paleo

got it fixed already? If not, try init 3 and then startx (when you do init 3, X is stopped so you shouldn't get an error about X already running). See what kind of error messages you get - if X seems to hang, press ctrl-alt-F1 which will get you back to the text console.

I have no idea what is going but I can confirm that my DVD drive also seems to ignore region settings when running in Linux. It's a DVD/CD-RW combo drive on a Toshiba satelite 1805 laptop.

that should be enough. If you have a windows installation, you can grab windows fonts as well. Just drop msgothic.ttc and msmincho.ttc into your /usr/share/fonts/ttf/japanese and then run fc-cache -fv (as root) and the fonts will be available. Have you tried typing in Japanese? That is, try create a new file with a Japanese name? If you can save the file and its name displays correctly, then it should be possible to convince your setup to show Japanese file names correctly even if they sit on a windows partition.

try replacing the ja_JP with ja_JP.UTF-8 Also, instead of modifying the /etc/sysconfig/i18n, you can copy the file to ~/.i18n (hidden file). This way, you don't have to mess with system wide settings and the changes to ~/.i18n take effect once you log out and in again instead of having to reboot. If this doesn't work, try modifying the /etc/fstab entry. Maybe iocharset=iso-2022-jp instead of iocharset=utf8 will do. Or maybe it has to be iocharset=iso2022jp (not sure).

I can't imagine mandrake 10 (or 9.2 for that matter) not satisfying the requirements for the gimp2. If you got compilation errors, it probably due to not installed devel packages. You probably need to install things like: freetype2-devel fontconfig-devel pango-devel The real names for the devel packages might be a bit different but you can find them with the search function of the software management tool.

Do you use any japanese on your system? try starting gaim from a terminal window like this: env LC_ALL=en_US gaim if this works, there is probably a problem with your /etc/sysconfig/i18n or ~/.i18n (the latter takes precedence). The file(s) have entries like LC_MESSAGES=en_US you might have some lines set to something like: LC_*=ja_JP change such lines and log in again and see if gaim starts in english.

I'm not sure why you want to compile freetype2, actually. Freetype2, fontconfig and pango should all be installed by default on Mandrake and gimp should be on the installation CDs and easy to install through mandrake's software manager. Is there any particular reason why you want to install these things manually?

I don't think there's going to be any problem. But if you want to be extra careful, here is what you can do: 1. copy the directory /usr/X11R6 to /usr/XFree86 2. copy the directory /etc/X11 to /etc/X11 now you have backups of the Xfree86 installation. 3. install x.org (using rpms, tarballs, compile, whatever) 4. see if it works. You can use the old config file which is /etc/Xfree86/XF86Config-4. Copy this file over to the new /etc/X11/ folder and rename it xorg.conf 5. if you have problems, you can simply get back to your Xfree86 using the backups. (copy /usr/XFree86 to /usr/X11R6; the same with /etc/XFree86) Of course, you'll need to know a little how to deal with no gui. But honestly, I don't think there's going to be a problem.

your printer is listed so it should work. Have you tried using the printer setup wizard after you've installed mandrake? The printer wizard is somewhere in mandrake control center. The printer should be installable that way.

How did you install freetype2? If you used the rpm method, install freetype2-devel using the same instructions with freetype2 replaced with freetype2-devel. If you didn't use an rpm, please tell us how you installed freetype2.

Did you get any further? What do you mean by saying you can't see kanji on the files? If you mean, you can't read the file names, on what kind of file-system are the files? If it's a windows partitions, you probably have to tell mandrake the right encoding for it. It's done in /etc/fstab. My entry is /dev/hda1 /mnt/windows ntfs noauto,iocharset=utf8,ro,umask=0 0 0 it's the iocharset part that matters. Some applications need to be told to use the right font before they can display kanji. Xmms is like this - you'll have to try which font works.

hi all, I've been trying to compile glibc2.3.2 on my mdk9.1 installation (just for fun and learning). I realized that I had to replace the kernel headers in /usr/include with the ones from /usr/src/linux-2.6.4/include in order to get the compile to finish when using the option --with-kernel-2.6.4 (I still have problems but that's a different topic). My question is this. Why are there kernel headers in /usr/include or rather, why is it that most programs as far as I know look for the kernel headers in /usr/src/linux/include when there are kernel headers in /usr/include? and why leave ancient headers in /usr/include when most programs don't look there for the kernel headers anyway? Should I expect compilation problems now that I've replaced them? thanks for any input.

I can't find anything on the security implications of root sudoing as a normal user (I guess it's not a normal thing to do)... Anyway, just found out that something simpler does it and seems safer, too: su -l {username} -c startx this way, I don't have any environmental variables around that indicate something funny. With the sudo method, some env variables clearly indicate that root is sudoing as a normal user. One catch is that on this method there is no login prompt available. On ctrl-alt-F1 I can see the X startup messages, and X is running on ctrl-alt-F2. The rest is blank. I guess not too bad. I can still bail out of X and get to a prompt with ctrl-alt-backspace or ctrl-alt-F1 and then ctrl-C. Edit: never mind. As they always say, rtfm. man inittab just solved things. no need to deal with rc.local. Edit2: one more thing. When I do this 'auto login' thing, the who command returns an empty list. No one's logged in. Is this bad? Things seem to be working but...

Hi all, I'm trying to figure out if it's possible to auto login into KDE without using a graphical login manager and the auto-login feature of it. Here is one thing that kind of works: sudo -u {username} -H bash -l -c startx if I put this is as the last line of rc.local, it will startx as the user {username}. It all seems fine except that the PATH remains that of root plus whatever other directories I specify for the user ($HOME/bin, in my case). I could get around this by modifying .bash_profile, BUT: I was wondering if there are other things that are bad about 'logging in' a user in this way. Are there things that a user could execute with root privileges on this way of logging in? Are there other ways of accomplishing what I want? Basically, I need a way of logging in automatically. In case you're wondering: I don't like graphical login; I once had a bad experience with it as a newbie and developed a phobia; plus, booting takes longer.

I just went through the trouble of compiling and installing XFree86 4.4rc3 and wanted to share the little I learned with the installation. First thing: don't do it if you have a working XFree86. It's probably not worth it. I regret having gone through the headaches. and there's the whole questionable licensing thing. If you do want to do it anyway: 1. backup the directories /usr/X11R6 to /usr/X11R6-4.3 and /etc/X11 to /etc/X11-4.3. This is important or else you might end up with a hosed system with a lot of trouble getting back a working X. Better yet, backup the whole system... 2. compile and install X. The instructions are here: http://www.xfree86.org/4.3.0/BUILD.html I decided not to bother messing with the config files. The documentation for doing it is dismal 3. the new X is in /usr/X11R6. Move this to /usr/x11R6-4.4 (or whatever) and then create a symlink /usr/X11R6 to /usr/X11R6-4.4. The same with /etc/X11. Move it to /etc/X11-4.4 and then symlink /etc/X11 to /etc/X11-4.4 (I actually just copied /etc/X11-4.3 to /etc/X11-4.4) 4. go into /usr/X11R6-4.4/lib 5. remove libfreetype*. I think there are three. Remove them. I also recompiled freetype2 but I'm not sure if it's needed. 6. remove libfont*. I think there are six or so. Remove them too. I had font rendering problems for qt-based apps (all kde apps) before figuring this one out. I even recompiled qt but I doubt it's necessary. There's definitely no need to recompile gtk. 7. run ldconfig 8. startx and hope everything works. (if you didn't stop the xfs service, do so and restart it before startx) 9. if it looks like there are things not working, you can compare /usr/X11R6-4.3 and /usr/X11R6-4.4 and see if you might want to remove/add things from /usr/X11R6-4.4 (this is how I figured out step 7.) 10. if it still doesn't work and you are tired of messing around, symlink /usr/X11R6 to /usr/X11R6-4.3 and /etc/X11 to /etc/X11-4.3 and you'll have the old X back. You can still tinker around with the new version of X since it's still in /usr/X11R6-4.4 and the configs in /etc/X11-4.4.

I'm using kernel 2.6.2 with mdk9.1 on a Toshiba Satelite 1805 (I think it has similar specs to yours) and apm is working so I would think it'll work on yours as well. Make sure that in /etc/lilo.conf the kernel options apm=on and acpi=off are set for the 2.6 kernel. Also you might want to recompile the kernel to make sure all the right options are selected.

never mind the ps/2 mouse. It's working now for some reason. I hope this isn't random.

Hi, I'm running mdk9.1 (with a lot of customization) on a Toshiba Satelite 1805 and I just installed kernel2.6.2 from kernel.org. After a bit of messing around, I got it running pertty well (some modules names changed from 2.4. series so I had to take care of that mostly and some modules I have to load manually somehow - I do it in rc.local). But I have a few problems with apm suspend (swsups doesn't work for me - never did; it always hangs when bringing up USB). 1. when I choose to restart network during suspend (in /etc/sysconfig/suspend), the process hangs. It cannot unload the eepro100 driver for my ethernet card (when this happens, I can't unload it manually either and I have to do a forced shutdown with alt-sysrq-S,U,B ). This is kind of random; sometimes the unloading works, at other times it will hang. I can get around this one by choosing not to restart the network but why can't I unload the module in the first place? Never had this with 2.4.22. 2. my ps/2 mouse goes nuts after resuming - the mouse pointer behaves as if it were getting fed by a random number generator. I can prevent this from happening by unloading the module psmouse before suspending. But then I have to load it again. This never happened with 2.4.22 either. With regard to 2, I figured out how to unload the module automatically. I just inserted the line rmmod psmouse in one of the suspend scripts. But I can't get it to load automatically. I thought it should do it when I insert modprobe psmouse somewhere but it doesn't. Here is a modified script: #!/bin/bash . /etc/sysconfig/suspend ARG=$1 function suspend { if [[ $RESTART_NETWORK == "yes" ]];then rmmod psmouse fi } function resume { if [[ $RESTART_NETWORK == "yes" ]];then /sbin/modprobe psmouse fi } case $ARG in suspend) suspend ;; resume) resume ;; esac As you can tell, I cannibalized the /etc/sysconfig/suspend-scripts/suspend.d/network script since I don't use the network options anyway (restarting network is a bad idea on this laptop). The suspend part works, the resume part does not. Why? Any ideas? I kinda wonder if I'm misunderstanding the function of the resume part...

I have a similar problem with psmouse driver compiled as a module. try modprobe psmouse as root once you're done booting (obviously, you need to be in text mode or have a way of opening a console window).

the addresses should be in the directory ~/.kde/share/apps/kabc/ (notice that it's kabc and not kab). The addresses are in the file std.vcf in that folder. There is also ~/.kde/share/congif/kaddressbookrc which stores the config of the kaddressbook

You could see if you can login from a text console. Press ctrl-alt-F1 (thru F6) to get to a text login. If that doesn't work, try booting into single user mode by pressing ESC at the lilo boot menu and then linux 1 you should then be able to change the root password. You can then go into multi-user mode init 3 see if you can login as a normal user or as root. If you cannot login as a normal user but as root, login as root, then su {name of normal user} and then passwd to change the password. All of this assuming that the only problem has to do with the password...

If you want kde3.1.4, I would get the ones from texstar. Go to PcLinuxOnline.com for the link or use easy urpmi for configuring urpmi for texstar. Edit Oops. sorry. Texstar's rpms are for 9.1 and not 9.2

A lilo.conf for multiple booting three different distros and windows. A lot of people seem to have difficulties figuring out how to set up lilo.conf for multiple booting different installations of linux. The trick is to mount the / partition (or /boot if you have a separate /boot partition) to somewhere in the directory tree of the distro from which you will run lilo. In my case, I mount the / of Slack under /Slack and the / of Knoppix under /KNOPPIX and then run lilo inside mandrake. (It also has the options to boot with the old kernel and with ide-scsi on or off) boot=/dev/hda map=/boot/map vga=normal default="linux" keytable=/boot/us.klt prompt nowarn timeout=100 message=/boot/message menu-scheme=wb:bw:wb:bw image=/boot/vmlinuz-2.4.21-0.25mdkcustom2 label=linux root=/dev/hda6 read-only optional vga=normal append="quiet devfs=mount hdc=ide-scsi apm=on acpi=off" initrd=/boot/initrd-2.4.21-0.25mdkcustom2.img image=/boot/vmlinuz-2.4.21-0.25mdkcustom2 label=linux-nonscsi root=/dev/hda6 read-only vga=normal append="quiet devfs=mount apm=on acpi=off" initrd=/boot/initrd-2.4.21-0.25mdkcustom2.img image=/boot/vmlinuz-2.4.21-0.13mdk label=linux-mdk root=/dev/hda6 read-only optional vga=788 append="quiet devfs=mount hdc=ide-scsi acpi=off" initrd=/boot/initrd-2.4.21-0.13mdk.img image=/boot/vmlinuz-2.4.21-0.13mdk label="failsafe" root=/dev/hda6 initrd=/boot/initrd-2.4.21-0.13mdk.img append="failsafe devfs=nomount hdc=ide-scsi acpi=off" read-only image=/KNOPPIX/boot/vmlinuz label=Knoppix root=/dev/hda14 append="devfs=mount hdc=ide-scsi acpi=off" image=/Slack/boot/vmlinuz-ide-2.4.22 label=Slack root=/dev/hda13 #initrd=/Slack/boot/initrd-2.4.21-0.13mdk.img append="devfs=mount hdc=ide-scsi acpi=off" read-only other=/dev/hda1 label="windows" table=/dev/hda

why not use konqueror? type man:mplayer for an html formatted version of the mplayer man page. Works with any man page. And it even links to other man pages. You can also use just man: and you can browse through the man pages in konqueror.

I remember having exactly the same problem. I don't remember what I did to solve it but I recently installed 9.1 on another computer and it could update everything w/o a problem. I just used urpmi --update --auto-select (not practical for you; I'm on broadband) Anyhow, did you try checking mozilla, libnspr and libnss manually - i.e. you would tell rpmdrake to install those things at once. The problem seems to be that rpmdrake fails to see the dependencies so does not automatically check all of them to be downloaded.