Linux Mint This forum is for the discussion of Linux Mint. |
Notices |
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
Are you new to LinuxQuestions.org? Visit the following links:
Site Howto |
Site FAQ |
Sitemap |
Register Now
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
|
|
|
01-03-2015, 01:16 PM
|
#1
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Rep:
|
how to recover 64-bit linux mint 17 after nvidia driver install disaster?
Help! I can't boot up! This is what happened.
I've been happy since I switched from 64-bit ubuntu to 64-bit linux mint several months ago. I have been developing a 3D engine with nvidia 331 drivers, and decided it was time to update, so I downloaded the new 64-bit drivers (346.22) and when through the usual process that worked before on ubuntu (which I wrote down when I upgraded nvidia drivers on ubuntu). Those steps (from my notes) were:
##########
ctrl-alt-F1
login if requested
sudo service mdm stop
login if requested
cd /home/max/download/nvidia
sudo sh NVIDIA-Linux-x86_64-346.22.run
answer "yes" to all questions asked
sudo shutdown 0 -h
reboot
##########
The nvidia install went perfectly. However, when bootup time came, an error happened when the xserver was trying to start up, and a crude old-DOS-style display appeared saying:
-----
Failed to start the X server. It is likely that it is not set up correctly. Would you like to view the X server output to diagnose the problem?
-----
I click the "yes" button.
It displays a bunch of stuff, and I scroll down until I see the first error, which I show below (starting a couple lines up from the error to provide context):
Initializing built-in extension DRI2
Loading extension GLX
vesa: Ignoring device with a bound kernel driver
(EE)
Fatal server error:
(EE) no screens found(EE)
(EE)
Please consult the x.org foundation for help.
When I finish, another crude DOS-like dialog appears that says:
-----
The X server is not disabled. Restart MDM when it is configured correctly".
-----
Well, I'm not sure how I can configure anything, since I find no way to boot up linux mint 17 again... not even a command prompt on a black window. I only seem to be able to get into grub during the "countdown" window by pressing "c".
##################################
So, have I been terminally screwed? Or does someone know a way out of this mess, to at least get the OS running again. Maybe if I could get a command prompt, I could try installing a slightly older driver and see if that helps, or go all the way back to 331.
What say ye linux experts? I need to get back up and running!
Thanks in advance for your help.
PS: The following link looks great, but didn't solve my problem... yet.
https://elementaryforums.com/index.p...een.7/#post-12
Last edited by maxreason; 01-03-2015 at 02:33 PM.
|
|
|
01-03-2015, 05:49 PM
|
#2
|
Senior Member
Registered: Jul 2006
Distribution: Debian Unstable
Posts: 1,177
|
Why not just use the nvidia packages that come in the repository?
Anyway, Grub should have an option for single user mode.
|
|
|
01-04-2015, 12:13 AM
|
#3
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Original Poster
Rep:
|
Yeah, tell me about it. Sigh. At this point I'd settle for a 1-year old driver.
Nonetheless, I managed to install a slightly older driver (340.xx), and managed to boot-up correctly after great effort. Except...
When I boot up normally, I get a 640x480 type display and it says it crashed. Restarting doesn't help.
However, if I boot into recovery mode, then just continue the bootup process from the menu, that boots up into a proper 1920x1200 display and the nvidia-settings applications looks right.
Not sure how this can happen, but maybe someone else does.
Getting closer, but... not there yet.
|
|
|
01-04-2015, 08:55 AM
|
#4
|
Senior Member
Registered: Jul 2006
Distribution: Debian Unstable
Posts: 1,177
|
I didn't realize Mint has old drivers. I guess it's bases on Ubuntu's LTS releases. I use Debian 'Sid', which seems to be up-to-date with those drivers.
|
|
|
01-04-2015, 09:32 AM
|
#5
|
Member
Registered: Apr 2012
Distribution: Gentoo
Posts: 616
Rep:
|
Did you try to run nvidia-xconfig?
You could try to boot in recovery mode, boot into proper resolution (as you've mentioned above) and get EDID bin file from NVidia-Settings. NVidia driver may be configured to use this EDID file by adding CustomEDID parameter to xorg.conf that would point to you file. It might solve incorrect resolution.
|
|
1 members found this post helpful.
|
01-05-2015, 06:07 PM
|
#6
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Original Poster
Rep:
|
Yes, but nvidia-settings says something like "it appears nvidia driver is not installed". Which I assume is why the bootup process displays the 640x480 desktop, which it calls "fallback" if I recall correctly.
It is strange... why doesn't a normal bootup find the driver, but the recovery bootup process does? After all, the only thing I do in the recovery bootup is to tell it to continue to boot up, nothing else.
|
|
|
01-05-2015, 06:38 PM
|
#7
|
Member
Registered: Apr 2012
Distribution: Gentoo
Posts: 616
Rep:
|
It might be due to different kernel versions used for normal and recovery/fallback modes. I.e. you have nvidia module for one of versions, but have no for another.
You didn't mentioned which kernel version you have in use and how much kernels do you have installed.
What is the output for
for both modes ( normal boot and recovery mode)?
And you might to check lsmod command output to see whether nvidia module loaded or it is nouveau? For both normal and recovery modes.
Last edited by Teufel; 01-05-2015 at 06:41 PM.
|
|
1 members found this post helpful.
|
01-05-2015, 11:19 PM
|
#8
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Original Poster
Rep:
|
In both modes (normal bootup and recovery mode), the "uname -r" command displays:
3.13.0-24-generic
And indeed, this agrees with the bootup screen where I select normal bootup or recovery bootup.
-----
As for your other question, the output of lsmod is different. The following are the files:
##### NORMAL BOOTUP MODE #####
----------
Code:
Module Size Used by
rfcomm 69160 0
bnep 19624 2
bluetooth 395423 10 bnep,rfcomm
binfmt_misc 17468 1
snd_hda_codec_hdmi 46207 1
dm_multipath 22873 0
scsi_dh 14882 1 dm_multipath
kvm 451511 0
crct10dif_pclmul 14289 0
crc32_pclmul 13113 0
ghash_clmulni_intel 13259 0
aesni_intel 55624 0
aes_x86_64 17131 1 aesni_intel
lrw 13286 1 aesni_intel
gf128mul 14951 1 lrw
glue_helper 13990 1 aesni_intel
ablk_helper 13597 1 aesni_intel
cryptd 20359 3 ghash_clmulni_intel,aesni_intel,ablk_helper
snd_hda_codec_realtek 61438 1
serio_raw 13462 0
edac_core 62291 0
fam15h_power 13119 0
edac_mce_amd 22617 0
k10temp 13126 0
snd_hda_intel 52355 5
snd_hda_codec 192906 3 snd_hda_codec_realtek,snd_hda_codec_hdmi,snd_hda_intel
snd_hwdep 13602 1 snd_hda_codec
snd_pcm 102099 3 snd_hda_codec_hdmi,snd_hda_codec,snd_hda_intel
snd_page_alloc 18710 2 snd_pcm,snd_hda_intel
snd_seq_midi 13324 0
snd_seq_midi_event 14899 1 snd_seq_midi
snd_rawmidi 30144 1 snd_seq_midi
snd_seq 61560 2 snd_seq_midi_event,snd_seq_midi
sp5100_tco 13979 0
i2c_piix4 22155 0
snd_seq_device 14497 3 snd_seq,snd_rawmidi,snd_seq_midi
snd_timer 29482 2 snd_pcm,snd_seq
snd 69238 21 snd_hda_codec_realtek,snd_hwdep,snd_timer,snd_hda_codec_hdmi,snd_pcm,snd_seq,snd_rawmidi,snd_hda_codec,snd_hda_intel,snd_seq_device,snd_seq_midi
soundcore 12680 1 snd
parport_pc 32701 0
ppdev 17671 0
mac_hid 13205 0
lp 17759 0
parport 42348 3 lp,ppdev,parport_pc
dm_mirror 22135 0
dm_region_hash 20862 1 dm_mirror
dm_log 18411 2 dm_region_hash,dm_mirror
hid_generic 12548 0
usbhid 52616 0
hid 106148 2 hid_generic,usbhid
pata_acpi 13038 0
usb_storage 62209 1
nouveau 1097199 2
firewire_ohci 40409 0
r8169 67581 0
firewire_core 68769 1 firewire_ohci
mxm_wmi 13021 1 nouveau
video 19476 1 nouveau
mii 13934 1 r8169
crc_itu_t 12707 1 firewire_core
i2c_algo_bit 13413 1 nouveau
pata_atiixp 13271 1
ttm 85115 1 nouveau
ahci 25819 7
e1000e 254433 0
libahci 32168 1 ahci
drm_kms_helper 52758 1 nouveau
ptp 18933 1 e1000e
drm 302817 4 ttm,drm_kms_helper,nouveau
pps_core 19382 1 ptp
wmi 19177 2 mxm_wmi,nouveau
----------
##### RECOVERY BOOTUP MODE #####
----------
Code:
Module Size Used by
rfcomm 69160 0
bnep 19624 2
bluetooth 395423 10 bnep,rfcomm
binfmt_misc 17468 1
parport_pc 32701 0
ppdev 17671 0
lp 17759 0
parport 42348 3 lp,ppdev,parport_pc
snd_hda_codec_hdmi 46207 1
dm_multipath 22873 0
scsi_dh 14882 1 dm_multipath
kvm 451511 0
crct10dif_pclmul 14289 0
crc32_pclmul 13113 0
ghash_clmulni_intel 13259 0
aesni_intel 55624 0
aes_x86_64 17131 1 aesni_intel
lrw 13286 1 aesni_intel
gf128mul 14951 1 lrw
glue_helper 13990 1 aesni_intel
ablk_helper 13597 1 aesni_intel
cryptd 20359 3 ghash_clmulni_intel,aesni_intel,ablk_helper
snd_hda_codec_realtek 61438 1
edac_core 62291 0
serio_raw 13462 0
edac_mce_amd 22617 0
fam15h_power 13119 0
k10temp 13126 0
snd_hda_intel 52355 5
nvidia 10557103 39
snd_hda_codec 192906 3 snd_hda_codec_realtek,snd_hda_codec_hdmi,snd_hda_intel
snd_seq_midi 13324 0
snd_hwdep 13602 1 snd_hda_codec
snd_seq_midi_event 14899 1 snd_seq_midi
snd_rawmidi 30144 1 snd_seq_midi
snd_pcm 102099 3 snd_hda_codec_hdmi,snd_hda_codec,snd_hda_intel
sp5100_tco 13979 0
snd_page_alloc 18710 2 snd_pcm,snd_hda_intel
i2c_piix4 22155 0
snd_seq 61560 2 snd_seq_midi_event,snd_seq_midi
snd_seq_device 14497 3 snd_seq,snd_rawmidi,snd_seq_midi
snd_timer 29482 2 snd_pcm,snd_seq
snd 69238 21 snd_hda_codec_realtek,snd_hwdep,snd_timer,snd_hda_codec_hdmi,snd_pcm,snd_seq,snd_rawmidi,snd_hda_codec,snd_hda_intel,snd_seq_device,snd_seq_midi
soundcore 12680 1 snd
mac_hid 13205 0
dm_mirror 22135 0
dm_region_hash 20862 1 dm_mirror
dm_log 18411 2 dm_region_hash,dm_mirror
hid_generic 12548 0
usbhid 52616 0
hid 106148 2 hid_generic,usbhid
pata_acpi 13038 0
usb_storage 62209 1
nouveau 1097199 0
firewire_ohci 40409 0
r8169 67581 0
firewire_core 68769 1 firewire_ohci
mii 13934 1 r8169
crc_itu_t 12707 1 firewire_core
mxm_wmi 13021 1 nouveau
pata_atiixp 13271 1
video 19476 1 nouveau
i2c_algo_bit 13413 1 nouveau
ttm 85115 1 nouveau
drm_kms_helper 52758 1 nouveau
e1000e 254433 0
ahci 25819 7
drm 302817 5 ttm,drm_kms_helper,nvidia,nouveau
libahci 32168 1 ahci
ptp 18933 1 e1000e
pps_core 19382 1 ptp
wmi 19177 2 mxm_wmi,nouveau
----------
I see a lot of "nouveau" in both, but I also see "nvidia" on the "drm" line in "recovery mode".
Also, when I run "nvidia-settings" in recovery mode, it runs happily and reports my GPU card (GTX680) and driver version (v340.65) and everything else I'm used to seeing.
----------
When I run nvidia-settings in normal mode (640x480), it says the nvidia driver is not running, and to change the X Windows configuration file (which it supposedly does when I cancel the dialog).
After I do, I find a message in a console window that says "nvidia-settings could not find the registry key file /usr/share/nvidia/nvidia-application-profiles-key-documentation.
When I look in /usr/share/nvidia I do find a file called /usr/share/nvidia/nviaid-application-profiles-340.65-key-documentation. So I make a copy without the version number, but that doesn't change the bootup process... still 640x480, etc.
----------
I hope this helps. What next?
Last edited by maxreason; 01-05-2015 at 11:51 PM.
|
|
|
01-06-2015, 04:46 AM
|
#9
|
Member
Registered: Apr 2012
Distribution: Gentoo
Posts: 616
Rep:
|
what it the output of
for both modes?
As for configuration tool: Nvidia-Settings is useless, at least right know.
You need to run
command as root. This command should configure nvidia driver, re-write xorg.conf and blacklists nouveau module.
If nvidia-xconfig complains about nvidia module, try
and run nvidia-xconfig again then reboot.
|
|
1 members found this post helpful.
|
01-06-2015, 03:53 PM
|
#10
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Original Poster
Rep:
|
In "recovery mode":
BOOT_IMAGE=/vmlinuz-3.13.0-24-generic root=UUID=4964f23f-7750-4974-9b39-f36361c98839 ro recovery nomodeset
In "normal bootup mode":
BOOT_IMAGE=/vmlinuz-3.13.0-24-generic root=UUID=4964f23f-7750-4974-9b39-f36361c98839 ro quiet splash vt.handoff=7
I tried the other stuff you mentioned (both ways), but no luck. I mean, it executes, but no behavior change. Is that "nomodeset" some kind of magic? Or that "vt.handoff=7"? Why else would they boot up so different?
Do these clues help?
Last edited by maxreason; 01-06-2015 at 03:57 PM.
|
|
|
01-06-2015, 04:29 PM
|
#11
|
Member
Registered: Apr 2012
Distribution: Gentoo
Posts: 616
Rep:
|
nomodeset might help, add it to kernel parameters and try. Usually it helps when you have no anything on the screen, but it might help this case too.
Quote:
nomodest
The newest kernels have moved the video mode setting into the kernel. So all the programming of the hardware specific clock rates and registers on the video card happen in the kernel rather than in the X driver when the X server starts.. This makes it possible to have high resolution nice looking splash (boot) screens and flicker free transitions from boot splash to login screen. Unfortunately, on some cards this doesnt work properly and you end up with a black screen. Adding the nomodeset parameter instructs the kernel to not load video drivers and use BIOS modes instead until X is loaded.
|
Added:
vt.handoff related to how does kernel maintains video on early stages. I do not know too much about it.
BTW, you might to boot into recovery mode without "recovery" parameter (enter Grub edit mode and remove this parameter and press F10)
And what nvidia-xconfig responded?
Last edited by Teufel; 01-06-2015 at 04:38 PM.
|
|
1 members found this post helpful.
|
01-06-2015, 06:25 PM
|
#12
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Original Poster
Rep:
|
how do i add nomodset (or anything else) to the kernel parameters?
|
|
|
01-06-2015, 08:35 PM
|
#13
|
Member
Registered: Apr 2012
Distribution: Gentoo
Posts: 616
Rep:
|
Temporarily you can do it when Grub2 menu appears, select menuetry you want to edit and press "e" key, it will drop you into editing mode. Navigate to line that starts with "linux" word and go to the end of it by pressing right arrow. This string might occupy two lines, if so you have to move to the very end. When you at the end type whitespace and nomodeset after it. When you are ready, press F10 key.
This change will work for current boot only.
To add it permanently, edit /etc/default/grub file (as root):
You have to find this string
Code:
#GRUB_CMDLINE_LINUX_DEFAULT=""
and change it to
Code:
GRUB_CMDLINE_LINUX_DEFAULT="nomodeset"
Note, that comment mark (#) must be removed.
Save the file and run in terminal as root:
Code:
grub-mkconfig -o /boot/grub/grub.cfg
it will apply your parameter to grub menu.
if you want just to test, how does it will be appear without re-written grub.cfg, run it this manner:
it will show grub menu in terminal window, how it would be written into grub.cfg. If all seems added, run the previous command to write it to file.
that's all
Last edited by Teufel; 01-06-2015 at 08:43 PM.
|
|
1 members found this post helpful.
|
01-06-2015, 10:20 PM
|
#14
|
Member
Registered: Dec 2007
Location: phobos, mars
Distribution: 64-bit linux mint v21.2
Posts: 270
Original Poster
Rep:
|
Strange. First of all, when I looked in the /etc/default/grub file, that line already said:
GRUB_CMDLINE_LINUX_DEFAULT="nomodeset"
However, below that was a line that was:
GRUB_CMDLINE_LINUX=""
So I changed the empty string to another "nomodeset".
However, when I ran:
sudo grub_mkconfig -o /boot/grub/grub.cfg
the output was
/usr/sbin/grub-mkconfig: 11: /etc/default/grub: nomodeset: not found
Line 11 is the GRUB_CMDLINE_LINUX_DEFAULT="nomodeset" line, not the one I added to (which was line 12). It didn't complain about line 12, but... maybe it only reports one error.
And note that this command prints out the same thing whether I booted up into recovery mode (1920x1200) or normal mode (640x480).
So, that's pretty strange, eh?
Now what?
|
|
|
01-07-2015, 05:13 AM
|
#15
|
Member
Registered: Apr 2012
Distribution: Gentoo
Posts: 616
Rep:
|
Yes, it is really strange that nomodeset found for recovery mode and not found for normal mode...
Well, there is another way to append it:
At first, run this command to create backup for /boot/grub/grub.cfg file:
Code:
sudo cp /boot/grub/grub.cfg /boot/grub/grub.cfg.bak
open /boot/grub/grub.cfg file with any text editor that you can run as root (sudo nano /boot/grub/grub.cfg or gksu geany /boot/grub/grub.cfg or whatever else do you have installed).
Navigate to
Code:
BEGIN /etc/grub.d/10_linux
section.
Here located your menu items. Edit that one that corresponds to normal boot, it should be first menuentry (if you have only one Linux system installed). Find string that starts with "linux". To be sure that it is the string you need, read it, there should be "quiet splash" parameters. Append nomodeset to the end of this string.
Do not edit menuentry for recovery mode.
Save changes and exit. Now reboot. Do not run grub-mkconfig, it may re-wrire your changes.
At reboot, when Grub menu will appear, select you normal boot entry, press "e" key and check does nomodeset mentioned in menuentry. To continue boot, press F10
|
|
1 members found this post helpful.
|
All times are GMT -5. The time now is 05:00 PM.
|
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.
|
Latest Threads
LQ News
|
|