Hi, My last post didn't get any replies so I am posting again hoping to attract some attention. The problem isn't too complicated so hopefully someone can help!

Thanks!

------------


Hi,

Normally when I boot into Red Hat 9, after it loads all the services it displays a text login prompt for two seconds before it loads the Red Hat graphical login screen.

Today when I booted my computer, all the services loaded as usual, it displayed the text login prompt for two seconds as usual, but then instead of displaying the graphical Red Hat login screen the screen went blank for about 5 or 6 seconds, then it went back to the text login screen for 2 seconds, then the blank screen, then it repeated this loop until I turned off my computer...

My question is how do I recover from this besides reloading RedHat? How do I get a command prompt and once I do what should I be looking for to help fix this?

Is there a log file stored with information on why the graphical screen is failing? (var/log/messages?)

I found out I can edit my inittab and change the level to 3 and then do tellinit -k, but I don't know how to edit my inittab?

Thanks.

FYI, feel free to bump your post after 24 hours as stated in the rules. On to your questions...

Did you make any changes to your XF86Config file? that holds all the configurations for the GUI, so you might want to check /var/log/XFree86.0.log (or similar).

You could try to get to a command prompt by pressing ctrl+alt+F[2-6]

once you get to a command line, you can edit your inittab by using any one of many text editors. (I typically use vim)

Once you are able to view the log, if you can't decipher the problem, just post the part in question...

Good luck.

Something must have changed - the XFree86 configuration, your kernel and/or modules configuration, or something to that effect.

Normally when the graphical login manager (GDM, XDM, etc) fails to start it will try three or four times and then display a text message saying it can't start XFree86 and then it will drop you to a console (text) login. Try pressin Ctrl+Alt+F2 to get a console login. If you do, log in as root and launch setup. You may have to specify the full path, which I don't recall right now - I think it is /usr/bin/setup but it could also be /sbin/setup. There you can choose X configuration and a host of other things. Enter the X configuration and see if you can make it work again.

The latest XFree86 log should be in /var/log/XFree86.0.log it might prove to contain useful information.

inittab is a file located in /etc. If you get a console login you can edit it using any of the text mode editors like nano. I don't really endorse the use of vim or emacs for newbies in emergencies as they can be very frustrating if you're new to them. nano is very straightforward, and you should launch it with nano -w /etc/inittab. The -w switch makes sure lines that are too long to fit onto the screen don't get broken when you save the file.


Håkan

To boot to the command shell, you can do some things at the grub menu.
At the grub menu highlight the Linux item
Type: e press enter
Highlight the line with kernel in it.
Type:e press enter
go to the end of that line and space over one, then type:
init 3 press enter
Now while highlighted on that kernel line, type b and enter
That should boot you up to the command line.

Once you get there and login as root, you can type: demesg to see if there are any errors.

Thanks superbondbond, hw-tph, and homey for the help so far!

So I was able to modify my kernal arguments to boot with init 3, and get into a command prompt.

My /var/log/messages file had numerous messages that said:

Feb 11 HH:MM:SS Computer gdm[1411]: gdm_slave_xioerror_handler:
Fatal X error - Restarting :0

I tried start x, which failed with the message:
Could not init font path element unix /:7100 removing from list!
Fatal server error:
could not open default font 'fixed'.

I tried running redhat-config-xfree86 but that failed with the following message:
xconf.py Fatal IO error 2 (connection reset by peer) on x server :17.0.

So now I need to know what those error messages mean??

Thanks!

I know its annoying when new users ask dumb questions and don't help themselves - at least I am reading the manual and making progress! :-)

So after reading the error messages and seeing a program with fonts I found the config file (/usr/X11R6/lib/X11/XF86Config) and found this:

# Multiple FontPath entries are allowed (they are concatenated together)
# By default, Red Hat 6.0 and later now use a font server independent of
# the X server to render fonts.
FontPath "unix/:7100"

I added the following lines:
FontPath "/usr/X11R6/lib/X11/fonts/"
FontPath "/usr/X11R6/lib/X11/fonts/local/"
FontPath "/usr/X11R6/lib/X11/fonts/misc/"

I then saved the file and entered startx and viola, it worked!

So my question is what is unix/:7100 and why was/is it failing to work as it had been??

The FontPath unix/:7100 is pointing to a local UNIX socket that is supplied by a font server, usually xfs or xfs-xtt. A font server does what it sounds like it does - it provides fonts.

You could try restarting xfs (often as simple as typing /etc/init.d/xfs restart) and see if you can catch any error messages that xfs spits out.


Håkan

After I had experienced this this weird font/xfree86 issue, a vulnerability was discovered in Xfree86!

This look related....

XFree86 Additional Font Information File Buffer Overflows Let Local Users Gain Root Privileges
http://www.snpx.com/cgi-bin/news5.cg...50533772?-2622