Unhandled Error by Vitest

[SOULPWRD]

Hi,

I've been struggling with the Unhandled error exception in our test files using for "vitest": "^1.6.0". The error appears only in the CircleCI test job (locally it works fine). I tried to debug what is causing the issue but couldn't find out the root cause because the reported Unhandled error is not bound to any failing test file. We had used pool: 'forks' before I changed it to the default behaviour (default is threads I think) and the problem has gone.

An additional context: We use Prisma calls in our integration test files. The error looks following (missing reference to the test file):

⎯⎯⎯⎯⎯⎯ Unhandled Error ⎯⎯⎯⎯⎯⎯⎯
Error: Worker exited unexpectedly
 ❯ ChildProcess.<anonymous> node_modules/tinypool/dist/esm/index.js:185:34
 ❯ ChildProcess.emit node:events:532:35
 ❯ ChildProcess._handle.onexit node:internal/child_process:294:12

My question/plea is if it's possible to somehow catch the Unhandled error within the Vitest context? If not, it would be really nice to have some kind of ability to either catch the error or report the file in which the error occurred.

[edwinlzs]

Been seeing the same on GitLab CI test jobs using gitlab runners as well (locally works totally fine):

⎯⎯⎯⎯⎯⎯ Unhandled Errors ⎯⎯⎯⎯⎯⎯
Vitest caught 1 unhandled error during the test run.
This might cause false positive tests. Resolve unhandled errors to make sure your tests are not affected.
⎯⎯⎯⎯⎯⎯ Unhandled Error ⎯⎯⎯⎯⎯⎯⎯
Error: Worker exited unexpectedly
 ❯ ChildProcess.<anonymous> ../../../../../../../../cx.app/node_modules/tinypool/dist/index.js:139:34
 ❯ ChildProcess.emit node:events:531:35
 ❯ ChildProcess._handle.onexit node:internal/child_process:294:12

Using "vitest": "^2.1.3" and running with 6 shards. Within a couple of shards, all the tests have passed, but the above error causes that shard job to fail. It will usually pass when I re-run the job, indicating that this is a flaky error.

Unfortunately there is not much to indicate what is causing the error, so as said above, having some way to catch and report what exactly such errors might be within Vitest would be very useful for debugging.

[AriPerkkio]

What pool option are you using? With vmThreads and vmForks there's an unresolved bug in Node +22.1.x that's crashing the worker unexpectedly: #6452

[edwinlzs]

Did not configure the pool option, therefore it should be using forks

[JamieGoodson]

We've started seeing this same error in our CI pipeline (vitest 2.0.5) with the default pool (which according to the docs is forks). When re-running the CI pipeline the error goes away so we believe vitest is flaky

[AriPerkkio]

Which Node version is this on?

[JamieGoodson]

v22.6.0

[AriPerkkio]

Please try if Node 22.0.x works, if possible. Related issue is nodejs/node#54735 - though that should also only come up with vmForks or vmThreads pool.