voidzero-dev / voidzero-dev/vite-plus

Concurrent `vp run dev` processes download the same Node.js runtime twice

Open
#2,642 0 comments 0 reactions 0 assignees View on GitHub
enhancement
Dominant language
Rust
Stars
5.8k
Forks
262
Avg merge
1d 34m
Merged PRs (30d)
135

Description

### Describe the bug

When running `vp run dev` from two separate repositories that require the same managed Node.js version, Vite+ downloads the same Node.js runtime independently for each process instead of coordinating the installation and reusing the shared runtime.

The two repositories are unrelated (API and web), but both resolve to the same Vite+ managed runtime location under the user-level Vite+ directory.

This suggests that the runtime itself is intentionally shared globally, but concurrent `vp` processes may both observe the runtime as missing and start their own download/install operation.

### Expected behavior

If multiple `vp` processes require the same Node.js version concurrently, only one process should download/install the runtime. Other processes should wait for the installation to complete and then reuse the same runtime.

Conceptually:

```text
API vp run dev ──┐
├── shared runtime lock ──> download Node 24.21.0 once
Web vp run dev ──┘ └─> both reuse it
```

### Actual behavior

Both projects trigger a Node.js download even though they resolve to the same shared managed runtime.

### Environment

OS: Windows 11

Node runtime reported by Vite+:

```text
VITE+ - The Unified Toolchain for the Web

Environment:
Version 24.21.0
Source lts

Tool Paths:
node C:\Users\\.vite-plus\js_runtime\node\24.21.0\node.exe
npm C:\Users\\.vite-plus\js_runtime\node\24.21.0\npm.cmd
npx C:\Users\\.vite-plus\js_runtime\node\24.21.0\npx.cmd
```

### Why this seems like a Vite+ issue

The managed Node runtime is stored in a user-level shared location rather than inside either repository. Therefore, downloading the same runtime independently for concurrent projects appears unnecessary and can waste bandwidth and startup time.

A related issue, #2449, shows that `vp run` provisions managed Node.js runtimes, but that issue concerns interrupted downloads rather than concurrent deduplication:
https://github.com/voidzero-dev/vite-plus/issues/2449

### Suggested solution

Add cross-process synchronization around managed Node.js runtime installation, keyed by runtime/platform/version. For example, a filesystem lock around the shared runtime directory could allow one process to perform the download while other processes wait and then reuse the installed runtime.

### Validations

- [x] Checked the existing Vite+ issues for a duplicate report.
- [x] Confirmed both projects resolve to the same global managed Node.js runtime.
- [x] This appears specific to Vite+'s managed runtime provisioning rather than the projects themselves.

Contributor guide

Open the contributing guide

Research direction

Start at the managed Node.js runtime provisioning path exercised by `vp run dev`, using the shared path shown in the report as the concurrency boundary. Read issue #2449 for related provisioning behavior and verify the Windows case with two repositories requesting Node 24.21.0. Done means concurrent processes perform one download/install and both reuse the completed runtime.

Written by the indexing model from the issue text.

Assessment

Tech stack
node.js, rust
Domain
devops, infrastructure
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Active
Clarity
Mostly clear
Newbie friendliness
66/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.