Skip to main content

NVIDIA drivers update guide

An instruction on updating NVIDIA drivers and solving any possible problems that may arise in the process.

Nazar avatar
Written by Nazar
Updated over 4 months ago

nvidia-driver-update default installation version

We update the NVIDIA drivers in our repository regularly.

After conducting our tests, we decided to choose the latest stable version for installation by default if no arguments are given to nvidia-driver-update. This version supports all the latest NVIDIA RTX cards and also includes all performance optimizations.

Hiveon OS update drivers

Update or not update?

Before we start, let’s answer some typical questions:

Is it necessary to update?

No! If everything works for you - don’t touch.

Why then update?

New drivers include support for new cards such as 3060, 3060 Ti, 3070, 3070 Ti, 3080, 3080 Ti and 3090 series. Sometimes special optimization in some miners has a positive effect on the hashrate, especially with the LHR GPUs.

Still not changing your mind?

Then we will start.

Updating procedure

We strongly recommend using Hive Shell for proper driver update procedure. Of course, if you can use a local console - use it or if your rigs are in the same LAN with your PC/laptop then you can use for example SSH-clients such as standalone app PuTTY or working in browser ShellInABox instead of Hive Shell.

To update the drivers, use the nvidia-driver-update tool and perform two steps described below.

Sep 1: Select the necessary option from the table below.

Option

Action

Default option for installation

download and install latest driver version

URL

download and install driver from URL (http/https/ftp)

123

download and install latest driver from series 123.*

123.45.06

download and install driver version 123.45.06

-s | --stable

download and install STABLE driver version

-f | --force

force install to bypass some DKMS build errors and etc

-l | --list

list available driver versions (compatible with current kernel)

-e | --ext

use external drivers repository

-c | --cuda

sync symlinks to CUDA RT libraries with installed driver

-n | --nvs

reinstall nvidia-settings only

-r | --remove

remove downloaded driver packages except currently installed

-h | --help

display help

--repo[=URL]

use custom repository (specified by URL) (HTTP(S)/FTP)

Step 2: Execute the necessary option the following format: nvidia-driver-update <option(s)> bee@hiveos.farm Command examples: nvidia-driver-update --force 455 nvidia-driver-update --force --ext --list nvidia-driver-update --list --repo=ftp://192.168.1.1/drivers/

The driver update procedure takes 5-10 min if you’re using an SSD/HDD and 20-30 min if you’re using a slow drive such as a USB 2.0 stick.

If the update procedure will finish successfully, you will see the following message:

nvidia dr image

Now you can reboot your rig. After reboot, check driver version and miner functionality.

Update/downgrade to desired version

You can update the driver to the desired version. With Hiveon OS it’s quite simple to update NVIDIA drivers. list the possible upgrades, type in the nvidia-driver-update --list command, and then type the Hiveon OS will update the NVIDIA drivers.

Troubleshooting

Cannot update the driver properly

  1. Reboot rig>Stop mining>Run nvidia-driver-update

  2. Enable maintenance mode without loading drivers first than run nvidia-driver-update

  3. Check GPUs power cables, risers, etc. and repeat steps 1-2

  4. Try to update image via hive-replace if steps 1-2 did not help

Not enough disk space

  • use disk-expand tool to enlarge disk to use all space

  • remove unused drivers at /hive-driver-pack (to do this, send the rm /hive-drivers-pack/NV* command to your worker - it will delete all the available NVIDIA driver archives)

  • uninstall miners by hpkg tool

Installation fail when unpacking driver

By default, the TMP directory is stored in RAM, so you need to enable logs to have some space: logs-on reboot

Nothing helps?

Troubleshooting steps to try:

  • contact the support at bee@hiveon.com

  • call community support on Telegram channel

  • reflash the image

  • remote update image via the hive-replace tool

Did this answer your question?