----- "Niki Kovacs" contact@kikinovak.net wrote:
Hi,
I installed CentOS 5 on a client's machine yesterday afternoon. He phoned me in the evening to tell me that his desktop "froze several times upon him", and he had to hard-reboot it.
The last time this happened to me was with some exotic wireless card whose driver had problems with SMP. But here, I have no wireless card.
So the only suspect is the NVidia Geforce 4000 card.
The only non-standard thing I did in my install was to install the proprietary NVidia driver from nvidia.com. I knew afterwards this was
not a good thing to do, since the RPMForge repos do have an nvidia-legacy drivers. I remembered this too late.
Anyway. I'm due to return to the client in a few hours time next morning. I wonder if reverting from "nvidia" to "nv" in xorg.conf could solve my problem.
As for updates, the system is up to date. But I admit I'm a bit clueless.
Any suggestions?
Niki _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
Did/Can you get the customer to try 'Ctrl-Alt-F1' or similar to see whether it may be that the Xserver has hung but not the OS ?
Also, if they know the IP can you get them to attempt a ping from another machine ?
----- "Niki Kovacs" contact@kikinovak.net wrote:
redhat@mckerrs.net a écrit :
Did/Can you get the customer to try 'Ctrl-Alt-F1' or similar
I suggested Ctrl-Alt-Backspace, but that didn't work.
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
Niki,
a lot of these type of issues are memory related. Run memtest to ensure that your RAM is OK.
Cheers.
redhat@mckerrs.net a écrit :
Niki,
a lot of these type of issues are memory related. Run memtest to ensure that your RAM is OK.
Excellent idea! Besides, the client gets the feeling that he "gets value for money" when he watches Memtest's busy blue screen. In the meantime I can have some coffee and explain that "the utility is hunting down the bug but it may take some time" :oD
Niki
Niki Kovacs wrote:
redhat@mckerrs.net a écrit :
Did/Can you get the customer to try 'Ctrl-Alt-F1' or similar
I suggested Ctrl-Alt-Backspace, but that didn't work.
Last time I saw something similar was when my harddisk get broken. Are there any opcode entries in /var/log/messages or /var/log/dmesg?
regards Olaf
Olaf Mueller a écrit :
Last time I saw something similar was when my harddisk get broken. Are there any opcode entries in /var/log/messages or /var/log/dmesg?
Sorry I don't understand what "opcode entries" means (I'm not a native english speaker).
Niki Kovacs wrote:
Olaf Mueller a écrit :
Last time I saw something similar was when my harddisk get broken. Are there any opcode entries in /var/log/messages or /var/log/dmesg?
Sorry I don't understand what "opcode entries" means (I'm not a native english speaker).
What I mean are error messages about your harddisk devices. Something like: kernel: hda: status error: status=0x58 { DriveReady SeekComplete DataRequest } Dec 30 16:43:13 AE86 kernel: ide: failed opcode was: unknown Dec 30 16:43:13 AE86 kernel: hda: status error: status=0x58 { DriveReady SeekComplete DataRequest } Dec 30 ...
Search for opc in /var/log/messages and /var/log/dmesg
Good luck Olaf
Olaf Mueller a écrit :
Last time I saw something similar was when my harddisk get broken. Are there any opcode entries in /var/log/messages or /var/log/dmesg?
Something crosses my mind: could it be related to a faulty video card? I vaguely (very vaguely) remember having had similar troubles with an old NVidia card a few years ago, and this is an old NVidia card.
I also remember something strange happening. Normally when I install a desktop, I begin with a minimal bootable system (everything unchecked in the installer). Then I 'yum groupinstall "X Window System"' and go on configuring X. Only when TWM is working OK, I install some desktop environment (GNOME, KDE, XFCE) and the apps on top.
After I had installed the driver and fired up TWM, I wanted to test 3D abilities, so I typed 'glxgears' in TWM's xterm, but I got an error message about some missing GLX extension. I didn't investigate this further, since it's an install for an office user, and I thought, let's install and configure all needed apps, and we'll see for Compiz later (even the most serious SOHO users love Compiz) :oD
Niki
----- "Niki Kovacs" contact@kikinovak.net wrote:
Olaf Mueller a écrit :
Last time I saw something similar was when my harddisk get broken.
Are
there any opcode entries in /var/log/messages or /var/log/dmesg?
Something crosses my mind: could it be related to a faulty video card? I vaguely (very vaguely) remember having had similar troubles with an old NVidia card a few years ago, and this is an old NVidia card.
I also remember something strange happening. Normally when I install a
desktop, I begin with a minimal bootable system (everything unchecked in the installer). Then I 'yum groupinstall "X Window System"' and go on
configuring X. Only when TWM is working OK, I install some desktop environment (GNOME, KDE, XFCE) and the apps on top.
After I had installed the driver and fired up TWM, I wanted to test 3D
abilities, so I typed 'glxgears' in TWM's xterm, but I got an error message about some missing GLX extension. I didn't investigate this further, since it's an install for an office user, and I thought, let's install and configure all needed apps, and we'll see for Compiz later
(even the most serious SOHO users love Compiz) :oD
Niki
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.
Niki,
It could be a faulty video card, however, I think the error message about no GLX extensions is standard when you are not running a proper 3D accelerated driver. Happy to be corrected.
Cheers.
Niki Kovacs wrote:
Olaf Mueller a écrit : After I had installed the driver and fired up TWM, I wanted to test 3D abilities, so I typed 'glxgears' in TWM's xterm, but I got an error message about some missing GLX extension. I didn't investigate this further, since it's an install for an office user, and I thought, let's install and configure all needed apps, and we'll see for Compiz later (even the most serious SOHO users love Compiz) :oD
Maybe to disable dri and glx could solve the problem? For that edit /etc/X11/xorg.conf (example entry):
Section "Module" #Load "glx" Load "freetype" #Load "dri" EndSection
regards Olaf
Olaf Mueller a écrit :
Section "Module" #Load "glx" Load "freetype" #Load "dri" EndSection
I think that was the culprit. After checking RAM and disk integrity, I reverted the driver back from "nvidia" to "nv" and uncommented the two lines you mentioned.
Looks like it's working now.
thanks very much,
Niki
-
Niki Kovacs -
Olaf Mueller -
redhat＠mckerrs.net