Just to give a feedback: A nightly build has been running on one of our affected systems for three weeks now, and we haven't had a crash since. So the fix seems to improve the behavior. Next step is to test it on a system with client devices on the other end, handling much more TLS connections and more frequent connects and disconnects.
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.