kernel-lt from ELRepo vs. GRUB: define default boot kernel


After upgrading my workstation to CentOS 7.5 (1804), I had to upgrade my
kernel from vanilla to kernel-lt from ELRepo. My NVidia GeForce 210
would only work with the driver provided by NVidia, which in turn
required a more recent kernel than 3.0.10. Anyway.

Right now here's all the kernels that I have on my workstation:

[root@alphamule:~] # rpm -qa | grep -i kernel

And here's what I'm currently running:

[root@alphamule:~] # uname -r

I have a Logitech Illuminated USB keyboard, and one of the quirks of
this piece of hardware is that it's unresponsive on boot time. Meaning
when GRUB displays the kernel menu, the arrow keys (or for the matter
any keys) won't work. Which means the only choice I have left is either
boot the default kernel in the list... or go to another office, grab a
standard USB keyboard and plug it in temporarily just to choose the
kernel to boot.

So right now I have two kernels on my machine, the 4.4.129 and the
4.4.131. How do I configure GRUB so that on the next reboot, it defaults
to the 4.4.131 kernel? I knew how to do this with LILO under Slackware,
but GRUB is a very different beast.




Re: kernel-lt from ELRepo vs. GRUB: define default boot

By Yan Li at 05/16/2018 - 11:48

On 05/16/2018 03:10 AM, Nicolas Kovacs wrote:
I'm not sure why you need kernel-lt. NVIDIA's proprietary binary drivers
always support the latest RHEL. RHEL workstations for 3D rendering is
one major reason why NVIDIA is providing drivers for the Linux platform.

For instance, this is NVIDIA driver 390.30 on my Dell Precision 7510:

% uname -a
Linux freesia 3.10.0-862.2.3.el7.x86_64 #1 SMP Wed May 9 18:05:47 UTC
2018 x86_64 x86_64 x86_64 GNU/Linux

% lsmod | grep nvidia
nvidia_drm 39676 1
nvidia_modeset 1087441 10 nvidia_drm
nvidia 14328472 878 nvidia_modeset
drm_kms_helper 176920 1 nvidia_drm
drm 397988 4 drm_kms_helper,nvidia_drm
ipmi_msghandler 46608 2 ipmi_devintf,nvidia
i2c_core 63151 6

% dmesg | grep nvidia
[ 10.515333] nvidia: loading out-of-tree module taints kernel.
[ 10.515339] nvidia: module license 'NVIDIA' taints kernel.
[ 10.572111] nvidia: module verification failed: signature and/or
required key missing - tainting kernel
[ 10.630900] nvidia-nvlink: Nvlink Core is being initialized, major
device number 239
[ 10.665785] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver
for UNIX platforms 390.30 Wed Jan 31 21:32:48 PST 2018
[ 10.672168] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
[ 10.672171] [drm] Initialized nvidia-drm 0.0.0 20160202 for
0000:01:00.0 on minor 0
[ 101.124034] nvidia 0000:01:00.0: irq 137 for MSI/MSI-X
[ 104.025728] nvidia-modeset: Allocated GPU:0
(GPU-29f026a6-c342-5e3c-c27a-5c3bf72bcce5) @ PCI:0000:01:00.0

What error do you see when you are trying to install the NVIDIA's driver
with a vanilla CentOS kernel?

Re: kernel-lt from ELRepo vs. GRUB: define default boot

By Niki Kovacs at 05/16/2018 - 13:41

Le 16/05/2018 à 17:48, Yan Li a écrit :
My GeForce 210 is slowly but steadily becoming a "legacy" card. Just try
to install the appropriate 340xx driver from manually, you
will notice that the installation fails with the stock CentOS kernel,
but it will succeed with the kernel-lt. As far as I'm concerned, I'm a
pragmatic guy, so whatever solves the problem is good, and I'm moving on.



Re: kernel-lt from ELRepo vs. GRUB: define default boot

By Anand Buddhdev at 05/16/2018 - 10:17

Bonjour Nicolas!

The easiest way is to use the "grubby" tool to set your default kernel.
Look at the man page of grubby for details.


Re: kernel-lt from ELRepo vs. GRUB: define default boot

By Gianluca Cecchi at 05/16/2018 - 10:37

This is helpful too and still valid for CentOS 7.5:
<a href="" title=""></a>