i am trying to deploy a flask app onto aliyun ecs following this tutorial: https://www.digitalocean.com/community/tutorials/how-to-serve-flask-applications-with-gunicorn-and-nginx-on-ubuntu-18-04 sorry if any of these questions are stupid, i am an absolute new comer to nginx and gunicorn.
i have questions about a few steps of the tutorial.
the tutorial creates a sample flask project for demonstration, but i have one from a git repository. after i git clone that repository onto my server, i have an extra directory inside the "myproject" directory, which differs from the tutorial. is that ok? or will that cause problems.
the tutorial saids to configure flask to run with host = 0.0.0.0 so that it listens to all ips of that network, however, after i write app.run(host="0.0.0.0"), my flask app still runs with a specified host as the following image.
running at the server's ip instead of 0.0.0.0
the tutorial instructs me to cd into the directory where app.py is located, and bind to gunicorn. and then use a browser into my server ip at port 5000. howver, i cannot load that page and it saids this site cannot be reached, [server ip] took too long to respond.
then i continued with the tutorial, and the after the step that told me to sudo systemctl status myproject, instead of what the expected output is, i get the expected output plus the following message tagged to the end.
active status plug error message in the end
finally, after completing the tutorial, i receive 502 bad gateway instead of my site.
If you want to extract the contents of your GitHub repo into the current directory use - git clone <YOUR REPO URL> . Look at the (.) at the end it instructs git to clone in the same directory instead of creating a new folder.
You can try this - app.run(host='0.0.0.0', port='5000')
Site cannot be reached because your port is blocked and you will have to enable the incoming and outgoing connections and enable the port as well (A quick google search will give you commands) (Also try to ping your server with port number and check if the port is blocked or not).
The systemctl thing is there to run your app every time you restart you server. Even if you don't do this and don't restart your server then the above three steps are sufficient.
I'm not sure if this is Flask specific, but when I run an app in dev mode (http://localhost:5000), I cannot access it from other machines on the network (with http://[dev-host-ip]:5000). With Rails in dev mode, for example, it works fine. I couldn't find any docs regarding the Flask dev server configuration. Any idea what should be configured to enable this?
While this is possible, you should not use the Flask dev server in production. The Flask dev server is not designed to be particularly secure, stable, or efficient. See the docs on deploying for correct solutions.
The --host option to flask run, or the host parameter to app.run(), controls what address the development server listens to. By default it runs on localhost, change it to flask run --host=0.0.0.0 (or app.run(host="0.0.0.0")) to run on all your machine's IP addresses.
0.0.0.0 is a special value that you can't use in the browser directly, you'll need to navigate to the actual IP address of the machine on the network. You may also need to adjust your firewall to allow external access to the port.
The Flask quickstart docs explain this in the "Externally Visible Server" section:
If you run the server you will notice that the server is only
accessible from your own computer, not from any other in the network.
This is the default because in debugging mode a user of the
application can execute arbitrary Python code on your computer.
If you have the debugger disabled or trust the users on your network,
you can make the server publicly available simply by adding
--host=0.0.0.0 to the command line:
$ flask run --host=0.0.0.0
This tells your operating system to listen on all public IPs.
If you use the flask executable to start your server, use flask run --host=0.0.0.0 to change the default from 127.0.0.1 and open it up to non-local connections.
If you run the server you will notice that the server is only
accessible from your own computer, not from any other in the network.
This is the default because in debugging mode a user of the
application can execute arbitrary Python code on your computer.
If you have the debugger disabled or trust the users on your network,
you can make the server publicly available simply by adding
--host=0.0.0.0 to the command line:
$ flask run --host=0.0.0.0
This tells your operating system to listen on all public IPs.
Reference: https://flask.palletsprojects.com/quickstart/
Try this if the 0.0.0.0 method doesn't work
Boring Stuff
I personally battled a lot to get my app accessible to other devices(laptops and mobile phones) through a local-server. I tried the 0.0.0.0 method, but no luck. Then I tried changing the port, but it just didn't work. So, after trying a bunch of different combinations, I arrived to this one, and it solved my problem of deploying my app on a local server.
Steps
Get the local IPv4 address of your computer.
This can be done by typing ipconfig on Windows and ifconfig on Linux
and Mac.
Please note: The above step is to be performed on the machine you are serving the app on, and on not the machine on which you are accessing it. Also note, that the IPv4 address might change if you disconnect and reconnect to the network.
Now, simply run the flask app with the acquired IPv4 address.
flask run -h 192.168.X.X
E.g. In my case (see the image), I ran it as:
flask run -h 192.168.1.100
On my mobile device
Optional Stuff
If you are performing this procedure on Windows and using Power Shell as the CLI, and you still aren't able to access the website, try a CTRL + C command in the shell that's running the app. Power Shell gets frozen up sometimes and it needs a pinch to revive. Doing this might even terminate the server, but it sometimes does the trick.
That's it. Give a thumbs up if you found this helpful.😉
Some more optional stuff
I have created a short Powershell script that will get you your IP address whenever you need one:
$env:getIp = ipconfig
if ($env:getIp -match '(IPv4[\sa-zA-Z.]+:\s[0-9.]+)') {
if ($matches[1] -match '([^a-z\s][\d]+[.\d]+)'){
$ipv4 = $matches[1]
}
}
echo $ipv4
Save it to a file with .ps1 extension (for PowerShell), and run it on before starting your app. You can save it in your project folder and run it as:
.\getIP.ps1; flask run -h $ipv4
Note: I saved the above shellcode in getIP.ps1.
Cool.👌
Add host='0.0.0.0' to app.run`.
if __name__ == '__main__':
app.run(host='0.0.0.0', port=5000)
If you get OSError: [WinError 10013] An attempt was made to access a socket in a way forbidden by its access permissions on Windows, you either don't have permission to use the port, or something else is using it which you can find with netstat -na|findstr 5000.
Check whether the particular port is open on the server to serve the client or not?
in Ubuntu or Linux distro
sudo ufw enable
sudo ufw allow 5000/tcp //allow the server to handle the request on port 5000
Configure the application to handle remote requests
app.run(host='0.0.0.0' , port=5000)
python3 app.py & #run application in background
If your cool app has it's configuration loaded from an external file, like in the following example, then don't forget to update the corresponding config file with HOST="0.0.0.0"
cool.app.run(
host=cool.app.config.get("HOST", "localhost"),
port=cool.app.config.get("PORT", 9000)
)
If you're having troubles accessing your Flask server, deployed using PyCharm, take the following into account:
PyCharm doesn't run your main .py file directly, so any code in if __name__ == '__main__': won't be executed, and any changes (like app.run(host='0.0.0.0', port=5000)) won't take effect.
Instead, you should configure the Flask server using Run Configurations, in particular, placing --host 0.0.0.0 --port 5000 into Additional options field.
More about configuring Flask server in PyCharm
You can also set the host (to expose it on a network facing IP address) and port via environment variables.
$ export FLASK_APP=app.py
$ export FLASK_ENV=development
$ export FLASK_RUN_PORT=8000
$ export FLASK_RUN_HOST=0.0.0.0
$ flask run
* Serving Flask app "app.py" (lazy loading)
* Environment: development
* Debug mode: on
* Running on https://0.0.0.0:8000/ (Press CTRL+C to quit)
* Restarting with stat
* Debugger is active!
* Debugger PIN: 329-665-000
See How to get all available Command Options to set environment variables?
Go to your project path on CMD(command Prompt) and execute the following command:-
set FLASK_APP=ABC.py
SET FLASK_ENV=development
flask run -h [yourIP] -p 8080
you will get following o/p on CMD:-
Serving Flask app "expirement.py" (lazy loading)
Environment: development
Debug mode: on
Restarting with stat
Debugger is active!
Debugger PIN: 199-519-700
Running on http://[yourIP]:8080/ (Press CTRL+C to quit)
Now you can access your flask app on another machine using http://[yourIP]:8080/ url
For me i followed the above answer and modified it a bit:
Just grab your ipv4 address using ipconfig on command prompt
Go to the file in which flask code is present
In main function write app.run(host= 'your ipv4 address')
Eg:
Create file .flaskenv in the project root directory.
The parameters in this file are typically:
FLASK_APP=app.py
FLASK_ENV=development
FLASK_RUN_HOST=[dev-host-ip]
FLASK_RUN_PORT=5000
If you have a virtual environment, activate it and do a pip install python-dotenv .
This package is going to use the .flaskenv file, and declarations inside it will be automatically imported across terminal sessions.
Then you can do flask run
This answer is not solely related with flask, but should be applicable for all cannot connect service from another host issue.
use netstat -ano | grep <port> to see if the address is 0.0.0.0 or ::. If it is 127.0.0.1 then it is only for the local requests.
use tcpdump to see if any packet is missing. If it shows obvious imbalance, check routing rules by iptables.
Today I run my flask app as usual, but I noticed it cannot connect from other server. Then I run netstat -ano | grep <port>, and the local address is :: or 0.0.0.0 (I tried both, and I know 127.0.0.1 only allows connection from the local host). Then I used telnet host port, the result is like connect to .... This is very odd. Then I thought I would better check it with tcpdump -i any port <port> -w w.pcap. And I noticed it is all like this:
Then by checking iptables --list OUTPUT section, I could see several rules:
these rules forbid output tcp vital packets in handshaking. By deleting them, the problem is gone.
I had the same problem, I use PyCharm as an editor and when I created the project, PyCharm created a Flask Server. What I did was create a server with Python in the following way;
basically what I did was create a new server but flask if not python
I hope it helps you
This finally worked for me.
import os
Then place this at the end of your python app.py or main file.
if __name__ == "__main__":
port = int(os.environ.get("PORT", 5000))
app.run(host='0.0.0.0', port=port)
go to project path
set FLASK_APP=ABC.py
SET FLASK_ENV=development
flask run -h [yourIP] -p 8080
you will following o/p on CMD:-
* Serving Flask app "expirement.py" (lazy loading)
* Environment: development
* Debug mode: on
* Restarting with stat
* Debugger is active!
* Debugger PIN: 199-519-700
* Running on http://[yourIP]:8080/ (Press CTRL+C to quit)
If none of the above solutions are working, try manually adding "http://" to the beginning of the url.
Chrome can distinguish "[ip-address]:5000" from a search query. But sometimes that works for a while, and then stops connecting, seemingly without me changing anything. My hypothesis is that the browser might sometimes automatically prepend https:// (which it shouldn't, but this fixed it in my case).
In case you need to test your app from an external network.
Simply serve it to the whole Internet with ngrok.com
which will deploy it like a dev server but in no time and locally, saved me a lot of time, and no, I'm not related to that company :)
Just make sure to change the port in your flask app:
app.run(host='0.0.0.0', port=80)
I'm currently running a locally hosted website, created in flask as part of a project. I wanted the server to be externally visible, so I checked the docs, and found this:
If you run the server you will notice that the server is only
accessible from your own computer, not from any other in the network.
This is the default because in debugging mode a user of the
application can execute arbitrary Python code on your computer.
If you have the debugger disabled or trust the users on your network,
you can make the server publicly available simply by adding
--host=0.0.0.0 to the command line:
$ flask run --host=0.0.0.0 This tells your operating system to listen
on all public IPs.
I assumed the same applied to python, and when started the server like this
if __name__ == "__main__":
app.debug=True
app.run(host='0.0.0.0')
When running this the terminal said Running on http://0.0.0.0:5000/, however clicking on that link just gave a "this site can't be reached" error. I tried to go the default address that it normally directed to (127.0. ...) and although this launched the site on my laptop (where flask is running) I still couldn't access the site from other devices on the same network, even when copy-pasting the url.
I also tried typing my laptops ipv4 (192.168. ...) followed by :5000 but still was unable to connect to the server.
What have I done wrong?
Run this in your terminal:
sudo ufw allow 5000
I have a remote machine at my workplace, when we developers run server/ or docker containers. everything was working fine but a while back somethign went wrong.
if I run the python flask app
from app import app
app.run(host='0.0.0.0', port=5050)
i get message
* Running on http://0.0.0.0:5050/
and I am able to access the above from my local machine using the remote server machine ip:5050 but if I run docker container docker run -itd <conta_image_name> -p 80:90 --add-host=localdomain.com:machine_ip_address i get error message saying IPv4 forwarding is disabled. Networking will not work.
Now this issue is in production so I really need someone to throw up some light, what might be wrong or let me know what more info I need to put.
I have fixed this issue myself following this: https://success.docker.com/article/ipv4-forwarding
Another solution is..
Try adding -net=host along with docker run command
https://medium.com/#gchandra/docker-ipv4-forwarding-is-disabled-8499ce59231e
I have an ubuntu EC2 server and want to run a flask server. I want to hit the server using my domain name, api.example.com, without having to include the port number. Right now, I can successfully access the server by doing api.example.com:5000/... but I can't figure out how to do api.example.com/....
Right now I'm just running the flask server directly, using python flask_server.py.
In flask_server.py:
if __name__ == '__main__':
app.run(host=0.0.0.0)
The run method takes a port optional argument:
if __name__ == '__main__':
app.run(host="0.0.0.0", port=80)
You can do this for testing, but for production I highly recommend you read the deployment options section in the documentation which details ways to run flask with various front end WSGI servers.
If you need help understanding how all these components work together and how to set them up; this gist has a nice summary.
Update: The host param needs to be a string.
You need sudo privilege in order to use port 80.
sudo python3 app.py
That will solve the problem.
Correct syntax to use Flask server on port 80 is:
if __name__ == '__main__':
app.run(host='0.0.0.0', port=80)
Keep in mind that you might need super user privileges.
This kind of workaround wound be acceptable if your were still building your application.
If, as I understood, you plan to deploy your app in production then you'd need to do it properly.
Here you find step by step information for Ubuntu, as requested:
https://www.digitalocean.com/community/tutorials/how-to-deploy-a-flask-application-on-an-ubuntu-vps