Python爬虫监控程序设计指南

import time
import logging
import psutil
from prometheus_client import start_http_server, Counter, Gauge, Histogram
​
class SpiderMonitor:
    def __init__(self, spider_name):
        self.spider_name = spider_name
        self.start_time = time.time()
        
        # 初始化监控指标
        self.requests_total = Counter(f'{spider_name}_requests_total', 'Total requests')
        self.requests_failed = Counter(f'{spider_name}_requests_failed', 'Failed requests')
        self.items_scraped = Counter(f'{spider_name}_items_scraped', 'Items scraped')
        self.memory_usage = Gauge(f'{spider_name}_memory_usage', 'Memory usage (MB)')
        self.request_latency = Histogram(f'{spider_name}_request_latency', 'Request latency (seconds)')
        
        # 启动指标服务器
        start_http_server(8000)
        logging.basicConfig(filename=f'{spider_name}.log', level=logging.INFO)
​
    def record_request(self, success=True, latency=0):
        self.requests_total.inc()
        if not success:
            self.requests_failed.inc()
        if latency > 0:
            self.request_latency.observe(latency)
    
    def record_item(self, count=1):
        self.items_scraped.inc(count)
    
    def update_resources(self):
        process = psutil.Process()
        self.memory_usage.set(process.memory_info().rss / 1024 / 1024)  # MB
    
    def log_error(self, error):
        logging.error(f"[{time.ctime()}] ERROR: {error}")
    
    def uptime(self):
        return time.time() - self.start_time

import requests
from monitor import SpiderMonitor
​
class MySpider:
    def __init__(self):
        self.monitor = SpiderMonitor("my_spider")
        self.session = requests.Session()
    
    def crawl(self, url):
        start = time.time()
        try:
            response = self.session.get(url, timeout=10)
            response.raise_for_status()
            
            # 处理数据
            items = self.parse(response)
            self.monitor.record_item(len(items))
            
            self.monitor.record_request(success=True, latency=time.time()-start)
            return items
        
        except Exception as e:
            self.monitor.record_request(success=False)
            self.monitor.log_error(f"URL: {url} - Error: {str(e)}")
            return []
    
    def parse(self, response):
        # 解析逻辑
        return [{"data": "sample"}]
    
    def run(self):
        while True:
            self.crawl("https://example.com/data")
            self.monitor.update_resources()
            time.sleep(5)
​
if __name__ == "__main__":
    spider = MySpider()
    spider.run()

import time
import subprocess
import smtplib
from email.mime.text import MIMEText
​
def check_heartbeat(spider_name):
    """检查最近15分钟是否有活动日志"""
    try:
        with open(f"{spider_name}.log") as f:
            logs = f.readlines()[-100:]
            return any(time.time() - get_log_time(line) < 900 for line in logs)
    except FileNotFoundError:
        return False
​
def get_log_time(log_line):
    # 从日志行提取时间戳
    timestamp_str = log_line.split("]")[0][1:]
    return time.mktime(time.strptime(timestamp_str))
​
def send_alert(subject, message):
    """发送邮件告警"""
    msg = MIMEText(message)
    msg['Subject'] = f"[SPIDER ALERT] {subject}"
    msg['From'] = 'monitor@example.com'
    msg['To'] = 'admin@example.com'
    
    with smtplib.SMTP('smtp.example.com') as server:
        server.send_message(msg)
​
def monitor_daemon():
    spider_name = "my_spider"
    consecutive_failures = 0
    
    while True:
        if not check_heartbeat(spider_name):
            consecutive_failures += 1
            if consecutive_failures >= 3:
                send_alert("Spider Down", 
                          f"{spider_name} has been inactive for 45+ minutes")
        else:
            consecutive_failures = 0
        
        time.sleep(300)  # 每5分钟检查一次
​
if __name__ == "__main__":
    monitor_daemon()