Nginx deployment Django Python virtual environment creation fool tutorial

Posted by tim_ver on Mon, 17 Jan 2022 04:09:59 +0100

The relationship between uwsgi and nginx is not discussed here, but learning is recommended

Generally speaking, Nginx can ensure load balance when your web page supports more requests,

Simple web pages can also meet the requirements with uwsgi deployment and django runserver, so the load capacity is

Nginx > uwsgi > local django runserver

(your server needs to install Python, pip and nginx and start the environment before installing uwsgi) - this step precedes the first one

The project enters the deployment state ALLOWED_HOST = [’*’]

DEBUG=False or something

1. First, you need to configure the virtual environment virtualenv on the server

Note (only Python 2.7 and later support virtualenv)

If Python 2

After navigating to your project folder

virtualenv -p /path/to/new/virtual/environment venv

If Python 3

python3 -m venv /path/to/new/virtual/environment

Running this command will create a new folder environment

And the virtual environment is named venv

There will be three folders in the created new folder, of which bin is the starting directory for starting the virtual environment in the future

If you need to delete the virtual environment, just delete the root directory of the virtual environment directly

As follows: use the source command to turn on and turn off using deactivate [venv name]. If you want to turn off the initial [base] environment, use conda deactivate

source /path/to/new/virtual/environment/bin/activate

Note that you must remember to install uwsgi for the virtual environment

pip install uwsgi

Next, it is also important to install the configuration file for the virtual environment (that is, the virtual environment installed in pycharm for your project)

cd to the local project folder or pycharm terminal

pip freeze > requirements.txt

If there is a problem (I have a problem with the package of - ip 24, just search and delete it on Baidu) you need to query each package with a problem

Then check the requirements Txt format

It's probably like this. The package = = version number must be exactly the same. PIP freeelist can't

asgi-redis==1.4.3
asgiref==3.3.4
asttokens==2.0.4
async-timeout==2.0.1

Then scp -r uploads the local project to the server

Then start the virtual environment on the server and check the pip version and python version

source /path/to/new/virtual/environment/bin/activate

Then install the dependency (there may be some problems in this step. It is mainly domain problems, so pychart can't be lazy when using it. It must be a project and an environment

Otherwise, too many packages always have package versions that are not supported by the domain. Try to use the image separately for the problematic package, and set http to safe source

What others wrote on csdn is too detailed (I'm sorry to copy)

pip install -r requirements.txt

If present

WARNING: Retrying (Retry(total=4, connect=None, read=None, redirect=None, status=None)) after connection broken by 'SSLError(SSLError(1, '[SSL: UNKNOWN_PROTOCOL] unknown protocol

Because PIP version = = 20.3.3 (latest version) has fully used SSL authentication. If the server is http, consider demoting pip to 20.2.2

2. Configure uwsgi ini

Under the root directory of django project (the same level as manage.py directory)

Test uwsgi connection

First create a py test file

# /path/to/django/project/test.py  
def application(env, start_response):
    start_response('200 OK', [('Content-Type','text/html')])
    return [b"Hello World"]

cd to the / path/to/django/project / command line, enter the following

uwsgi --http :8001 --wsgi-file test.py

The browser inputs your server URL plus port 8001. If you see Hello World on the page, it indicates that uwsgi configuration is successful

If you are prompted with Internal Server Error

Check the version of uwsgi to see if it is installed in the virtual environment
Check test Is py running the correct one

Next, formally configure the uwsgi of the Django project INI file. To start this file, you need to modify some previous commands, because the entry used is module

uwsgi --ini uwsgi.ini

New uwsgi INI file

[uwsgi]

socket = 127.0.0.1:6667
chdir = /path/to/django/project/[Yours django Project name]
module = [Yours django Project name].wsgi
master = true         
processes = 10
threads = 2
max-requests = 2000
chmod-socket = 664
pidfile= /path/to/django/project/[Yours django Project name]/[Yours django Project name].pid
vacuum = true
py-autoreload = 1
daemonize = /path/to/django/project/[Yours django Project name]/log/uwsgi.log

Common options:

http: protocol type and port number

Processes: number of open processes

Workers: the number of open processes, which is equivalent to processes (the official website says spawn the specified number of workers / processes)

Chdir: chdir to specified directory before apps loading

WSGI file: load WSGI file (load. WSGI file)

Stats: enable the stats server on the specified address

Threads: run threads. Due to the existence of GIL, I think this is really useless. (run each worker in prethreaded mode with the specified number of threads)

Master: enable master process

Daemon: make the process run in the background and print the log to the specified log file or udp server (daemon uwsgi). In fact, the most common is to output the operation records to a local file. (you need to create a log folder manually). Otherwise, an error is reported

pidfile: specify the location of the pid file and record the pid number of the main process.

vacuum: automatically clean up the environment when the server exits, and delete unix socket files and pid files (try to remove all of the generated files / sockets)

uwsgi --ini uwsgi.ini

3. Create nginx Conf configuration file

It is also in the root directory of django project (the same level as manage.py)

This is my configuration

events {
    worker_connections  1024;
}

http {
    include       /usr/local/nginx/conf/mime.types;
    default_type  application/octet-stream;
    sendfile        on;
    #tcp_nopush     on;
    #keepalive_timeout  0;
    keepalive_timeout  65;
    #gzip  on;
    server {
        listen 10110;
        server_name localhost;
        charset     utf-8;
        access_log      /path/to/django/project/[Yours django Project name]/log/nginx_access.log;
        error_log       /path/to/django/project/[Yours django Project name]/log/nginx_error.log;
        large_client_header_buffers 4 16k;     # Maximum number and size of buffers to read large client request headers
        client_max_body_size 10000m;             #Set the maximum request body size that nginx can handle.
        client_body_buffer_size 128k;          #Buffer size of the request body.
        proxy_connect_timeout 600;
        proxy_read_timeout 600;
        proxy_send_timeout 600;
        proxy_buffer_size 64k;
        proxy_buffers   4 32k;
        proxy_busy_buffers_size 64k;
        proxy_temp_file_write_size 64k;


        location /static {
                alias /path/to/django/project/[Yours django Project name]/static; # New static file collection address
        }

        location / {
                uwsgi_send_timeout 600;        # Specify the timeout for transmitting the request to uWSGI, and the timeout for transmitting the request to uWSGI after the handshake is completed.
                uwsgi_connect_timeout 600;
                uwsgi_read_timeout 600;
                include     /usr/local/nginx/conf/uwsgi_params;
                uwsgi_pass  127.0.0.1:8208;
        }
    }
}

Keep it simple

listen 88;　　　　　　　　　　　　　　　 # Port different from uwsgi setting
server_name www.wcwnina.com;　　　　　# Remember to add the mapping between IP and domain name in the / etc/hosts file of the system!

location / {
    include uwsgi_params;　　　　　　　# With nginx Conf same directory
    uwsgi_pass 192.168.1.2:8000;　　 # Consistent with socket in uwsgi configuration
}

location /static {
    alias /path/to/django/project/[Yours django Project name]/static;
}

Restart nginx after creation (modification)

nginx -s reload

4. Collect static files

It is also in the root directory of django project (the same level as manage.py)

python manage.py collectstatic

5. Run uwsgi again

It is also in the root directory of django project (the same level as manage.py)

uwsgi --ini uwsgi.ini

If you encounter problems, you need to check nginx in the log folder of the project directory_ error. log

You can use the following to see the last 10 lines and new information, and then specify Baidu according to specific problems

tail -f nginx_error.log

This article omits the installation requirements of pip and the problems encountered in installing other packages

You can start with the domain proxy pip proxy of python nginx pip Version (I won't be more difficult...)

I hope your process can be smooth sailing, otherwise the specific problems will be specific

If you encounter a problem, you need to restart, find the pid of the process, close it with kill-9, and then try again

ps -ef | grep [project name] or [: port number] ps: see the: before the port number

lsof -i: port number

kill -9 [pid]

Topics: Python Django Nginx Flask

Programmer Think

Nginx deployment Django Python virtual environment creation fool tutorial

1. First, you need to configure the virtual environment virtualenv on the server

2. Configure uwsgi ini

3. Create nginx Conf configuration file

4. Collect static files

5. Run uwsgi again

Hot Topics