BUG/MINOR: backend: hold correctly lock when killing idle conn

The wrong lock seems to be held when trying to remove another thread
connection if max fd limit has been reached (locking the current thread
instead of the target thread lock).

This could be backported up to 2.0.
1 file changed