rockybalboa

Sorry it's taken so long, new job, lots of hours. The only probs I had before with my Nv video card was getting the right drivers installed. At that time there were two sets of instructions floating around that were somewhat different, I had the wrong ones. And "surprise surprise" I got turned on to this group when looking for the answer, and since then this is my primary source for answers and learning.Yack, Yack,, Yack The kernel stuff I did was very early in my ongoing newbie phase. Meaning I did nothing regarding keeping notes and didn't have a very consistent setup for comparison. My guess is that my experience doesn't support upgrading. As I remember it was a big learning (ego) thing for me to compile my own kernel. Good experience, kind of like driving, fun when you're young, but later you only do when you need to. So I'll take my undocumented and inconclusive evidence and go install the new KDE. (The sh*t some of us will do in the name of entertainment) Thanks for the responses, Dan

When I loaded up MDK 10.0 (before I wiped it out) someone told me to install a different kernel. Upon doing so I was amazed at how much better the whole thing ran. Should I do the same since I installed MDK 10.1 Official? The install runs but I'm not real impressed. I'm currently running 2.6.8.1-12. Is there somewhere that keeps stats and comments on each kernel? I googled it and didn't find much. Also, when I installed my nVidia Geforce3 Ti250 in 10.0 it was a nightmare. Can I expect more of the same when I install my card properly this time? I'm currently running a generic driver but it's not real good looking. Thanks in advance for any suggestions. Dan

Hey all, Please don't flame me for this. I was in a heavy MDK learning mode after installing 10.1. I ran into problems (school, new job, bad motherboard) and gave up using (learning) my MDK install. Sorry, but I have 15 years Winblows experience and less than one on MDK and I was in emergency mode. This was a dual boot install with WinXP, that was working great at one time. Anyway, I want my MDK back. Problem is I lost the partition tables or whatever so it looks like a reinstall would be easiest. Of course I'm thinking in Micro$oft mode and have been told that you should never have to reinstall Linux. So if I do a reinstall, should I use the 10.1 ISOs that I downloaded right after 10.1 went to full public release? Or have they changed enough that I would benefit from fresh ones? Data transfer is not a problem since I have a high speed connection. Is there one last hope of reloading and saving my former configuration? As I said I was off it for 3+ months and after installing my new mobo, I screwed up the drive partitions on one (the Linux one) of my drives. So I have no way of telling what got damaged/lost, etc. Thanks to anyone that can advise me, Dan

Ahoy there, I'm just trolling for more opinions/suggestions. My MDK install has been sitting for a week now. It (the file browser) crashed while I was moving a vital paper, so now Tux will stand in the corner until I find a way to clean his mess. I'm still hoping to get a more solid OS going out of this. Also, is KDE 3.3 a good upgrade over 3.2? Thanks Dan

Since installing MDK 10 OE (my first 'nix system), I have had numerous problems and I am just getting around to trying to finding answers. Sorry, but most of these are in comparison to my XP partition. 1) It seems to me that the whole OS runs pretty slow, however intensive things still run OK (ie. music, mpegs) 2) I continually have file browser crashes no matter if I use Konqueror, Nautilus, or various others. I'll get about 2-3 minutes into browsing almost any directory and suddenly the whole thing will crash to the desktop. 3) I have tried regular Mozilla, Firefox (personal favorite from Winblows) and Opera, and have the same crash problem as above. However, this happens at random, sometimes 2 minutes sometimes 30-40. 4) Occasionally when I start up the desktop will hang while loading, requiring me to restart the machine. I'm using KDE, but I have duped it with Gnome also. I'm running this all through an Athlon Tbird on an Abit KT7A board, nothing else spectacular about the hardware, standard 3 year old Winblows gaming machine. After all this I'm curious about the benefits of MDK 10.1 how could (should?) I save what I have previously installed. When I set this up I was a total Linux virgin, I followed the instructions meticulously, but could I have missed something in the install? I'm open to any suggestions right now. I love it when it works, but it's not been very reliable. Thanks Dan P.S. I need to add that I am expecting this to eventually replace my Win system, but I'm a student and can't afford the risk of losing data.

I have been trying to use the MDK update too through the MCC. I followed the instructions on "Easy URPMI.org" because it looked like I was having a problem connecting to the mirrors. When inputing the mirror info Easy URPMI spits out through a terminal, I get the following errors: Does anyone know what my problem is and how to fix it?I have been following the urpmi.removemedia -a instruction (from easyurpmi.org), since the first failed attempt. After the first couple of tries with the same results, I went in through the MCC and removed the mirror entries, still got the same outcome. I'm sure that I didn't intentionally mess with these things prior to what I mentioned above. Thanks to anyone who can offer help. Dan P.S. For the record, I searched a couple hours for my own answer and didn't get far.

Sorry this may be in the wrong section but here goes. I'm running MDK 10 I was wondering if installing more fonts increases load time or uses extra memory like Windoze? Years ago, when alot of things were still free, I buried my Winbrick 95 system by installing a whole bunch of extra fonts. I laugh now because it was early in Win95 release and it took me a couple days to figure out what I had done wrong. I was going to add some TTF fonts to my MDK setup. Depending on the answer to the above, would it be advantageous to unload some of the fonts that were installed by default? Thanks, Dan [moved from Installing Mandrake by spinynorman]

I guess it's because I have read listings from Gnome fanatics, KDE fanatics, fluxbox.., etc. and felt like I was missing something. But it doesn't look like it. And frankly after the description it looks like a big pain in the arse. I doubt I'll follow through with this idea. Thanks for the advice, you guys saved me headaches, DW

When I installed MDK 10 oe I used default for most graphic options. I could be wrong on alot of this; but, I believe that the desktop I wound up with is an amalgamation of KDE 3.(2?), anyway it's KDE and it seems to have a bunch of Gnome stuff thrown in. What I want to do is setup separate desktops one pure KDE, one Gnome, and then one of the lightweight options (IceWM, etc..), so I can see which one best suits my use. Honestly, I can't tell all the specific KDE, Gnome features. Does anyone know how to set this up? This just for comparative purposes, so I'd rather not go into editing pre-existing items. I want to play with the options, but it will be too new to me to mess around changing or scripts. Thanks DW

Another mess solved! Deleting the resume entries did the trick. Everything seems to be working ok. The first couple of boots went really slow. What does this message mean drakboot[4355]: check_geometry_using_the_partition_table failed for (hde1, 14346044): 892,254,62 vs 1023,2,62 with geometry 155061/16/63 It's from the bootlog. While trying to pinpoint the reason for the slow boot (and curiosity's sake) I found it. My greatest concern is that after all this partition juggling there might be something screwed up that will sneak up later. I particularly don't like the idea that it's talking about the boot sector. Thanks again for the help! DW

Ouch you caught me. I meant for the resume line to read hde11. Although it doesn't give me any error on boot, despite the fact that it's clearly wrong. I can't figure out how I should be getting the error involving hdg5. I am not trying to suspend. If you are refferring to the resume lines, they were there when I first opened lilo.conf. Haven't tried hardware suspend mode and don't intend to. Or does the suspend have to do with the fact that it's a journaling file system? Additionally, I get the resume errors listed earlier when ever I try to boot the -13 kernel. What would happen if I edited out the resume part of the lines? Yes, I ran the swapon command. It gave no output so it either didn't work or it worked without a problem. I'll go back and edit the errant lines. I doubt it will change my problem, but that's a highly uneducated guess.

Back I did as you said. At swapoff /dev/hdg9 I got "invalid argument" I assumed you meant append in stead of resume at resume=/dev/hde11 Because I had no resume entries. Any way I can now get all options in my Lilo.conf to boot "except the -13 kernel" I gives me resuming from /dev/hdg9 Resuming from device hdg9 Kernel panic: Resume:Unable to find suspended-data signature (-misspelled? Not sure what else to do. It looks as if all the partitions are still good. Is this possible a LILO problem? If so can I install GRUB with the current state of affairs? I'm confused! Here are my new LILO.conf and fstab in case they will help. /dev/hdg1 / ext3 defaults 1 1 /dev/hdg5 /boot ext3 defaults 1 2 none /dev/pts devpts mode=0620 0 0 /dev/hdg7 /home ext3 defaults 1 2 none /mnt/cdrom supermount dev=/dev/hdc,fs=udf:iso9660,ro,--,iocharset=iso8859-1 0 0 none /mnt/floppy supermount dev=/dev/fd0,fs=ext2:vfat,--,umask=0022,iocharset=iso8859-1,sync,codepage=850 0 0 /dev/hde1 /mnt/win_c vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde5 /mnt/win_d vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde6 /mnt/win_e vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde7 /mnt/win_f vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde8 /mnt/win_g vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde9 /mnt/win_h vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde10 /mnt/win_i vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hdg8 /mnt/windows vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 none /proc proc defaults 0 0 /dev/hdg6 /usr ext3 defaults 1 2 /dev/hdg9 /var ext3 defaults 1 2 /dev/hde11 swap swap defaults 0 0 # File generated by DrakX/drakboot # WARNING: do not forget to run lilo after modifying this file boot=/dev/hde map=/boot/map default="windows" keytable=/boot/us.klt prompt nowarn timeout=300 message=/boot/message menu-scheme=wb:bw:wb:bw disk=/dev/hde bios=0x80 disk=/dev/hdg bios=0x81 image=/boot/vmlinuz label="linux" root=/dev/hdg1 initrd=/boot/initrd.img append="devfs=mount acpi=ht resume=/dev/hde11 splash=silent" vga=788 read-only image=/boot/vmlinuz-2.6.3-7mdk label="263-7" root=/dev/hdg1 initrd=/boot/initrd-2.6.3-7mdk.img append="devfs=mount acpi=ht resume=/dev/hde11 splash=silent" read-only image=/boot/vmlinuz label="failsafe" root=/dev/hdg1 initrd=/boot/initrd.img append="failsafe acpi=ht resume=/dev/hde11 devfs=nomount" read-only other=/dev/hde1 label="windows" table=/dev/hde image=/boot/vmlinuz-2.6.3-13mdk label="263-13" root=/dev/hdg1 initrd=/boot/initrd-2.6.3-13mdk.img append="devfs=mount acpi=ht resume=/dev/hdg11 splash=silent" vga=788 read-only

OK I understand how to execute the instructions. But the GUI DiskDrake says that hdg9 is currently the /var partition. When I try to unmount it I get an error that says the partition is currently busy. Does /var have to have it's own partition or can it share. If it can share can I just delete it or do I have to move it. Do I need to learn how to force umount for it?

OH OH, I'm in deep now. You were both right, either command allows me to boot MDK fully. Now the problem with the hard disks is worse. I think I'm showing partitions that are not there or some aren't getting recognized or something. My initial thought was right, I had created a swap partition but had failed to format and mount it. I did so right after getting MDK started. DiskDrake or whatever it is won't mount my floppy so I created a backup of my partition table (corrupted by now?) and pushed it over to my WinDrive on IDE1. I also coppied the fstab and other info. I'm right back to the original error. However; when I go to the DiskDrake hdg9 doesn't even show up, all I see are hdg1,5,7,6,8. I mounted the swap file(on hde) and when I went to exit the utility said it was going to write partition table of hdg. So I canceled it, half of hdg is a 40 GB FAT32 partition full (27GB) of irreplacable music files. Now I'm afraid to even reboot the damn thing. The following are copies of my fstab and lilo.conf /dev/hdg1 / ext3 defaults 1 1 /dev/hdg5 /boot ext3 defaults 1 2 none /dev/pts devpts mode=0620 0 0 /dev/hdg7 /home ext3 defaults 1 2 none /mnt/cdrom supermount dev=/dev/hdc,fs=udf:iso9660,ro,--,iocharset=iso8859-1 0 0 none /mnt/floppy supermount dev=/dev/fd0,fs=ext2:vfat,--,umask=0022,iocharset=iso8859-1,sync,codepage=850 0 0 /dev/hde1 /mnt/win_c vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde5 /mnt/win_d vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde6 /mnt/win_e vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde7 /mnt/win_f vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde8 /mnt/win_g vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde9 /mnt/win_h vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hde10 /mnt/win_i vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 /dev/hdg8 /mnt/windows vfat umask=0022,iocharset=iso8859-1,codepage=850 0 0 none /proc proc defaults 0 0 /dev/hdg6 /usr ext3 defaults 1 2 /dev/hdg9 /var ext3 defaults 1 2 # File generated by DrakX/drakboot # WARNING: do not forget to run lilo after modifying this file boot=/dev/hde map=/boot/map default="windows" keytable=/boot/us.klt prompt nowarn timeout=300 message=/boot/message menu-scheme=wb:bw:wb:bw disk=/dev/hde bios=0x80 disk=/dev/hdg bios=0x81 image=/boot/vmlinuz label="linux" root=/dev/hdg1 initrd=/boot/initrd.img append="devfs=mount acpi=ht resume=/dev/hdg9 splash=silent" vga=788 read-only image=/boot/vmlinuz-2.6.3-7mdk label="263-7" root=/dev/hdg1 initrd=/boot/initrd-2.6.3-7mdk.img append="devfs=mount acpi=ht resume=/dev/hdg9 splash=silent" read-only image=/boot/vmlinuz label="failsafe" root=/dev/hdg1 initrd=/boot/initrd.img append="failsafe acpi=ht resume=/dev/hdg9 devfs=nomount" read-only other=/dev/hde1 label="windows" table=/dev/hde image=/boot/vmlinuz-2.6.3-13mdk label="263-13" root=/dev/hdg1 initrd=/boot/initrd-2.6.3-13mdk.img append="devfs=mount acpi=ht resume=/dev/hdg5 splash=silent" vga=788 read-only I also ran the out puts of fdisk -l for each drive First one hde (IDE1 Master) Disk /dev/hde: 80.0 GB, 80026361856 bytes 255 heads, 63 sectors/track, 9729 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/hde1 * 1 893 7172991 b W95 FAT32 /dev/hde2 894 9729 70975170 f W95 Ext'd (LBA) /dev/hde5 894 1914 8201151 b W95 FAT32 /dev/hde6 1915 3291 11060721 b W95 FAT32 /dev/hde7 3292 4668 11060721 b W95 FAT32 /dev/hde8 4669 6843 17470656 b W95 FAT32 /dev/hde 6976 8352 11060721 b W95 FAT32 /dev/hde10 8353 9729 11060721 b W95 FAT32 /dev/hde11 6844 6941 787153 82 Linux Swap Second one (IDE2 Master) Disk /dev/hde: 80.0 GB, 80026361856 bytes 255 heads, 63 sectors/track, 9729 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/hdg1 * 1 254 2040223+ 83 Linux /dev/hdg2 255 9602 75087810 5 Extended /dev/hdg5 255 266 96358+ 83 Linux /dev/hdg6 267 1641 11044656 83 Linux /dev/hdg7 1642 4336 21647556 83 Linux /dev/hdg8 4451 9602 41383408+ b W95 FAT32 /dev/hdg9 4337 4450 915673+ 83 Linux I am at a loss. Before I shut down I am going to try to figure out how to move 27GB of stuff off the Win32 partition on hdg. After that I'll reboot, with no idea what to expect. Any more ideas? I could use them and a stiff scotch. Thanks guys, DW

I'm hoping you mean the lilo prompt in the rescue console. That's the only way I've been able to get to it. I'll give it a try and report back. DW

After looking around it looks as though I need to figure out how to resize a partition in the rescue console too. After I get the thing running again I can search for the free space I "thought" I had provided for the swap. Am I right on this, or can I find free space from the console? DW

Well as far as I can tell, it looks as if I failed to create a new swap partition. Watching the console output during the mount process, all the partitions show up except the swap I thought I had created. I tried them all anyway and kept getting the same error (showing the new partiton assignment of course) Is it possible to create a new swap from the rescue console? The rest seems to be intact. chroot mnt found the /root partition during the mnt process and I'm currently in Windex, LILO booted to it as default. So either I create a new swap or I may be reloading MDK (that would suck bad). Any way, Steve, your advice was right on the money and I thank you for the help. I'm hoping you'll tell me there is a way to create the swap. In the meantime I'm going internet trolling, I can't imagine I'm the only idiot that ever deleted his swap partition. Thanks DW

I decided to move my swap partition to my other hard drive using MDK Control Center utility. All seemed to go well until I rebooted (isn't it always like this?) Anyway, MDK won't boot. First the pertinent specs: Abit KT7-A MoBo- RAID controller cannot be disabled, but I have it set as pass through (No RAID) 2x Western Digital 80G HDD hda - Master IDE1-boot sector (using LILO) Win XP setup multiple partitions. Moved swap to hda5 hdb - Master IDE2- MDK10 running /root, /boot, /home, (was) /swap, /tmp, /FAT32 Trying to reboot gives the following errors; resuming from /dev/hdb9 Resuming from device hdb9 Kernel panic: Resume: Unable to find suspended-data signature (-mispelled? Tried booting failsafe mode and all other options, all modes give the above errors. I doubt this is broken hardware. I haven't been having any problems until now so I'm guessing it's all due to the swap moving thing. I booted install disk to repair (recover?) and can't make allot out of the options offered. HEEEEELLLLLPPPPPP! I just got this thing all put together the way I liked it and now I do a bonehead and can't start it. Thanks for all the advice past and future DW

Found my solution! I finally got the installer to work. I installed it to one of my user account directories. While installing I got the following errors (I have a few guesses what they mean, but they are "rough" guesses so I won't post them) anyway here are the errors firefox installer: line 56: 4426 segmentation fault ./${BINNAME}-b $@ At this point I deleted all the related directories and started over again. The next attempt ran without giving up the error. During the second try I got the following errors (I noted and ignored them because the installation continued to run): **(firefox-installer-bin:4488):WARNING**:Invalid UTF8 string passed to pango_layout_set_text() **(firefox-installer-bin:4488):WARNING**:Invalid UTF8 string passed to pango_layout_set_text() **(firefox-installer-bin:4488):WARNING**:Invalid UTF8 string passed to pango_layout_set_text() -----the above repeated for about 12 lines------------- ***eeeeee =[Exception..."Component returned failure code: "0x80520006(NS_ERROR_FILE_TARGET_DOES_NOT_EXIST) [nsIFile.remove]" nsresult: "0x80520006(NS_ERROR_FILE_TARGET_DOES_NOT_EXIST)" location: "JS frame::file:///home(my user filepath)/firefox-installer/components/nsExtendionManager.js::ensureExtensionFiles:: line 245" data:no] ***loading the extensions data source ***loading the extensions data source In the end I wound up with a running version of FF 0.9. I followed the mozilla.org instructions to the letter. All I can say at this point is that it launches fine from my user account. I haven't tried to do any configuration to it yet. But I'll close this post since I got the install problem fixed. Nothing against the Moz forums, but I couldn't find any of this info in the forums. I will post it there next. Thanks for reading and advising DW

I have tried for two days to install FF 0.9. I have used both versions (installer and non). All I usually get is a non executable mess. I have followed all the info I can find on this. But most of it is not tailored to MDK. I was just looking to compare notes with anyone else who may have had install problems. I may try again later and will post specifics if I can record any of them. I'm quite sure it's all FFs problem because the rest of my system continues to run without problems. Thanks DW P.S. I forgot to ask my first question. The install suggests running the browser for the first time as root. My problem is that the browser launches automatically as part of the install routine. If I install as root, no matter how I change the permissions it will never launch from one of the user accounts.

Install went well. Just for the record I kept LILO, it worked fine. Then updated kernel, LILO added the new kernel so I have the safe option of booting the old kernel had things not gone well. This is the second kernel update that I have done and LILO did it's job flawlessly both times. Thanks for all the advice DW

Weird cure, as usual I rebooted today and the problem had cured itself I checked and I was running the -13 kernel source Everything seems to be working fine. On of my next guesses would have been the fact that I installed with the NVidia installer without removing the RPM first, which was probably a bad idea. Oh well. Having problems getting arts to run when logout/login but I think that's for another posting. Thanks for the advice, DW

The .run file came from NVIDIA Linux I think it's their installer. I might not have mentioned it but I did the original install the way you recommended. How ever, the NV driver complained about wrong kernel source and downloaded it's own. I'm going to backup and remove all the stuff and start again. I know how to remove the rpm pkg, any idea how to remove the .run pkg and the NV kernel source? I just want to get it back to the way it was just after the -13 kernel upgrade, then I can start again. Hopefully X isn't corrupted. I'll post back with the solution when I find it. Thanks, Dan

Not sure where to post this but what the hell. I loaded the NVIDIA drivers (5336) on a fresh install of MDK 10. As would be expected it was a bumpy ride but I finally got a working display. But now it seems that along the way the KDE log in/out interface are having a battle between them. When I start/restart it automatically starts in command mode, even though I've already set the option to boot X at start. I have to do startx at the cmd line to get X started. But after this X loads and works fine. When I log out X either hangs or crashes to command line and gives the following errors. Warning: DCOP Communication problem ICE default IO error handler doing an exit () pid=6571 errno=2 ICE default IO error handler doing an exit () pid=6564 errno=0 ICE default IO error handler doing an exit () pid=6551 errno=0 kde init: Communication error with launcher, Exiting start kde: Running shutdown scripts.... Application 'mdkapplet' lost it's connection to the display :0.0; most likely the X server was shutdown or you killed/destroyed the application waiting for the X server to shut down There are alot more errors regarding sound etc. but the ones above seem to be the critical ones. I'm tempted to just stick the disks in and reload the whole thing, but that is too much like the M$ way. Anyway, if some of you gurus can make heads or tails of this I'd appreciate it I can't start looking for a fix until I get a better idea of what's going on here. I might mention that I was running MDK 10 with the same drivers last week, just did the reinstall due to a new hard drive. I was running kernel 2.6-9, this setup I loaded the -13 version right after install. Another problem I may have created is that I loaded the new NV drivers by RPM first, X wouldn't work so I loaded the NV .run package and that's how I got the display I currently have woking. Thanks to anyone who might advise me on this. DW moved from installing mandrake to hardware - tyme

How can I tell which NVidia drivers I am currently running? I went through a mess getting a set to work on my MDK 10 install. Now I can't remember which ones I finally settled on. Thanks in advance. DW