Repository navigation
Conversation
gpu-node/gpu-node-check.sh checks a GPU node (driver, firmware, container runtime, CPU governor, disk, k3s version) and gpu-node/proxmox-gpu-lxc.sh checks a Proxmox host and its LXC container. Both only report unless --apply is given. The README walks through dedicated machines, VMs and Proxmox LXC containers for NVIDIA and AMD, one vendor per panel as setup-gpu-nodes.sh sets it up.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Adds two check scripts for GPU nodes, next to
setup-gpu-nodes.sh(#700):gpu-node/gpu-node-check.shruns on the GPU node. It checks the driver, firmware, container runtime (NVIDIA toolkit), CPU governor, free disk and the k3s version against the panel's, and says when an AMD GPU is integrated so it goes in the integrated list.gpu-node/proxmox-gpu-lxc.sh CTIDruns on a Proxmox host. It checks the host driver and kernel modules k3s needs, and the container's settings (privileged, nesting, device entries for/dev/dri+/dev/kfdor the NVIDIA devices, cgroup and mounts).gpu-node/README.mdwalks through a dedicated machine or VM and a Proxmox LXC container, for NVIDIA and AMD, one vendor per panel assetup-gpu-nodes.shsets it up.Both scripts only report unless
--applyis passed, exit 1 while something still fails, and never touch GPU power or clock settings.Testing
tests/scripts/proxmox_gpu_lxc_test.sh(9 tests, added to the Panel scripts workflow),gpu_nodes_test.shand shellcheck pass.--applyof both scripts is not yet tested on real hardware (a Proxmox host or a fresh node). The config linesproxmox-gpu-lxc.sh --applywrites are the ones our AMD container runs with, but the write path itself has only run in the test. Please treat--applyas a first version.