wide consumers by the prostrate and corrupt US legal system? Vista’s draconian DRM.

Here’s an excellent run-down of what will probably result in the break-up of Microsoft, or the US loss of software hegemony: A Cost Analysis of Vista Content Protection, by Peter Gutmann.

3. Everything you say regarding Linux based Desktop environments’ need to pay attention to gui usability, and to test, is spot on. Gnome still feels clumsy, and KDE does fine, once cleaned up by somebody like the Ubuntu project. Gnome and KDE are vastly improved from even a year or two ago, as are many of the other window managers out there.

4. Diversity in Linux distributions is a strength, not a weakness. You badly underestimate the scale of development unleashed by open sourced software. In addition, what you’ve said is akin to arguing that the West has been hurt by the free exchange of knowledge that characterizes its universities; that a closed, peer-reviewless, proprietary system of academic inquiry would serve humankind the best. This is demonstrably false.

The US constitution of 1787 sought to balance the need to reward innovation with the need for open academic inquiry. From the the broken US patent system to the DMCA, to the moribund state of US Anti-trust enforcement, this system is now in dire need of reform.

The plethora of linux distributions, ranging from 50 MB fully functional gui’d desktop environments to multi-gig distros such as Ubuntu, openSuse, et al., to embedded systems vetted for stability — the Linux ecosystem’s diversity is a testament to the virtues of free inquiry, the absurdity of patenting software, and the novel scale of intellectual work made possible by the internet.

5. The command-line: this is just silly. The command line exists on Windows boxes. One can hack the registry or talk to the OS via the Windows API, or, conversely, one can never see a jot of code on a Windows box, same as on Linux boxes. If Linux desktop environment and application users too often are subjected to command-line fixes, that just supports your argument that desktop oriented Linux environments still need work. That work is progressing at an astonishing speed, thanks again to the scale of development made possible by the internet and open access to code.

What I have found, however, is that the Windows command-line environment is gravely impoverished compared to Unix/Linux. Yes, I have done extensive work on both platforms. The Microsoft command line environment has come a long way, but it still lacks the depth and richness that the Unix/Linux environment possesses.

The article that this post is a reply too: Five crucial things … .

I also posted the trackback above as a reply in the blog.

Printing from a fresh install of Kubuntu Dapper failed on an install that I did, and I’ve already replaced Kubuntu with openSuse on my laptop, and I’ll probably use Mepis for terminal server applications or desktops. Issuing a distribution with something like printing broken (and by ‘broken’ I mean anything that smells like an open-ended, hours long troubleshooting situation, even if that is somehow by design) is unacceptable.

Just to memorialize them, here are two helpful posts from LinuxQuestions.org:

See the tail end of this post here:

http://www.linuxquestions.org/questi…d.php?t=331050

I spent literally months f*-king around with Ubuntu trying to get it to print. The post referenced above details my sorry saga. Finally gave up and moved on to other distros. Found Mepis to work flawlessly. Did an A-B comparison, found numerous differences in config settings but identical processes, daemons, versions of executables, etc. Conclusion: Its all in the config settings. So in theory if these config settings work on Mepis, they should work on Ubuntu. BINGO BABY !!!

My bottom line advice for Ubuntu users is: If you want out of the box printing, you are hunting up the wrong distro. Try Mepis instead. If you insist on using Ubuntu - Kubuntu - Xubuntu, save yourself some torture and just pull actually useful config files off Mepis.

Warning: There is a risk that a s/w upgrade could toast your config files. I recommend archiving a copy of the Mepis /etc/cups folder and treating it as a data file to keep backed up. In the event printing mysteriously breaks, try reapplying the Mepis config files and see if printing is restored.

good luck.

And the second post:

Dapper printer problems
There seems to some real problems with printer setup in Dapper and a number of solutions that work for some but not others. I’ll just detail what it finally took to install my HP Deskjet as a local USB printer.
Code:

sudo foomatic-cleanupdrivers sudo adduser cupsys shadow sudo dpkg-reconfigure cupsys

Select the proper printer interface. For the rest, the defaults worked for me.
You should be part of the lpadmin group — if not, add yourself.
(The printer should be turned on of course.)
At this point the KDE tools still can’t find more than four HP printer drivers, so open a browser window at http://localhost:631/. The password problem should now be fixed.

Click on Administration and the printer should be listed.
Click to add the correct printer.
Click Set As Default.

I’d still like to know how to get the KDE tools to see all the printer drivers if anyone has a solution.

This post describes how to make Firestarter’s gui load at startup:

Launching Firestarter minimized to the tray on login

Having performed the above configuration of permissions, the system can further be set up to load Firestarter when you log in with your regular user account. Firestarter will in that case load directly into the system tray without user intervention, after which the main interface can be accessed by clicking the tray icon.

Using GNOME:
The GNOME sessions manager

Open up your GNOME menu, select Preferences followed by Sessions. Switch to the Startup programs tab, pictured right.

Click Add and enter
sudo firestarter –start-hidden
as the startup command. Click OK and you’re done.

To stop Firestarter from loading on login, simply remove its entry from the startup programs listing.

Using KDE:

Open a terminal and execute the following two commands:

echo -e ‘#’!'/bin/shnsudo firestarter –start-hidden’ > ~/.kde/Autostart/firestarter
chmod a+x ~/.kde/Autostart/firestarter

Firstarter will now load automatically when KDE starts. To stop Firestarter from loading when you log in, remove the ~/.kde/Autostart/firestarter file.

Jens Benecke discusses the relative of merits of KDE and OS X.

For example, he has experienced the same frustration I have witnessed with iPhoto’s (an OS X application) way of organizing things:

Also, I think Apple overdid integration at some point. Consider iTunes and iPhoto. I don’t like the fact that they basically reinvent the file manager. This forces you to learn a different interface to manage your files. It also kills the possibility to organize your files the way you want it. For example, I play in a band. I have a folder “Band” in my home folder. This folder contains “Originals”, “Gigs”, “Repertoire”, “New” and so on. “Originals” contains MP3s of the songs we play. Repertoire and New contain one dir for each song, and in this directory, there are scores (PDF, Garageband, Lilypond, etc), lyrics, further MP3s, recordings, pictures and so on. In the “Gigs” folder there are pictures and video/audio recordings of performances, demo tapes of performances etc.

So, basically I have organized my files according to purpose, not file type. The seperation of iTunes / iPhoto / iMovie etc. forces me to seperate my files according to file type. I could theoretically reorganize everything (about 2000 files), then create “intelligent folders” for each and everything in every application (why are these folders not shared?) and then tag the files so they get sorted into the right folders. Problem with that is: the tagging mechanism is not standardized. I can apply any number of free-form tags to photos. I can’t do that with MP3s and PDFs, for example. Plus, these tags would probably get lost when synchronizing with our band’s FTP server directory, which is shared by a dozen people. Directory information doesn’t.

This post is a very slightly abbreviated version of Ubuntu Blog’s excellent post. I did this on Kubuntu Breezy 5.10, and it worked fine. Now, I need to find out how to do this via a startup script, a la nfs. Does the ’sudo modprobe fusemount’ have to be done each session anew? If so, how does one do a sudo in a login script?

The following guide will step you through the process of mouting file systems over ssh on Ubuntu 5.10 Breezy systems.

sshfs:
sshfs is a filesystem client based on the SSH File Transfer Protocol. Since most SSH servers already support this protocol it is very easy to set up: i.e. on the server side there’s nothing to do. On the client side mounting the filesystem is as easy as logging into the server with ssh.

Install sshfs by doing a:
$sudo apt-get install sshfs

This will also install fuse-utils and libfuse2, which are required.

Now, let us create a local directory where you want the files mounted. You should create the directory and make yourseld the owner of the directory:

$sudo mkdir /media/dir-name
$sudo chown your-username /media/dir-name

Where “dir-name” is the name of the directory on your local computer where you want to access the files from the remote computer. Say I want the files on the server to be available at /media/home-pc. I want this because the “server” in this case is the desktop I have at home, which allows me to access it through ssh. Let us use “home-pc” as an example for this guide. “your-username” is your username on the local computer.

Go to System->Administration->Users and Groups, select the group “fuse” and then add yourself to this group.

If you prefer to do it the easy way, on the commandline, then use
$sudo adduser your-username fuse

[…]

For some reason, the /usr/bin/fusermount binary is installed in such a way that users cannot execute it. To fix this do a:
$sudo chmod +x /usr/bin/fusermount

[Mount as follows (do this on client machine)]:
$sshfs example.com:/stuff /media/home-pc

Where “/stuff” is the folder on the computer with the domain name “example.com”, which I want to mount and access on my local computer at the location /media/home-pc. Remember that the /media/home-pc directory must exist and be owned by you, the user. I already mentioned how to do this in the initial part of this guide.

If you get the following error:

fusermount: fuse device not found, try ‘modprobe fuse’ first

You will have to load the fuse module by doing:
$sudo modprobe fuse

You can add fuse to the modules that are loaded on startup by editing the file /etc/modules and adding a line with only the word “fuse” in it, at the end.

Then issue the sshfs command above again.

To unmount the directory once your work is done, use the command:
$fusermount -u

for example, in my case, I would use
$fusermount -u /media/home-pc

[Update, 2 December 2005]
On a Kubuntu Breezy 5.10 box, when attempting to launch nxclient (/usr/ ), I received this error message: “nxclient unable to create the x authorization cookie.” I found the fix here. This entry in that thread contains the fix (italicized, emphasis mine):

spikkle
10-02-2005, 10:09 AM
I was able to make it work by taking the following steps:

Install all FreeNX packages from here from the deb packages you are allowed to download:
http://blackbird.kaarsemaker.net/

Install the nxclient deb package from:
http://www.nomachine.com/download_client_linux.php

If you get a message to the effect of “Unable to create X authority” or something like that, then for me the problem was that it was looking in an (outdated) incorrect location for the ‘xauth’ program. Run these commands:

cd /usr/X11R6/bin
sudo ln -sf /usr/bin/xauth

This will make a link to the xauth program in the location it expects it to be in. At this point the client worked fine for me.

Good luck.

Just prior to that, I encountered an error which prevented the nxclient from loading at all. I ran the client from the command line so that I could see error messages, and this appeared: “nxclient: error while loading shared libraries: libstdc++-libc6.2-2.so.3: cannot open shared object file: No such file or directory.” I found the fix here.

The relevant post read:

The file you are looking for is in this package libstdc++2.10-glibc2.2

Get it from universe.

So, in a shell, I typed: sudo apt-get install libstdc++2.10-glibc2.2

One last note (sorry to do this in reverse order, but I just wanted to get this down where I’d be able to find it later): I needed to copy a key for the nxclient to use from the server to the client box. Freenx / nxserver use ssh to connect, and ssh requires an encrypted key pair, one for each box that participates in the session. When setting up the nxclient, you need to go to [the appropriate directory on the server] and copy the key over to your client’s /usr/NX/share directory (location may differ depending on which distribution you’re using).

One last, last note: I set up nxserver using Rick Stout’s excellent nxserver fedora setup instructions.

Another (really! the last!) note: in the last sentence of this post is a link about where the keys are located. Also, my update repeats what I had said earlier, in part — I was setting up another Kubuntu Breezy box, and had apparently run into this problem before, and forgotten.

[End update]

This post contained a fix for the “nxclient cannot open shared object” error that I received when trying to open in Kubuntu Breezy/5.10.

The relevant post reads:

Re: HOWTO: Can I use FreeNX with Ubuntu? - 4 Weeks Ago
In my original question I asked:

Quote:
Originally Posted by tlepes
Anyone here got nxclient working on Breezy? I used Seveas’s guide on the WIKI https://wiki.ubuntu.com/FreeNX but get this error when runnin nxclient on the client machine:

Code:

/usr/NX/bin/nxclient: error while loading shared libraries: libstdc++-libc6.2-2.so.3: cannot open shared object file: No such file or directory

Searching Synaptic I only find up to C/C++ 6.0 (methinks). Am I doing something wrong? Do I need to try compiling nxclient from source?

I got a fast response from “bob2″ in the #ubuntu channel on freenode IRC. He pointed me to a web page with search results for the missing library file “libstdc++-libc6.2-2.so.3″ that showed what packages it was. Issuing the following command solved my problems getting nxclient to run:

Code:

sudo apt-get install libstdc++2.10-dbg libstdc++2.10-glibc2.2

On further reflection, I suppose I could have gotten the missing library file by just installing one or the other, as opposed to installing both libstdc++2.10-dbg and libstdc++2.10-glibc2.2 packages.

That package search was very cool. I poked around a bit on the website and figured that bob2 must have used the “Search the contents of packages” section of this page: http://packages.ubuntu.com/#search_contents

I also made this comment earlier, which I still believe may be a source of trouble for people later on:

Quote:
Originally Posted by tlepes
I noticed that in synaptic the nxclient does not show anything listed for “dependencies”. Maybe this is a problem with the deb in Seveas’ repository?

Now lets see if we can get this mutha humming…

Other resources that might help (right now I’m getting an authentication error):
ltsp freenx wiki setup page

lug.mtu freenx wiki page

post on nx’s authentication keys and the directories to get from on server and to put in on client. The relevant excerpt:

> >> You will need to tell your NX windows client to use they private key
> >> that FreeNX created during installation. FreeNX creates a
> >> private/public key pair, and stores the private key into
> >> /var/lib/nxserver/nxhome/.ssh/client.id_dsa.key, and the public key
> >> into /var/lib/nxhome/.ssh/authorized_keys2.
> >>
> >> You’ll have to install the private key stored in
> >> /var/lib/nxserver/nxhome/.ssh/client.id_dsa.key into your Windows NX
> >> client. Don’t have a clue on how to do this, but for the Linux NX
> >> client is as simple as copying that file into
> >> /usr/NX/share/client.id_dsa.key.

Note that on my FC4 server the key directory was named:
/var/lib/nxserver/home/.ssh

Desination directory on host (Kubuntu/Breezy 5.10) was named:
/usr/NX/share

Another pretty good page for troubleshooting nx ssh authentication problems

BASH Guide

Internet FAQ Archive

sysadmin notes

I stumbled across an fairly good article, titled
Why Linux isn’t too fat & MS hurts customers, that enumerates reasons to switch from Microsoft to Linux (or, rather, to anything open source).

An excerpt:

The other big benefit is that companies can control their own destinies more finely by using Linux. When you use Windows, you get whatever version of Windows was shipping when you bought the PC.

Often, most of a company is using an older version of Office. Then someone gets a new PC with the latest Office, and starts sending around Word documents. People with the older versions of Office can’t read those, and pressure quickly rises to upgrade everyone to the latest Office. Microsoft could easily make the old and new file formats compatible, but they deliberately do the opposite to force unnecessary upgrades on the market.

The correct approach when your company stumbles into this situation is to take the opportunity to introduce OpenOffice into your company. Microsoft makes a big song and dance about “supporting what customers want,” but they really don’t support what customers want. Microsoft acts to protect its monopoly, rather than supporting what customers want.

At work, I tried to make the point that it might not be a good idea to encase the intellectual output of a thousand employees into formats whose proprietary owner not only has a record of extortionate capriciousness, but one who has broken the law to maintain its monopoly control of markets (I’m referring to Microsoft, of course), but I was ignored. One more reason to move on.

This looks like a pretty good set of instructions for installing linux to a usb pen drive.

Diet-PC is a tool for building thin clients. thinstation looks like a interesting thin client distribution/tool. FreeNX is an implementation of nomachine’s nx server.

[Edit: I was looking over my entries, and noticed that this article, clearly not written by me, was neither blockquoted nor attributed — a rather egregious oversight on my part. The article is by Daniel Petri, and can be found here.]

Usually, if you want to see all the used and listening ports on your computer, you’d use the NETSTAT command.

Note: The NETSTAT command will show you whatever ports are open or in use, but it is NOT a port scanning tool! If you want to have your computer scanned for open ports see this page instead (link will follow shortly).

Open Command Prompt and type:

C:WINDOWS>netstat -an |find /i “listening”
TCP 0.0.0.0:135 0.0.0.0:0 LISTENING
TCP 0.0.0.0:445 0.0.0.0:0 LISTENING
TCP 0.0.0.0:1025 0.0.0.0:0 LISTENING
TCP 0.0.0.0:1084 0.0.0.0:0 LISTENING
TCP 0.0.0.0:2094 0.0.0.0:0 LISTENING
TCP 0.0.0.0:3389 0.0.0.0:0 LISTENING
TCP 0.0.0.0:5000 0.0.0.0:0 LISTENING

You can redirect it to a text file by adding >c:openports.txt to the command, if you want to:

netstat -an |find /i “listening” > c:openports.txt

You can also change “listening” to “established” to see what ports your computer actually communicates with:

C:WINDOWS>netstat -an |find /i “established”
TCP 192.168.0.100:1084 192.168.0.200:1026 ESTABLISHED
TCP 192.168.0.100:2094 192.168.0.200:1166 ESTABLISHED
TCP 192.168.0.100:2305 209.211.250.3:80 ESTABLISHED
TCP 192.168.0.100:2316 212.179.112.230:80 ESTABLISHED
TCP 192.168.0.100:2340 209.211.250.3:110 ESTABLISHED

Note: In Windows XP and Windows Server 2003, you can type NETSTAT -O to get a list of all the owning process ID associated with each connection:

C:WINDOWS>netstat -ao |find /i “listening”
TCP pro1:epmap pro1.dpetri.net:0 LISTENING 860
TCP pro1:microsoft-ds pro1.dpetri.net:0 LISTENING 4
TCP pro1:1025 pro1.dpetri.net:0 LISTENING 908
TCP pro1:1084 pro1.dpetri.net:0 LISTENING 596
TCP pro1:2094 pro1.dpetri.net:0 LISTENING 596
TCP pro1:3389 pro1.dpetri.net:0 LISTENING 908
TCP pro1:5000 pro1.dpetri.net:0 LISTENING 1068

You can use PULIST from the W2K Resource Kit (Download Free Windows 2000 Resource Kit Tools) to find the PID and see what process uses it and who started it. For example, you found out that your computer had an open connection to a remote IP address on TCP port 80, and you don’t have any Internet Explorer or other browser windows open. You want to find out what process is using that session.

C:WINDOWS>netstat -no

Active Connections

Proto Local Address Foreign Address State PID
TCP 192.168.0.100:2496 212.179.4.7:80 ESTABLISHED 1536

You can then use PULIST with the FIND command:

C:WINDOWS>pulist |find /i “1536″

Process PID User
LUCOMS~1.EXE 1536 DPETRIdanielp

In this case, LUCOMS~1.EXE is run by DANIELP (myself) and as it happens, it’s the Symantec Live Update process.

You can also look in Task Manager for the respective PID.

1.

To set up Task Manager to show the PID column open Task Manager by using CTRL+SHIFT+ESC.
2.

Go to the Processes tab, click View and then Select Columns.

3.

In the Select Columns windows click to select PID and then click Ok.

4.

You can sort the PID column to display the PIDs in descending or ascending order.

To see all open, established, closing and other used ports type:

C:WINDOWS>netstat -a

Active Connections

Proto Local Address Foreign Address State
TCP pro1:epmap pro1.dpetri.net:0 LISTENING
TCP pro1:microsoft-ds pro1.dpetri.net:0 LISTENING
TCP pro1:1025 pro1.dpetri.net:0 LISTENING
TCP pro1:1084 pro1.dpetri.net:0 LISTENING
TCP pro1:2094 pro1.dpetri.net:0 LISTENING
TCP pro1:3389 pro1.dpetri.net:0 LISTENING
TCP pro1:5000 pro1.dpetri.net:0 LISTENING
TCP pro1:1084 srv1.dpetri.net:1026 ESTABLISHED
TCP pro1:2094 srv1.dpetri.net:1166 ESTABLISHED
TCP pro1:2365 srv1.dpetri.net:epmap TIME_WAIT
TCP pro1:2366 srv1.dpetri.net:1026 TIME_WAIT
UDP pro1:epmap *:*
UDP pro1:microsoft-ds *:*
UDP pro1:isakmp *:*
UDP pro1:1026 *:*
UDP pro1:1027 *:*
UDP pro1:1028 *:*
UDP pro1:1038 *:*
UDP pro1:1043 *:*
UDP pro1:1085 *:*
UDP pro1:1086 *:*
UDP pro1:1242 *:*
UDP pro1:ntp *:*
UDP pro1:1649 *:*
UDP pro1:1900 *:*
UDP pro1:2095 *:*
UDP pro1:2217 *:*
UDP pro1:ntp *:*
UDP pro1:1900 *:*

Again, in XP/2003 you can use the -O switch:

C:WINDOWS>netstat -ao

Active Connections

Proto Local Address Foreign Address State PID
TCP pro1:epmap pro1.dpetri.net:0 LISTENING 860
TCP pro1:microsoft-ds pro1.dpetri.net:0 LISTENING 4
TCP pro1:1025 pro1.dpetri.net:0 LISTENING 908
TCP pro1:1084 pro1.dpetri.net:0 LISTENING 596
TCP pro1:2094 pro1.dpetri.net:0 LISTENING 596
TCP pro1:3389 pro1.dpetri.net:0 LISTENING 908
TCP pro1:5000 pro1.dpetri.net:0 LISTENING 1068
TCP pro1:1084 srv1.dpetri.net:1026 ESTABLISHED 596
TCP pro1:2094 srv1.dpetri.net:1166 ESTABLISHED 596
UDP pro1:epmap *:* 860
UDP pro1:microsoft-ds *:* 4
UDP pro1:isakmp *:* 680
UDP pro1:1026 *:* 1040
UDP pro1:1027 *:* 1040
UDP pro1:1028 *:* 680
UDP pro1:1038 *:* 908
UDP pro1:1043 *:* 624
UDP pro1:1085 *:* 596
UDP pro1:1086 *:* 596
UDP pro1:1242 *:* 1040
UDP pro1:ntp *:* 908
UDP pro1:1649 *:* 596
UDP pro1:1900 *:* 1068
UDP pro1:2095 *:* 976
UDP pro1:2217 *:* 1856
UDP pro1:ntp *:* 908
UDP pro1:1900 *:* 1068

An article in three parts on Linux clusters. Recently I’ve discovered OpenMosix, and I’m interested in doing something like building a Graphics lab using nxserver. In a related vein, I wonder how load balancing works, or if it is possible, in the Linux terminal server variants (ltsp, freenx, etc.)?

Here’s an article on how to build a Linux Live CD. It looks like the pclinuxos distribution might be useful for automating and learning how to do this.

Here’s an article on FreeNX, a GPL’d terminal server: FreeNX article.