I am working on Binance Websocket to listen to account events MARKET events. I copy every order from masters to salve it works just fine when I run it and it copies from all master. However, when I run it in background using nohup Linux service it stops reading any events after 3 days of running the script. I tried to capture the STDERR but nothing is piped from the script. Even though it is appeared in htop as a running process but without doing its job. I tried to use asyncio framework but I got the same problem. I need to listen to multi-master at the same time and in a long-run term.
import csv
import math
import threading
import time
from datetime import datetime
from binance import ThreadedWebsocketManager, Client
masters = [
[
'api1',
'secret1',
'test1'
],
[
'api2',
'secret2',
'test2'
],
]
def CopyOrder(order, client, name):
# some code to copy order from master
def main(api_key: str, api_secret: str, name: str):
print("Start listening on", name) # Debug
twm = ThreadedWebsocketManager(api_key=api_key, api_secret=api_secret)
cl = Client(api_key=api_key, api_secret=api_secret, testnet=False)
twm.start() # Start the websocket manager
def handle_socket_message(msg):
if msg['e'] == 'executionReport':
# check if it is order or not
CopyOrder(order=msg, client=cl, name=name)
twm.start_user_socket(callback=handle_socket_message)
twm.join()
if __name__ == "__main__":
for master in masters:
# listen to all masters for new orders
thread = threading.Thread(target=main, args=(master[0], master[1], master[2]))
thread.start()
Quoting from Binance https://binance-docs.github.io/apidocs/spot/en/#websocket-market-streams
A single connection to stream.binance.com is only valid for 24 hours; expect to be disconnected at the 24 hour mark
I am working with Binance Websocket too and I noticed the same random disconnection, sometimes after a day, sometimes after a few days, so I now have an uptime check in place and I manage a disconnection/reconnection