-
Notifications
You must be signed in to change notification settings - Fork 10.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
bm_chttp2_transport --benchmark_filter=BM_TransportStreamSend/134217728$ GRPC_POLL_STRATEGY=poll #18416
Comments
Seen the same issue on run_tests.py stack:
|
Possibly related, though slightly different error:
|
|
On linux I'm seeing a similar problem:
|
This is currently our top flake - it's failing >50% builds on macos (measured over the last week). It's flaky mostly on Mac, but it seems that it can be reproduced on Linux as well. |
@AspirinSJL can you take a look? |
@yashykt owns chttp2, so he's probably the right person to look at this. |
The issue is that we are relying on the closures being scheduled on the exec_ctx for correctness. If the closures end up being scheduled on the executor thread, then we might end up destroying the Stream object before the grpc stream is freed. The fix is to simply delete the Stream after the grpc stream is done being destroyed. |
I think #19342 partially solves this. There still seems to be a race. |
This was fixed by #20091 |
https://source.cloud.google.com/results/invocations/17e1f1a3-a281-468a-9dd7-2c803fb25266/targets/github%2Fgrpc%2Fcpp_linux_dbg_native/tests
It's not clear what the actual failure was, since there aren't any error messages in the log:
The text was updated successfully, but these errors were encountered: