本文主要集中在huggingface保活,在serv00部署的所有服务都可以参考下面方式进行保活
推荐使用Cloudflare更加稳定
由于在serv00部署的服务使用pm2保活存在进程被清理的情况,所以我之后在serv00中加入了restart.sh 自动重启脚本配合cron来保活,但是这种方法依然被清理进程。既然在serv00内部保活存在定期服务被清理的情况,我们可以选择通过外部连接serv00并执行restart.sh,从而保活serv00
该脚本需要结合restart.sh使用,你…
1.开放一个端口并开启应用权限
2.将域名托管到serv00并创建Website
域名托管不在这里叙述,可以看这部分,也可以使用CF进行托管
创建Website
3. 创建mysql数据库
记住你创建的数据库,等会需要用
4.下载Alist
进入 public_html
文件下
cd /usr/home/你的serv名字/domains/域名/public_html
下载alist包
wget https://github.com/uubulb/alist-freebsd/releases/download/v3.36.0/alist
5.进入文件管理器后台创建并编辑 data/config.json
文件
在下载alist
同级目录创建data
文件,并进入data文件在其下创建config.json
文本
把代码放入其中
{ "force": false, "site_url": "", "cdn": "", "jwt_secret": "random_generated", "token_expires_in": 48, "database": { "type": "sqlite3", "host": "", "port": 0, "user": "", "password": "", "name": "", "db_file": "data\\data.db", "table_prefix": "x_", "ssl_mode": "", "dsn": "" }, "meilisearch": { "host": "http://localhost:7700", "api_key": "", "index_prefix": "" }, "scheme": { "address": "0.0.0.0", "http_port": 5244, "https_port": -1, "force_https": false, "cert_file": "", "key_file": "", "unix_file": "", "unix_file_perm": "" }, "temp_dir": "data\\temp", "bleve_dir": "data\\bleve", "dist_dir": "", "log": { "enable": true, "name": "data\\log\\log.log", "max_size": 50, "max_backups": 30, "max_age": 28, "compress": false }, "delayed_start": 0, "max_connections": 0, "tls_insecure_skip_verify": true, "tasks": { "download": { "workers": 5, "max_retry": 1 }, "transfer": { "workers": 5, "max_retry": 2 }, "upload": { "workers": 5, "max_retry": 0 }, "copy": { "workers": 5, "max_retry": 2 } }, "cors": { "allow_origins": [ "*" ], "allow_methods": [ "*" ], "allow_headers": [ "*" ] }, "s3": { "enable": false, "port": 5246, "ssl": false } }
修改
database": { "type": "sqlite3", "host": "", "port": 0, "user": "", "password": "", "name": "",
"site_url": "",
为你你自己的:格式:https://xxx.com
type
改为mysql
,余信息填写你前面创建的mysql数据库"http_port": 5244,
改为你自己放行的端口
5.进行alist调试并获取密码
赋权并执行alist
chmod +x alist && ./alist server
获取你的密码并保存(只出现一次
)
进入后台并更改你的账号密码
初始账号:admin
密码:前面获取到的
下面是保活的重点
6.创建start.sh文件
在alist同级目录创建start.sh
文件,这个是启动文件,执行./alist server
命令(因为alist本身不自带,需要自行创建 )其余项目类似
1.开放一个端口并开启应用权限
[image]
[image]
2.将域名托管到serv00并创建Website
域名托管不在这里叙述,可以看这部分,也可以使用CF进行托管
创建Website
[image]
3.创建mysql数据库
记住你创建的数据库,等会需要用
[image]
4.安装One-API
进入 public_html 文件下
cd /usr/home/你…
1.开放一个端口并开启应用权限
[image]
[image]
2.将域名托管到serv00并创建Website
域名托管不在这里叙述,可以看这部分,也可以使用CF进行托管
创建Website
[image]
3.创建mysql数据库
记住你创建的数据库,等会需要用
[image]
4.安装halo
进入你的域名目录并创建halo文件
cd /usr/home/你的serv…
#!/bin/bash cd /usr/home/Serv00账号名/domains/域名/public_html chmod +x alist ./alist server
路径对应你自己的,并将其中的serv00
和域名
换为你自己的,下面同理
7.创建restart.sh文件
在alist同级目录创建restart.sh
文件,这个文件是保活文件,用于检测start.sh
是否运行,没运行则重启服务
#!/bin/bash # 设置脚本路径 SCRIPT_PATH="/usr/home/Serv00账号名/domains/域名/public_html/start.sh" WORK_DIR="/usr/home/Serv00账号名/domains/域名/public_html" # 检查脚本是否在运行 if ! pgrep -f "$SCRIPT_PATH" > /dev/null then # 如果脚本没有运行,则重新启动它 cd "$WORK_DIR" nohup ./start.sh > /dev/null 2>&1 & echo "Restarted start.sh at $(date)" >> "$WORK_DIR/restart_log.txt" fi
8.赋权所创建的脚本
chmod +x start.sh restart.sh
9.在CF上保活
9.进入huggingface创建一个空间(不稳定)
放入以下代码:
FROM python:3.8-slim-buster WORKDIR /app RUN apt-get update && apt-get install -y openssh-client RUN pip install paramiko schedule flask COPY vps_monitor.py . EXPOSE 8080 CMD ["python", "-u", "vps_monitor.py"]
编辑 README.md,
在README.md
添加app_port: 8080
创建一个名为vps_monitor.py
的文本
放入以下代码并保存
import paramiko import schedule import time import os import sys from flask import Flask, jsonify, render_template_string from threading import Thread import logging app = Flask(__name__) vps_status = {} # 设置日志 logging.basicConfig( level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s', handlers=[ logging.StreamHandler(sys.stdout), logging.StreamHandler(sys.stderr) ] ) logger = logging.getLogger() def get_vps_configs(): configs = [] index = 1 while True: hostname = os.environ.get(f'HOSTNAME_{index}') if not hostname: break config = { 'index': index, 'hostname': hostname, 'username': os.environ.get(f'USERNAME_{index}'), 'password': os.environ.get(f'PASSWORD_{index}'), 'script_path': os.environ.get(f'SCRIPT_PATH_{index}') } configs.append(config) logger.info(f"Config {index}: {config['hostname']}, {config['username']}, {config['script_path']}") index += 1 return configs def check_and_run_script(config): logger.info(f"Checking VPS {config['index']}: {config['hostname']}") client = None try: client = paramiko.SSHClient() client.set_missing_host_key_policy(paramiko.AutoAddPolicy()) client.connect( hostname=config['hostname'], username=config['username'], password=config['password'], port=22 ) script_path = config['script_path'] script_name = os.path.basename(script_path) check_command = f"ps -eo args | grep {script_name} | grep -v grep" stdin, stdout, stderr = client.exec_command(check_command) output = stdout.read().decode('utf-8').strip() if output and script_path in output: status = "Running" logger.info(f"Script is running on {config['hostname']}") else: logger.info(f"Script not running on {config['hostname']}. Executing restart script.") restart_command = f"nohup /bin/sh {script_path} > /dev/null 2>&1 &" stdin, stdout, stderr = client.exec_command(restart_command) status = "Restarted" logger.info(f"Script restarted on {config['hostname']}") vps_status[config['index']] = { 'index': config['index'], 'hostname': config['hostname'], 'status': status, 'last_check': time.strftime('%Y-%m-%d %H:%M:%S'), 'username': config['username'] } except Exception as e: logger.error(f"Error occurred while checking VPS {config['index']} - {config['hostname']}: {str(e)}") vps_status[config['index']] = { 'index': config['index'], 'hostname': config['hostname'], 'status': f"Error: {str(e)}", 'last_check': time.strftime('%Y-%m-%d %H:%M:%S'), 'username': config['username'] } finally: if client: client.close() def check_all_vps(): logger.info("Starting VPS check") vps_configs = get_vps_configs() for config in vps_configs: check_and_run_script(config) # 创建表格头 table = "+---------+-----------------------+----------+-------------------------+----------+\n" table += "| Index | Hostname | Status | Last Check | Username |\n" table += "+---------+-----------------------+----------+-------------------------+----------+\n" # 添加每个VPS的状态 for index in sorted(vps_status.keys()): status = vps_status[index] table += "| {:<7} | {:<21} | {:<8} | {:<23} | {:<8} |\n".format( status['index'], status['hostname'][:21], status['status'][:8], status['last_check'], status['username'][:8] ) table += "+---------+-----------------------+----------+-------------------------+----------+\n" logger.info("\n" + table) @app.route('/') def index(): html = ''' <h1>VPS Status Overview</h1> <table border="1"> <tr> <th>Index</th> <th>Hostname</th> <th>Status</th> <th>Last Check</th> <th>Username</th> </tr> {% for index in vps_status.keys()|sort %} {% set data = vps_status[index] %} <tr> <td>{{ data['index'] }}</td> <td><a href="/status/{{ data['index'] }}">{{ data['hostname'] }}</a></td> <td>{{ data['status'] }}</td> <td>{{ data['last_check'] }}</td> <td>{{ data['username'] }}</td> </tr> {% endfor %} </table> ''' return render_template_string(html, vps_status=vps_status) @app.route('/status/<int:index>') def vps_status_detail(index): if index in vps_status: return jsonify(vps_status[index]) else: return jsonify({"error": "VPS not found"}), 404 @app.route('/health') def health_check(): return jsonify({"status": "healthy", "uptime": time.time() - start_time}), 200 def run_flask(): app.run(host='0.0.0.0', port=8080) def main(): global start_time start_time = time.time() logger.info("===== VPS monitoring script is starting =====") flask_thread = Thread(target=run_flask) flask_thread.start() logger.info("Flask server started in background") vps_configs = get_vps_configs() logger.info(f"Found {len(vps_configs)} VPS configurations") logger.info("Running initial VPS check") check_all_vps() schedule.every(15).minutes.do(check_all_vps) logger.info("Scheduled VPS check every 15 minutes") logger.info("===== VPS monitoring script is running =====") heartbeat_count = 0 while True: schedule.run_pending() time.sleep(60) heartbeat_count += 1 if heartbeat_count % 5 == 0: # 每5分钟输出一次心跳信息 logger.info(f"Heartbeat: Script is still running. Uptime: {heartbeat_count} minutes") if __name__ == "__main__": main()
9.在后台填写变量
USERNAME_X HOSTNAME_X PASSWORD_X SCRIPT_PATH_X
X为数字编号 如(1,2,3),每个主机4项要相同才能对应
如:
USERNAME_1 HOSTNAME_1 PASSWORD_1 SCRIPT_PATH_1
USERNAME
为serv00用户名,HOSTNAME
为主机名(如:s3.serv00.com
),PASSWORD
为serv00密码,SCRIPT_PATH
为执行脚本的绝对路径(如:/usr/home/serv00用户名/domains/域名/public_html/restart.sh
)
running
后点击logs
可以查看进度
点击app
可看到服务器正常启动,这时你可以访问你的网址进入你部署的项目
在huggingface部署的重启脚本避免了后台被清除,因为他是远程访问的
他可以保活账号并且保活运行的程序,可以多个账号一起保活,前面的4个变量序号一致就行
参考项目