vitest-dev/vitest · error · Error

The child process was torn down or never initialized. This i

Error message

The child process was torn down or never initialized. This is a bug in Vitest.

What it means

Internal assertion in the forks worker: the private _fork getter throws if the underlying ChildProcess was never created (start() not called or failed) or was cleared (set to undefined after stop()). The getter is used by on/off/send, so any message interaction with a non-running forked process trips it.

Source

Thrown at packages/vitest/src/node/pools/workers/forksWorker.ts:108

      fork.stdout?.unpipe(this.stdout)
      this.stdout.setMaxListeners(this.stdout.getMaxListeners() - 1)
    }

    if (fork.stderr) {
      fork.stderr?.unpipe(this.stderr)
      this.stderr.setMaxListeners(this.stderr.getMaxListeners() - 1)
    }

    this._fork = undefined
  }

  deserialize(data: unknown): unknown {
    return data
  }

  private get fork() {
    if (!this._fork) {
      throw new Error(`The child process was torn down or never initialized. This is a bug in Vitest.`)
    }
    return this._fork
  }
}

View on GitHub (pinned to d568f8ce37)

Solutions

1. Report as a Vitest bug with reproduction (focus on cancellation/teardown timing).
2. Retry — transient teardown races during Ctrl+C can surface this once.
3. If using the @experimental ForksPoolWorker directly, never send before start() or after stop().

try {
  await vitest.start()
} catch (e) {
  if (e.message === 'The child process was torn down or never initialized. This is a bug in Vitest.') {
    console.error('Forks worker teardown race — retry; if persistent, report as a bug.')
  }
  throw e
}

When it happens

Trigger: ForksPoolWorker.on/off/send is called when this._fork is undefined — either before start() ran or after stop() set it to undefined at packages/vitest/src/node/pools/workers/forksWorker.ts:99-108. Indicates the runner sent a message to a torn-down worker.

Common situations: A race where the pool/runner posts a message after worker.stop() completed; stop() running during cancellation while a queued message is still in flight; a regression in PoolRunner.postMessage gating (it should drop messages when STOPPED).

Related errors

• The worker thread was torn down or never initialized. This i
• [vitest-pool-runner]: Cannot start a stopped runner
• Unexpected message ${JSON.stringify(message, null, 2)}
• Cannot find the environment. This is a bug in Vitest.
• The browser server was not initialized${project.name ? ` for

AI-assisted analysis of vitest-dev/vitest@d568f8ce37 (2026-08-03). Data as JSON: /data/errors/409372ccec371539.json. Report an issue: GitHub.