How could I check the loaded drivers and their versions? - centos

The operating system is CentOS 6.5, we could use the following commands for hardware information:
cat /proc/cpuinfo
cat /proc/meminfo
lspci
lsusb
And, how could I get the drivers of these hardware and their versions? Does it need root access? Thank you!
UPDATED

Related

Error when running mongo image - docker-entrypoint.sh: line 381

After installing Ubuntu v20 and then installing docker:
$ docker network create test-network
$ docker pull mongo
$ docker run --network test-network --name mongodb \
-e MONGO_INITDB_ROOT_USERNAME=admin \
-e MONGO_INITDB_ROOT_PASSWORD=pawwrord \
mongo
I got an error like this:
/usr/local/bin/docker-entrypoint.sh: line 381: 25 Illegal instruction (core dumped) "${mongodHackedArgs[#]}" --fork
Do you know what the problem is? I just need some guidance to investigate the problem.
UPDATE
I don't have any problem with other docker hub images.
Specifically only when I want to run mongo, I got this error.
MongoDB 5.0 requires a Sandy Bridge or newer CPU. Get a newer processor or use an older version of MongoDB.
https://jira.mongodb.org/browse/SERVER-54407
I have exactly same problem with yours, i try to run on my local computer work but not in VM.
thanks to this I simply changed mongo's version to 4.4 from latest which is 5.0 and running well. Maybe need newer CPU
It's a shortcoming in the available instructions of your CPU. Mongo 5.0 requires CPU instructions only available in Sandy Bridge era or more recent.
I had this problem using a Proxmox VM, which defaults to a simpler virtual CPU. Simply changing the virtual CPU to type "host" fixed the problem, by allowing the VM CPU to include all the instructions of the host CPU.
If you are running directly on metal, you'll need a newer CPU. Good luck!

Nmap multiple os information

Pardon in advance if this is a very noob question..
So I am learning to use nmap command and currently, I am doing a scan to check for the OS details of an image.
The command I have used is nmap -O -sV <ip address> and I am slightly confused by the OS details it outputted as I am trying to derive the exact OS version, to check if there are any vulnerability in the version.
It is as follows:
Running: Linux 3.X|4.X
OS CPE: cpe:/o:linux:linux_kernel:3 cpe:/o:linux:linux_kernel:4
OS details: Linux 3.2 - 4.9
While I am unable to access/ log into the said image, based off the above result, can I presume that there are 2 different Linux OS versions, in this case, one is running on Linux 3.2 and the other is on Linux 4.9?
Is it possible that an image can have multiple OS (as seen above, assuming what I mentioned above is correct)?
Many thanks in advance.
Nmap does not always know the exact OS on the host you are scanning, so it tries to geuss it.

mount: unknown filesystem type 'vmhgsf'

I'm trying to mount my windows shared folder in centOS using command:
~mount -t vmhgfs .host:/shared-folder /var/www/html/
Unfortunatelly I get :
~monut: unknown filesystem type 'vmhgfs'
error. I tried to use:
~/usr/bin/vmhgfs-fuse /mnt
but mountpoint is not empty...
Is there any way to mount this folder on VMware player?
Cyb
Try this:
vmhgfs-fuse .host:/shared-folder /var/www/html/
you might need to use sudo on this
Working from a MacBook Pro running Big Sur and using VMware to host the virtual machine with CentOS 7 operating system. Had issues with loading in shared folders after VMware tools were installed. What worked for me is to use this exact command:
sudo /usr/bin/vmhgfs-fuse .host:/ /mnt/hgfs -o subtype=vmhgfs-fuse,allow_other
Hope this saves others the trouble of tracking down this solution.
The below is working perfectly fine for me. Might be useful for someone.
Already I have mapped the required folders in "SharedFolder" settings. But it was not showing up.
Additionally running this command shares the windows directories.
sudo /usr/bin/vmhgfs-fuse .host:/ /home/user/win -o subtype=vmhgfs-fuse,allow_other

How to disable Nouveau kernel driver on CentOS

I am installing CUDA on my GPU machines. While at it, I need to disable Nouveau Kernel Driver.
I did find a solution here: https://askubuntu.com/questions/841876/how-to-disable-nouveau-kernel-driver
But update-initramfs is not found on CentOS.
I am looking for an equivalent of sudo update-initramfs -u in CentOS
If your goal is to install the latest nvidia driver to run with cuda. The best way to disable nouveau is indeed to rebuild the initramfs, as written by Gediz. Since, as spotted here (https://forums.centos.org/viewtopic.php?t=68800), it is only a 5 steps process I think it deserves to be right here :
grubby ––update-kernel=ALL ––args="rd.driver.blacklist=nouveau nouveau.modeset=0"
mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r).img.bak
echo "blacklist nouveau" > /etc/modprobe.d/nouveau-blacklist.conf
dracut /boot/initramfs-$(uname -r).img $(uname -r)
reboot
I believe Nouveau driver can be easily unloaded using modprobe:
modprobe -r nouveau
Also there is an option -b which blacklists it.
-b, --use-blacklist Apply blacklist to resolved alias.
In the web address you attached there is :
option nomodeset
I guess it is a kernel option not to load !ANY display drivers.You wont always need to update initramfs, only if module is included in initramfs you need to update it.
You can check it using one of initramfs-tools by :
lsinitramfs /boot/initrd.img-4.9.0-5-amd64 |less
However if you need to update or rebuild initramfs there is a way shown in CentOs Wiki :
https://wiki.centos.org/TipsAndTricks/CreateNewInitrd

Remove USB Host Controller driver from Linux

Can you please let me know how do i disable/remove the USB Host Controller driver from the kernel. Is there a way I can unhook the usb driver module from the kernel.
Linux 2.6.24.7-r1116 / Fedora
sudo lsmod | grep ohci
sudo lsmod | grep ehci
If these commands show nothing - then your host controller drivers are compilled into the kernel. Otherwise rmmod all modules, that my commands show.
This is a programmer question. I've used this techinque while debugging hotplug of my USB device in my device driver.