ASUS M5A97-R2.0 Motherboard
AMD 6300 8.5 GHz 6 core Processor
32 GB of DDR3 1600 RAM (CAS 9-9-9-24)
1 TB HD
250GB SSD
SATA Optical Drive
ASUS NVidia Gefore 640GT Video card w/2GB DDR3 RAM



My buddy and I have IDENTICAL machines that are kept up-to-date and we are getting the SAME problem.

For those who may wonder what the kernel version is if you run uname -a the kernel is: 3.10.0-327.36.3.el7.x86_64 which is the latest and greatest.

We just ran yum update and there is a new NVIDIA driver -- actually drivers plural -- listed:

---> Package kmod-nvidia.x86_64 0:367.57-1.el7.elrepo will be updated
---> Package kmod-nvidia.x86_64 0:367.57-2.el7.elrepo will be an update

It then goes into listing and endless number of" --> Processing Dependency:..." before it hurls on your shoes. Anyone else having this problem in updating their NVIDIA driver, and if you did were you able to solve the problem, and if so HOW??

For the time being we are sitting tight until the next kernel is released and hope the problem will then be solved. We don't want to toast two production machines which would be a disaster.

please post the output of
it sounds like you have incompatible repos installed and have not set up "yum-priorites" plugin

the kmod drivers ARE!!!!! very dependant on having the MATCHING!!!!! kernel
if the kernel dose not match the kmod it can not work


if i am using the kmod ( been using the nvidia.run for YEARS)
i add a block to updating the kernel ( an exception in the yum config to block it )
then manually update the kmod
this WILL grab the matching kernel as a dependent

Likewise we have been using kmod for years too and this is a first. We know it must be the kernel, but for some odd reason when we ran yum update it listed not one but TWO drivers. We tried to manually install one -- say kmod-nvidia.x86_64 0:367.57-1.el7.elrepo -- then run it but it comes back and it still insists to update BOTH drivers. If we try updating say the other driver -- kmod-nvidia.x86_64 0:367.57-2.el7.elrepo -- well it seems to be dependent on finding kmod-nvidia.x86_64 0:367.57-1.el7.elrepo.As I noted the current version of the kernel that we are running is 3.10.0-327.36.3.el7.x86_64. What version are you running? The the latest version of the Nvidia driver we are running is 367.57, so the two new drivers 367.57-1 and 367.57-2 seem to be bug fixes. At first I thought it might be just *my* workstation that was having the problem, so I called my buddy who has an exact duplicate, but is on a different set of mirrors, yet when *he* ran yum update he had the *exact same problem*, so the problem seems to be system wide. Since these are production machines (his statistics, mine microbiological research though I've been retired for years because I have Parkinson's, I still follow epidemiological research that I pass on to colleagues) we don't want to blow them up, so we are sitting tight until the next kernel release. I'd be curious to know what version of the kernel you are running and what version of the Nvidia driver you have. Thanks

the software i need to use is on SL6 the USGS/JPL code dose not run well on 7 yet

as to the nvidia driver i gave up using the kmods a long time ago
too many problems with a new kernel and NO!!!! matching kmod for a few days

i asked for the repo list

ELrepo and the cent base repos should be fine BUT
if you are mixing in say nux and/or linuxtech

do you have Priorities installed and configured
( 7 should not need this and much as 5 and 6 do )
https://wiki.centos.org/PackageManag...Yum/Priorities

OK here is what I have been up to since last I posted. Before I go on THANK YOU ALL FOR YOUR INPUT!!! OK ONWARD!!!

Using the suggestions found herein I installed yum-plugin-priorities for CentOS-7 [Latest version is:yum-plugin-priorities.noarch 1.1.31-34.el7 @base]

Result: NO CHANGE

Checked to see what graphics card I had (I forgot what it was): GeForce GT 640
Checked to see what nVidia's latest driver for that card is: Latest nVidia Driver is 367.57,
Checked to see what Driver Version is CURRENTLY installed: 367.57 [installed via kmod-nvidia yum update command]
Checked to see what what drivers kmod-nvidia wants to install via yum update: 367.57-1.el7. & 367.57-2.el7.

These are in fact *it seems* to be bug fixes [367.57-1 and 367.57-2] for the current nVida driver: 367.57. As suggested the problem is *probably* an incompatibility between the current kernel 3.10.0-327.36.3.el7.x86_64 which was released one to two weeks ago and the new kmod-nvidia drivers released just a few days ago. It probably does not help that el-repo released TWO nvidia drivers at the exact same time one being dependent on the other. With LUCK the problem will be cleared up when CentOS releases its next kernel update and el-repo releases a NEW kmod-nvidia driver to match the new kernel.

Unless someone can figure out a work around when you go to do your weekly CentOS 7 Updates (I do mine either Fri. or Sat.) the ONLY way you will be able to get those updates installed is to run yum update --skip-broken command. It will still go through the long song and dance of " --> Processing Dependency:..." but at the end of it will skip trying to install the kmod nvida drivers and ask you if you want to update the other packages (yes of course).

In all the years I've been using kmod-nvidia this is the first time it has really been scrambled. I've tried as many things as I could think of but got no where. If anyone has the latest kernel installed [3.10.0-327.36.3.el7.x86_64] and if have run the yum update command I'd be curious -- a professional hazard of being a cat -- to see if you encounter the same problem both I and my buddy encountered, and if you did, were you able to solve the problem, and if so HOW!!!

Thanks

I have exactly the same problem as stated above, while updating package kmod-nvidia.x86_64 0:367.57-1.el7.elrepo

During updating, I get lots of lines like:
....
--> Processing Dependency: kernel(drm_vblank_init) = 0x9c4a7a63 for package: kmod-nvidia-367.57-2.el7.elrepo.x86_64
--> Processing Dependency: kernel(drm_universal_plane_init) = 0x3707f7bd for package: kmod-nvidia-367.57-2.el7.elrepo.x86_64
--> Processing Dependency: kernel(drm_plane_index) = 0x370ee46b for package: kmod-nvidia-367.57-2.el7.elrepo.x86_64
--> Processing Dependency: kernel(drm_plane_cleanup) = 0x55db6aac for package: kmod-nvidia-367.57-2.el7.elrepo.x86_64
......

I am running a clean, up to date CentOS 7.2.1511. Never had any problems with this kmod-nvidia package before.
I surely would appreciate a solution for this.
For now, I just don't and can't update the package.

I have installed (always using yum -update):
kmod-nvidia 367.57-1.el7.elrepo System Environment/Kernel nvidia kernel module(s)
nvidia-detect 367.44-1.el7.elrepo Applications/System NVIDIA graphics card detection utility
nvidia-x11-drv 367.57-1.el7.elrepo User Interface/X Hardware Support NVIDIA OpenGL X11 display driver files
xorg-x11-drv-nouveau 1.0.11-2.el7 User Interface/X Hardware Support Xorg X11 nouveau video driver for NVIDIA graphics chipsets
yum-plugin-nvidia 1.0.2-1.el7.elrepo Development/Tools Yum plugin to prevent update of NVIDIA drivers on unsupported hardware

Nvidia hardware:
Nvidia GeForce GT 720, running Driver version 367.57

Just as a curiosity, has anyone tried first UN-INSTALLING the current version of the Nvida driver, then RE-INSTALLING to see if the problem goes away??? If you did were your results the same once you tried to re-install the drivers?

On a side note this may be the last kernel update and Nvidia driver set of up dates. RHEL 7.3 in theory was released on 3 November 2016 with Kernel 3.10.0-514, thus the RHEL version will be RHEL 7.3.1611. That means that the CentOS 7.3 release can be too far behind, and it too should be CentOS 7.3.1611. With luck, my server will simply roll over to 7.3.1611 during the next kernel update, along which will come a NEW kmod-nvida driver, and with luck, that will solve the problem.