A common idiom in UNIX is to create a dir and then cd into it to do something. This is is a ksh compliant function to make a dir and cd into it... note that if you are dumb and give it more than one arguement you'll probably be sad.

To use put this in your .profile or .bashrc or where ever you put your shell tweaks. I've been told that this thing is easier to do in zsh.

--snip--snip--snip--
cdmkdir () {
mkdir "$*";
cd "$*";
}
--snip--snip--snip--

--timball

You can force the kernel to reuse the TIME_WAIT buckets though, by setting the /proc/sys/net/ipv4/tcp_tw_recycle entry to 1. I have tested this, and it works: very few TIME_WAIT entries will be present when using this setting. It could be however, that using this could affect the reliability of TCP connections. I believe problems may arrise in NAT environments.

This makes those annoying TIME_WAIT connections go away faster...

--timball

TIME_WAIT sockets in Debian GNU/Linux

Jailtime.org: Virtual filesystems for Xen

This site provides a variety of downloads and howto’s to facilitate use of the Xen Virtual Machine Monitor. Xen allows you to run multiple operating systems on a single piece of physical hardware. Here you will find Linux distributions that can run as Xen guests out of the box, obviating the need to create your own custom filesystems. The filesystems on this site have already been tweaked to deal with Xen’s idiosyncracies, and are also designed to be lightweight and minimally divergent from the original distribution.

Lazy man's quick and dirty way to setup some stripped down xen installs.

--timball

Start [Jailtime.org - Downloadable Images for Xen]

Screensaver linux distro. What will they think of next.

Linux screensaver for Windows

From the initial question posed by mydoghasworms, to the complete and utter lack of evidence to show acquaintance with the subject matter of the people posting, to the rather horrifying upward moderation of their comments, Slashdot has hit an all-time low with this article.

The fact of the matter is that the Linux Standards Base initially neither provided anything useful in the way of defining the structure and operation of a Linux machine, nor did it provide constructive justifications for the things that it *did* specify. Now in what seems like their nth revision, they *still* haven't gotten a clue, and have strayed even further into uselessness--a uselessness which now apparently incorporates a certification body! (Because there's nothing like bureaucratic overhead to make an operating system function smoothly!)

If you'd like some really stellar examples of the kind of mistakes they were making, the initial standards documents actually contained very little in the way of where files should go, and instead focused heavily on requiring people to use RPM. No joke. Their latest revision has entire sections devoted to what amounts to a re-documentation of the X library APIs. It's all well and good to attempt to freeze APIs--development teams for the various projects that make up the typical Linux machine do it all the time--but for an outside vendor to do this in the absence of any context for facilitating the upward change which makes Linux so attractive to people who actually know what they're doing, well, it just doesn't bloody work! A Linux machine effectively ceases to be a Linux machine once you've metaphorically vivisected it and pinned all it's various organs out on the block and then sealed it in ambergris for all time. It becomes a _dead thing_.

Probably a million times more useful is the Filesystem Hierarchy Standard located at http://www.pathname.com/fhs which, in a very straightforward and engineer-like way worries about the filesystem layout first, and trusts that vendors have engineers smart enough to be able to write shell scripts to determine whether or not something is going to work if/when it's installed.

The main difference between these two approaches being that the LSB folks seem to think that installation of software is something between sticking your hand and arm as deep as you can into a nest of sleeping scorpions in total darkness and possibly airlift-delivery of a playground deep into the heart of New York City. They apparently feel that the host operating system should be required to lie there perfectly still and "just take it". The FHS team's approach is one much more like "stuffing" a volkswagen bus. They appear to feel that as long as everything is laid out reasonably, that one can get feedback from the people already inside the bus about whether or not there's room for more people and whether or not they should come in through the door or the back window.

Stupidest Slashdot Topic Ever

] The purpose behind the buyout was to convert as much
] Linux code over as possible to another open source
] operating system project which is sponsored at
] www.gadugi.org. This project is hosted by the Cherokee
] Nation and is sovereign under US Federal Laws. This
] project is merging the Linux Kernel with the Open Source
] NetWare project and distributing the operating system.
] The site is operational and the full code repository will
] be posted with the merged operating system after the
] Cherokee Nation Public License is published in January.
] Anyone who wishes to participate can email the site and
] get an account.

Jeff Merkey has lost most of his brain.

MARC: msg 'Linux 2.6.9 and the GPL Buyout'

This is how you turn a set of fedora core isos into a bootable DVD.

--timball

Fedora Core DVD

Collected here are some of the ogg recordings from OLS2004 (http://www.linuxsymposium.org). Hilights include:

- "The Cursor Wiggles Faster: Measuring Scheduler Performance - Rick Lindsley"
- "I would hate user space locking if it weren't that sexy... - Inaky Perez-Gonzalez"
- "Linux Kernel Hotplug CPU Support - Rusty Russell"
- "Perl 6 - Damian Conway"

And the Keynote by Andrew Morton (new 2.6 series maintainer)

OLS Audio

I found a webpage that succinctly describes why I hate all gentoo users.

Welcome to Gentoo is Rice, the Volume goes to 11 here.

] Little cms intends to be a small-footprint, speed
] optimized color management engine in open source form.
] Since the initial release, back in 1998, lcms has been
] continuously evolving, mostly thanks to contributions
] done by skilled programmers, who generously donated their
] valuable time to review the engine and the documentation.
] For now, little cms has been ported to a big number of
] platforms and is currently used in many open source and
] commercial products, as well as being distributed in
] major linux enviroments.

*yay* color management is here to stay!

LittleCms, Great color at small footprint