CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

General support questions
Post Reply
tenkiforecast
Posts: 4
Joined: 2020/07/26 16:34:03

CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by tenkiforecast » 2020/07/27 14:30:14

Hello,

This is my first post here, so please forgive me for any formatting/protocol issues.

I've been having frequent difficulties with a new CentOS 7 machine set up for work. This machine was purchased a few years ago, and due to various reasons was recently formatted to be swapped in to replace an older machine. Coworkers put together an installation checklist for the CentOS machines to ensure configuration control because the programs we use on the machine are very finicky with the installation, user setup, and permissions. Following this checklist, I have installed a set of 4 computers prior to this machine according to the installation checklist without significant issue.

Hence my confusion and frustration when this specific computer, one with identical hardware to another functioning computer, began failing on boot at "Started GNOME Display Manager."

Consultation with coworkers early on resulted in a recommendation to re-install CentOS from scratch. I did so, and after following the checklist and bringing the computer configuration up to match the sister computer, the computer once again starting hanging on boot at "Started GNOME Display Manager." After searching to try and find guides as to why the boot was failing, I verified the installation of LightDM and disabled the Gnome Display Manager (systemctl enable lightdm.service; systemctl disable gdm.service). Upon restart, the boot again hung with the following displayed text:

"[ OK ] Removed slice User Slice of mfe.. Systems..utdown.... DKMS.n.on....ta.576/200385536 blocks"

I observed the same issue after re-enabling the Gnome Display Manager, and it no longer hung at "Started Gnome Display Manger"

I am able to access the computer only through rescue mode, and any attempt to switch to multi-user or graphical mode results in a hung computer. I am out of ideas. I do not know what is happening when an identical configuration on one computer results in a hung boot while the CentOS machines work perfectly well on 4 other computers. I have beaten my head against the wall trying to figure out what is happening. What is going on, and how can I get the computer to start normally?

Information about this computer configuration:
  • Graphics card is Quadro P600, uses NVIDIA drivers
  • I9 Processor on an X-299 Motherboard
  • CentOS 7 is up to date. After re-establishing a network connection in rescue mode, I was able to run 'yum update' to verify recent programs and files
  • Nouveau video driver is disabled.
  • Grub command line includes: quiet nomodeset modprobe.blacklist.nouveau nouveau.modeset=0 rd.driver.blacklist=nouveau plymouth.ignore-udev
  • A file to blacklist the nouveau driver exists at /etc/modprobe.d/blacklist.conf
  • Research indicated that the proper driver is kmod-nvidia, and configuration has the elrepo catalog installed.
  • kmod-nvidia version installed is kmod-nvidia-450.57
  • CentOS 7 Kernel version is 3.10.0-1127.13.1
Please let me know if I need to provide more details, and how I could provide said information.
Thank you.

User avatar
TrevorH
Forum Moderator
Posts: 29051
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by TrevorH » 2020/07/27 16:05:03

Did you check to see if the machine is actually up and running and if only the display is broken? Try to ssh to it and see if it works. If it does then you just have display driver problem. In any case, say more about what drivers you are using - many repos have a kmod-nvidia but only the ELRepo copy is recommended here.
CentOS 6 will die in November 2020 - migrate sooner rather than later!
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 is dead, do not use it.
Full time Geek, part time moderator. Use the FAQ Luke

tenkiforecast
Posts: 4
Joined: 2020/07/26 16:34:03

Re: CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by tenkiforecast » 2020/07/27 17:01:28

Trevor,

Thanks for the reply. I let the computer boot up, and remote access is working correctly.

The kmod-nvidia driver installed came from the ElRepo Repository. The ouptut from re-installing kmod-nvidia is below.
Reinstalling:
kmod-nvidia x86_64 450.57-1.el7_8.elrepo elrepo 19 M


The following output is from lspci -v

65:00.0 VGA compatible controller: NVIDIA Corporation GP107GL [Quadro P600] (rev a1) (prog-if 00 [VGA controller])
Subsystem: NVIDIA Corporation Device 11bd
Flags: fast devsel, IRQ 255, NUMA node 0
Memory at d7000000 (32-bit, non-prefetchable) [size-16M]
Memory at c0000000 (64-bit, prefetchable) [size-256M]
Memory at d0000000 (64-bit, prefetchable) [size-32M]
I/O ports at b000 [size-128]
Expansion ROM at d8000000 [disabled] [size-512K]
Capabilities: [60] Power Management version 3
Capabilities: [68] MSI: Enable- Count=1/1 Maskable- 64bit+
Capabilities: [78] Express Legacy Endpoint, MSI 00
Capabilities: [100] Virtual Channel
Capabilities: [250] Latency Tolerance Reporting
Capabilities: [128] Power Budgeting <?>
Capabilities: [420] Advanced Error Reporting
Capabilities: [600] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?>
Capabilities: [900] #19
Kernel modules: nouveau, nvidia_drm, nvidia


For reference, the below output is from lshw -C display

*-display UNCLAIMED
description: VGA compatible controller
product: GP107GL [Quadro P600]
vendor: NVIDIA Corporation
physical id: 0
bus info: pci@0000:65:00.0
version: a1
width: 64 bits
clock: 33MHz
capabilities: pm msi pciexpress vga_controller cap_list
configuration: latency=0
resources: memory:d7000000-d7ffffff memory:c0000000-cfffffff memory:d0000000-d1ffffff ioport:b000(size=128) memory:d8000000-d807ffff


It looks like the computer isn't registering any display driver.

User avatar
TrevorH
Forum Moderator
Posts: 29051
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by TrevorH » 2020/07/27 17:33:05

Thanks for the reply. I let the computer boot up, and remote access is working correctly.
So that tends to confirm that this is just a display issue and the machine is otherwise working ok. You can at least use ssh to read some logs and I'd start with /var/log/messages, /var/log/secure and /var/log/Xorg.0.log (might be others similarly named too, check dates).

Can you confirm that you also have the nvidia-x11-drv package installed? If not then install it.

Also can you interrupt the boot process at the grub menu, edit the current kernel entry and append a space followed by the digit 3 then use ctrl-x to continue the boot. That should get you to a command line login. Try running systemctl isolate graphical.target from there and see if it does anything different. You could also try that again but this time run setenforce 0 before the isolate and see if that changes things.
CentOS 6 will die in November 2020 - migrate sooner rather than later!
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 is dead, do not use it.
Full time Geek, part time moderator. Use the FAQ Luke

tenkiforecast
Posts: 4
Joined: 2020/07/26 16:34:03

Re: CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by tenkiforecast » 2020/07/27 18:24:44

The attempt to access the normal display using the 'init 3' did not work. I've only been able to access the attached display using maintenance mode, otherwise the display breaks. The current message is:
Started Session c1 of user mfe... System..netd.n.... DKMS.n.on....ta.576/200385536 blocks

I was able to ssh in and look at some of the log files, and one was full of error messages from the x11vnc service set up on the computer. I disabled the x11vnc service to reduce the size of the message, and that somehow allowed the login screen to show up when I told the computer to boot into multi-user mode. However, graphical mode still failed.

I was able to verify nvidia-x11-drv was installed and up to date.

Upon restart, I noticed something in the messages log file:

Jul 27 18:07:18 himalia systemd: Started GNOME Display Manager.
Jul 27 18:07:18 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Jul 27 18:07:18 himalia gdm: GdmDisplay: display lasted 0.165584 seconds
Jul 27 18:07:18 himalia systemd: Received SIGRTMIN+21 from PID 399 (plymouthd).
Jul 27 18:07:18 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Jul 27 18:07:18 himalia gdm: GdmDisplay: display lasted 0.055270 seconds
Jul 27 18:07:18 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Jul 27 18:07:18 himalia gdm: GdmDisplay: display lasted 0.054775 seconds
Jul 27 18:07:18 himalia systemd: Removed slice User Slice of mfe.
Jul 27 18:07:18 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Jul 27 18:07:18 himalia gdm: GdmDisplay: display lasted 0.055554 seconds
Jul 27 18:07:18 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Jul 27 18:07:18 himalia gdm: GdmDisplay: display lasted 0.052107 seconds
Jul 27 18:07:19 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Jul 27 18:07:19 himalia gdm: GdmDisplay: display lasted 0.047839 seconds
Jul 27 18:07:19 himalia gdm: GdmLocalDisplayFactory: maximum number of X display failures reached: check X server log for errors
Jul 27 18:07:19 himalia systemd: Started Postfix Mail Transport Agent.


The abnormal part of the Xorg.0.log is listed below:
[ 23.240] (EE) NVIDIA: Failed to initialize the NVIDIA kernel module. Please see the
[ 23.240] (EE) NVIDIA: system's kernel log for additional error messages and
[ 23.240] (EE) NVIDIA: consult the NVIDIA README for details.
[ 23.240] (EE) No devices detected.
[ 23.240] (EE)
Fatal server error:
[ 23.240] (EE) no screens found(EE)
[ 23.240] (EE)
Please consult the The X.Org Foundation support
at http://wiki.x.org
for help.
[ 23.240] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
[ 23.240] (EE)
[ 23.243] (EE) Server terminated with error (1). Closing log file.


I don't think the kernel knows how to access the device drivers.

User avatar
TrevorH
Forum Moderator
Posts: 29051
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by TrevorH » 2020/07/28 10:12:00

Jul 27 18:07:19 himalia kernel: Request for unknown module key 'The ELRepo Project (http://elrepo.org): ELRepo.org Secure Boot Key: f365ad3481a7b20e3427b61b2a26635b83fe427b' err -11
Is secure boot enabled? If it is then you need to import the ELRepo signing key into your BIOS setup. They have a wiki page on how...
CentOS 6 will die in November 2020 - migrate sooner rather than later!
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 is dead, do not use it.
Full time Geek, part time moderator. Use the FAQ Luke

tenkiforecast
Posts: 4
Joined: 2020/07/26 16:34:03

Re: CentOS 7 Boot Hangs at "Started GNOME Display Manager" or "Removed slice User Slice of mfe"

Post by tenkiforecast » 2020/07/28 23:21:58

Followed the steps. That fixed it, I was able to get to the GUI.

Thanks for the help!

Post Reply

Return to “CentOS 7 - General Support”