Why Is There Yogurt In Your Cap?

OpenSuse 10.2 ZMD Removal

June 27th, 2007

No one likes ZMD. Heck, I don’t think the people who wrote ZMD like it. Luckily for us, the OpenSuse maintainers are getting rid of it for 10.3. Unfortunately, 10.3 isn’t here yet, so we need to get rid of it ourselves. This is how to do that.

There’s two ways to do it: The Easy Way, and the Easier Way.

The Easy Way:
rpm -e zen-updater
rpm -e rug
rpm -e zmd
rpm -e libzypp-zmd-backend
rpm -e sqlite-zmd

And the Easier Way:
rpm -e zen-updater rug zmd libzypp-zmd-backend sqlite-zmd

Now you may have a few questions, such as how are you easially going to update your system? Can I use something else to manage my packages? How old are you? Well, the answers to these questions are simple: Yes, Yes, and 31. The simplest way to update your system is to use the built in OpenSuseUpdater utility. You can launch OpenSuseUpdater by doing the following from a terminal:

opensuseupdater

Don’t be discouraged if it doesn’t run the first time you start it. I had to launch it, launch it again, exit my X session, and launch it one last time to get it to run properly on two of my systems. After that, it’s run perfectly for me. You should also be able to force it to launch on login by doing the following:
Computer->Sessions->Startup Programs->Add->opensuseupdater

As far as managing packages goes, I prefer to use the Smart system. The OpenSuse wiki has a page dedicated to setting that up, but that’s not going to stop me from typing up what I had to do to get it working. But that’s for another time.

Kill the Beagle

May 2nd, 2007

I dislike beagle. It can slow my system (any system) down to a crawl. Why it is enabled by default in the newer Suse/OpenSuse distros is beyond me. I dislike it almost as much as I dislike using the Zen updater (which I’ll remove later). So I get rid of it:

rpm -e kerry
rpm -e kio_beagle
rpm -e kdebase3-beagle
rpm -e beagle-evolution
rpm -e beagle-gui
rpm -e beagle-index
rpm -e beagle
rpm -e beagle-firefox

Alternately, you can do it all on a single line:

rpm -e kerry kio_beagle kdebase3-beagle beagle-evolution beagle-gui beagle-index beagle beagle-firefox

Bye-bye, Beagle.

I’ve decided: today is the day that I setup OpenSuse 10.2 on my new system. I just built this PC two weeks ago, and it’s been running Windows XP the entire time. My “old” PC got converted into my full-time OpenSuse box, but I like to have Linux on all of my boxes, at least in a dual-boot setup. Here’s what’s in my new system:

AMD Athlon64X2 6000+
Gigabyte GA-3PXSL-RH MB
PNY Geforce 8800 GTS 320MB
2GB DDR2-667 RAM
Seagate 250GB SATA2 HDD
WD 160GB SATA2 HDD
ASUS 18X DVD+-RW w/Lightscribe SATA
Plextor 48x CDRW IDE
Creative Audigy2

I booted off the DVD, and right away ran into problems. As soon as the bootloader loaded the kernel, my screen goes black. I figure it is probably something to do with my video card, so I reboot the system and press F3 at the DVD boot screen and change my screen to “VESA”. Kernel loads, I see text scrolling on the screen, and then the system locks. I reboot and do the same thing, expecting different results for some reason, but it still doesn’t work. So off to Google I go, searching for the last thing that is listed on my screen before the lockup, “net: registered protocol family 2 opensuse” . The first result has the answer; add maxcpus=0 as a kernel parameter during setup. That does the trick, and the system boots into the Suse installer. I chose all the normal additional things I do during setup (Kernel-sources, C/C++ Dev tools, findutils-locate and nano), and installation proceeds as I would expect it too. I let it update all of the installed software, so it’s less to do later. I get all the way through to where it is detecting hardware, and my screen goes black again. Crap. Reboot, Suse notices that the installation didn’t finish, picks up where it left off, and the gently proceeds to do the same thing.
Read the rest of this entry »

Feisty

April 6th, 2007

For some reason, I am a Suse Linux guy (for the most part). Over the years, I’ve used many distributions, but I keep going back to Suse. But Ubuntu intrigues me. Every time there is a new release I load it up on a spare machine and play with it. So when I saw the beta 7.04 was out, I downloaded it. I installed it. I updated it.

Then I tried to change my display resolution. My choices were 1024×768, 800×600, and the ever-present 640×480. Um, what? Ok, I figure that it didn’t load the proper display driver. I check the /etc/X11/xorg.conf file, and I see the open source “radeon” driver is loaded, which is fine for my purposes. Then I see what I figure is the problem: all my color depths have modes of the above-listed resolutions, and nothing higher. So I manually create the modeline for 1600×1200 and add it to all the color-depths. I restart X, expecting things to be right with the world. But it isn’t. It’s all still 1024×768, with no higher choices.

I head back to the xorg.conf file. This time, I notice that my monitor was detected as a standard VESA monitor with crappy horizontal sync and vertical refresh rates. I have a Dell 20.1 LCD, hardly what I would consider a strange display, considering that other distributions find it perfectly. I check the xorg.conf file on one of my other systems and input the correct values for my monitor in this one, restart X, and all is good in the world.

Why, since Ubuntu aims to be a simple-to-use distro, did I have to do this? I had to do this by hand because Ubuntu 7.04 “Feisty Fawn” Beta didn’t come with a utility to allow me to choose my monitor. If I had been an inexperienced user, I would have been screwed. Yes, I know it’s a beta, but we’re two weeks from the official release, and I doubt that this will be changed.

Other than the display issue, Ubuntu 7.04 seems to be very solid and usable. But I don’t think I’ll be switching from Suse permanently any time soon.

Update: Since I wrote this, the final version of “Feisty Fawn” was released. With the same monitor problem.

If you are like me, you’ve added different NICs to your laptop/PC and had them end up with strange names like “eth2″ when they should be named “eth1″, or whatever is next in line. Well, I know how to fix that (on SLED 10, at least). There is a file, /etc/udev/rules.d/30-net_persistent_names.rules that dictates what the NICs get named. Mine looked like this: (I’ve bolded the ones I want to remove)

SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:16:41:e2:dd:6b", IMPORT="/lib/udev/rename_netiface %k eth0"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:00:86:50:aa:ca", IMPORT="/lib/udev/rename_netiface %k eth1"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:0f:b5:89:29:43", IMPORT="/lib/udev/rename_netiface %k eth2"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:19:d2:34:dc:00", IMPORT="/lib/udev/rename_netiface %k eth3"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:04:e2:fe:a3:98", IMPORT="/lib/udev/rename_netiface %k eth4"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:13:f7:2d:9a:07", IMPORT="/lib/udev/rename_netiface %k eth5"

The eth1, 3 and 4 NICs are not present, so we need to take them out of the file. You’ll have to go by MAC address of the NIC here. I went through and removed the non-existent NIC lines, and changed the assignments of the present NICs. Now my file looks like this:

SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:16:41:e2:dd:6b", IMPORT="/lib/udev/rename_netiface %k eth0"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:0f:b5:89:29:43", IMPORT="/lib/udev/rename_netiface %k eth1"
SUBSYSTEM=="net", ACTION=="add", SYSFS{address}=="00:13:f7:2d:9a:07", IMPORT="/lib/udev/rename_netiface %k eth2"

After you make any changes to this, you’ll need to reboot the system, as the kernel reads this in at boot time. Additionally, if you run VMWare of any sort, you’ll need to run vmware-config.pl and reconfigure your virtual network settings to match up with any changes you made.

Fedora Directory Server Setup

November 13th, 2006

Did you ever want to setup your own LDAP directory server, and just couldn’t seem to figure it out? Do you have trouble getting your Java VM to export $JAVA_HOME? Have you ever had some really hot pizza and burned then top of your mouth, and yet you kept eating it? Well then, you’re just like me! Read below to see what I have gone through (and still going through) to get Fedora Directory Server up and running.

Read the rest of this entry »

Proudly powered by WordPress. Theme developed with WordPress Theme Generator.
Copyright &Copy; Why Is There Yogurt In Your Cap?. All rights reserved.