Closed KingMob closed 2 years ago
FTR, I just had the same error on another PR. Maybe we need to bump the timeout value here a bit?
I'll rerun the test, but I don't think we should bump the timeout before checking for a regression. The slow-stream
in the test writes every 50 ms, but the idle timeout is 200 ms, so it shouldn't timeout. Plus, both tests failed at the "Client is slow to write"
subsection, which suggests it's repeatable and specific.
I'll rerun, but let's open a new issue to check this.
Fixes #620