blinking nvidia logo...

[mit]

i have got troubles

i am very new to linux, so bear with me

i have followed the advice listed here thus far and i have learned a lot about linux

i am using a geforce 4 ti 4200 on an athlon 1600+ running redhat 7.1

as far as i can tell, i have installed all the necessary files correctly
and my x86config is in the right place

when i boot up into linux the Xserver gives me the nvidia logo, but i stays up for three seconds or so and then goes to black, only to give me the nvidia logo again

i am only able to use (ctrl - alt -del) at this point
when i boot up i cannot reach the prompt before i get the nvidia logo

like i said, i'm new to this, but as far as i can tell the problem is with my x86config

-mit

[MacSlow]

Hm... can you paste the contents of the file /var/log/XFree86.0.log here?

When booting into Linux you should be able to issue some boot/kernel parameters before the actual boot-process is initiated. Every Linux-distribution offers such functionality. Take a look at your distributions manuals on how to do that and either make sure you boot up in "single-user" mode or with init-level 3 (this would be better/more convienent than "single-user").

Once you got this done you should easily be able to "cat /var/log/XFree86.0.log")

Best regards...

MacSlow

[mit]

good news
i went through the red hat cd and did the whole rescue mode thing

and i set it up to boot into init mode 3

and i managed to retrive my XF86Config

now i am able to boot back to the text login prompt

so, i guess we have to find where i went wrong on my XF86Config

-mit

ps. by the way, thanks to all you guys who post here and help us newbies out, and thanks in advance to you guys who help me

[MacSlow]

Hm... I took a look at your xf86config file. From the first look of it, it appears to be ok. But a couple of questions arise.

1.) Are you sure that this file is the one used by XFree86 when it starts? You can verify this by taking a look at /var/log/XFfree86.0.log. grep through it with this command:

grep "Using config file" /var/log/XFree86.0.log

This will present you the full path of the config-file XFree86 uses.

2.) Do you explicitly select any of the multiple server-layouts defined? Or is it just started as is?
If that config file, you posted, is actually used, you could try any of these commands to select a particular layout (from the command-line):

X -layout AGP
X -layout PCI
X -layout Both
X -layout AGPTwinView
X -layout BothTwinView

See if any of those work. If you don't use the -layout option server-layout "AGP" will become the default one and get used.

3.) Did you write that xf86-config file yourself or was a special tool used?

4.) Did you check /var/log/XFree86.0.log for any error messages or strange entries?

Best regards...

MacSlow

[mit]

XFree86 Version 4.2.0 / X Window System

(protocol Version 11, revision 0, vendor release 6600)
Release Date: 18 January 2002
If the server is older than 6-12 months, or if your card is
newer than the above date, look for a newer version before
reporting problems. (See http://www.XFree86.Org/)
Build Operating System: Linux 2.4.9-13smp i686 [ELF]
Module Loader present
Markers: (--) probed,
(**) from config file,
(==) default setting,
(++) from command line,
(!!) notice,
(II) informational,
(WW) warning,
(EE) error,
(NI) not implemented,
(??) unknown.
(==) Log file: "/var/log/XFree86.0.log",
Time: Mon Sep 30 11:13:25 2002
(==) Using config file: "/etc/X11/XF86Config-4"
Parse error on line 193 of section ServerLayout in file /etc/X11/XF86Config-4
"Identifier" is not a valid keyword in this section.
(EE) Problem parsing the config file
(EE) Error from xf86HandleConfigFile()
Fatal server error:

no screens found
When reporting a problem related to a server crash, please send
the full server output, not just the last messages.
This can be found in the log file "/var/log/XFree86.0.log".
Please report problems to xfree86@xfree86.org.


here is the contents of my XFree 86 log

what does it mean "no screens found"?

[bwkaz]

Here's the area of the file that's giving the problem (I think; it's hard to read between all the ^M characters and the lack of indentation -- anyway,...):

#2 cards, one agp twinview and one pci (3 monitors total)
#Section "ServerLayout"
Identifier "BothTwinView"
Screen "Screen AGP TwinView"
Screen "Screen PCI" LeftOf "Screen AGP TwinView"
InputDevice "Mouse1" "CorePointer"
InputDevice "Keyboard1" "CoreKeyboard"
EndSection

I'm pretty sure that the Section line should not be commented. Either uncomment that line, or comment everything until EndSection.

[mit]

thanks for the reply, bwkaz, but it didn't work

i downloaded the nv_check.sh script and ran it

it says that the NVdriver kernel module is not loaded. It says i have to compile it for my kernel, set it up to insert the module on boot, then run the script again to recheck

then it says if i have trouble compiling the module, i have to add the argument -D_LOOSE_KERNEL_NAMES to the makefile

i sort of understand what is going on, but i really don't know the procedure

-mit

[bwkaz]

I think just doing a modprobe NVdriver as root, then running the nv_check.sh script, will fool it into thinking that the right things are running. Though I'm not sure.

To set it up to automatically load (you won't have to do this now, but you will eventually), it depends on your distribution. RedHat uses the /etc/modules.conf file. You have to add a line there that says either alias /dev/nvidia* NVdriver or alias char-major-195 NVdriver, and the line you add depends on whether or not RedHat have finally enabled devfs on their systems. If [ -c /dev/.devfsd ] ; echo $? prints a 0, then you're using devfs. If it prints a 1, then you're not using it. If you're using devfs, then use the /dev/nvidia* alias, and if not, use the char-major-195 alias. Just for future reference...