On Thursday 01 July 2004 14:42, Patrick Shanahan wrote:

SuSE 9.0 Pro - stock, utd Intel P3-733 NVidia TNT-2(32)

Yesterday, while trying to load a broken jpg into gimp2 and 30 minutes of 100 per cent system usage with no keyboard response, I felt it necessary to take charge with the kill switch. I was unable to kill the offending app or change to a text console.

Rebooting I arrived at the NVidia splash screen, then the gray screen with the 'X' cursor and ....... no further action.

I checked that XF86Config was not changed/damaged, did 'init 3' and reinstalled the nvidia driver (5336). No faults.

Is this the latest driver?

'init 5' and same hang at gray screen.

'init 3' and startx gives me kde and a working system, but....

tail /var/log/XFree86.1.log: (II) NVIDIA(0): AGP 2X successfully initialized (EE) NVIDIA(0): Failed to allocate external video decoder object (EE) NVIDIA(0): *** Aborting ***

Fatal server error: AddScreen/ScreenInit failed for driver 0

What have I done, as I have not knowingly made any recent system changes, and what do I do to recover.

Not a solution, but: Perhaps using the X built-in NVidia driver (nv) wil get you a working system again. Just change 'Driver "NVidia"' to 'Driver "nv"' (maybe backup first). And to see what packages you've installed lately with YaST, look at /var/log/YaST2/y2logRPM. No log entry if you installed something without YaST. Cheers, Leen

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Patrick Shanahan wrote: | * Leendert Meyer [07-01-04 09:46]: | |>Is this the latest driver? | | | yes | No: http://www.nvidia.com/object/linux_display_ia32_1.0-6106.html - -- Blue Moose IT Support www.bluemooseit.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFA5UOGh6qZaDhhuVMRAqf8AJ4i6P7elyhiqG0NuRmsphfEecJNEgCeKVsO c3tkk7coFxhbXYD4seTMwbY= =Fl/z -----END PGP SIGNATURE-----

* Patrick Shanahan [07-02-04 08:00]:

Thankyou, I have installed the 6106 update (via YOU) and ?normal? function has returned. I do *not* understand what changed, thou. I had used the 5336 set since January w/o a problem ???

Downside, 9.0 has performed so well and the derogatory comments flying wrt 9.1, I will probably postpone even further the 9.1 install >:^(

Update: forced to reboot to regain hotplug after better-half pulled camera memory chip from reader. Situation is the same as before. Boot reaches level 5 and NVidia splash shows, turns to gray screen with 'X' cursor which is movable, but does not start kde or kde does not visualize on the screen. /var/log/boot.msg: (note: one real long line included) Starting CRON daemondone <notice>exit status of (nscd hwscan cron) is (0 0 0) <notice>start services (xdm splash_late) Starting service kdm<notice>startproc: execve (/opt/kde3/bin/kdm) [/opt/kde3/bin/kdm ], [ LC_MONETARY= CONSOLE=/dev/console TERM=linux SHELL=/bin/sh LC_NUMERIC= QTDIR=/usr/lib/qt3 LC_ALL=http_proxy=http://127.0.0.1:8118/ progress=50 INIT_VERSION=sysvinit-2.82 KDEROOTHOME=/root/.kdm REDIRECT=/dev/tty1 COLUMNS=148 PATH=/usr/local/sbin:/sbin:/bin:/usr/sbin:/usr/bin LC_MESSAGES=vga=0x31A RUNLEVEL=5 LC_COLLATE= PWD=/ LANG=en_US PREVLEVEL=N LINES=51 HOME=/ SHLVL=2 XCURSOR_THEME=blueprint-cursor-theme no_proxy=localhost WINDOWMANAGER=/usr/X11R6/bin/kde LC_CTYPE=en_US splash=silent sscripts=51 LC_TIME= _=/sbin/startproc DAEMON=/opt/kde3/bin/kdm ] done <notice>exit status of (xdm splash_late) is (0 0) Master Resource Control: runlevel 5 has been reached Skipped services in runlevel 5: smbfs acpid <notice>killproc: kill(688,3) after issuing init 3, startx will start kde, but init 5 returns the same result. SuSE 9.0 Pro up2date Linux wahoo 2.4.21-231-default #1 Mon Jun 28 15:39:34 UTC 2004 i686 i686 i386 GNU/Linux nvidia driver is 6106 -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Saturday 03 July 2004 17.12, Patrick Shanahan wrote:

/var/log/boot.msg: (note: one real long line included)

How about /var/log/kdm.log? Does that have anything interesting?

On Saturday 03 July 2004 17:12, Patrick Shanahan wrote:

<notice>killproc: kill(688,3)

Tried it with RUN_PARALLEL="no" in /etc/sysconfig/boot? (I /hate/ it when process A is started, followed by process B, followed by an error message that could be from process A, B, or perhaps even C.) Cheers, Leen

* Leendert Meyer [07-03-04 10:32]:

On Saturday 03 July 2004 17:12, Patrick Shanahan wrote:

...

after issuing init 3, startx will start kde, but init 5 returns the same result.

Perhaps it is a display manager issue. [gkx]dm does not run in runlevel 3, but does in 5.

Should I change the display manager from kdm to xdm or ?? and try 'init 3', 'init 5' again? Or should I reinstall kdm, can I? I am grasping at straws here <grin>. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

* Anders Johansson [07-03-04 10:49]:

On Saturday 03 July 2004 17.38, Patrick Shanahan wrote:

...

* Anders Johansson [07-03-04 10:27]:

...

How about /var/log/kdm.log? Does that have anything interesting?

Only two missing font path elements that have appeared since I installed 9.0 and successfully used previous nvidia ver's. ??

Can you tell from "ps aux" or "top" if anything is actually doing anything (I'm thinking kdm related here)?

No instance of kdm. About 20 instances of kdeinit, mostly related to normal processes I run, kicker, kdesktop, ... top shows X running, but nothing that ..looks.. odd.

...

One other thing that I have noticed that ?might? be related. I am unable to start a second session (:1), even with 'xhost + local:'.

xhost only controls one X server. If you start a new session you are running a completely new instance of X that is completely unaffected by your previous xhost settings

How do you start the new session, and what is the error message?

pat@wahoo:~> startx -- :1 This is a pre-release version of XFree86, and is not supported in any way. Bugs may be reported to XFree86@XFree86.Org and patches submitted to fixes@XFree86.Org. Before reporting bugs in pre-release versions, please check the latest version in the XFree86 CVS repository (http://www.XFree86.Org/cvs). XFree86 Version 4.3.0.1 Release Date: 9 May 2003 X Protocol Version 11, Revision 0, Release 6.6 Build Operating System: SuSE Linux [ELF] SuSE Build Date: 12 February 2004 Before reporting problems, check http://www.XFree86.Org/ to make sure that you have the latest version. 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/XFree86.1.log", Time: Sat Jul 3 10:58:01 2004 (==) Using config file: "/etc/X11/XF86Config" (EE) NVIDIA(0): Failed to allocate external video decoder object (EE) NVIDIA(0): *** Aborting *** Fatal server error: AddScreen/ScreenInit failed for driver 0 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/XFree86.1.log". Please report problems to http://www.suse.de/feedback. XIO: fatal IO error 104 (Connection reset by peer) on X server ":1.0" after 0 requests (0 known processed) with 0 events remaining. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Saturday 03 July 2004 17:59, Patrick Shanahan wrote:

(EE) NVIDIA(0): Failed to allocate external video decoder object (EE) NVIDIA(0): *** Aborting ***

Yeah, got that one too when I was running 'yast x11', after I had installed the previous NVidia drivers today. I dropped this driver though, running XFree86's nv driver again (1920x1440 ;P, NVidia only allows up to 1600x1200). Cheers, Leen

* Leendert Meyer [07-03-04 11:15]:

On Saturday 03 July 2004 17:59, Patrick Shanahan wrote:

...

(EE) NVIDIA(0): Failed to allocate external video decoder object (EE) NVIDIA(0): *** Aborting ***

Yeah, got that one too when I was running 'yast x11', after I had installed the previous NVidia drivers today. I dropped this driver though, running XFree86's nv driver again (1920x1440 ;P, NVidia only allows up to 1600x1200).

My NEC 19inch lcd only goes to 1280x1024 8-( update: changed boot/parallel to 'no', but have not rebooted yet. Did 'init 5' to get errors from ~/.xsession-errors and kde started correctly. Going to reboot to see if it will work again. Surely the change from parallel yes to no did not affect going from runlevel 3 to 5. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

* Leendert Meyer [07-03-04 11:22]:

On Saturday 03 July 2004 17:42, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 10:32]:

...

On Saturday 03 July 2004 17:12, Patrick Shanahan wrote:

...

after issuing init 3, startx will start kde, but init 5 returns the same result.

Perhaps it is a display manager issue. [gkx]dm does not run in runlevel 3, but does in 5.

Should I change the display manager from kdm to xdm or ?? and try 'init 3', 'init 5' again? Or should I reinstall kdm, can I? I am grasping at straws here <grin>.

If you do that, and the problem does not go away, than it is at least not specific to 1 display manager.

BTW, try 'startx', followed by 'startx -- :1' from another console. I like to know if you can run two instances at the same time. My guess is 'no'.

You are correct. See other post in this thread (answering Anders). I just rebooted after changing parallel yes to no and changing from auto-logon to not. The 'welcome' screen shows and X hangs right there. I suppose it is doing the same thing that it did with the auto-logon but doesn't display the 'welcome' screen. Tried again 'startx -- :1', with same errors reported previous. Something about failed to allocate decoder object ... -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Saturday 03 July 2004 18:39, Patrick Shanahan wrote:

* Leendert Meyer [07-03-04 11:22]:

...

On Saturday 03 July 2004 17:42, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 10:32]:

...

On Saturday 03 July 2004 17:12, Patrick Shanahan wrote:

...

after issuing init 3, startx will start kde, but init 5 returns the same result.

Perhaps it is a display manager issue. [gkx]dm does not run in runlevel 3, but does in 5.

Should I change the display manager from kdm to xdm or ?? and try 'init 3', 'init 5' again? Or should I reinstall kdm, can I? I am grasping at straws here <grin>.

If you do that, and the problem does not go away, than it is at least not specific to 1 display manager.

BTW, try 'startx', followed by 'startx -- :1' from another console. I like to know if you can run two instances at the same time. My guess is 'no'.

You are correct. See other post in this thread (answering Anders).

I just rebooted after changing parallel yes to no

This keeps the 'Starting ...' and their eventual error messages next to eachother. (Booting is slightly slowed down, but I can't tell on my pc)

and changing from auto-logon to not.

The 'welcome' screen shows and X hangs right there. I suppose it is doing the same thing that it did with the auto-logon but doesn't display the 'welcome' screen.

Tried again 'startx -- :1', with same errors reported previous. Something about failed to allocate decoder object ...

My believe that it's the nvidia driver gets stronger. But perhaps it's just permissions being wrong somewhere (/dev/* ?). leen@cc22149-a:~> grep nvidia /etc/modprobe.conf alias char-major-195* nvidia leen@cc22149-a:~> ls -l /dev/nvidia* crw-rw-rw- 1 root root 195, 0 2004-07-03 10:09 /dev/nvidia0 crw-rw---- 1 root video 195, 0 2004-04-06 15:27 /dev/nvidia00 crw-rw---- 1 root video 195, 1 2004-04-06 15:27 /dev/nvidia01 crw-rw---- 1 root video 195, 2 2004-04-06 15:27 /dev/nvidia02 crw-rw---- 1 root video 195, 3 2004-04-06 15:27 /dev/nvidia03 crw-rw-rw- 1 root root 195, 1 2004-07-03 10:09 /dev/nvidia1 crw-rw-rw- 1 root root 195, 2 2004-07-03 10:09 /dev/nvidia2 crw-rw-rw- 1 root root 195, 3 2004-07-03 10:09 /dev/nvidia3 crw-rw-rw- 1 root root 195, 4 2004-07-03 10:09 /dev/nvidia4 crw-rw-rw- 1 root root 195, 5 2004-07-03 10:09 /dev/nvidia5 crw-rw-rw- 1 root root 195, 6 2004-07-03 10:09 /dev/nvidia6 crw-rw-rw- 1 root root 195, 7 2004-07-03 10:09 /dev/nvidia7 crw-rw-rw- 1 root root 195, 255 2004-07-03 10:09 /dev/nvidiactl This is from another machine that was never touched by the NVidia drivers: um-01:~ # ls -l /dev/nvidia* crw-rw---- 1 root video 195, 0 Apr 6 15:27 /dev/nvidia0 crw-rw---- 1 root video 195, 0 Apr 6 15:27 /dev/nvidia00 crw-rw---- 1 root video 195, 1 Apr 6 15:27 /dev/nvidia01 crw-rw---- 1 root video 195, 2 Apr 6 15:27 /dev/nvidia02 crw-rw---- 1 root video 195, 3 Apr 6 15:27 /dev/nvidia03 crw-rw---- 1 root video 195, 1 Apr 6 15:27 /dev/nvidia1 crw-rw---- 1 root video 195, 2 Apr 6 15:27 /dev/nvidia2 crw-rw---- 1 root video 195, 3 Apr 6 15:27 /dev/nvidia3 crw-rw---- 1 root video 195, 255 Apr 6 15:27 /dev/nvidiactl I'm getting near unknown territory. Cheers, Leen

* Leendert Meyer [07-03-04 12:23]:

This keeps the 'Starting ...' and their eventual error messages next to eachother. (Booting is slightly slowed down, but I can't tell on my pc)

Did not seem to make a diff...

leen@cc22149-a:~> ls -l /dev/nvidia* crw-rw-rw- 1 root root 195, 0 2004-07-03 10:09 /dev/nvidia0 crw-rw---- 1 root video 195, 0 2004-04-06 15:27 /dev/nvidia00 crw-rw---- 1 root video 195, 1 2004-04-06 15:27 /dev/nvidia01 crw-rw---- 1 root video 195, 2 2004-04-06 15:27 /dev/nvidia02 crw-rw---- 1 root video 195, 3 2004-04-06 15:27 /dev/nvidia03 crw-rw-rw- 1 root root 195, 1 2004-07-03 10:09 /dev/nvidia1 crw-rw-rw- 1 root root 195, 2 2004-07-03 10:09 /dev/nvidia2 crw-rw-rw- 1 root root 195, 3 2004-07-03 10:09 /dev/nvidia3 crw-rw-rw- 1 root root 195, 4 2004-07-03 10:09 /dev/nvidia4 crw-rw-rw- 1 root root 195, 5 2004-07-03 10:09 /dev/nvidia5 crw-rw-rw- 1 root root 195, 6 2004-07-03 10:09 /dev/nvidia6 crw-rw-rw- 1 root root 195, 7 2004-07-03 10:09 /dev/nvidia7 crw-rw-rw- 1 root root 195, 255 2004-07-03 10:09 /dev/nvidiactl

I have the same perms and owners.... -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

* Leendert Meyer [07-03-04 13:20]:

On Saturday 03 July 2004 19:34, Patrick Shanahan wrote:

...

undefined symbol: init_kwifimanager_kcmodule mcop warning: user defined signal handler found for SIG_PIPE, overriding xscreensaver: couldn't get password of "pat" xscreensaver: couldn't get password of "root" xscreensaver-gl-helper: /usr/lib/libGL.so.1: no version information available (required by xscreensaver-gl-helper) kdesu: ERROR: User 501 does not exist

Who is user 501? Does this user really not exist?

I am: pat@wahoo:~> grep pat /etc/passwd pat:x:501:100:Patrick Shanahan:/home/pat:/bin/bash I don't understand that "kdesu: ERROR: User 501 does not exist". I had a 'kdesu' session open when I rebooted.

The only error message in /var/log/{message,warn,XFree86*.log} at about the *same* time is the one you reported earlier? I mean this one:

...

...

(EE) NVIDIA(0): Failed to allocate external video decoder object (EE) NVIDIA(0):  *** Aborting ***

Yes. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

Alle 19:23, sabato 3 luglio 2004, Leendert Meyer ha scritto:

On Saturday 03 July 2004 18:39, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 11:22]:

...

On Saturday 03 July 2004 17:42, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 10:32]:

...

On Saturday 03 July 2004 17:12, Patrick Shanahan wrote:

...

after issuing init 3, startx will start kde, but init 5 returns the same result.

Perhaps it is a display manager issue. [gkx]dm does not run in runlevel 3, but does in 5.

Should I change the display manager from kdm to xdm or ?? and try 'init 3', 'init 5' again? Or should I reinstall kdm, can I? I am grasping at straws here <grin>.

If you do that, and the problem does not go away, than it is at least not specific to 1 display manager.

BTW, try 'startx', followed by 'startx -- :1' from another console. I like to know if you can run two instances at the same time. My guess is 'no'.

You are correct. See other post in this thread (answering Anders).

I just rebooted after changing parallel yes to no

This keeps the 'Starting ...' and their eventual error messages next to eachother. (Booting is slightly slowed down, but I can't tell on my pc)

...

and changing from auto-logon to not.

The 'welcome' screen shows and X hangs right there. I suppose it is doing the same thing that it did with the auto-logon but doesn't display the 'welcome' screen.

Tried again 'startx -- :1', with same errors reported previous. Something about failed to allocate decoder object ...

My believe that it's the nvidia driver gets stronger. But perhaps it's just permissions being wrong somewhere (/dev/* ?).

I am reading now this thread and I have to say that old Nvidia drivers are sometimes faulty. At least they did not work very well on my laptop giving me the problem similar to what you describe. I solved with a shell script which waited for 5 seconds after X hung up, then killed it, then restarted it and everything worked. A better solution though could be using the lastest nvidia driver, which are not giving me that bad behaviour now. If you have the latest SuSE kernel and have the same problem I got installing those drivers, see my other posts.

Cheers,

Leen

Greetings, Praise

* Leendert Meyer [07-03-04 14:19]:

On Saturday 03 July 2004 20:50, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 13:20]:

...

Who is user 501? Does this user really not exist?

I am: pat@wahoo:~> grep pat /etc/passwd pat:x:501:100:Patrick Shanahan:/home/pat:/bin/bash

I don't understand that "kdesu: ERROR: User 501 does not exist". I had a 'kdesu' session open when I rebooted.

Aha. Probably another problem, unrelated to NVidia.

I killed the process before the last reboot and the error is no longer present.

Perhaps it's better to try NVidia's Linux Discussion Forum. On http://www.nvidia.com/object/linux.html is a link on that forum. I tried to search, but it refused because I was not logged in. Oh, I /love/ those forums...

I will, but I don't believe that it is the driver. Two points, it stopped working when it had been working and it still works via startx. When I changed from auto-login, the windowmanager (kdm and xdm)came up asking for user/passwd, accepted the entry and ...stoped...

Another suggestion: you could try earlier NVidia drivers.

I have used 5336 since the last of January, and updated yesterday to 6106 when I thought the problem was solved &8-(

I'm afraid I can't help you any further with this one. :(

thanks for your time and effort.. I owe you a LARGE cold one 8-). you also Anders -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

* Patrick Shanahan [07-03-04 14:30]:

* Anders Johansson [07-03-04 14:28]:

...

On Saturday 03 July 2004 18.39, Patrick Shanahan wrote:

...

The 'welcome' screen shows and X hangs right there. I suppose it is doing the same thing that it did with the auto-logon but doesn't display the 'welcome' screen.

Just idle speculation now, but you got it to work before by upgrading to the latest drivers. Then it failed again on reboot. Perhaps there is a problem in the hardware initialization. Perhaps you should try rmmoding the nvidia module and let it reload and hopefully reinitialize

trying now....

:::: rmmod nvidia init 5 success. Now, is it necessary to issue 'rmmod nvidia' from ?? boot.local to allow for reboots or should the single instance just performed correct the situation, or ....???? note: Still unable to start a second X session... -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Saturday 03 July 2004 21.41, Patrick Shanahan wrote:

success. Now, is it necessary to issue 'rmmod nvidia' from ?? boot.local to allow for reboots

No, that wouldn't make much sense because unless you put the nvidia module in your initrd, it's not loaded at that point. It's not loaded until X starts, and that's normally when kdm runs. If this works, I can't imagine what the problem might be. Like I said, it's voodoo.

note: Still unable to start a second X session...

No clue at all about that. Tell you what though, with all the trouble you've been having in the past and present, I wouldn't be half surprised if it turned out to be some deeper hardware trouble. Some unreliable chipset of sorts

* Leendert Meyer [07-03-04 14:57]:

I think I'm not done yet...

That's good for me <grin>.

1) Does the error appear in /var/log/kdm.log?

entire log entry for a failed attempt: XFree86 Version 4.3.0.1 Release Date: 9 May 2003 X Protocol Version 11, Revision 0, Release 6.6 Build Operating System: SuSE Linux [ELF] SuSE Build Date: 12 February 2004 Before reporting problems, check http://www.XFree86.Org/ to make sure that you have the latest version. 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: Sat Jul 3 16:43:12 2004 (==) Using config file: "/etc/X11/XF86Config" (II) Initializing extension GLX Could not init font path element /usr/X11R6/lib/X11/fonts/local, removing from list! Could not init font path element /usr/X11R6/lib/X11/fonts/CID, removing from list! This is a pre-release version of XFree86, and is not supported in any way. Bugs may be reported to XFree86@XFree86.Org and patches submitted to fixes@XFree86.Org. Before reporting bugs in pre-release versions, please check the latest version in the XFree86 CVS repository (http://www.XFree86.Org/cvs).

2) If KDM accepts the login/passwd combo, then it could be related to your user account... What happens if you create a *new* user account, and login as the new user???

ok, created a new user and the new user acc't works correctly. It is able to log onto kde and xfce from init 5. But if I log-out the new user and try to log-in as pat, I hang again... after entering the password. What is the next step, change password for pat ?? -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Sunday 04 July 2004 00:54, Patrick Shanahan wrote:

* Leendert Meyer [07-03-04 14:57]:

...

I think I'm not done yet...

That's good for me <grin>.

...

1) Does the error appear in /var/log/kdm.log?

entire log entry for a failed attempt:

<snipped> No error, looks good. The error seems to occur later, after logging in.

...

2) If KDM accepts the login/passwd combo, then it could be related to your user account... What happens if you create a *new* user account, and login as the new user???

ok, created a new user and the new user acc't works correctly. It is able to log onto kde and xfce from init 5. But if I log-out the new user and try to log-in as pat, I hang again... after entering the password.

What is the next step, change password for pat ??

I'm trying to figure out if the error is caused by something related to your own user account (home dir, perhaps ownership). If the error does not occur with the new user, but does occur when you login as pat, then something in your home dir must cause the error? I'm now going to try the new drivers myself. Maybe I get the same problem as you. ;) Cheers, Leen

* Patrick Shanahan [07-03-04 17:54]:

* Leendert Meyer [07-03-04 14:57]:

...

2) If KDM accepts the login/passwd combo, then it could be related to your user account... What happens if you create a *new* user account, and login as the new user???

ok, created a new user and the new user acc't works correctly. It is able to log onto kde and xfce from init 5. But if I log-out the new user and try to log-in as pat, I hang again... after entering the password.

What is the next step, change password for pat ??

Changing *my* (pat) password does not help .... -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Sunday 04 July 2004 00.54, Patrick Shanahan wrote:

...

2) If KDM accepts the login/passwd combo, then it could be related to your user account... What happens if you create a *new* user account, and login as the new user???

ok, created a new user and the new user acc't works correctly. It is able to log onto kde and xfce from init 5. But if I log-out the new user and try to log-in as pat, I hang again... after entering the password.

What is the next step, change password for pat ??

I thought you said plain kdm hung if you disabled autologin?! If that's true, the problem is unlikely to be in your user account

On Sunday 04 July 2004 01:31, Leendert Meyer wrote:

I'm trying to figure out if the error is caused by something related to your own user account (home dir, perhaps ownership). If the error does not occur with the new user, but does occur when you login as pat, then something in your home dir must cause the error?

I'm now going to try the new drivers myself. Maybe I get the same problem as you. ;)

No luck. The driver installed flawlesly. Sh*t. See if there is a difference in file permissions/ownership in your home dir. Just compare the config files in /home/pat with /home/. Uhm, does it help if you choose a different desktop for user pat? Cheers, Leen

* Leendert Meyer [07-03-04 18:54]:

See if there is a difference in file permissions/ownership in your home dir.

seem to be. Of course, the <new-user> doesn't have near as much. .kde/share new-user had go+rw different

Just compare the config files in /home/pat with /home/.

samo, samo, for same apps.

Uhm, does it help if you choose a different desktop for user pat?

I 'assume' you mean window manager. I tried Xfce and had the same. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Sunday 04 July 2004 03:48, Patrick Shanahan wrote:

Then removed (mv) ~/.kde but X still hangs after I hit the <enter> key after name/password on login screen. I tried 4 different previously unused window managers with same result.

If you select 'Failsafe' as window manager, what happens then?

The <test-user> I created can login (after init 5) to kde, gnome, and xfce and out and in w/o a problem.

The common factor here is X, perhaps Qt. .DCOPserver is KDE related (AFAIK), therefore not to blame (I'm not sure though). Perhaps ~/.ICEauthority or ~/.Xauthority? What happens if you rename those files, or move them out of the way? (logout first). Cheers, Leen

On Sunday 04 July 2004 04:16, Patrick Shanahan wrote:

* Leendert Meyer [07-03-04 21:05]:

...

If you select 'Failsafe' as window manager, what happens then?

below

...

...

The <test-user> I created can login (after init 5) to kde, gnome, and xfce and out and in w/o a problem.

The common factor here is X, perhaps Qt.

.DCOPserver is KDE related (AFAIK), therefore not to blame (I'm not sure though).

Perhaps ~/.ICEauthority or ~/.Xauthority?

What happens if you rename those files, or move them out of the way? (logout first).

logged out, renamed both files (.ICExx & .Xautxxx) init 5 logon screen, name, password, window manager = 'failsafe'

welcome screen with mouse movement, nothing else. sameo...

Finally. I can reproduce the same error message. I can start a first X session normally with NVIDIA's driver. A second X session fails with the same error message: (EE) NVIDIA(0): Failed to allocate external video decoder object (EE) NVIDIA(0): *** Aborting *** Note: from /var/log/kdm.log I tried to start the 2nd X session with <KDE-menu> -> Switch User With a new created user, and issuing a 'startx -- :1' I get the same error message! (~/.X.err). I'll try older NVIDIA drivers now. Hangon. Cheers, Leen

On Saturday 03 July 2004 21:24, Praise wrote:

Alle 19:23, sabato 3 luglio 2004, Leendert Meyer ha scritto:

...

On Saturday 03 July 2004 18:39, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 11:22]:

...

On Saturday 03 July 2004 17:42, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 10:32]:

...

On Saturday 03 July 2004 17:12, Patrick Shanahan wrote: > after issuing init 3, startx will start kde, but init 5 returns > the same result.

Perhaps it is a display manager issue. [gkx]dm does not run in runlevel 3, but does in 5.

Should I change the display manager from kdm to xdm or ?? and try 'init 3', 'init 5' again? Or should I reinstall kdm, can I? I am grasping at straws here <grin>.

If you do that, and the problem does not go away, than it is at least not specific to 1 display manager.

BTW, try 'startx', followed by 'startx -- :1' from another console. I like to know if you can run two instances at the same time. My guess is 'no'.

You are correct. See other post in this thread (answering Anders).

I just rebooted after changing parallel yes to no

This keeps the 'Starting ...' and their eventual error messages next to eachother. (Booting is slightly slowed down, but I can't tell on my pc)

...

and changing from auto-logon to not.

The 'welcome' screen shows and X hangs right there. I suppose it is doing the same thing that it did with the auto-logon but doesn't display the 'welcome' screen.

Tried again 'startx -- :1', with same errors reported previous. Something about failed to allocate decoder object ...

My believe that it's the nvidia driver gets stronger. But perhaps it's just permissions being wrong somewhere (/dev/* ?).

I am reading now this thread and I have to say that old Nvidia drivers are sometimes faulty. At least they did not work very well on my laptop giving me the problem similar to what you describe. I solved with a shell script which waited for 5 seconds after X hung up,

In what file did you do that?

then killed it, then restarted it and everything worked. A better solution though could be using the lastest nvidia driver, which are not giving me that bad behaviour now.

I could reproduce Pat's error message with the latest kernel and nvidia driver. Perhaps waiting a few seconds after X hangs up is a work-around. Cheers, Leen

On Sunday 04 July 2004 13:32, Leendert Meyer wrote:

On Sunday 04 July 2004 13:23, Anders Johansson wrote:

...

On Sunday 04 July 2004 04.16, Patrick Shanahan wrote:

...

logged out, renamed both files (.ICExx & .Xautxxx) init 5 logon screen, name, password, window manager = 'failsafe'

welcome screen with mouse movement, nothing else. sameo...

Have you tried clearing out /tmp ?

Anders: Will try that right now.

A 1st session is no problem, a 2nd session gives the error message. Cheers, Leen

On Sunday 04 July 2004 13.49, Leendert Meyer wrote:

On Sunday 04 July 2004 13:32, Leendert Meyer wrote:

...

On Sunday 04 July 2004 13:23, Anders Johansson wrote:

...

On Sunday 04 July 2004 04.16, Patrick Shanahan wrote:

...

logged out, renamed both files (.ICExx & .Xautxxx) init 5 logon screen, name, password, window manager = 'failsafe'

welcome screen with mouse movement, nothing else. sameo...

Have you tried clearing out /tmp ?

Anders: Will try that right now.

A 1st session is no problem, a 2nd session gives the error message.

Yeah, that /tmp thing was a try for Pat's original problem (can't log in as user A, can log in as user B). I have absolutely no clue about the other problem.

Alle 13:29, domenica 4 luglio 2004, Leendert Meyer ha scritto:

On Saturday 03 July 2004 21:24, Praise wrote:

...

Alle 19:23, sabato 3 luglio 2004, Leendert Meyer ha scritto:

...

On Saturday 03 July 2004 18:39, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 11:22]:

...

On Saturday 03 July 2004 17:42, Patrick Shanahan wrote:

...

* Leendert Meyer [07-03-04 10:32]: > On Saturday 03 July 2004 17:12, Patrick Shanahan wrote: > > after issuing init 3, startx will start kde, but init 5 > > returns the same result. > > Perhaps it is a display manager issue. [gkx]dm does not run in > runlevel 3, but does in 5.

Should I change the display manager from kdm to xdm or ?? and try 'init 3', 'init 5' again? Or should I reinstall kdm, can I? I am grasping at straws here <grin>.

If you do that, and the problem does not go away, than it is at least not specific to 1 display manager.

BTW, try 'startx', followed by 'startx -- :1' from another console. I like to know if you can run two instances at the same time. My guess is 'no'.

You are correct. See other post in this thread (answering Anders).

I just rebooted after changing parallel yes to no

This keeps the 'Starting ...' and their eventual error messages next to eachother. (Booting is slightly slowed down, but I can't tell on my pc)

...

and changing from auto-logon to not.

The 'welcome' screen shows and X hangs right there. I suppose it is doing the same thing that it did with the auto-logon but doesn't display the 'welcome' screen.

Tried again 'startx -- :1', with same errors reported previous. Something about failed to allocate decoder object ...

My believe that it's the nvidia driver gets stronger. But perhaps it's just permissions being wrong somewhere (/dev/* ?).

I am reading now this thread and I have to say that old Nvidia drivers are sometimes faulty. At least they did not work very well on my laptop giving me the problem similar to what you describe. I solved with a shell script which waited for 5 seconds after X hung up,

In what file did you do that?

I wrote my own init script that started after kdm. Unfortunately I do not have it any more because I do not need it, but it was very quick & dirty though.

...

then killed it, then restarted it and everything worked. A better solution though could be using the lastest nvidia driver, which are not giving me that bad behaviour now.

I could reproduce Pat's error message with the latest kernel and nvidia driver. Perhaps waiting a few seconds after X hangs up is a work-around.

Cheers,

Leen

You can also kill it by hand via network, if the problem is like mine. Praise

* Leendert Meyer [07-04-04 08:21]:

I've just subscribed to the nvidia forum I mentioned earlier. The problem is known. I'll see if there is a solution, but I fear there's none so far.

Thanks, I see your post and others similar. I will keep an eye on it. ps. my subscription to suse-linux-e apparently ran out sometime shortly after 0255GMT and 1600GMT today ??? don't forget the cold one <grin>. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Sunday 04 July 2004 21:03, Leendert Meyer wrote:

On Sunday 04 July 2004 19:37, Patrick Shanahan wrote:

...

* Leendert Meyer [07-04-04 08:21]:

...

I've just subscribed to the nvidia forum I mentioned earlier. The problem is known. I'll see if there is a solution, but I fear there's none so far.

Thanks, I see your post and others similar. I will keep an eye on it.

There is a solution, but a was not able to compile the driver.

http://www.minion.de/ http://www.sh.nu/download/nvidia/linux-2.6/

You need this driver: http://www.sh.nu/download/nvidia/linux-2.6/NVIDIA-Linux-x86-1.0-4620-pkg0.r un

Perhaps there is someone with knowledge of kernel-2.6 Makefile's who can assist here.

Here is what I tried: cc22149-a:/tmp/NVIDIA-Linux-x86-1.0-4620-pkg0 # make O=/usr/src/linux-obj/i386/default install cd usr/src/nv; make install make[1]: Entering directory `/tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv' make[2]: Entering directory `/usr/src/linux-2.6.5-7.95-obj/i386/default' make -C /usr/src/linux-2.6.5-7.95 O=/usr/src/linux-2.6.5-7.95-obj/i386/default modules CC [M] /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.o /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: In function `nvos_malloc_pages': /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:384: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: In function `nvos_create_alloc': /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:457: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:467: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: At top level: /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:1101: warning: initialization from incompatible pointer type /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: In function `nv_alloc_file_private': /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:1108: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:1119: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: In function `nv_kern_open': /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:1179: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: In function `nv_kern_ctl_open': /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:1826: warning: use of cast expressions as lvalues is deprecated /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c: At top level: /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:1919: error: conflicting types for `nv_set_hotkey_occurred_flag' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:337: error: previous declaration of `nv_set_hotkey_occurred_flag' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2106: error: conflicting types for `nv_find_kernel_mapping' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:330: error: previous declaration of `nv_find_kernel_mapping' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2181: error: conflicting types for `nv_find_agp_kernel_mapping' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:331: error: previous declaration of `nv_find_agp_kernel_mapping' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2216: error: conflicting types for `nv_get_phys_address' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:332: error: previous declaration of `nv_get_phys_address' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2268: error: conflicting types for `nv_alloc_pages' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:340: error: previous declaration of `nv_alloc_pages' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2444: error: conflicting types for `nv_free_pages' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:341: error: previous declaration of `nv_free_pages' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2542: error: conflicting types for `nv_lock_rm' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:334: error: previous declaration of `nv_lock_rm' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2562: error: conflicting types for `nv_unlock_rm' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:335: error: previous declaration of `nv_unlock_rm' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2582: error: conflicting types for `nv_post_event' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:350: error: previous declaration of `nv_post_event' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2620: error: conflicting types for `nv_get_event' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:351: error: previous declaration of `nv_get_event' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2663: error: conflicting types for `nv_agp_init' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:343: error: previous declaration of `nv_agp_init' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2725: error: conflicting types for `nv_agp_teardown' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:344: error: previous declaration of `nv_agp_teardown' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2766: error: conflicting types for `nv_agp_translate_address' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:345: error: previous declaration of `nv_agp_translate_address' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2798: error: conflicting types for `nv_int10h_call' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:338: error: previous declaration of `nv_int10h_call' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2807: error: conflicting types for `nv_start_rc_timer' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:347: error: previous declaration of `nv_start_rc_timer' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.c:2828: error: conflicting types for `nv_stop_rc_timer' /tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.h:348: error: previous declaration of `nv_stop_rc_timer' make[5]: *** [/tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv/nv.o] Error 1 make[4]: *** [_module_/tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv] Error 2 make[3]: *** [modules] Error 2 make[2]: *** [modules] Error 2 make[2]: Leaving directory `/usr/src/linux-2.6.5-7.95-obj/i386/default' make[1]: *** [module] Error 2 make[1]: Leaving directory `/tmp/NVIDIA-Linux-x86-1.0-4620-pkg0/usr/src/nv' make: *** [kernel_module_install] Error 2 Who can assist here, please? Cheers, Leen

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Why don't you try changing your UID. I think in one of the previous thousand posts for this thread you got a user unknown error or something like that. Since a new user can log in this may have something to do with it. I realize this won't *really* solve the problem, but could be a workaround for now... - --Sayf Patrick Shanahan wrote: | * Leendert Meyer [07-04-04 14:04]: | |>On Sunday 04 July 2004 19:37, Patrick Shanahan wrote: |> |> |>>ps. my subscription to suse-linux-e apparently ran out sometime shortly |>>after 0255GMT and 1600GMT today ??? |> |>I had no problems AFAIK. | | | I don't know what happened. I had several posts bounced; not | subscribed, so I repaid my subscription. I have retrieved the posts | during the period I was *absent*. Dues aren't that much. | | I have not tried twn, but the talk on the nvidia forum indicates the | window manager is not the problem. | | I have cleared /tmp completely several times during the reboots to no | avail. I also mv'ed .Xauthority and .ICEauthority to no avail. | | I *still* do not understand the ability to open X with another user, | and to logoff and back on as the other user, but not myself. | | I *still* cannot open a second X session, even from the new user, | either from the Kicker button or with 'startx -- :N' from level 3 in | an f1 - f6 console. | | btw, I am still on v 9.0. I saw that you were recompiling. | | uname -a | Linux wahoo 2.4.21-231-default #1 Mon Jun 28 15:39:34 UTC 2004 i686 i686 i386 GNU/Linux | nVidia TNT2 AVG2 | Coppermine 733 | - -- Blue Moose IT Support www.bluemooseit.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFA6QMrh6qZaDhhuVMRAruVAKDPr9EcYhDRa5JnGllWY3yC49+3BACfWLnd pxg6iMIAGdZTEpQx+5r74UI= =XOap -----END PGP SIGNATURE-----

* Leendert Meyer [07-05-04 03:08]:

On Monday 05 July 2004 03:09, Patrick Shanahan wrote:

...

I have not tried twn, but the talk on the nvidia forum indicates the window manager is not the problem.

Indeed. But 1 click on the empty Welcome screen will tell you.

see below

...

I *still* cannot open a second X session, even from the new user, either from the Kicker button or with 'startx -- :N' from level 3 in an f1 - f6 console.

*This* is the same overhere: I can't get a 2nd X session too.

again, see below

...

btw, I am still on v 9.0. I saw that you were recompiling.

Then you should try asap that driver I was trying to recompile. As you have kernel 2.4, you can try the original driver. Get it directly from:

http://www.sh.nu/download/nvidia/linux-2.4/

The 1.0-4620 is the latest that was reported to work.

And it does. I cleared /tmp again. Downloaded and installed v 4620 (nvidia). I now can 'init 5' and achieve the log-on screen and log-on as myself, log-off and back on as myself or another user, and again. And, I can open a second and third X session w/o complaint from the system <grin>. Apparently, the entire problem is the nvidia drivers with my video card and, perhaps, kernel ver 2.4.21-231-default. Since you have access to the nvidia forum, would you post this info there? Thanks for everyone's efforts and concerns. If there are further tests or combinations of test conditions that need to be made, I will attempt. -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

* Leendert Meyer [07-05-04 10:11]:

And, from what I have read, the performance difference compared to 1.0-6106 is minimal.

So, TNT doesn't blow up your X sessions anymore! :P

<big-smile>

BTW, a few minutes ago I was able to compile 1.0-4620 on my pc, so I'm about to test it...

Please, keep up advised. thanks again, remember the cold one(s). -- Patrick Shanahan Registered Linux User #207535 http://wahoo.no-ip.org @ http://counter.li.org HOG # US1244711 Photo Album: http://wahoo.no-ip.org/photos

On Saturday 10 July 2004 01.50, Patrick Shanahan wrote:

Major power outage during thunderstorm (2+ inches rain in 15 minutes).

When power return ~1 hour later, had to power cycle cable-modem and router. Booted to level 1, then level 3, then level 5. Reached logon screen, entered name/passwd/kde .... and nothing. Did init 3 and startx and kde works.

Did init 5, logged into <text> user and everything works normally, including starting another X session.

This is exactly what I had before except now on ver 4620 nvidia drivers.

May I suggest getting a UPS? Power cuts and/or power spikes on unprotected hardware can often cause very strange effects

On Saturday 10 July 2004 15:14, Patrick Shanahan wrote:

* Anders Johansson [07-09-04 18:59]:

...

May I suggest getting a UPS? Power cuts and/or power spikes on unprotected hardware can often cause very strange effects

We experience outages less than once per year and less than 60 minutes. I also have made a habit of shutting down and disconnecting when lightening is expected.

Good habit! Do you use reiserfs? A very convenient fs. Saved me quite some headaches! Cheers, Leen

*** Reply to message from Patrick Shanahan on Sat, 10 Jul 2004 09:39:59 -0500***

No, ext3. But it has also saved some headaches. Journal is good.

ditto.. using JFS here and love it... If you are using Reiserfs on 9.1 backup religeously... IT's Suse's default install for 9.1, but there seems to be some little nasty (bug) in it. I decided to just eat one of my installs. And when something bites me I usually look for a happier solution. <Grin> Does anyone know yet what is the disonence between Suse9.1 and Reiserfs? I figure some sort of fix will show up before long. 9.1 can be installed on most hardware ( tho my kids Cow box refuses to boot from a bootable cd) er, "cow box" refers to Gateway computer co. For some reason they put cow spots on things, like the shipping packaging. Unexplainable.. they refuse any information because she doesn't have the origional w2k on the box.. Apparenly when you buy something from them you are never spupsosed to upgrade.. wierd.( she has xp on it, but doesn't want to connect to the internet... as she has had a lot of fun w/ all those virii.. well fun if you like torture!!.) We managed to get a 9.1 on it w/ lots of problems <sigh> Everything doesn't work yet. I made boot floppies and modules disks from Suse's dvd... and hope that some combination will let her install 9.1 pro ( I gave her the cd's ) ;) I have a hunch it's either some weird proprietary setup or it could be just old hardware. oh yeah, boot from cd is first in her bios ... but something is stopping that from happening.. hence the floppies. shades of 7.2 ;) -- j -- nemo me impune lacessit it's just an afterthought; okay ? : Programmer for Ark. Experience in binary searches.

On Saturday 10 July 2004 19.06, Leendert Meyer wrote:

...

Does anyone know yet what is the disonence between Suse9.1 and Reiserfs?

What's a disonence? KDict could not find it.

Dissonance? Reiserfs has a number of nasty bugs, I've had it completely wipe partitions on me after crashes. Nowadays I much prefer XFS

On Saturday 10 July 2004 19:33, jfweber@bellsouth.net wrote:

*** Reply to message from Anders Johansson on Sat, 10 Jul 2004 19:08:34 +0200***

...

Dissonance?

jfweber: You originally wrote "disonence", ;) now I understand why KDict could not find that word... ;P

usually a disconnect , or mis match of parts that used to work together.. It's been adapted from it's origional meaning by techies <grins> I *think* the origional meaning had to do w/ sound and what

A false tone. Cheers, Leen

On Sat July 10 2004 9:53 am, jfweber@bellsouth.net wrote:

*** Reply to message from Patrick Shanahan on Sat, 10 Jul 2004 09:39:59 -0500***

Apparenly when you buy something from them you are never spupsosed to upgrade.. wierd.( she has xp on it, but doesn't want to connect to the internet... as she has had a lot of fun w/ all those virii.. well fun if you like torture!!.) We managed to get a 9.1 on it w/ lots of problems <sigh> Everything doesn't work yet. I made boot floppies and modules disks from Suse's dvd... and hope that some combination will let her install 9.1 pro ( I gave her the cd's ) ;) I have a hunch it's either some weird proprietary setup or it could be just old hardware.

oh yeah, boot from cd is first in her bios ... but something is stopping that from happening.. hence the floppies. shades of 7.2 ;)

Hi, have you tried booting from the second CD? It seems like it's more of a problem with the various CD formats. Rich

nemo me impune lacessit

it's just an afterthought; okay ? : Programmer for Ark. Experience in binary searches.

-- C. Richard Matson

*** Reply to message from "C. Richard Matson" on Sat, 10 Jul 2004 13:10:38 -0700***

Hi, have you tried booting from the second CD? It seems like it's more of a problem with the various CD formats. Rich

oops brain cramp, ignore earlier answer. yes we did try Suse's 2nd cd but the thing doesn't want to boot anything, not even my trusty tag linux ddisk.. -- j -- nemo me impune lacessit it's just an afterthought; okay ? : I'm already visualizing the duct tape over your mouth.