danleff

Don't know. You need to give more information. Was it working in RedHat before? If so, have you changed anything, say installing new packages, upgrade the kernel....? Is this a broadband connection to your ISP through a NIC (ethernet) card?

You are correct. You apparently got the following; Linux:~ # ndiswrapper -l Installed ndis drivers: airplus driver present You should have gotten; Linux:~ # ndiswrapper -l Installed ndis drivers: airplus driver present, hardware present You also got; Linux:~ # dmesg | grep ndiswrapper ndiswrapper: module not supported by Novell, setting U taint flag. ndiswrapper version 1.2 loaded (preempt=no,smp=no) There is no use going on, if the ndiswrapper -l command does not show that the hardware is detected. From the directions; Quote: Where 'present' means that you have a card that can be used with the driver installed. In this case, marvell driver mrv8335 is used. If you see 'cannot locate lspci. Unable to see if hardware is present', you need to install the pciutils package. Did you install, or have installed, the pciutils package? If yes, or you now installed the pciutils package, try unloading the driver with; ndsiwrapper -e AIRPLUS Then install the driver again with; ndiswrapper -i AIRPLUS.INF Do ndiswrapper -l and see if the hardware is present now.

OK, what is the output when you type ndiswrapper -l? You should get a message like; Installed ndis drivers: xxxxxx driver present, hardware present Where xxxxxx is the name of the driver. Consult the SuSE 10 specific ndiswrapper instructions, located here. If the message of ndiswrapper -l is OK, then follow the rest of the directions. If not, post what you found. Finally, do you have the wireless tools package installed in your SuSE installation? Consult the begining of the instructions from the link above and let us know what point you are at.

Ooops! I took your beginning line on your last post as it was working. Let's try an experiment from my experience on another distro. Try running as root user; cardctl info. See if it produces any output at all. In my Thinkpad, it showed; MANFID = 0271,0012 Then make ndiswrapper use the pcmcia device with the -d command manufacturer's id; ndiswrapper -d 0271:0012 net5211 Where 0271:0012 is the code that was returend in your cardctl command, and net5211 is the name of your ndiswrapper module that you get when doing ndiswrapper -l (the list command). Of course, what quetion I did not ask, is what exact Windows driver did you use to install with ndiswrapper -i? Was it the Windows XP driver from your driver cd, or something else? I also assume you did modprobe ndiswrapper after installing the drivers, to load the module? Did you follow the generic ndiswrapper instructions, or the SuSE specific ones? Were the .inf and .sys files both in the source directory when you ran the install command for ndsiwrapper?

Take a look at this FC 5 installation guide. It should answer any questions thast you have.

raymondobasildon, nice reply which I hope helps someone. Just a note, look at the date of the last post from r444, which was over two years ago. Looks like he was onthe right track though.

Glad you got it working. It looks like the card bus services were not loading or installed on your first installation. Nice work!

Let's see if this will be simple. It's at these times that I wish I had one of these modems to test. Since just loading the module seems to work, try the following; Go to Yast-->System-->Kernel-->MODULES_LOADED_ON_BOOT and add "usbnet" (without the quotations). Bear with me. I am on a new system and don't have my SuSE installation available to me yet. I assume this connection is broadband and not DSL?

Once you modprobe the usbnet module, go to Yast and set up the connection. Choose DHCP and to enable the connection on boot. Follow though with the menu windows until the process is finished. Your new connection should be up after the next boot.

...and this is where folks get into trouble. If you add an extended boot partition with Bootit, then you changed the logical order of your partitions. Take a look at the Bootit/Linux readme on the subject, located here. Quote: BootIt NG can not boot Linux by itself. It requires the help of Lilo or Grub, because they are needed to actually boot the Linux kernel. When you boot Linux from the BootIt NG menu, BootIt NG transfers control over to Lilo or Grub, and they take it from there. * BootIt NG occupies the MBR of the first hard drive. For that reason, Lilo and Grub can not be installed in the MBR or they will overwrite BootIt NG. Instead they need to be installed in the boot sector (first sector) of a Linux partition. The Linux partition to install Lilo or Grub to is typically the root partition (/), but can also be another Linux partition such as the the /boot partition if you have one. The important thing is that whichever partition you install the boot loader to, that same partition should be designated as the partition to boot from when you set up Linux in BootIt NG. That's how BootIt NG finds Lilo or Grub so that it can pass control of the boot process over to them. So, you can't tell linux to use the MBR, but you must have a distinct /boot partition, or have grub installed on the root partition. Remember, if you made any changes to the partition table after installing any flavor of Linux, then the logical order has changed and grub is looking in the wrong place.

This is why I generally ask what make and model system one has. The hidden restore partititon in many systems cause this exact issue. Nice work!

Nice to see that you like Mandrake and would like to switch over. Let's consider some things first. What version of Windows are you running, XP? If so, are you running windows XP on the NTFS filesystem? The current linux partititoning tools don't always handle NTFS filesystems correctly. Are you sure that the issue is a virus and not a problem with the hard drive itself? What leads you to believe that there is a virus infecting Windows? Realize that grub is now set up to boot in a dual boot environment. What this means, is if you delete the Windows partition and take the whole disk for Mandrake, then grub's entry to boot Mandrake from it's correct partition is not correct and needs to be altered. if you really have a virus on the Windows partition, then you want to reformat that partition. One solution, would be to install Mandrake, or better yet, Mandriva 2006 to format and take over the NTFS partition. This way, you still keep you data on the current Linux partition. Or, it is possible to resize the Windows partition to recover some space, keep the current grub mbr inact and leave a small amount of space on the Windows partition, so that grub does not get confused. See this article and pay attention to the section about resizing the partition with Knoppix. Just to make sure that you don't have any other potential issues, Please note what make and model system that you have. Or if you built your system yourself. Why, do I ask? If you have a Compaq or HP system with Windows preinstalled, there are sometimes issues with recovery partitions on these systems that may be resident, or that may be present as part of the bios information.

Yes, when you re-install Windows, the MBR is written over. boot disc 1 of fedora and type at the menu screen, linux rescue When you get a command prompt, type, chroot /mnt/sysimage This should mount the root filesystem (as long as you don't get any error returned) then type; grub-install /dev/hda A good explanation if the method can be found here.

Can you post the exact make and model of your laptop? What led you to use the pci=routeirq settings on boot? What happens if you ammend the boot kernel command from pci=routeirq to acpi=off? What version of ndiswrapper are you using, the one that came with SuSE, or via an updated source package? Is this SuSE 10 final release, or a beta version that you are using?

The easy explanation? The LKA is a true hardware modem and the other (LKU) has a different chipset, or is not a true hardware (onboard chipset driver) modem and/or is not supported by Linux. Any idea what chipsets are in each modem? Quote: The Linux is Fedora Core 2 and 3 So, which is it? Or, do you have two different installations of Fedora? Realize that some of these modems are designed for Windows, in that they are soft modems designed for Windows. Support improves with each release of Fedora, so it may be that native support for the non-functional modem is in Fedora Core 5.

I don't use ATI cards (well I have an old one in one of my systems, but that does not apply to this issue). Let's look at what you have and the potential issues. I assume the following. 1. You elected to do a text installation, as the graphical installation did not work. 2. You can now boot into FC5, but now have only a command line interface, as the graphical interface is not configured. 3. You know how to read and edit, if needed, the xorg file. What entry is in the xorg file for the card, ati, or radeon? You can probably get the gui up by changing the section "Device...driver" to "vesa", but this only may give you the gui and not use all the features of your card. Make sure that you monitor is not at issue, as well. Look at the specs. for the monitor and make aure that the horizontal and vertical modes match, or come close to your monitors capabilities. Or, you can do, as root user, system-config-display Look under the hardware tab. Make sure that the monitor type says "LCD Panel" and matches the resolution capability of your monitor. Consult the users manual, if needed. Make sure that the video card section also matches the card in your system. The other issue, which has come up is that this is a pci card. Don't have one of these in my systems, so I'm sure there are possible specific issues and solutions for this setup.

Tru installing the alsamixergui package from the software manager (Applications --> add/remove software}. That should make things easier. jk12, thanks for starting a new thread. Refer to that thread for answers, which should make following your issue easier.

Yep, that's the ticket! Good job!!!! The problem also is that new posters are reading this thread and trying too many individual solutions suggested here at will. Let's close this out and ask users with new problems to start a new post with there probles.

Quote: The problem with the soundblaster is equally aggrivating, and I can't seem to get any help or thoughts on that problem over at fedora forums. Justbill, where did you get the soundblater card? There are dozens of variants of this card. A fair number of them are legacy Dell or Compaq models, which do not behave like true Sondblaster cards. I bought one of these "cheap" of a surplus site a while back and found it to be one of these Dell legacy cards. If you think it might be one of these cards (an OEM plain packaged card, rather than a retail version) then this may be the issue. What is the model number on the card itself. I can't remember the series, but does it start with SB, CT or something else? If you are daring, the new kernel is located here. This is what the article refers to, where you don't need the patches to compile the NVIDIA drivers right off.

Quote: Not on FC5, though. It has been a fiasco, put the card in the pci slot, and turn off the onboard graphics, and FC5 wont boot. So that eliminates running the 8178 installer from nvidia, and it looks as though the current kernel may not support nvidia (however I may have misunderstood what I read). I have not done a pci video card in a while, but once the system is configured for agp or another card, I bet you have to run the hardware config. utility via the console. As root user, try the command system-config-display You are correct. There is a bug, I read, in the base kernel that does not allow installation of the NVIDIA drivers without a patch. A new kernel will be added to the Fedora repository soon to fix this issue. See this quide. Looks like ATI support is lacking also. Too bad. Confused2, try this at the initial boot welcome screen, as added text commands; linux resolution=1024x768 nofb See this post. [Edited by danleff on 2006-03-28 12:58:11]

Don't get grub's values mixed up with partition designations. Here is the output of my fdisk; Device Boot Start End Blocks Id System /dev/sda1 * 1 6202 49817533+ 7 HPFS/NTFS /dev/sda2 6203 8752 20482875 83 Linux /dev/sda3 8753 8879 1020127+ 82 Linux swap / Solaris (note that I have a sata drive, so this is why the sda, rather than hda, as in an ide drive) You can quicly see what the root partition is with the command df, which should output the root drive mounted, say, Filesystem 1K-blocks Used Available Use% Mounted on /dev/sda2 19840924 3222140 15594644 18% / In my case the / is mounted on sda2. Also, note that the * denotes the boot partition in fdisk, in my case, sda1.

What is important to know, is exactly where is grub? Since I always place everything on one partition, so that grub is on the Linux root partition, I am going by what seems correct for you. O'Reilly's article assumes the same. You noted earlier that your original boot.lnx file was 1 kb. If your new grub.bin file is the same size, and by the last post that you made, grub is on hdbv2, not hdb1. So, if the above are true, try the dd command to copy the first 512 bytes to a grb.bin file and place this in the windows root directory, or c:. Do you have 512 bytes now?

Mathew's post is over a year old, so I doubt that he is still around. However, nice link!

Sometimes you get the Verifying DMI Pool Data.... message on a soft reboot after making any changes like this to the bios. What happens when you do a hard reboot, in other words, allow the system to completely shut down, wait 3 minutes, then try the boot. This allows the system to completely reset. However, I suspect that there are other issues, if this does not work. I have not asked yet, but what make and model system or motherboard is this?

zenarcher is absolutely correct. Ndiswrapper is used as a module, so that if you update the kernel, it needs to be installed/reconfigured for the new kernel.