Python Web项目的服务器部署

news2025/3/13 12:54:43

一.部署运行

1.虚拟环境的安装：（一行一行运行）

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh
bash miniconda.sh -b -p /opt/miniconda3
echo 'export PATH="/opt/miniconda3/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc
conda init bash
source ~/.bashrc
conda config --set auto_activate_base false

2. 系统依赖安装：（一行一行运行）

sudo yum update -y
sudo yum install -y epel-release
sudo yum install -y gcc gcc-c++ make
sudo yum install -y mariadb-server mariadb-devel nginx
sudo yum install -y git wget unzip
sudo yum install -y cairo-devel freetype-devel libpng-devel libjpeg-devel

3.项目环境安装：（灵活修改，以下参考）

mkdir -p /var/www/ershouche
cd /var/www/ershouche
conda create -n ershouche_env python=3.8 -y
conda activate ershouche_env

cat > requirements.txt << 'EOF'
# Web framework
flask>=2.0.1,<2.1.0
Werkzeug>=2.0.1,<2.1.0
Jinja2>=3.0.1,<3.1.0
itsdangerous>=2.0.1,<2.1.0
MarkupSafe>=2.0.1,<2.1.0
# Database
pymysql>=1.0.2
dbutils>=3.0.3
# Data processing and visualization
numpy>=1.20.0,<1.23.0
pandas>=1.3.0,<1.4.0
matplotlib>=3.4.0,<3.5.0
wordcloud>=1.8.1
statsmodels>=0.13.0,<0.14.0
# Web scraping
beautifulsoup4>=4.9.0
selenium>=3.141.0,<4.1.0
requests>=2.25.0,<2.27.0
lxml>=4.6.0
# Tools
python-dotenv>=0.19.0
pillow>=8.3.0
# WSGI server
gunicorn>=20.1.0
EOF

conda install -c conda-forge -y numpy pandas matplotlib pillow statsmodels lxml
pip install -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/

4.安装宝塔（上传文件，配置数据库等方便）

通用安装脚本：

if [ -f /usr/bin/curl ];then curl -sSO https://download.bt.cn/install/install_panel.sh;else wget -O install_panel.sh https://download.bt.cn/install/install_panel.sh;fi;bash install_panel.sh ed8484bec

上传文件略，数据库可以：

mysql -uroot -p
SELECT User, Host FROM mysql.user;
CREATE USER 'root'@'%' IDENTIFIED BY '123456';
GRANT ALL PRIVILEGES ON *.* TO 'root'@'%' WITH GRANT OPTION;
FLUSH PRIVILEGES;
navicat连接搞定。

5.直接运行项目：

5.项目运行
conda activate ershouche_env
cd py脚本启动目录
python app.py

二.性能优化

方案：

软件：Supervisor、Nginx

python库：Gunicorn、Gevent

vim pro.sh

#!/bin/bash

# 启用调试模式
set -e

# 获取用户输入
read -p "请输入服务器IP地址: " SERVER_IP
read -p "请输入项目工作目录: " WORK_DIR
echo "请输入项目启动命令（以EOF结束）："
RUN_COMMAND=""
while IFS= read -r line; do
    [[ $line == "EOF" ]] && break
    RUN_COMMAND+="$line"$'\n'
done

# 提取工作目录的最后一个目录名
PROJECT_NAME=$(basename "$WORK_DIR")

# 安装必要的软件包
echo "安装必要的软件包..."
sudo yum install -y epel-release
sudo yum update -y
sudo yum install -y nginx supervisor

# 安装 Python 包
echo "安装 Python 包..."
pip install gunicorn gevent

# 配置 Gunicorn
echo "配置 Gunicorn..."
cat > $WORK_DIR/gunicorn_config.py << EOF
bind = "127.0.0.1:8000"
workers = 4
worker_class = "gevent"
worker_connections = 1000
timeout = 120
loglevel = "debug"
errorlog = "/var/log/$PROJECT_NAME/gunicorn_error.log"
accesslog = "/var/log/$PROJECT_NAME/gunicorn_access.log"
capture_output = True
EOF

# 创建日志目录
sudo mkdir -p /var/log/$PROJECT_NAME
sudo chown -R $(whoami):$(whoami) /var/log/$PROJECT_NAME

# 创建启动脚本
echo "创建启动脚本..."
cat > $WORK_DIR/start.sh << EOF
#!/bin/bash
cd $WORK_DIR
$RUN_COMMAND
EOF
chmod +x $WORK_DIR/start.sh

# 配置 Supervisor
echo "配置 Supervisor..."
sudo tee /etc/supervisord.d/$PROJECT_NAME.ini << EOF
[program:$PROJECT_NAME]
command=$WORK_DIR/start.sh
directory=$WORK_DIR
user=root
autostart=true
autorestart=true
startsecs=10
startretries=3
stopwaitsecs=10
killasgroup=true
stopasgroup=true
redirect_stderr=true
stdout_logfile=/var/log/$PROJECT_NAME/supervisor_stdout.log
stderr_logfile=/var/log/$PROJECT_NAME/supervisor_stderr.log
EOF

# 配置 Nginx
echo "配置 Nginx..."
sudo tee /etc/nginx/conf.d/$PROJECT_NAME.conf << EOF
server {
    listen 80;
    server_name $SERVER_IP;

    access_log /var/log/nginx/$PROJECT_NAME_access.log;
    error_log /var/log/nginx/$PROJECT_NAME_error.log;

    location /static {
        alias $WORK_DIR/static;
        expires 30d;
    }

    location / {
        proxy_pass http://127.0.0.1:8000;
        proxy_set_header Host \$host;
        proxy_set_header X-Real-IP \$remote_addr;
        proxy_set_header X-Forwarded-For \$proxy_add_x_forwarded_for;
        proxy_set_header X-Forwarded-Proto \$scheme;
        
        proxy_connect_timeout 60s;
        proxy_read_timeout 60s;
        proxy_send_timeout 60s;
    }
}
EOF

# 启动服务
echo "启动服务..."
sudo systemctl restart supervisord
sudo systemctl enable supervisord
sudo systemctl restart nginx
sudo systemctl enable nginx

# 检查服务状态
echo "检查服务状态..."
sudo supervisorctl status $PROJECT_NAME
sudo systemctl status nginx

echo "部署完成！您可以通过 http://$SERVER_IP 访问您的应用。"

三. 其他操作

端口占用的处理：找到占用端口杀死它。

比如80：

sudo lsof -i :80
sudo kill <PID>

终止特定命令行启动的进程

ps aux | grep "python app.py"
pkill -f "python app.py"

注意事项

权限要求：操作系统资源（如低端口号）通常需 sudo。
精确终止：优先用 kill + PID（避免 pkill/fuser 误杀）。
网络分析：lsof -i 比 netstat 更详细（支持显示远程 IP）。
风险控制：fuser -k 和 pkill -f 可能影响其他进程，建议先确认匹配结果。