Can't use emit inside app.route

[Perrutte]

So what is happeing right now is when i try to use socketio.emit, inside any app.route it doesn't give any error it just doesn't work.
I'm using a Linux EC2 machine in aws, and i ran "python flasky.py", so i execute my flasky file and run the app on port 5000 with
socketio_.run(app, host='0.0.0.0') and everything works.
But when i reboot the machine and wsgi starts, sudently all the emits inside the routes stop working, and i'm assuming it's because on my wsgi.py i have app.run(), instead of socketio_.run(app, host='0.0.0.0'), so i'ved tried to change the wsgi.py to socketio_.run(app, host='0.0.0.0') but when i do that all the emits (inside or outside a route) stop working.

flasky.py

app = create_app(os.getenv('FLASK_CONFIG') or 'default')
socketio_ = SocketIO()
socketio_.init_app(app,cors_allowed_origins="*", async_mode='eventlet')

if __name__ == "__main__":
    socketio_.run(app, host='0.0.0.0')

wsgi.py

if __name__ == "__main__":
    app.run()

So on conclusion, if i run my flasky.py app everything works on port 5000.
If i let wsgi.py run by it self when i reboot the machine, all socket emits stop working.

Thank you in advance!

note: I'm not putting here the imports, but i'm importing everything correctly

[miguelgrinberg]

First of all, please do not write issues when you are troubleshooting a problem. Issues is where I track bugs and enhancement for this package.

Second, your wsgi.py is wrong, it will never work for Socket.IO. You have to use socketio.run() to start your Eventlet web server.

If that does not work, then you need to look at the server logs for clues. See the troubleshooting section of the documentation to learn how to enable Socket.IO logs on the server.

[Perrutte]

Yes, the message was sent correctly because this logs are from port 5000, when i switch to port 80 i dont get any logs and that event "mbway Error|11207", its not being emited for some reason, all the other sockets are being emited because they are inside @socketio. I just dont understand why is not working i still think its because the 2 services that im running i sended them the previous msgs

[Perrutte]

A few months ago we had the same issue and the a collegue of fine fixed it by
using only one ExecStart which was this one :
#ExecStart=/usr/local/bin/python3.7 -m gunicorn --workers 1 -k eventlet --log-file /home/admin/API/app-log.log --access-logfile /home/admin/API/access.log --bind unix:/home/admin/API/app.sock --bind unix:/home/admin/API/socket.io -m 777 wsgi:app

So with this ExecStart this "emit Mbway|Error" was working and we could emit outside routes.
This worked for a while but it started to give some really bad issue where the app was basicly freezing because of long pulling sockets i assume, the app was basicly crashing and we had to reboot it 2-3 times a day. So we went back to what we had before which was 2 Services insted of 1.

app-socket.service : ExecStart=/usr/local/bin/python3.7 -m gunicorn --worker-class eventlet -w 1 --log-file /home/admin/API/app-log.log --access-logfile /home/admin/API/access.log --bind unix:/home/admin/API/socket.io -m 777 wsgi:app

app.service: ExecStart=/usr/local/bin/python3.7 -m gunicorn --workers 3 --log-file /home/admin/API/app-log.log --access-logfile /home/admin/API/access.log --bind unix:/home/admin/API/app.sock -m 777 wsgi:app

[Perrutte]

Update: Sorry i mess up the words what i meant to say was, with this one
#ExecStart=/usr/local/bin/python3.7 -m gunicorn --workers 1 -k eventlet --log-file /home/admin/API/app-log.log --access-logfile /home/admin/API/access.log --bind unix:/home/admin/API/app.sock --bind unix:/home/admin/API/socket.io -m 777 wsgi:app
Everything with sockets work (Inside Routes, outside Routes, works everywhere) in port 5000 and port 80. The down side is this is constently making our app crash because of "Long pooling" and "Timeouts". We assume it was because of the number of workers so we decided to split this ExecStart in two.

app-socket.service : ExecStart=/usr/local/bin/python3.7 -m gunicorn --worker-class eventlet -w 1 --log-file /home/admin/API/app-log.log --access-logfile /home/admin/API/access.log --bind unix:/home/admin/API/socket.io -m 777 wsgi:app

app.service: ExecStart=/usr/local/bin/python3.7 -m gunicorn --workers 3 --log-file /home/admin/API/app-log.log --access-logfile /home/admin/API/access.log --bind unix:/home/admin/API/app.sock -m 777 wsgi:app

With these 2 services, our app never crashes. We can emit inside Routes only on port 5000, when we switch to port 80 it stops working and doesn't provide any error logs.

So i'm just not sure what to do, i don't really understand why it isn't working on port 80, and it's so strange because with only 1 ExecStart eveything works.

Thank you in advance for all the help and fast responses.

[miguelgrinberg]

You insist in not showing actual details. You mentioned several times port 5000 and port 80. There is absolutely nothing in the commands or logs that you show that indicate a port number. At this point I'm going to ask you to stop this, since it is going nowhere.

If you have a problem making your application work on port 80, then go to stack overflow and ask a question about that. Your problem is not with Flask-SocketIO.

[Perrutte]

Ok. Just weird how if i run the app with 1 worker with Eventlet eevrything works, and if i run 2 others services doing the same thing it doesn't.
Thank you for your time and all the help.