Join GitHub today
GitHub is home to over 40 million developers working together to host and review code, manage projects, and build software together.Sign up
Still attempting to reproduce on a node configured to give decent stack traces.
My gut feeling is this is related to blocking IO and we are retrying (likely writing to a socket) in a tight loop in a situation that is not recoverable in the peer_write thread.
Note: just speculation, no evidence to indicate this is actually the underlying issue.
Specifically our call to
Pinging @hashmap for another pair of eyes on this.
I also bet that if the
Bunch of threads stuck in a loop doing the following -
So this is the point where the
Need to look at the code again but I think we may not be handling error conditions correctly when we call