Files
aler_dashboard_git/alert_dashboard/python_crawler/mail.py
T

151 lines
5.0 KiB
Python
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
import poplib
import ssl
import email
from email.header import decode_header
import time
import pymysql
# 关闭SSL校验,解决老Python握手重置
#ssl._create_default_https_context = ssl._create_unverified_context
# ========== 邮箱配置 自行修改 ==========
MAIL_HOST = "pop.163.com"
MAIL_PORT = 995
MAIL_USER = "lyudream@163.com"
MAIL_PWD = "DVZxyF5NKCkR6fL5"
# ========== 本地MariaDB配置 ==========
DB_HOST = "127.0.0.1"
DB_USER = "root"
DB_PWD = "hp93000"
DB_NAME = "alert_mail_stat"
# 解码邮件标题、正文
def decode_str(s):
value, charset = decode_header(s)[0]
if charset:
value = value.decode(charset)
return value
# 获取邮件正文
def get_email_content(msg):
content = ""
if msg.is_multipart():
for part in msg.walk():
content_type = part.get_content_type()
content_disposition = str(part.get("Content-Disposition"))
if content_type == 'text/plain' and 'attachment' not in content_disposition:
payload = part.get_payload(decode=True)
charset = part.get_charset()
if charset is None:
charset = 'utf-8'
content = payload.decode(charset, errors='ignore')
break
else:
payload = msg.get_payload(decode=True)
charset = msg.get_charset()
if charset is None:
charset = 'utf-8'
content = payload.decode(charset, errors='ignore')
return content
# 简单解析告警关键字(适配Alertmanager中文邮件模板:告警名称、实例、级别)
def parse_alert_info(text):
alert_name = ""
instance = ""
severity = "unknown"
lines = text.splitlines()
for line in lines:
if "告警名称" in line:
alert_name = line.split("")[-1].strip()
if "实例" in line:
instance = line.split("")[-1].strip()
if "级别" in line:
s_val = line.split("")[-1].strip()
if s_val.lower() in ["critical", "严重"]:
severity = "critical"
elif s_val.lower() in ["warning", "警告"]:
severity = "warning"
return alert_name, instance, severity
# 主抓取逻辑
def crawl_mail():
# 连接数据库
db = pymysql.connect(host=DB_HOST, user=DB_USER, password=DB_PWD, database=DB_NAME, charset='utf8mb4')
cursor = db.cursor()
# 连接POP3邮箱
pop_conn = poplib.POP3_SSL(MAIL_HOST, MAIL_PORT)
pop_conn.user(MAIL_USER)
pop_conn.pass_(MAIL_PWD)
total_num, total_size = pop_conn.stat()
print(f"当前邮箱总邮件数量: {total_num}")
if total_num == 0:
print("暂无新邮件")
pop_conn.quit()
db.close()
return
# 从最新一封开始遍历
for idx in range(total_num, 0, -1):
# 获取邮件原始内容
resp, raw_lines, octet = pop_conn.retr(idx)
raw_msg = b'\r\n'.join(raw_lines).decode('utf-8', errors='ignore')
msg = email.message_from_string(raw_msg)
# 生成唯一标识去重
mail_uid = msg.get("Message-ID", str(time.time()))
# 查重,避免重复入库
cursor.execute("SELECT id FROM alert_log WHERE mail_uid=%s", (mail_uid,))
if cursor.fetchone():
print(f"邮件{idx}已存在数据库,跳过")
continue
# 解析标题、发件人
subject = decode_str(msg.get("Subject", ""))
from_addr = msg.get("From", "")
# ===================== 临时关闭发件人过滤,全部邮件放行测试 =====================
# if "prometheusalert" not in from_addr.lower():
# print(f"邮件{idx}发件人不匹配,跳过,发件人:{from_addr}")
# continue
# 调试打印所有邮件基础信息
print("\n========================================")
print(f"【调试】正在处理第 {idx} 封邮件")
print(f"发件人:{from_addr}")
print(f"邮件标题:{subject}")
print("========================================")
# 正文解析告警字段
body = get_email_content(msg)
alert_name, instance, severity = parse_alert_info(body)
if not alert_name:
print(f"邮件{idx}未识别到【告警名称】关键字,跳过")
continue
# 区分触发/恢复告警
alert_type = 1
if "恢复" in subject or "resolved" in subject.lower():
alert_type = 2
now = time.strftime("%Y-%m-%d %H:%M:%S")
sql = """
INSERT INTO alert_log
(mail_uid, alert_type, alert_name, instance, severity, starts_at, ends_at, content, receive_time)
VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s)
"""
cursor.execute(sql, (mail_uid, alert_type, alert_name, instance, severity, now, None, body, now))
db.commit()
print(f"入库成功 | 告警:{alert_name} 实例:{instance}")
pop_conn.quit()
cursor.close()
db.close()
print("\n本次抓取全部完成")
if __name__ == "__main__":
crawl_mail()