Ubuntu 18.04 Server : Nvidia driver module loaded but device not found

Graphics / Linux Linux
1

Hi
I want to install nvidia driver for a gigabyte rtx 3070 card. My system is linux ubuntu 18.04 server. My goal is to make a headless installation to get a deep learning server. The computer is not connected to a screen and managed by ssh. I am a big newb so i start from clean install. I format my disk twenty times.
I try a lot of methods: apt-get, nvidia installed downloaded from nvidia website, ubuntu-drivers autoinstall and several drivers : nvidia-headless-460, nvidia-headless-460-server, nvidia-drivers-460, cuda-repo etc. I also try to activate/deactivate secure boot in my bazooka b550 bios.

When I am running:

uname -r
4.15.0-142-generic

sudo lshw -C display
*-display
description: VGA compatible controller
product: NVIDIA Corporation
vendor: NVIDIA Corporation
physical id: 0
bus info: pci@0000:2b:00.0
version: a1
width: 64 bits
clock: 33MHz
capabilities: pm msi pciexpress vga_controller bus_master cap_list rom
configuration: driver=nvidia latency=0
resources: irq:52 memory:fb000000-fbffffff memory:d0000000-dfffffff memory:e0000000-e1ffffff ioport:e000(size=128) memory:fc000000-fc07ffff

dmesg | grep nvidia
[ 1.608143] nvidia: loading out-of-tree module taints kernel.

[ 1.608562] nvidia: module license ‘NVIDIA’ taints kernel.

[ 1.634383] nvidia-nvlink: Nvlink Core is being initialized, major device number 240

[ 1.635106] nvidia 0000:2b:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem

[ 1.683842] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 460.73.01 Thu Apr 1 21:32:31 UTC 2021

[ 1.685664] [drm] [nvidia-drm] [GPU ID 0x00002b00] Loading driver

[ 1.686118] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:2b:00.0 on minor 0

[ 4.520745] nvidia-uvm: Loaded the UVM driver, major device number 237.

lsmod | grep nvidia
nvidia_uvm 983040 0

nvidia_drm 49152 0

nvidia_modeset 1224704 1 nvidia_drm

nvidia 34074624 2 nvidia_uvm,nvidia_modeset

drm_kms_helper 172032 1 nvidia_drm

drm 401408 3 drm_kms_helper,nvidia_drm

nvidia-smi
No devices were found

sudo lspci -vvv | grep nvidia
pcilib: sysfs_read_vpd: read failed: Input/output error
Kernel driver in use: nvidia
Kernel modules: nvidiafb, nouveau, nvidia_drm, nvidia

Nouveau driver is disabled beacause
lsmod | grep nouveau reports nothing

What can i do to get working install ?
Help me. Thanx.
nvidia-bug-report.log.gz (496.9 KB)

2

The gpu is crashing:

May 05 13:32:07 dlserver kernel: NVRM: GPU 0000:2b:00.0: RmInitAdapter failed! (0x26:0xffff:1290)
May 05 13:32:07 dlserver kernel: NVRM: GPU 0000:2b:00.0: rm_init_adapter failed, device minor number 0
May 05 13:32:07 dlserver kernel: NVRM: GPU 0000:2b:00.0: RmInitAdapter failed! (0x24:0xffff:1248)
May 05 13:32:07 dlserver kernel: NVRM: GPU 0000:2b:00.0: rm_init_adapter failed, device minor number 0

Please try reseating the card in its slot, check power connectors, check if the card works in another system.