Tip

A few months back, I wrote about my trials and tribulations with Samba in KDE + Feisty, but, now that I had a fresh install of Gutsy to play with, I tested the sharing applet once again. Once again, however, it failed out of the box, although I must say that it was not as fucking bad as the last time.

If you ever find yourself running into a monotonous

"The following packages have been kept back"

error message when you run sudo apt-get upgrade, then the "fix" is to either run apt-get dist-upgrade which is quicker, but sounds more ... dangerous or to run apt-get install [package list] where package list is a list of all packages that have been kept back.

Following my recent adventures with my motherboard, I wanted to set up a temperature sensor on my Kubuntu desktop, so that I could keep an eye on my motherboard temperature. Searching the repository came up with a number of potential options including ksensors. However, when I installed ksensors I couldn't find a temperature option; just memory usage and stuff.

Earlier today, I found that one of my boxen had hung. I tried restarting it, but was met with a keening siren alarm from my motherboard. Giving it a rest and restarting was either producing the same siren or a hang after a few minutes along with some nifty monitor graffiti.

I had to clean everything inside my PC today due to heavy dust accumulation and in the process remove the processor fan to clean it and the heatsink below it. However, while I usually mark the "right side up" for the fan, I'd forgotten to do so this time. While it made sense to keep the AMD logo on top, I was still not confident to start the PC without being completely sure. Google spouted advice hither and thither, but the general consensus appeared to be as follows:

• CPU fans should blow air into the heatsink.
• Case fans should bring cool air into the cabinet.

I quite often want to / have to leech all files of a certain type from a website. The best way to do this is by using wget. The following command will download all files of type mp3 from a website:
wget -r -l1 -t1 -nd -N -np -A.mp3 http://example.com/mymp3s

Do this from within the folder where you want to download the files to.

This is actually more for my own record so that I don't have to man and google next time, but I'll try and explain all the switches when I'm motivated to do so :)

Enjoy!

CentOS 4.5 (final)
Plesk 8.2

When I ran YUM to update my server, I ran into a dependency error:
Error: Missing Dependency: glibc-common = 2.3.4-2.25 is needed by package glibc-dummy-centos-4

One of my linux boxen (my fastest with the best hardware) died a couple of days ago :( Motherboard issues. My two other Kubuntu boxen are stocked with hard drives and therefore, when I wanted to retrieve data from the hard drive of my busted PC, I didn't want to risk fucking around with the already overloaded PSUs of said boxen. However, my sole Windows box was pretty free, but obviously Windows does not understand Ext3.

If you ever run into an error along the lines of 1155 instmsi30.exe not found specifying a directory, usually your CD-ROM drive, you need to update your Windows installer.

I suck at regex. But the fact that Eclipse leaves trailing whitespace everywhere is pretty fucking annoying. Although, this will apparently be solved in 3.3, that's still a month or so away (IIRC). Anyways, while I normally just use a $ regex search (whitespace switches don't like working in Eclipse's find dialog) to repeatedly weed out the bastards, today, I came across a more robust expression. Here goes:

Find: [\t ]+(?:(\n)|$)