[santusht06]

Description

fixes #23392

When a config/env change lands while restartServer is in flight (e.g. fast consecutive edits or switching branches with git checkout), server.restart() returns the active _restartPromise directly without scheduling a follow-up restart once the current one completes. As a result, the server stays running the previous config snapshot.

This PR adds an internal _pendingRestart flag to server.restart() so that if another restart request arrives while a restart is in flight, restartServer loops until no pending requests remain before resolving the promise.

Added a test case in hooks.spec.ts covering concurrent server.restart() calls.

[github-actions]

This PR has been automatically flagged as likely to be created by a bot, LLM, or agent, and will be automatically closed. These contributions harm the maintenance of the project. Please read our AI policy for more information.

If you believe this is a mistake, please reply to this comment and we will review it.

[santusht06]

Hi maintainers,

This is a legitimate manual fix. I investigated #23392, reproduced the race condition with the author's reproduction repo (silentoplayz/vite-restart-race-repro), and traced where in-flight restarts drop subsequent requests in packages/vite/src/node/server/index.ts.

The fix uses a _pendingRestart flag in server.restart() to queue and loop through subsequent restarts before settling _restartPromise. I also added a unit test in hooks.spec.ts and verified that all test suites, typechecks, and linters pass.

Please let me know if any changes are needed.