Skip to content

Possible RN update for RKE2 and K3s #127

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mikemckiernan merged 4 commits into from
Nov 13, 2024
Merged

Possible RN update for RKE2 and K3s #127

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1edcbb7
Possible RN update for RKE2 and K3s
mikemckiernan Nov 12, 2024
8aae1a7
Use v1.17.1 for RKE2 and K3s
mikemckiernan Nov 12, 2024
9c8f68b
Use RST markup instead of Markdown
mikemckiernan Nov 12, 2024
3cc9c18
Show full command
mikemckiernan Nov 13, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions gpu-operator/getting-started.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -513,6 +513,8 @@ For Rancher Kubernetes Engine 2 (RKE2), refer to
`Deploy NVIDIA Operator <https://docs.rke2.io/advanced#deploy-nvidia-operator>`__
in the RKE2 documentation.

Refer to the :ref:`v24.9.0-known-limitations`.

MicroK8s
========

Expand Down
13 changes: 13 additions & 0 deletions gpu-operator/release-notes.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,8 @@ New Features

For more information regarding NVIDIA security vulnerability remediation policies, refer to https://www.nvidia.com/en-us/security/psirt-policies/.

For Rancher RKE2 and K3s, refer to the :ref:`v24.9.0-known-limitations`.

* Added support for the NVIDIA Data Center GPU Driver version 550.127.05.
Refer to the :ref:`GPU Operator Component Matrix`
on the platform support page.
Expand Down Expand Up @@ -127,6 +129,17 @@ Fixed Issues
Previously, the toolkit could configure the ``runc`` handler with the ``nvidia`` runtime handler even if ``runc`` was not the default runtime and cause CRI-O to crash.
In this release, the toolkit determines the default runtime by running ``crio status config`` and configures that runtime with the ``nvidia`` runtime handler.

.. _v24.9.0-known-limitations:

Known Limitations
-----------------

* On Rancher RKE2 and K3s, NVIDIA Container Toolkit v1.17.0 fails to start.
The toolkit attempts to use the ``containerd`` command to determine the container runtime configuration on the host.
Copy link
Contributor

@tariq1890 tariq1890 Nov 12, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To be specific , it attempts to run the following commands depending on the container runtime used for the cluster

i. containerd config dump
ii. crio status config

On these platforms, the ``containerd`` executable is not on the PATH and results in an error.

NVIDIA recommends installing v1.17.1 of the toolkit when you install or upgrade the Operator.
You can specify the ``--set toolkit.version=v1.17.1-ubuntu20.04`` or ``v1.17.1-ubi8`` argument to Helm.

.. _v24.6.2:

Expand Down