Page 1 of 1


Posted: 2020/11/21 16:02:52
by frg
Hi there,

This is my first post on this forum as I'm not a regular CentOS user, but are more familiar to Ubuntu :oops: . A problem similar to the one I'm posting here I found in the following (old) post:


I think I've actually reached to almost the end of the driver installation process but when rebooting the screen blacks out and finally freezes. Pressing F2-key during booting, the verbose boot screen appears, and the process stops/freezes after the "Starting certmonger: [ OK ]" line.

Hardware setup:
Dell PowerEdge R730 server
NVIDIA GRID K2 GK104GL graphics card

Software setup: (ref. ... index.html)
VMware ESXi 6.5.0
vSphere Client version
CentOS 6.10

However when ssh-ing into the virtual machine and issuing the "cat /var/log/Xorg.0.log" command the following log appears:

Code: Select all

[   126.970] 
X.Org X Server 1.17.4
Release Date: 2015-10-28
[   126.970] X Protocol Version 11, Revision 0
[   126.970] Build Operating System: x86-02 3.10.0-957.1.3.el7.x86_64 
[   126.970] Current Operating System: Linux VM03-CentOS6 2.6.32-754.35.1.el6.x86_64 #1 SMP Sat Nov 7 12:42:14 UTC 2020 x86_64
[   126.971] Kernel command line: ro root=/dev/mapper/vg_vm03centos6-lv_root rd_LVM_LV=vg_vm03centos6/lv_root rd_NO_LUKS LANG=en_US.UTF-8  KEYBOARDTYPE=pc KEYTABLE=no rd_NO_MD SYSFONT=latarcyrheb-sun16 crashkernel=128M rd_LVM_LV=vg_vm03centos6/lv_swap rd_NO_DM rhgb quiet nouveau.modeset=0 rd.driver.blacklist=nouveau video=vesa:off vga=normal
[   126.971] Build Date: 07 November 2020  12:41:27PM
[   126.971] Build ID: xorg-x11-server 1.17.4-18.el6.centos 
[   126.971] Current version of pixman: 0.32.8
[   126.971] 	Before reporting problems, check
	to make sure that you have the latest version.
[   126.971] Markers: (--) probed, (**) from config file, (==) default setting,
	(++) from command line, (!!) notice, (II) informational,
	(WW) warning, (EE) error, (NI) not implemented, (??) unknown.
[   126.971] (==) Log file: "/var/log/Xorg.0.log", Time: Fri Nov 20 12:49:50 2020
[   126.976] (==) Using config file: "/etc/X11/xorg.conf"
[   126.976] (==) Using config directory: "/etc/X11/xorg.conf.d"
[   126.976] (==) Using system config directory "/usr/share/X11/xorg.conf.d"
[   126.988] (==) ServerLayout "Layout0"
[   126.988] (**) |-->Screen "Screen0" (0)
[   126.988] (**) |   |-->Monitor "Monitor0"
[   126.988] (**) |   |-->Device "Device0"
[   126.988] (**) |-->Input Device "Keyboard0"
[   126.988] (**) |-->Input Device "Mouse0"
[   126.988] (==) Automatically adding devices
[   126.988] (==) Automatically enabling devices
[   126.988] (==) Not automatically adding GPU devices
[   126.988] (==) Max clients allowed: 256, resource mask: 0x1fffff
[   126.995] (**) FontPath set to:
[   126.995] (==) ModulePath set to "/usr/lib64/xorg/modules"
[   126.995] (WW) Hotplugging is on, devices using drivers 'kbd', 'mouse' or 'vmmouse' will be disabled.
[   126.995] (WW) Disabling Keyboard0
[   126.995] (WW) Disabling Mouse0
[   126.995] (II) Loader magic: 0x5564de716020
[   126.995] (II) Module ABI versions:
[   126.995] 	X.Org ANSI C Emulation: 0.4
[   126.995] 	X.Org Video Driver: 19.0
[   126.995] 	X.Org XInput driver : 21.0
[   126.995] 	X.Org Server Extension : 9.0
[   127.013] (--) PCI:*(0:0:15:0) 15ad:0405:15ad:0405 rev 0, Mem @ 0xe8000000/134217728, 0xfe000000/8388608, I/O @ 0x00001070/16, BIOS @ 0x????????/32768
[   127.013] (--) PCI: (0:11:0:0) 10de:11bf:10de:100a rev 161, Mem @ 0xfc000000/16777216, 0xd8000000/134217728, 0xe4000000/33554432, I/O @ 0x00005000/128, BIOS @ 0x????????/65536
[   127.013] (WW) "glamoregl" will not be loaded unless you've specified it to be loaded elsewhere.
[   127.013] (II) "glx" will be loaded by default.
[   127.013] (II) LoadModule: "glx"
[   127.043] (II) Loading /usr/lib64/xorg/modules/extensions/
[   127.239] (II) Module glx: vendor="NVIDIA Corporation"
[   127.240] 	compiled for 4.0.2, module version = 1.0.0
[   127.240] 	Module class: X.Org Server Extension
[   127.240] (II) NVIDIA GLX Module  367.134  Tue Dec 17 06:34:03 PST 2019
[   127.240] (II) LoadModule: "nvidia"
[   127.256] (II) Loading /usr/lib64/xorg/modules/drivers/
[   127.295] (II) Module nvidia: vendor="NVIDIA Corporation"
[   127.295] 	compiled for 4.0.2, module version = 1.0.0
[   127.295] 	Module class: X.Org Video Driver
[   127.296] (II) NVIDIA dlloader X Driver  367.134  Tue Dec 17 06:11:18 PST 2019
[   127.296] (II) NVIDIA Unified Driver for all Supported NVIDIA GPUs
[   127.296] (++) using VT number 1

[   127.305] (EE) No devices detected.
[   127.305] (EE) 
Fatal server error:
[   127.305] (EE) no screens found(EE) 
[   127.305] (EE) 
Please consult the CentOS support 
 for help. 
[   127.305] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[   127.305] (EE) 

If anybody could please guide me in the right direction to solve the issue I would really appreciate it! Please also note that my general linux experience is VERY limited, and if any answers to this is posted I would also be happy if your suggested procedure or input commands are specified in detail (imagine yourself feeding an infant with a teaspoon :) ).


Re: Pr

Posted: 2020/11/21 16:07:20
by TrevorH
You should probably start by installing a version of CentOS that is not about to go out of support in less than 2 weeks. CentOS 6 is dead, do not use it.

Second, this is a VM so it cannot see your real hardware. It only sees what VMWare emulates for it and that will not be an Nvidia card. To use the nvidia card in a guest you will have to attach it to the VM (dedicate it to it). Otherwise what your VM sees will be whatever VMWare emulates for the VM in question.

Re: Pr

Posted: 2020/11/21 17:41:37
by frg
Hi Trevor,

and thanks for your quick response!

Well, my choice of CentOS version is solely based on compatibility issues, as given in e.g. the following link: ... 0Final.png

It might well be that I can use a CentOS 7 version too, but I really don't know how far in the OS version line these graphic cards are compatible. Therefore my choice ended up being at 6.10.

For getting accessibility to both cuda and graphics capabilities I know I have to choose the virtual direct graphics acceleration (vDGA) alternative where the pci device is passed through the ESXi and where one of the physical GPU's on the card is dedicated to the VM. So I think I got this right already.

Re: Pr

Posted: 2020/11/21 19:22:09
by TrevorH
It's pointless using CentOS 6 when it's about to die.

Re: Pr

Posted: 2020/11/21 20:09:24
by frg
So I cannot use the OS even if it is not supported (which it by the way is for the rest of November)? To my knowledge I'm stuck to using an OS which is COMPATIBLE WITH MY HARDWARE. Anyhow, thanks for sharing the info.

Anyone else who can contribute to my issue?

Re: Pr

Posted: 2020/11/21 21:09:29
by TrevorH
So I cannot use the OS even if it is not supported (which it by the way is for the rest of November)?
And what will you do in 9 days time? It's pointless putting a lot of effort into a new install that's going to have a lifespan measured in may-fly lifetimes.

Re: Pr

Posted: 2020/11/21 21:25:54
by frg
Well, the benefit with using a virtual environment like VMware is that one can deploy almost any OS in it. And I don't expect the installed Centos 6.10 to crash on 1st December (or will it?). So my simple answer to your question is that I will simply continue using it (and I will not need to update it after installing the only program I'm establishing the OS to run).