~Blog~

…since I posted the last entry. This is because I wasn’t that motivated for writing anyhthing. Sorry about that…

Well, what happened in the meantime? I successfully started studying at the FHNW in Windisch and I’m really enjoying it. The professors are really capable and I already learned a lot of interesting stuff. Unfortunately I didn’t pass one module that dealt with electro-technics, but I’m looking forward to pass the rest. On the other hand I passed others very well, like the system administration module. Today I had the objective programming final exam and I think I made it as well. Now I concentrate on the two math exams, which make me the belly ache. So far, my first education comment.

Now some technical news:

The interessting other thing I want to tell you is the upgrading procedure of my macbook pro from harddisk to ssd. I own a 160G HDD and I wanted to move to a brand-new Intel x25 80G SSD Drive. My first approach was inserting the OSX Disk and boot it. Afterwards I tried to restore a just created timemachine backup. It all looked well until I tried to launch some application after the “successful” restore. Most of them could not be started and there was no way to get them working, even reinstalling them. I suppose the timemachine cant deal the fact, that the restore partition is smaller that the partition the backup was taken from.

However, my second approach was cloning the partition using disk utility from the OSX boot CD. Once again, OSX chocked on the big to small partition issue again. My only chance to clone it was resizing the partition to a size smaller that 80G. I did this using

diskutil resizeVolume “disk_identifier” 72G

but it always failed with the error “The partition cannot be resized. Try reducing the amount of change in the size of the partition.” Now I had a bellyful of the Disk and downloaded the GParted Boot Disk. Funny, but with the third party tool It finally worked. After resizing I booted the OSX disk again and i could also clone it. What a tough nut to crack.

Generally I’m really impressed from the speed of Intel’s masterpiece.

Belonging to this posting I want to present you an easier way to fix the cursor bug caused by the radeon opensource driver.

• Add two new entries to your /etc/apt/sources.list

deb http://ppa.launchpad.net/tormodvolden/ppa/ubuntu jaunty main

deb-src http://ppa.launchpad.net/tormodvolden/ppa/ubuntu jaunty main

• Add PPA Key from Tormod Volden (sudo apt-key adv –recv-keys –keyserver keyserver.ubuntu.com 0×4b1e287796dd5c9a)
• Upgrade Drivers (sudo aptitude update && sudo aptitude upgrade)
• Reboot Box and have fun

Afterwards you can remove the last added entries from the sources.list if you want.

I want to refer to the display bug I have with the opensource radeon driver. Alex Deucher commited 9 hours ago a bugfix for the annoying mouse cursor bug.  I compiled the driver and replaced the driver from repositories with the homebrew one. It seems like the problem has really been solved. I keep observe this driver-version for some days and hope the problem is gone now.

I wont keep back the way how I created the driver:

• git-clone git://anongit.freedesktop.org/git/xorg/driver/xf86-video-ati
• ./autogen.sh –prefix=/usr
• make
• make install

During the «make»-step I had to add some additional developement packages like mesa-common-dev, xorg-dev and xserver-dev.

Afterwards the files radeon_drv.la, radeon_drv.so, ati_drv.la and ati_drv.so located under /usr/lib/xorg/modules/drivers were updated.

You could also try to replace only the binaries in the last-mentioned directory. I offer you the 4 files for download (No warranty!). http://combatx.net/temp/radeon.tar

After a long-winded filesystem check yesterday i could successful recover my slug. I was unsure if some important files were corrupted and moved to /lost+found. To ensure the itegrity of all packages from dpkg, I installed the tool called  «debsums» from the repositories. I execuded “debsums -sa” to see, which files dont match to the md5 sum delivered in every deb-package.  I asserted that no files (in which I didnt made consciously a change) were differing from the original md5 sum. Phew!

The other thing, my umts-modem, isnt actually working like it should. Its recognized correctly by the harware abstraction layer, but the network manager cant etablish a connection. I already found a solution by unload the driver and reload it manually (modprobe -r option && modprobe -a option). Afterwards it (somtimes) worked like it should, I try to create an automatic solution with an udev rule, but more on this later.

Hey dear followers.. My NSLU2 with Debian installed on it had a blackout some time ago and for example the /tmp directory was no more usable (Errormessage: “Stale NFS file handle”). First I tried to do a fsck on reboot, but I had no success, the check could not be done. The conclusion was that I plugged the external drive, where the corrupt filesystem is stored on, into a personal computer and made a fsck with it. I was astonished that the whole procedure took about 6 hours and many files are now located in the /lost+found directory, unfortunately i spotted some config files there aswell. I’m curious if its still working.. I will see.

The other thing I did today was reinstall my eeepc. All worked fine and I restored the most of my settings. Its alot faster now than before, all the obsolesce stuff isnt installed anymore. I couldnt test the UMTS-card yet, but I think it will work fine.