Black screen after kernel upgrade (NVIDIA)

Sector11 · 2017-06-10 16:33:29

Head_on_a_Stick wrote:

from the OP:

sudo apt-get install -t jessie-backports nftables linux-image-amd64 # instructions here: [url]https://wiki.debian.org/Backports[/url]
sudo cp /usr/share/doc/nftables/examples/syntax/workstation /etc/nftables.conf # "workstation" is a whitelist firewall, other examples available
sudo systemctl start nftables
sudo systemctl enable nftables

Did my thing yesterday (Fri 9 June 2017) and ironed out a couple of things. Like:

Sector11 wrote:

Head_on_a_Stick wrote:
Try commenting out this line from /etc/nftables.conf:
# flush ruleset
Sorry but I'm in OpenBSD atm so I can't test directly.
Done ... where do I start now?

commenting out that line ...

Making sure all info needed is in this post because:
Booted to a flashing cursor this morning ... could not do anything.
- started again and selected advanced options and I see a v4 kernel in there ... OK, step one in original instructions.
- - is that my problem or is it the now commented out line?

Going to re-enable that line and reboot.

Sector11 · 2017-06-10 17:32:52

OK, I was rebooting into advanced and selecting kernel 3.16 and answered another post, then tried some stuff.
Did this in my terminal then came here, so this is before seeing this:

 10 Jun 17 @ 14:15:55 ~
   $ dpkg --list | grep linux-image
ii  linux-image-3.16.0-4-amd64             3.16.43-2                            amd64        Linux 3.16 for 64-bit PCs
ii  linux-image-4.9.0-0.bpo.3-amd64        4.9.25-1~bpo8+1                      amd64        Linux 4.9 for 64-bit PCs
ii  linux-image-amd64                      4.9+80~bpo8+1                        amd64        Linux for 64-bit PCs (meta-package)
 
 10 Jun 17 @ 14:16:06 ~
   $ uname -r
3.16.0-4-amd64
 
 10 Jun 17 @ 14:16:36 ~
   $ sudo systemctl start nftables
[sudo] password for sector11: 
 
 10 Jun 17 @ 14:16:59 ~
   $ sudo systemctl enable nftables
 
 10 Jun 17 @ 14:17:06 ~
   $ sudo nft list ruleset
table inet filter {
	chain input {
		type filter hook input priority 0; policy accept;
		iif lo accept
		ct state established,related accept
		icmpv6 type { nd-router-advert, nd-neighbor-advert, nd-neighbor-solicit} accept
		counter packets 20 bytes 4836 drop
	}
}
 
 10 Jun 17 @ 14:17:16 ~
   $

I have no kernel 3.18.

That's true :-) you need at least 3.18 (4.0.2 in unstable right now)

And since you asked:

 10 Jun 17 @ 14:17:16 ~
   $ cat /proc/version
Linux version 3.16.0-4-amd64 (debian-kernel@lists.debian.org) (gcc version 4.8.4 (Debian 4.8.4-1) ) #1 SMP Debian 3.16.43-2 (2017-04-30)
 
 10 Jun 17 @ 14:23:53 ~
   $

Obviously v4.9 is NOT working for me. sad

HoaS wrote:

This has actually come up already in this very thread, I'm not having a good week here

You are having a great week! Repeat after me: I am having a great week!

Not all systems are equal, and mine is obviously "mentally challenged" - that's not your fault!
And it's not keyboard error in this case either.

Sector11 · 2017-06-10 17:35:11

HoaS = 4.5.0-0.bpo.2-amd64
S11 = 4.9.0-0.bpo.3-amd64

What gives with the difference?

Can I still get 4.5.0-0.bpo.2-amd64 and try that? NO!

Normal boot takes me to: 4.9.0-0.bpo.3-amd64
and a black screen with a flashing cursor.

Last edited by Sector11 (2017-06-10 17:45:07)

Head_on_a_Stick · 2017-06-10 17:49:57

Sector11 wrote:

Normal boot takes me to: 4.9.0-0.bpo.3-amd64
and a black screen with a flashing cursor.

You should be able to switch to a TTY and login from there wink

Don't you usually disable kernel mode setting for your NVIDIA card?

Presuming the console login works with the new kernel, check the command line options that have been applied and compare them with those used for 3.16:

cat /proc/cmdline

The options to turn off kernel mode setting are:

nomodeset nouveau.modeset=0

https://wiki.debian.org/KernelModesetting#Disabling_KMS

Sector11 · 2017-06-10 18:00:15

Head_on_a_Stick wrote:

Sector11 wrote:
Normal boot takes me to: 4.9.0-0.bpo.3-amd64
and a black screen with a flashing cursor.
You should be able to switch to a TTY and login from there
Don't you usually disable kernel mode setting for your NVIDIA card?
Presuming the console login works with the new kernel, check the command line options that have been applied and compare them with those used for 3.16:
cat /proc/cmdline
The options to turn off kernel mode setting are:
nomodeset nouveau.modeset=0
https://wiki.debian.org/KernelModesetting#Disabling_KMS

Pardon the full quote but leaving this here:

 10 Jun 17 @ 14:57:31 ~
   $ cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-3.16.0-4-amd64 root=UUID=9c8f5b02-2631-4eee-afec-71c40f5ba863 ro quiet nomodeset nouveau.modeset=0

Will try a reboot and change tty to see if that works. BRB

Sector11 · 2017-06-10 19:05:46

RE: NVIDIA Card, yes, look up one post it's there:

nomodeset nouveau.modeset=0

Rebooted - went to TTY2 - login FAILED

Suggested I read /var/log/Xorg.0.log
Suggested I read /etc/X11/xorg.config
modeprobe: FATAL: module nvidia not found
(EE) Fatal server error all over the screen ....

- I did not read them instead:
- - started SMXI and ripped out nvidia drivers - installing nouveau ...
- - shut down - pause - two - three
- - power up ...
- - and here I is!

conky no longer reports GPU temp but everything else is just fine - so far!

BUT ... time will tell.

Let's do some CLI Foo!

 10 Jun 17 @ 15:49:56 ~
   $ cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-4.9.0-0.bpo.3-amd64 root=UUID=9c8f5b02-2631-4eee-afec-71c40f5ba863 ro quiet
 
 10 Jun 17 @ 15:50:00 ~
   $ dpkg --list | grep linux-image
ii  linux-image-3.16.0-4-amd64             3.16.43-2                            amd64        Linux 3.16 for 64-bit PCs
ii  linux-image-4.9.0-0.bpo.3-amd64        4.9.25-1~bpo8+1                      amd64        Linux 4.9 for 64-bit PCs
ii  linux-image-amd64                      4.9+80~bpo8+1                        amd64        Linux for 64-bit PCs (meta-package)
 
 10 Jun 17 @ 15:56:26 ~
   $ uname -r
4.9.0-0.bpo.3-amd64
 
 10 Jun 17 @ 15:56:36 ~
   $ sudo systemctl start nftables
[sudo] password for sector11: 
 
 10 Jun 17 @ 15:56:55 ~
   $ sudo systemctl enable nftables
 
 10 Jun 17 @ 15:57:07 ~
   $ sudo nft list ruleset
table inet filter {
	chain input {
		type filter hook input priority 0; policy accept;
		iif lo accept
		ct state established,related accept
		icmpv6 type { nd-neighbor-advert, nd-neighbor-solicit, nd-router-advert} accept
		counter packets 24 bytes 5802 drop
	}
}
 
 10 Jun 17 @ 15:57:51 ~
   $ cat /etc/nftables.conf
#!/usr/sbin/nft -f

##flush ruleset  ## try commenting this out

table inet filter {
	chain input {
		type filter hook input priority 0;

		# accept any localhost traffic
		iif lo accept

		# accept traffic originated from us
		ct state established,related accept

		# activate the following line to accept common local services
		#tcp dport { 22, 80, 443 } ct state new accept

		# accept neighbour discovery otherwise IPv6 connectivity breaks.
		ip6 nexthdr icmpv6 icmpv6 type { nd-neighbor-solicit,  nd-router-advert, nd-neighbor-advert } accept

		# count and drop any other traffic
		counter drop
	}
}

Guess I should try with

flush ruleset

again huh. smile

Last edited by Sector11 (2017-06-10 19:12:30)

Sector11 · 2017-06-10 19:11:48

All this just to get a firewall going that I probably don't need.

- upgraded a kernel - just because the (probably not needed) firewall needed it
- lost nvidia driver (easy to reinstall if I want)
- and the worst part is this: gave HoaS the feeling he was having a bad week!

BAD S11, BAD, BAD, BAD! Noooooooooooooooooo! RECIND that!!!!!
Sounds too much like The Orange Monkey.©jr

Head_on_a_Stick · 2017-06-10 19:33:27

Sector11 wrote:

modeprobe: FATAL: module nvidia not found

D'oh! ops

sudo apt install module-assistant
sudo m-a prepare
sudo apt install -t jessie-backports nvidia-driver

No need for that silly SXMI script wink

Sector11 · 2017-06-10 20:17:21

OH MY!

The following extra packages will be installed:
  dkms glx-alternative-mesa glx-alternative-nvidia glx-diversions libegl-nvidia0 libegl1-glvnd-nvidia libgl1-glvnd-nvidia-glx
  libgl1-nvidia-glvnd-glx libgldispatch0-nvidia libglx-nvidia0 libglx0-glvnd-nvidia libnvidia-eglcore libnvidia-glcore libnvidia-ml1
  libvdpau1 nvidia-alternative nvidia-driver-bin nvidia-driver-libs nvidia-egl-common nvidia-egl-icd nvidia-installer-cleanup
  nvidia-kernel-common nvidia-kernel-dkms nvidia-kernel-support nvidia-legacy-check nvidia-modprobe nvidia-support nvidia-vdpau-driver
  update-glx xserver-xorg-video-nvidia
Suggested packages:
  python3-apport
Recommended packages:
  linux-headers-686-pae linux-headers-amd64 linux-headers-generic linux-headers nvidia-settings nvidia-persistenced nvidia-driver-libs-i386
  libopengl0-glvnd-nvidia libgles-nvidia1 libgles-nvidia2 libnvidia-cfg1 nvidia-vulkan-icd
The following NEW packages will be installed:
  dkms glx-alternative-mesa glx-alternative-nvidia glx-diversions libegl-nvidia0 libegl1-glvnd-nvidia libgl1-glvnd-nvidia-glx
  libgl1-nvidia-glvnd-glx libgldispatch0-nvidia libglx-nvidia0 libglx0-glvnd-nvidia libnvidia-eglcore libnvidia-glcore libnvidia-ml1
  nvidia-alternative nvidia-driver nvidia-driver-bin nvidia-driver-libs nvidia-egl-common nvidia-egl-icd nvidia-installer-cleanup
  nvidia-kernel-common nvidia-kernel-dkms nvidia-kernel-support nvidia-legacy-check nvidia-modprobe nvidia-support nvidia-vdpau-driver
  update-glx xserver-xorg-video-nvidia
The following packages will be upgraded:
  libvdpau1
1 upgraded, 30 newly installed, 0 to remove and 188 not upgraded.
Need to get 27.8 MB of archives.
After this operation, 113 MB of additional disk space will be used.
Do you want to continue? [Y/n]

Obviously: Y

Said [No] and

 10 Jun 17 @ 17:15:27 ~
   $ aremv nvidia-driver
[sudo] password for sector11: 
Reading package lists... Done
Building dependency tree       
Reading state information... Done
Package 'nvidia-driver' is not installed, so not removed
The following packages will be REMOVED:
  glx-alternative-mesa* glx-alternative-nvidia* glx-diversions* libuuid-perl* nvidia-installer-cleanup* update-glx*
0 upgraded, 0 newly installed, 6 to remove and 16 not upgraded.
5 not fully installed or removed.
After this operation, 165 kB disk space will be freed.
Do you want to continue? [Y/n] 
(Reading database ... 224279 files and directories currently installed.)
Removing glx-alternative-nvidia (0.7.3~bpo8+1) ...
Purging configuration files for glx-alternative-nvidia (0.7.3~bpo8+1) ...
Removing glx-alternative-mesa (0.7.3~bpo8+1) ...
Removing update-glx (0.7.3~bpo8+1) ...
Removing glx-diversions (0.7.3~bpo8+1) ...
Purging configuration files for glx-diversions (0.7.3~bpo8+1) ...
Removing libuuid-perl (0.05-1+b1) ...
Removing nvidia-installer-cleanup (20151021+1~bpo8+1) ...
Purging configuration files for nvidia-installer-cleanup (20151021+1~bpo8+1) ...
Processing triggers for man-db (2.7.0.2-5) ...
Setting up libvdpau1:amd64 (1.1.1-1~bpo8+1) ...
Processing triggers for libc-bin (2.19-18+deb8u9) ...

Head_on_a_Stick · 2017-06-10 20:26:17

^ Yeah, use whatever is recommended by

nvidia-detect

https://packages.debian.org/jessie-back … dia-detect

Needs to be from jessie-backports though, to match the kernel.

Sector11 · 2017-06-10 20:47:16

nvidia-detect

before after or get it with:

sudo apt install -t jessie-backports nvidia-driver

Sorry for OT

Head_on_a_Stick · 2017-06-10 20:51:22

Sector11 wrote:

before after or get it with:

Use it beforehand and then install the package it suggests.

If you take the time to read the message given in the screenshot that you posted, you will see that it explains exactly what the problem is and exactly which package you need to install wink

The `nvidia-detect` command should just confirm this.

Last edited by Head_on_a_Stick (2017-06-10 22:26:36)

Head_on_a_Stick · 2017-06-10 20:58:58

I have split this out into it's own thread for any other NVIDIA users who switch kernel versions.

Sector11 · 2017-06-10 21:41:45

OK, thanks.

#1 2017-06-10 16:33:29

Black screen after kernel upgrade (NVIDIA)

#2 2017-06-10 17:32:52

Re: Black screen after kernel upgrade (NVIDIA)

#3 2017-06-10 17:35:11

Re: Black screen after kernel upgrade (NVIDIA)

#4 2017-06-10 17:49:57

Re: Black screen after kernel upgrade (NVIDIA)

#5 2017-06-10 18:00:15

Re: Black screen after kernel upgrade (NVIDIA)

#6 2017-06-10 19:05:46

Re: Black screen after kernel upgrade (NVIDIA)

#7 2017-06-10 19:11:48

Re: Black screen after kernel upgrade (NVIDIA)

#8 2017-06-10 19:33:27

Re: Black screen after kernel upgrade (NVIDIA)

#9 2017-06-10 20:17:21

Re: Black screen after kernel upgrade (NVIDIA)

#10 2017-06-10 20:26:17

Re: Black screen after kernel upgrade (NVIDIA)

#11 2017-06-10 20:47:16

Re: Black screen after kernel upgrade (NVIDIA)

#12 2017-06-10 20:51:22

Re: Black screen after kernel upgrade (NVIDIA)

#13 2017-06-10 20:58:58

Re: Black screen after kernel upgrade (NVIDIA)

#14 2017-06-10 21:41:45

Re: Black screen after kernel upgrade (NVIDIA)

Board footer