Add shared-storage validation and cold-start metrics
Open
@shobham-nv is already working on this.
Since Jul 15, 2026.
- Dominant language
- Go
- Stars
- 218
- Forks
- 72
- Avg merge
- 1d 12h
- Merged PRs (30d)
- 427
Description
Description
Add or reuse nvcf_shared_storage_cold_start_seconds to measure shared-storage cold-start latency by backend mode. Include the bounded labels secret_backend, shared_data_backend, and smb_server_expected, and emit one sample when a function deployment first becomes ready.
Document the encryption-at-rest evidence required for non-local clusters and clearly state when a local functional test does not validate that control.
Definition of Done
- Tests verify one cold-start sample per deployment and the expected backend labels.
- Validation guidance covers legacy SMB, native sensitive-data, and fully SMB-free modes.
- The runbook explains p50, p95, and p99 evaluation.
- The runbook requires encryption-at-rest evidence or an explicit non-production caveat.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.