-
-
Notifications
You must be signed in to change notification settings - Fork 37.1k
flaky test-regress-GH-814_2 #34527
Copy link
Copy link
Closed
Labels
flaky-testIssues and PRs involving tests that fail intermittently in CI.Issues and PRs involving tests that fail intermittently in CI.fsIssues and PRs related to file-system APIs and the fs module.Issues and PRs related to file-system APIs and the fs module.libuvIssues and PRs related to the libuv dependency or the uv binding.Issues and PRs related to the libuv dependency or the uv binding.
Description
Activity
Metadata
Metadata
Assignees
Labels
flaky-testIssues and PRs involving tests that fail intermittently in CI.Issues and PRs involving tests that fail intermittently in CI.fsIssues and PRs related to file-system APIs and the fs module.Issues and PRs related to file-system APIs and the fs module.libuvIssues and PRs related to the libuv dependency or the uv binding.Issues and PRs related to the libuv dependency or the uv binding.
What steps will reproduce the bug?
Run test/pummel/test-regress-GH-814_2.js on fedora-last-latest-x64 in CI.
How often does it reproduce? Is there a required condition?
Not every time. Running https://ci.nodejs.org/job/node-stress-single-test/151/ right now to see how often.
What is the expected behavior?
The test should pass.
What do you see instead?
The test crashes with -9, which I believe means libuv is complaining about a bad file descriptor.
Additional information
Something similar to this was going on with readstream last year and I think @addaleax fixed it. Maybe
fshas the same/similar issue, perhaps when explicitfs.close()is not used?