FTP error on NVIDIA driver install...
 

Okay, I've read all the posts, searched here and at nvnews forums for the answer but cannot find it. I am following the instructions in the FAQ in this forum (Newbie forum) about installing the NVIDIA drivers. I get to the part where the nvidia installer runs and says something like it couldn't find the correct compiled headers and it would try to download them from ftp.nvidia.com. Then another error message comes up saying it couldn't connect to ftp.nvidia.com because of DNS errors and to try again later. This has happened all throughout today. After this it kicks me back out to text-mode.

Is there a step I'm missing?? I printed the instructions from the FAQ and all seems to go fine until this happens.

Any help is appreciated, or if you find where this has been answered elsewhere just point me in the right direction.

rberry88

Might just very well be a problem with their service today, you should email them to see if it is indeed a problem on their end.

Cool

When it asks you if you want to download a module from FTP, just choose no, and it will build it on your machine.

If that doesnt work, you need these installed:

GCC
kernel source
XFree86-devel
XFree86-headers

Update:

Okay it seems that if I follow what the FAQ says in runlevel 3 I no longer get the ftp error, however, whether I choose yes or no to dowload from ftp//download.nvidia.com it still gives me an error that says it cannot compile a new header for the kernal that is currently running. I've gone into 'install new programs' and made sure I had

GCC
kernal source
Xfree86-devel
Xfree86-headers

all are loaded (check marks in the boxes).

Do I maybe need to make sure that the kernal that is listed in YAST2 for my graphic card is a certain one before I do this. And if so which one would I choose for a Nvidia Geforce FX 5600?? I see a selection for 5800 and possibly a generic for DDR and 256. The one that currently shows up is 0x312.

rberry88
:newbie:

You know you're a newbie when you add to your 'call for help' messages before you get a reply....

The exact error, if it helps, that I get from the nvidia installer is as follows:

--
Error: Unable to find the kernel header files for the currently running kernel. Please make sure you have installed the kerner header files for your kernel; on Red Hat Linux Systems, for example, be sure you have the 'kernal-source' rpm installed. If you know the correct kernel header files are installed, you may specify the kernel include path with '--kernel-include-path' command line option.

--

I have everything in the 'install software' section installed except for two items that say they are more multi-processor machines. Is there a way to find out if the correct kernel is in my system??

rberry88 :Pengy:

I took a different approach and attempted to install the driver using the run command instead of the nvidia installer (posted in the README of the driver) and this is the output:

<i>-----

You Appear to be compiling the Nvidia kernel module with a compiler different from the one that was used to compile the running kernel. This may be perfectly fine, but there are cases where this can lead to unexpected behaviour & system crashes.

If you know what you are doing and want to override this check, you can do so by setting IGNORE__CC__MISMATCH.

In any other case set the CC environment variable to the name of the compiler that was used to compile the kernel.

-----</i>

How would I go about doing the last item mentioned, setting the CC environment variable??

Should I give up or anyone have any ideas??

rberry88
:scratch: :study: :Pengy: