NVIDIA / NVIDIA/nvcf

Replace self-managed SMB shared storage with Kubernetes-native backends

Open
#70 0 comments 0 reactions 1 assignee View on GitHub

@mikeyrcamp is already working on this.

Since Aug 28, 2026.

roadmap
Dominant language
Go
Stars
218
Forks
72
Avg merge
1d 12h
Merged PRs (30d)
427

Description

Description

Replace the in-cluster SMB server and SMB CSI-backed volumes used by self-managed NVCF with Kubernetes-native secret and shared-data backends. Preserve existing workload file paths and retain SMB as an upgrade-safe fallback while deployments migrate.

Sensitive workload files should use projected Kubernetes Secrets. Bulk and shared data should use persistent volumes provisioned from an administrator-provided ReadWriteMany (RWX) StorageClass.

Definition of Done

  • Sensitive files under /var/secrets and /var/run/nvcf/info can be delivered through projected Kubernetes Secrets.
  • /var/task, model cache data, and other bulk shared data can use persistent volumes from an administrator-provided RWX StorageClass.
  • Operators can configure the secret and shared-data backends independently.
  • Existing workload file paths remain compatible.
  • Deployments that omit the new settings retain the existing SMB behavior until migration is complete.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.