vps_monitor.py

import paramiko
import schedule
import time
import os
import sys
from flask import Flask, jsonify, render_template_string
from threading import Thread
import logging
from datetime import datetime, timedelta

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
        
        username = os.environ.get(f'USERNAME_{index}')
        password = os.environ.get(f'PASSWORD_{index}')
        
        script_paths = []
        script_index = 1
        while True:
            script_path = os.environ.get(f'SCRIPT_PATHS_{index}_{script_index}')
            if not script_path:
                break
            script_paths.append(script_path.strip())
            script_index += 1
        
        for script_path in script_paths:
            config = {
                'index': index,
                'hostname': hostname,
                'username': username,
                'password': password,
                'script_path': script_path
            }
            configs.append(config)
        
        index += 1
    return configs

def check_and_run_script(config):
    logger.info(f"Checking VPS {config['index']}: {config['hostname']} - {config['script_path']}")
    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)
        key = f"{config['hostname']}:{script_name}"
        
        # 获取上次保存的 PID（如果有）
        last_pid = vps_status.get(key, {}).get('pid', None)
        
        # 检查进程是否在运行
        if last_pid:
            check_command = f"ps -p {last_pid} -o pid,etime,args"
        else:
            check_command = f"ps aux | grep '{script_path}' | grep -v grep"
        
        stdin, stdout, stderr = client.exec_command(check_command)
        output = stdout.read().decode('utf-8').strip()
        
        if output and (last_pid or script_path in output):
            parts = output.split()
            if last_pid:
                pid = last_pid
                runtime = parts[1] if len(parts) > 1 else "Unknown"
            else:
                pid = parts[1] if len(parts) > 1 else "Unknown"
                runtime = parts[9] if len(parts) > 9 else "Unknown"
            status = "Running"
            logger.info(f"Script {script_name} is running. PID: {pid}, Runtime: {runtime}")
        else:
            logger.info(f"Script {script_name} not running. Attempting to restart.")
            restart_command = f"nohup /bin/sh {script_path} > /dev/null 2>&1 & echo $!"
            stdin, stdout, stderr = client.exec_command(restart_command)
            new_pid = stdout.read().decode('utf-8').strip()
            
            if new_pid.isdigit():
                pid = new_pid
                runtime = "Just started"
                status = "Restarted"
                logger.info(f"Script {script_name} restarted. New PID: {pid}")
            else:
                pid = "N/A"
                runtime = "N/A"
                status = "Restart Failed"
                logger.error(f"Failed to restart script {script_name}")
        
        vps_status[key] = {
            'index': config['index'],
            'status': status,
            'last_check': time.strftime('%Y-%m-%d %H:%M:%S'),
            'username': config['username'],
            'script_name': script_name,
            'runtime': runtime,
            'pid': pid
        }
        
    except Exception as e:
        logger.error(f"Error occurred while checking VPS {config['index']} - {config['hostname']} - {script_name}: {str(e)}")
        key = f"{config['hostname']}:{script_name}"
        vps_status[key] = {
            'index': config['index'],
            'status': f"Error: {str(e)}",
            'last_check': time.strftime('%Y-%m-%d %H:%M:%S'),
            'username': config['username'],
            'script_name': script_name,
            'runtime': "N/A",
            'pid': "N/A"
        }
    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              | Script Name      | Status   | Last Check              | Username | Runtime  | PID   |\n"
    table += "+---------+-----------------------+------------------+----------+-------------------------+----------+----------+-------+\n"
    
    # 添加每个VPS的状态
    for key, status in vps_status.items():
        hostname, script_name = key.split(':')
        table += "| {:<7} | {:<21} | {:<16} | {:<8} | {:<23} | {:<8} | {:<8} | {:<5} |\n".format(
            status['index'],
            hostname[:21],
            script_name[:16],
            status['status'][:8],
            status['last_check'],
            status['username'][:8],
            status['runtime'][:8],
            status['pid'][:5]
        )
        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>Script Name</th>
            <th>Status</th>
            <th>Last Check</th>
            <th>Username</th>
            <th>Runtime</th>
            <th>PID</th>
        </tr>
        {% for key, data in vps_status.items() %}
        <tr>
            <td>{{ data.index }}</td>
            <td><a href="/status/{{ key }}">{{ key.split(':')[0] }}</a></td>
            <td>{{ data.script_name }}</td>
            <td>{{ data.status }}</td>
            <td>{{ data.last_check }}</td>
            <td>{{ data.username }}</td>
            <td>{{ data.runtime }}</td>
            <td>{{ data.pid }}</td>
        </tr>
        {% endfor %}
    </table>
    '''
    return render_template_string(html, vps_status=vps_status)

@app.route('/status/<path:key>')
def vps_status_detail(key):
    if key in vps_status:
        return jsonify(vps_status[key])
    else:
        return jsonify({"error": "VPS or script 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")

    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()