You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Tracebacks similar to the following are produced intermittently while performing receptor operations:
Traceback (most recent call last):
File "/home/graham/git/receptor/receptor/connection/base.py", line 62, in receive
await self.buf.put(msg)
File "/home/graham/git/receptor/receptor/messages/framed.py", line 245, in put
await self.consume(data)
File "/home/graham/git/receptor/receptor/messages/framed.py", line 267, in consume
await self.finish()
File "/home/graham/git/receptor/receptor/messages/framed.py", line 283, in finish
FramedMessage(msg_id=self.current_frame.msg_id, header=json.load(self.bb))
File "/usr/lib64/python3.7/json/__init__.py", line 296, in load
parse_constant=parse_constant, object_pairs_hook=object_pairs_hook, **kw)
File "/usr/lib64/python3.7/json/__init__.py", line 343, in loads
s = s.decode(detect_encoding(s), 'surrogatepass')
UnicodeDecodeError: 'utf-8' codec can't decode byte 0x8c in position 4080: invalid start byte
I don't have a reliable reproducer yet but it seems to happen more often when there are a larger number of communicating peers.
Activity
jhjaggars commentedon Apr 8, 2020
https://github.com/project-receptor/receptor/blob/devel/receptor/connection/base.py#L109-L111
I think this section might be the issue.
ghjm commentedon Apr 8, 2020
Is this as simple as changing
ensure_futuretowait_for? Or is there more to it?jhjaggars commentedon Apr 9, 2020
It used to be
await.I changed it due to high latency, but didn't consider out of order messages (duh).