Keeping server UP with python simpleHTTP server

Question

I am trying to make a blog and sustain it, on this blog the content will only be in html/css or some images, so I don't need any sophisticated web server, I've tried to write a script that will hold the blog active, but sometimes and that's really frequent, it just goes down and I've tried to get logs of it and at first it was something about broken pipe but I tried to fix that using signal handling, and I don't know how to keep the server up all the time. here is the code:

#!/usr/bin/python
from http.server import HTTPServer, BaseHTTPRequestHandler
import os
from os import curdir, sep, path
from signal import signal, SIGPIPE, SIG_DFL
import logging

HOST = #ip_address
PORT = 8089

class SimpleHTTPRequestHandler(BaseHTTPRequestHandler):

    def do_GET(self):
        self.send_response(200)
        self.end_headers()
        try:
            print(self.path)
            path = self.path
            if ('?' in path):
                path = path.split('?')[0]
            self.path = path
            print(self.path)
            if (self.path == '/'):
                self.path = '/index.html'
            self.path = 'output' + self.path
            if(os.path.isdir(self.path)):
                if(not os.path.exists(self.path + 'index.html')):
                    self.send_response(403)
                    self.wfile.write(str.encode("Listing of directories not permited on this server"))
                else:
                    self.path = self.path + 'index.html'

            f = open(curdir + sep + self.path, 'rb')
            self.wfile.write(f.read())
            f.close()
        except IOError:
            print("File "+self.path+" not found")
            self.send_response(404)


httpd = HTTPServer((HOST, PORT), SimpleHTTPRequestHandler)
httpd.serve_forever()

signal(SIGPIPE, SIG_DFL)

Roberto Gonçalves · Accepted Answer · 2020-01-03T20:09:57.060

You can solve it by applying a handle like this:

import os
import SocketServer

HOST = 'localhost'
PORT = 8089

class ForkingEchoRequestHandler(SocketServer.BaseRequestHandler):

    def do_GET(self):
        self.send_response(200)
        self.end_headers()
        try:
            print(self.path)
            path = self.path
            if ('?' in path):
                path = path.split('?')[0]
            self.path = path
            print(self.path)
            if (self.path == '/'):
                self.path = '/index.html'
            self.path = 'output' + self.path
            if(os.path.isdir(self.path)):
                if(not os.path.exists(self.path + 'index.html')):
                    self.send_response(403)
                    self.wfile.write(str.encode("Listing of directories not permited on this server"))
                else:
                    self.path = self.path + 'index.html'

            f = open(curdir + sep + self.path, 'rb')
            self.wfile.write(f.read())
            f.close()
        except IOError:
            print("File "+self.path+" not found")
            self.send_response(404)
        except socket.error:
            pass


class ForkingEchoServer(SocketServer.ThreadingMixIn, SocketServer.TCPServer):
    pass

if __name__ == '__main__':
    import socket
    import threading

    address = (HOST, PORT) # let the kernel give us a port
    server = ForkingEchoServer(address, ForkingEchoRequestHandler)
    # Start a thread with the server -- that thread will then start one
    # more thread for each request
    server_thread = threading.Thread(target=server.serve_forever)
    # Exit the server thread when the main thread terminates
    server_thread.daemon = True
    server_thread.start()
    print('Server loop running in thread: "%s"' % server_thread.name)
    while True:
        pass
    server.shutdown()

But, maybe this might be a problem since there is a bug in Socket Server, this means that it can't process multiple requests at the same time

Additionally, if you want your HTTP server to process more than one request at a time, you need your server class to use one of the SocketServer.ForkingMixIn and SocketServer.ThreadingMixIn classes. Check the documentation of the SocketServer module for details.

Here is a more specific question that lists most cases about BaseHTTPServer

EDIT

After a chat discussion it was possible to resolve the problem with NGINX using a simple configuration file:


user  nginx;
worker_processes  1;

error_log  /var/log/nginx/error.log warn;
pid        /var/run/nginx.pid;


events {
    worker_connections  1024;
}


http {
    include       /etc/nginx/mime.types;
    default_type  application/octet-stream;

    log_format  main  '$remote_addr - $remote_user [$time_local] "$request" '
                      '$status $body_bytes_sent "$http_referer" '
                      '"$http_user_agent" "$http_x_forwarded_for"';

    access_log  /var/log/nginx/access.log  main;

    sendfile        on;
    #tcp_nopush     on;

    keepalive_timeout  65;

    #gzip  on;

    include /etc/nginx/conf.d/*.conf;


    server {
        listen 80;
        root /var/www/cristicretan.ro/output/;
        index index.html;
        server_name cristicretan.ro www.cristicretan.ro;
        location / {
        try_files $uri $uri/ =404;
        }
    }
}

Can you give me an example of how to use SocketServer.ThreadingMixIn? — J. Homer, Dec 31 '19 at 10:15
Will that solve my problem that my server keeps on going down? — J. Homer, Jan 03 '20 at 14:04
I've tested with multiple requests in a linux server for almost 8 hours and it was alive the whole time. You need to test it too. — Roberto Gonçalves, Jan 03 '20 at 14:06
How could I add some logging output? I swapped the "localhost" with my ip to test it online but it doesn't work — J. Homer, Jan 03 '20 at 14:08
Also, I would recommend another webserver, like Flask, it is more reliable than simplehttp. — Roberto Gonçalves, Jan 03 '20 at 14:08
Why it doesn't work when I try to swap the 'localhost' with the correct ip address? — J. Homer, Jan 03 '20 at 18:53
Interesting...I'll take a look it it! Have you searched about Flask? It's very similar to simple http but it's more resilient. SimpleHttp have some scalability issues and it is not meant to be a production server. Here is a [example of flask serving files](https://stackoverflow.com/questions/20646822/how-to-serve-static-files-in-flask/) — Roberto Gonçalves, Jan 03 '20 at 19:04
I will take a look, but I really want something simple, like a simple blog created with pelican — J. Homer, Jan 03 '20 at 19:06
We should move it to a chat, then we can discuss better about it! — Roberto Gonçalves, Jan 03 '20 at 19:17
Let us [continue this discussion in chat](https://chat.stackoverflow.com/rooms/205315/discussion-between-roberto-goncalves-and-j-homer). — Roberto Gonçalves, Jan 03 '20 at 19:17

score 0 · Answer 2 · answered Jan 02 '20 at 01:20

0

I highly suggest gunicorn to manage your running process. You can also handle multiple requests and concurrency really well with this tool. I linked to the custom application portion since you're writing your own

answered Jan 02 '20 at 01:20

Zoidberg

425
3
10

How do you use that in my context? I mean why is that better? – J. Homer Jan 02 '20 at 09:25

score 0 · Answer 3 · answered Jan 02 '20 at 09:20

We can either ignore the BrokenPipe and continue or catch the exception.

For ignoring, SIGPIPE is used. But according to the python signal documentation,

Do not set SIGPIPE’s disposition to SIG_DFL in order to avoid BrokenPipeError. Doing that would cause your program to exit unexpectedly also whenever any socket connection is interrupted while your program is still writing to it.

So the best way is to catch the BrokenPipeError

import sys, errno
try:
    ### IO operation ###
except IOError as e:
    if e.errno == errno.EPIPE:
        ### Handle error ###

I did that too, and for handling I used "pass" but it still goes down.. — J. Homer, Jan 02 '20 at 09:26

Keeping server UP with python simpleHTTP server

3 Answers3