<-- | Map | Index | Search

Welcome to the LinuxFocus November/December 2003 issue

It is already late in the evening and I am still wondering what to write in the editorial. How about SCO? The products of SCO are press releases these days. It is clear that the main ambition is to drive up the stock price, make as much money as possible and then disappear into irrelevance. The link to Linux? Well if you look at SCO more closely then the whole thing has not much to do with Linux. They just needed some kind of target.

What else did we have the last few weeks? There was the ruling on software patents in Europe. To have no patents at all would have been better but otherwise we can be quite happy with the result.

All this does not really bother us at the moment .... Oh yes, this year I had to replace harddisks in almost all PCs and servers that I have! The first two were replaced already in March and I thought "OK, that can happen" but now I have replaced the 5th disk. How many computers does that guy have? Well I don't run a server farm. I have 7 disks on 5 computers. All the disks were less than 2 years old. The very old ones are still good. I had to replace the new disks.

The reliability of software is improving all the time but it seems that the hardware is now the weakest part. Harddrive capacities are sky rocketing. As a result the quality suffers.
I still remember the Quantum Fireball disk, 1Gb, in the first Linuxfocus server. It was running continuously 24h a day and it ran for 8 years with no problems.

Replacing a disk at home is not a problem. Usually you notice that some application hangs or saving a file fails. You type "dmesg" and you see an IDE seek error: It is time to buy a new disk. This is annoying but it becomes a real problem if it happens on a server far away. Even with redundancy and RAID you still need to replace the disk. The service hours are usually much more expensive than the disk. I would be ready to pay three times as much if the disk has at least 8-10 years average life time. Anybody who has a solution? Would be nice to write an article in LinuxFocus about it. I am sure I am not the only one who likes to have a really reliable computer.

-- Guido Socher

LinuxFocus.org Articles

Graphics

GIMP: Make your own brushes , by Eric Lamarque

This article explains the various brush types that Gimp has and shows how to create own brushes.

Going 3D with Blender: A room with toys , by Katja Socher

In this third part of our series about Blender we set our train in a kid's room and model some more toys.

Software development

Concurrent programming - Message queues (3) , by Leonardo Giordani

This is the last article in the series on concurrent programming.

Book review: The Art of UNIX Programming , by Edgar Hernández Zúñiga

This a review of a new book from Eric S. Raymond.

Articles at Linux Gazette

Note: The real linuxgazette is now at linuxgazette.net and not at linuxgazette.com

System administration

An ISDN - Fax with Linux , by Sebastian Bauer

This article describes how to use your linux box as an ISDN fax machine.

UNIX Basics

Writing man-pages , by Guido Socher

Finally your finished you long awaited program and you want to distribute it. For others to use the program you need documentation. It's time to write a man-page.

Hardware

Temperature monitoring with Linux , by Stefan Blechschmidt

This article shows how you can monitor temperature with a very simple and inexpensive circuit.

The LinuxFocus Tip

This month tip shows how to display the output of syslog on a virtual console. Add this at the end of /etc/syslogd.conf:

# this is an additional condition and does not 
# affect any other logs:
*.*                            /dev/tty8

The condition "*.*" means everything and tty8 is the 8'th console. You can just press alt-F8 (or from X11 crtl-alt-F8, to go back to X11 press alt-F7) and look at the logging output. In most distributions console 8 is not in use for anything however which console is really free can be found in /etc/inittab file. The logger configuration can be found in /etc/syslogd.conf file.

This is very comfortable solution. You don't have to dive into /var/log/* files tree to see what's going on in your box. The logs on console are very helpful in tuning your system or finding out what is wrong when something does not work.

After you have changed /etc/syslogd.conf you have to restart syslog with the command:

service syslog restart      (for Redhat, Mandrake) 
or 
killall -HUP syslogd