Closed iacore closed 5 months ago
I believe this is fixed in the zig-0.11
branch. And the same fix has been applied to the nonblocking_threadpool
branch which I'm currently testing and should be merging into master next week.
The issue is a double-free of the connection by the connect manager and the connection pool.
I think the issue is fixed now. Thanks!
How to reproduce (roughly)
server.stop()
How to reproduce this issue exactly. It's not 100% reproducible, and I don't know why.
error trace