[elrepo] Messages during install of kmod-nvidia
Phil Perry
phil at elrepo.org
Mon Nov 20 14:35:19 EST 2017
On 20/11/17 12:15, Kretschmer, Jens wrote:
> Hi,
> we use RHEL 7.4 and we are getting the following messages during
> installation of kmod-nvidia on several machines:
> <redacted>[root]> yum reinstall kmod-nvidia nvidia-detect nvidia-x11-drv
> Loaded plugins: langpacks, nvidia, product-id, search-disabled-repos,
> subscription-manager
> This system is not registered with an entitlement server. You can use
> subscription-manager to register.
> Resolving Dependencies
> --> Running transaction check
> ---> Package kmod-nvidia.x86_64 0:384.98-1.el7_4.elrepo will be reinstalled
> ---> Package nvidia-detect.x86_64 0:384.98-1.el7.elrepo will be reinstalled
> ---> Package nvidia-x11-drv.x86_64 0:384.98-2.el7.elrepo will be reinstalled
> --> Finished Dependency Resolution
> Dependencies Resolved
> =====================================================================================================================================
> Package Arch
> Version Repository Size
> =====================================================================================================================================
> Reinstalling:
> kmod-nvidia x86_64
> 384.98-1.el7_4.elrepo LocalELREPO 12 M
> nvidia-detect x86_64
> 384.98-1.el7.elrepo LocalELREPO 23 k
> nvidia-x11-drv x86_64
> 384.98-2.el7.elrepo LocalELREPO 43 M
> Transaction Summary
> =====================================================================================================================================
> Reinstall 3 Packages
> Total download size: 55 M
> Installed size: 223 M
> Is this ok [y/d/N]: y
> Downloading packages:
> -------------------------------------------------------------------------------------------------------------------------------------
> Total
> 87 MB/s | 55 MB 00:00:00
> Running transaction check
> Running transaction test
> Transaction test succeeded
> Running transaction
> Installing :
> kmod-nvidia-384.98-1.el7_4.elrepo.x86_64
> 1/3
> Broadcast message from systemd-journald@<redacted> (Wed 2017-11-15
> 10:11:28 CET):
> dracut[11094]: /lib/modules/3.10.0-693.el7.x86_64//modules.dep is
> missing. Did you run depmod?
> Message from syslogd@<redacted> at Nov 15 10:11:28 ...
> dracut:/lib/modules/3.10.0-693.el7.x86_64//modules.dep is missing. Did
> you run depmod?
> Working. This may take some time ...
> /lib/modules/3.10.0-693.el7.x86_64//modules.dep is missing. Did you run
> depmod?
> /sbin/weak-modules: line 116: /boot/initramfs-3.10.0-693.el7.x86_64.tmp:
> No such file or directory
> /sbin/weak-modules: line 132: /boot/initramfs-3.10.0-693.el7.x86_64.tmp:
> No such file or directory
> /sbin/weak-modules: line 137: /boot/initramfs-3.10.0-693.el7.x86_64.tmp:
> No such file or directory
> Unable to decompress /boot/initramfs-3.10.0-693.el7.x86_64.tmp: Unknown
> format
> /sbin/weak-modules: line 175:
> /tmp/weak-modules.Lvy1aX/new_initramfs.img: No such file or directory
> rm: cannot remove '/tmp/weak-modules.Lvy1aX/new_initramfs.img': No such
> file or directory
> mv: cannot stat '/boot/initramfs-3.10.0-693.el7.x86_64.tmp': No such
> file or directory
> Done.
> Installing :
> nvidia-x11-drv-384.98-2.el7.elrepo.x86_64
> 2/3
> Installing :
> nvidia-detect-384.98-1.el7.elrepo.x86_64
> 3/3
> Verifying :
> kmod-nvidia-384.98-1.el7_4.elrepo.x86_64
> 1/3
> Verifying :
> nvidia-x11-drv-384.98-2.el7.elrepo.x86_64
> 2/3
> Verifying :
> nvidia-detect-384.98-1.el7.elrepo.x86_64
> 3/3
> Installed:
> kmod-nvidia.x86_64 0:384.98-1.el7_4.elrepo nvidia-detect.x86_64
> 0:384.98-1.el7.elrepo nvidia-x11-drv.x86_64 0:384.98-2.el7.elrepo
> Complete!
> I know that the kmod packages are built against a certain kernel
> version, installed for that kernel version and weak linked against all
> compatible kernel versions. But I think that those messages should not
> appear nevertheless.
> The kernel 3.10.0-693.5.2.el7.x86_64 is the only one installed.
> The kernel module is loaded and apart from the messages during the
> installation there are no visible errors.
> <redacted>[root]> lsmod |grep nvidia
> nvidia_drm 44108 1
> nvidia_modeset 841676 5 nvidia_drm
> nvidia 13117469 288 nvidia_modeset
> drm_kms_helper 159169 1 nvidia_drm
> drm 370825 4 drm_kms_helper,nvidia_drm
> i2c_core 40756 4 drm,i2c_i801,drm_kms_helper,nvidia
> <redacted>[root]> modinfo nvidia | grep version
> version: 384.98
> rhelversion: 7.4
> srcversion: C169B0DD461631A8B5EF92E
> vermagic: 3.10.0-693.el7.x86_64 SMP mod_unload modversions
> Someone on the CentOS Mailing List had the same problem, but did not
> post a solution. [1]
> Br,
> Jens
> [1] https://lists.centos.org/pipermail/centos/2017-September/166520.html
>
>
Hi,
Please can you show the output from:
rpm -qa kernel\*
find /lib/modules -name nvidia*\.ko
What happens if you uninstall (yum erase kmod-nvidia) and yum install
kmod-nvidia again? Do you see the same error each time? Is it reproducible?
If it's reproducible I suspect it's probably due to weak-modules trying
to run dracut on the kernel the module was compiled against, which isn't
installed on your system.
More information about the elrepo
mailing list