使用Python开发一个Prometheus exporter

UOUUYd · 2025-2-21 20:04:23

需求：需要监控指定的邮箱发送邮件是否正常，使用 Prometheus 官方的 Python sdk，代码如下：
代码

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40

from prometheus_client import start_http_server, Gauge
import smtplib
import argparse
import time

# 创建一个 Gauge 类型的指标
EMAIL_AUTHENTICATION_SUCCESS = Gauge('email_authentication_success', 'Indicates if the email authentication was successful', ['email'])

def check_email_authentication(email, password, smtp_server, smtp_port):
try:
      server = smtplib.SMTP_SSL(smtp_server, smtp_port)
      server.login(email, password)
      # 如果登录成功，则更新指标为 1
      EMAIL_AUTHENTICATION_SUCCESS.labels(email=email).set(1)
      server.quit()
except Exception as e:
      # 如果登录失败，则更新指标为 0
      EMAIL_AUTHENTICATION_SUCCESS.labels(email=email).set(0)
      print(f"Failed to authenticate email {email}: {e}")

if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Email Authentication Prometheus Exporter')
parser.add_argument('--email', required=True, help='Email address to authenticate')
parser.add_argument('--password', required=True, help='Password for the email account')
parser.add_argument('--smtp_server', default='smtp.gmail.com', help='SMTP server address')
parser.add_argument('--smtp_port', type=int, default=587, help='SMTP server port')

args = parser.parse_args()

# 启动 Prometheus HTTP 服务
start_http_server(8000)

print("Prometheus exporter is running on port 80. Use CTRL+C to exit.")

# 让程序保持运行
while True:
      # 进行邮箱验证
      check_email_authentication(args.email, args.password, args.smtp_server, args.smtp_port)
      # 等待一小时，频率太快容易被ban？
      time.sleep(3600)

展示

邮件登录正常：

断开网络，或者修改邮箱密码：

Prometheus 界面查询

使用方式

安装依赖包

1	pip3 install prometheus-client

1 2	# 启动 exporter python3.8 email_exporter.py --email qq@qq.com --password xxx --smtp_server smtp.feishu.cn --smtp_port 465

部署到 Kubernetes

Dockerfile

1
2
3

FROM python:3.9-slim
ADD email-exporter.py /
RUN pip3 install prometheus-client

yaml

部署到 Kubernetes，借助 Prometheus 服务发现将 service 的 metrics 数据接入

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51

apiVersion: apps/v1
kind: Deployment
metadata:
  name: email-exporter
  namespace: ops
spec:
  replicas: 1
  selector:
matchLabels:
   app: email-exporter
  template:
metadata:
   labels:
      app: email-exporter
spec:
   containers:
      - image: registry.cn-hangzhou.aliyuncs.com/iuxt/email-exporter:1
      name: email-exporter
      command:
         - "bash"
      args:
         - "-c"
         - "python3 -u /email-exporter.py --email qq@qq.com --password xxx --smtp_server smtp.feishu.cn --smtp_port 465"
      resources:
         limits:
            cpu: 200m
            memory: 200Mi
         requests:
            cpu: 200m
            memory: 200Mi
---
apiVersion: v1
kind: Service
metadata:
  annotations:
prometheus.io/env: dev
prometheus.io/path: /
prometheus.io/port: "80"
prometheus.io/project: idp
prometheus.io/scrape: "true"
  name: email-exporter
  namespace: ops
spec:
  ports:
- port: 80
   protocol: TCP
   targetPort: 80
  selector:
app: email-exporter
  type: ClusterIP

其他问题

python 容器化部署不产生控制台日志：

1 2	# 在部署python的容器服务时，docker logs/kubectl logs获取不到的主要原因是print的标准控制台输出，产生了buffer，导致控制台输出不及时。解决该缓存问题或者禁止buffer即可 python -u xx.py