Just curious anyone else having this same issue. Code runs fine but 40min into it, it suddenly stopped, no error message. Just sth like:
RuntimeWarning: coroutine ‘StreamConn.close’ was never awaited
conn.close()
Event loop is closed
Event loop is closed
Event loop is closed
Event loop is closed
Event loop is closed
Event loop is closed
Event loop is closed
Event loop is closed
folllowed by “Fatal Python error: Cannot recover from stack overflow.”
…
I have this problem constantly. I use c# but problem is same. Runs for 1+ minutes, occasionally a couple hours. Then fails with no status reported.
Worse. Closing, waiting a few seconds, and reopen doesn’t always work. Sometimes, have to walk away and try quite a bit later to get connected again.
The C# SDK doesn’t re-connect by default to the streaming server. In fact, it’s just a low-level wrapper not a full-powered client. You have to implement re-connection manually, subscribe on OnError event and connection state changes event and check what happens. And report such a problems on GitHub - I check it more often.
I tried a lot and found that the processing time on the client end has sth to do with this. Even though I was unable to solve this completely, but for example, when I show plots in my code, it failed very quickly. But when I commented out the showplot function, it goes further, but still fails at some point. With this situation, it’s impossible to bot trade ( I tried hard, but find it difficult). Remember seeing someone mentioned that the cache is full because the data is not consumed fast enough, trying to find a way to ignore while processing of the current data is still underway (like the reading of frames from a camera),