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
Context
I will link a previous issue containing many informations about the current problem. I thought the problem came from the process manager for redirecting errors but after reading the code of uvicorn I found the problem.
Behavior
When a WebSocket request is received (and accepted), the log is redirected in the stderr.log file...
INFO: None - "WebSocket /ws/" [accepted]
After the response the supervisor team, I have search the logger used for websocket communication...
Problem
Did you have a reason to use ("uvicorn.error") for WebSocket requests?
Why we can't use the logger for success requests like HTTP? Because it's seem unicorn correctly understand the communication and accept it, as specified in the log [accepted]. So is behavior completely break the logging system to manage errors...
The text was updated successfully, but these errors were encountered:
Context
I will link a previous issue containing many informations about the current problem. I thought the problem came from the process manager for redirecting errors but after reading the code of uvicorn I found the problem.
Behavior
When a WebSocket request is received (and accepted), the log is redirected in the stderr.log file...
After the response the supervisor team, I have search the logger used for
websocket
communication...uvicorn/protocols/websockets/websockets_impl.py
Problem
Did you have a reason to use
("uvicorn.error")
for WebSocket requests?Why we can't use the logger for success requests like HTTP? Because it's seem unicorn correctly understand the communication and accept it, as specified in the log
[accepted]
. So is behavior completely break the logging system to manage errors...The text was updated successfully, but these errors were encountered: