Hi all,
The title screams it all...I installed an nVidia driver on a box and now, at boot the screen gets deactivated . Usually, I'd Ctrl-Alt-F1 into a CLI but since the screen is disabled...hehe. So, I boot in with a live CD (Knoppix or something) and usually I get to the nub of things, text only but still...
So, where do I tickle the system to get it to NOT load the (wrong) driver?
Thanks
PS - I could just reinstall, it is a fresh system, but, I'd miss out on a great opportunity to learn something new
I install fgrlx driver on vivid os the driver look like is istalled correct on system and built the necessary kernel modules because the dpkg log not report any errors, i configured the driver by simple commands like:
sudo aticonfig --initial
sudo aticonfig --screen-layout=on
i reboot the machine , but i face an strange issue the boot splash screen is hangup,i cant enter my lvm password to boot the system, i was have a chance to boot into rescue mode when boot to rescue mode the display not read by the system i try to export display by
export DISPLAY=:0
I try to configure /etc/X11/xorg.conf there is not good resaults the boot splash still hang up
I edited the grub , i uncomment "quiet splash" modified by "nomodeset"
i can boot without boot splash but im faced other issue i cant switch to virtual tty
i dont know is this a bug in the driver or in the kernel
any suggest
Hey guys. So I just went out and bought a Sager 8268-s (clevo p150sm-a) With an i7 4910mq and a Nvidia 980m gt. As I am a computer tech by trade, every time i see Windows i get horrible PTSD so i decided to try Linux instead. After many hours of trying to get a dual boot working finally got Linux Mint 17 Cinnamon installed on my computer, but when I went to install the Nvidia driver it landed me in fallback mode which I could not escape. So then I tried Ubuntu 14.04 LTS thinking maybe Mint was incompatible with the driver and the same thing happened.
So here is my method:
(sorry i don't know how to make a code box :/ )
First I log in to get root access:
$ su
Then I Update the xorg:
$ sudo apt-add-repository ppa:xorg-edgers/ppa
Then update my system:
$ sudo apt-get update && sudo apt-get dist-upgrade
Then i go into Virtual terminal (Ctrl+Alt+F1) and disable the display manager:
$ sudo service mdm (or lightdm for Ubuntu) stop
Then i install the drivers which i have done multiple ways:
$ sudo apt-get install nvidia-current nvidia-settings
or
$ sudo apt-get install nvidia-343 nvidia-settings
or by downloading the driver from Nvidia (first designating the file as executable)
$ sudo sh nvidia.run && sudo apt-get install nvidia-settings
Then after install, I create the config file:
$ sudo nvidia-xconfig
Then I reboot. After I get the low res mode in Ubuntu or fallback mode in Mint and Driver manager doesn't detect any proprietary drivers and I have to reinstall to get the OS working again. Any suggestions would be greatly appreciated.
Hi,
These NVIDIA products are real nightmares. I recently downloaded a copy a sabayon linux for my desktop.
The live dvd booted with the following error message throughout the screen
Quote:
[ 3129.434776] NVRM: No NVIDIA graphics adapter found!
[ 3129.434855] NVRM: NVIDIA init module failed!
[ 3129.479276] NVRM: The NVIDIA GeForce 7025 / nForce 630a GPU installed in this system is
NVRM: supported through the NVIDIA 304.xx Legacy drivers. Please
NVRM: visit http://www.nvidia.com/object/unix.html for more
NVRM: information. The 340.58 NVIDIA driver will ignore
NVRM: this GPU. Continuing probe...
[ 3129.479285] NVRM: No NVIDIA graphics adapter found!
[ 3129.479362] NVRM: NVIDIA init module failed!
[ 3129.519879] NVRM: The NVIDIA GeForce 7025 / nForce 630a GPU installed in this system is
NVRM: supported through the NVIDIA 304.xx Legacy drivers. Please
NVRM: visit http://www.nvidia.com/object/unix.html for more
NVRM: information. The 340.58 NVIDIA driver will ignore
NVRM: this GPU. Continuing probe...
However after a long time it started up properly. After installing the os I found that the problem still persists. Firtsly it took about 10 mins or so to boot. And then apart from terminal 7 (which is the graphical terminal) all other terminals shows the above output. It seems like it has fallen through an infinite loop. I tried downloading some drivers from the nvidia site but simply could not install them because it mandates the X to stop working. If I do so I won't be able to install it from other terminals.
I have attached the output of "dmesg".
Any help is appreciated.
These instructions originate from https://www.linkedin.com/pulse/20140...nvidia-drivers, but I modified them slightly for my desktop environment (64-bit HP Envy Phoenix 810-135qe w/Intel i7-4770 and HP ZR2440w monitor, nVidia GeForce GTX645, Red Hat Enterprise Linux 7.1 w/Gnome3).
The following procedures replace the nouveau drivers with kmod-nvidia drivers from ElRepo.
Install the repository ElRepo from: http://elrepo.org/tiki/tiki-index.php
Install the package nvidia-detect from within Applications>SystemTools>Software
From within the terminal, enter the command nvidia-detect.
Example:
[anthony@hp810135qe ~]$ nvidia-detect
Output is...
Probing for supported NVIDIA devices...
[10de:11c4] NVIDIA Corporation GK106 [GeForce GTX 645 OEM]
This device requires the current 346.59 NVIDIA driver kmod-nvidia
For my system the following software packages will apply (yours maybe slightly different, the key is to note the ###.## version identified for your system using nvidia-detect as explained above): kmod-nvidia-346.59-1.el7.elrepo (64-bit), nvidia-x11-drv-32bit-346.59-1.el7.elrepo (64-bit), nvidia-x11-drv-346.59-1.el7.elrepo (64-bit)
Now do the following steps in the sequence indicated:
1. Disable X Windows
a) open a terminal and "su" or "sudo -i"
b) ln -fs /lib/systemd/system/multi-user.target /etc/systemd/system/default.target
Note: Another way to disable X Windows is to do the following from within the Terminal # systemctl set-default multi-user.target
To later change back to booting into graphical mode, substitute 'graphical.target' for 'multi-user.target' in the above commands.
c) reboot
2. Remove Nouveau
a) rpm -e xorg-x11-drivers xorg-x11-drv-nouveau
Note: Another way to remove Nouveau is to do the following from within the Terminal
$ yum remove xorg-x11-drivers xorg-x11-drv-nouveau xorg-x11-glamor
See http://elrepo.org/tiki/kmod-nvidia which discusses that glamor still causes a conflict with nVidia in RHEL 7
3. Blacklist Nouveau
a) edit /etc/modprobe.d/blacklist.conf and add line:
blacklist nouveau
Note: A way to edit this file is by doing the following:
# vi /etc/modprobe.d/blacklist.conf
(use INSERT key to make changes in vi; add the line
blacklist nouveau
Hit Esc key to escape INSERT mode, then hit SHIFT Z Z to save and exit with the changes. Hitting :q ENTER leaves vi without saving changes.
b) edit /etc/default/grub and append to GRUB_CMDLINE_LINUX:
rd.driver.blacklist=nouveau
Note: A way to edit this file is by doing the following:
# vi /etc/default/grub and append to GRUB_CMDLINE_LINUX:
(use INSERT key to make changes in vi; add the line
rd.driver.blacklist=nouveau
Hit Esc key to escape INSERT mode, then hit SHIFT Z Z to save and exit with the changes. Hitting :q ENTER leaves vi without saving changes.
IMPORTANT NOTE IF YOU HAVE AN ENCRYPTED BOOTUP DRIVE -
If you have an encrypted root drive, remove "rhgb" from GRUB_CMDLINE_LINUX. This will allow you to interact with the encryption passphrase prompt, since Plymouth doesn't seem to run without a framebuffer friendly video driver loaded. IF YOU DON'T THIS, WHEN YOU BOOTUP YOU WILL GET STUCK AT A BLINKING CURSOR AND BLACKSCREEN THAT HAS THE FOLLOWING ERROR MESSAGES:
[5.757276] scsi 7:0:0:0: alua: not attached
[5.759625] sd 7:0:0:0: [sdg] no caching made page found
[5.759668] sd 7:0:0:0: [sdg] assuming drive cache: write through
WHEN THIS HAPPENS IT MEANS PLYMOUTH IS NOT ABLE TO SHOW YOU THE PROMPT FOR ENTERING THE PASSPHRASE FOR THE ENCRYPTED DRIVES. WITHOUT THE PROMPT THE SYSTEM CAN'T ACCEPT THE PASSPHRASE AND YOU ARE STUCK. IF THIS HAPPENS TO YOU, YOU CAN DO CTRL-ALT-DELETE TO REBOOT - THIS ALLOWS YOU TO HIT 'e' AT THE KERNEL LINE IN GRUB TERMINAL TO EDIT (MOST LIKELY THE TOP KERNEL) AND THEN REMOVE 'rhgb' (which stands for Red Hat Graphical Bootsplashcreen or something like that). WHEN DONE HIT THE KEYS 'CTRL X' TO REBOOT - YOU'LL THEN SEE TEXT MESSAGES SCROLLING AND GET AN OPPORTUNITY TO ENTER THE PASSPHRASE. ENTER THE PASSPHRASE AND YOU SHOULD BE GOOD TO GO.
4. Reconfigure the grub.cfg files
Two options for booting nowadays are BIOS and EFI
-If you chose BIOS boot run this command:
# grub2-mkconfig -o /boot/grub2/grub.cfg
Note: I also have this one being auto-updated-configured in my system (?why? It's EFI).
-If EFI boot on CentOS:
# grub2-mkconfig -o /boot/efi/EFI/centos/grub.cfg
-If EFI boot on RHEL:
# grub2-mkconfig -o /boot/efi/EFI/redhat/grub.cfg
Note: Mine is an EFI system. But I did both this and the BIOS grub.cfg above since it too is being updated by RHEL updates concurrently with the EFI grub.cfg
5. Reboot and Install NVidia Driver - I use ElRepo's kmod-nvidia for this step.
Steps by me to install ElRepo's kmod-nvidia (You should still be in Text mode at this time):
If in multi-user.target mode (text, runlevel 3), then do the following
$ su -
enter root password
# yum install kmod-nvidia - version from above
# yum install nvidia-x11-drv-32bit - version from above (this is for 32bit compatibility)
# yum install nvidia-x11-drv - version from above (this is for Open GL X11 display driver and to be done if kmod didn't already include doing it with the kmod-nvidia installation above)
# exit
$
If still in graphics.target mode (you shouldn't be but if you are) (GUI, runlevel 5),
Use Applications > Software > kmod-nvidia - version from above
Install kmod-nvidia - version from above
Install nvidia-x11-drv-32bit - version from above (this is for 32bit compatibility)
Install nvidia-x11-drv - version from above (this is for Open GL X11 display driver)
6. At this point rebuild initramfs. This recreates initrd w/o nouveau. The existing initrd still has the embedded nouveau drivers if this step is not performed) :
Back up the current initramfs keeping nouveau
# mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img
Build the new one without nouveau
# dracut /boot/initramfs-$(uname -r).img $(uname -r)
7. Re-Enable X Windows
# systemctl set-default graphical.target
8. Reboot
If you have an encrypted bootup drive, when you bootup you may notice that Plymouth doesn't look so good with its text in a lower resolution and you'll see lines of text scrolling until you get to a stopped cursor waiting for you to enter the passphrase to unencrypt your drive. Maybe eventually there'll be a solution for this to keep things moving with a graphical bootsplash all throughout. Until then, this is a minor inconvenience for me.
Hi everybody!
I have visited this forum a lot in the last years to learn information about Linux, so above all thank you very much to the linuxquestions.org community for building this valuable resource!
Now my question:
I have a brand new Nvidia GTX750 graphics card. Nouveau does not work correctly with this card (very low resolution and red dots all around the screen appear), so I need to install propietary drivers.
I have coped with this when installing distros in my hard drive, I just install the propietary drivers following the reccomended user guidelines and that's all, but I'm now starting to work with Tails (amnesic Linux distribution) and 'unfortunately' as it's an amnesic system I'll have to install them everytime I boot.
I was wondering if there's a way to install them permanently on my Live CD/USB so I am ready to work as soon as I boot. I have not really any clue on how a LiveUSB is composed, and as far as I have searched on the Internet I have not found any information about this matter.
Sorry for the bad grammatics, I have a pretty bad level of English
TL;DR: How can I permanently install Nvidia propietary drivers on an amnesic Linux Live CD/USB?
Regards,
tete7
Booting to CentOS Linux problem. I have an HP desktop PC with an AMD CPU and win8.1 installed on it with UEFI as the boot process. The CentOS7 DVD indicates that I have installed the CentOS successfully. When I boot the PC it does take me to a menu offering CentOS and Win8. Win 8 will boot correctly. However, when I try to boot to CentOS I get a screen that says “Kernel not found”. The boot loading process continues until it ultimately brings up a terminal login screen. When I try to login with my password and usr name it tells me that they are wrong. I suspect the Kernel not found is the real problem but I can't get into the CentOS program to see if GRUB or any of the other programs are missing or defective. So far I have not disabled the UEFI since I didn't think I needed to. Any suggestions?
Hi Guys, I installed a dual boot config. on a 160 gb hdd. The first choice in grub menu is Linux Lite (latest distro) and the second boot choice is Mint 17.1 Rebecca. I'm encountering a situation that when I boot into LL the system boots OK. Then when I try a restart and boot into Mint it seems to be doing OK until the end of the boot process at which time the screen flashes the Mint logo and then goes dark-then flashes then goes dark every second or so-endlessly. After a while I'm sometimes able to get Mint to finish booting by hitting the esc key. Any ideas how to make this screen flashing situation stop and do a normal boot?
Update: Solved the problem by changing the resolution settings to coincide with those of LL....flashing stopped!
LL and Mint IMHO are two of the best distros for a Newbie to "get their feet wet." - I know, I am that Newbie.
This morning I was watching some online TV shows on CBS.com. After watching two shows, I tried to watch a third. Instead of a video, I got a screen saying I needed to install Adobe Flash. Obviously, Adobe Flash was already installed--that's how I had watched the first two videos. I decided maybe the problem was CBS wanted a later version of Flash. That's when I ran the YUM update.
Everything seemed to download OK except for firefox. It must have been downloading from a mirror somewhere on the other side of the planet because the download rate was something like 7kb/s. The ETA was something like two hours. I decided to end that terminal session and started the YUM update over again. This time it recognized it had already downloaded all the other packages and the download for firefox was several hundred kilobytes per second. After that the update seemed to go OK.
Unfortunately, when I rebooted and chose the new kernel in the grub list, the boot appears to hang just before I'm supposed to get a login screen. Normally, booting takes only a minute or so, but now all I get is a blinking cursor in the upper left corner, even after waiting 10 minutes. I hit the power switch to turn the computer off and rebooted.
This time I hit the escape key to watch the boot progress. These are the last three lines:
[ OK ] Started Command Scheduler
Starting Terminate Plymouth Boot Screen...
Starting wait for Plymouth Boot Screen to quit...
Then it hangs. Hitting the escape key has no effect. CTRL-ALT-DEL reboots.
According to grub, the newly updated kernel version is: Fedora (3.19.5-200.fc21.x86_64) 21 (Twenty One)
I tried another YUM update. Here's the results:
[root@XXXXX ~]# yum update
Loaded plugins: fastestmirror, langpacks
Loading mirror speeds from cached hostfile
* fedora: fedora.mirrors.tds.net
* rpmfusion-free: mirror.us.leaseweb.net
* rpmfusion-free-updates: mirror.us.leaseweb.net
* rpmfusion-nonfree: mirror.us.leaseweb.net
* rpmfusion-nonfree-updates: mirror.us.leaseweb.net
* updates: repo.atlantic.net
No packages marked for update
[root@XXXXX ~]#
I had been using the video driver available on nVidia.com on the previous kernel. (That kernel still works because I'm able to type this.) It's very likely not the latest version of video driver since it has been a while since I installed it. Could this have something to do with the hang?
What should I try to do next? Thanks for reading this.
Hello all,
I was working through LFS 7.7 Chp 2.4 and mounted a wrong partition to /mnt/lfs.
I did the following commands while trying to mount the correct LFS partitions:
mount -v -t ext4 /dev/sda8 /mnt/lfs
mount -v -t ext4 /dev/sda9 /mnt/lfs/home
but I was supposed to mount sda11 and sda 12.
Now the system won't boot past the splash screen. I tried booting into a live usb and typed in the following commands
mount -v -t ext4 /dev/sda8 /
mount -v -t ext4 /dev/sda9 /home
but the system still won't boot past the splash screen.
Both gparted and the partition manager for the installer don't show a mount point.
How do I create a mount point from USB or is there another problem/solution?
Thanks
Hello.
I am a newcomer to Linux, coming from Windows, desktop PC user. I already have a few months' experience with Ubuntu and Mint. I can handle some basic terminal commands, but I prefer the GUI whenever possible.
I'm not an IT specialist, far from it. Just a normal average computer user who can read a few things if they are understandable enough, and wants a decent operating system. So please speak to me in simple human terms, I can handle high tech jargon only so much. Thanks.
I have 2 main questions:
1. I've encountered problems when booting from LiveCD (written to USB) with both Mint and Lubuntu. The boot menu appeared but when I pressed any of the “try live” or “install” options, the screen froze with garbled checkerd pixels. Web searching for solutions, I found the thing with accessing special boot options and adding kernel parameters like 'nomodeset' and 'noaccel' and doing that I could proceed with installation. Then, after installing proprietary drivers, everything was fine.
This is not the first time this happened. As far as I understand this is related to the nouveau driver.
Does this happen only with some distros, like, for example, Ubuntu and its derivatives, or is it a larger problem from the main Linux kernel? From what I've read on some forums, such problems happened with other graphic cards as well, and it seems to be an old problem as old as 2011 if not older.
Why nothing could have been done to fix these issues so far? Couldn't all distros use some option from boot menu to either go with simple safe vesa graphics mode or a text based helpful install that might guide the user afterwards in downloading the proprietary drivers if s/he desires?...
And what is the main cause of the problem? Nvidia not doing FOSS drivers? Nouveau not being flexible enough? Linux kernel not keeping up? Particular distros that don't care about adding an extra boot option? A combination of all these? What is to be done? Would switching to other distro help in this regard? how would I know which distros use nouveau and which don't?
2. I'd like to look into other distros as well. What I need is stability, meaning as bug free as possible, as few apps hanging or crashing as possible, while still being user friendly. But no rolling realeases, please. I want to update the system without fear that I won't boot into desktop – again!
I need distros that come with multimedia codecs, Flash and stuff out of the box and also an easy option to install the proprietary video card graphics driver. I am all for FOSS, but for now I'm also being realistic, and unfortunately have to go with proprietary drivers.
What recommendations do you have?
Thank you.