服务器监控应用（3）：监控告警通知开发 ​

import json
from datetime import datetime


def action_check_host_status(recipient_list=None, times=None):
    from django.conf import settings
    from django.core.mail import send_mail
    from .models import MonitorServer

    # 可以通过参数传递通知的频率
    times = times or [1, 10, 60, 60 * 4, 60 * 24]

    if not recipient_list:
        return 'No recipient_list, please set it.'

    if hasattr(settings, 'DEFAULT_FROM_EMAIL') and settings.DEFAULT_FROM_EMAIL:
        from_email = settings.DEFAULT_FROM_EMAIL
    else:
        # 如果未设置发件人邮箱，设置为空，直接退出
        return 'Email configuration not set'

    current_date = datetime.now()
    alarm_list = []
    hosts = MonitorServer.objects.filter(
        secret_key__isnull=False,
        secret_value__isnull=False,
        data__isnull=False,
        active=True
    )
    for host in hosts:
        # 转换成分钟
        m = int((current_date - host.update_date).total_seconds() / 60)
        # 多个时间点发送
        if m in times:
            msg = f'警告：节点 {host.name} 离线 {m} 分钟'
            alarm_list.append(msg)
        else:
            continue
    if all([alarm_list, from_email, recipient_list]):
        subject = f'⚠️服务监控告警 {current_date.strftime("%Y-%m-%d %H:%M:%S")}'
        message = '\n'.join(alarm_list)
        ok_num = send_mail(subject, message, from_email, recipient_list)
        return f"Send email ok: {ok_num}"

    return "Not alarm !!!"

@shared_task
def check_host_status(recipient_list=None, times=None):
    """
    定时检查服务监控的节点状态
    定时任务需要设置1分钟执行一次
    @param times: 通知频率，默认[1, 10, 60, 60 * 4, 60 * 24]
    @param recipient_list: 收件人的邮件地址，必填，否则不检查
    @return:
    """
    response = TaskResponse()
    msg = action_check_host_status(recipient_list=recipient_list, times=times)
    response.data = {'msg': msg}
    return response.as_dict()

# ****************************************** 邮箱配置开始 ****************************************
# 配置管理邮箱，服务出现故障会收到到邮件，环境变量值的格式：name|test@test.com 多组用户用英文逗号隔开
ADMINS = []
admin_email_user = os.getenv('IZONE_ADMIN_EMAIL_USER')
if admin_email_user:
    for each in admin_email_user.split(','):
        a_user, a_email = each.split('|')
        ADMINS.append((a_user, a_email))

# 邮箱配置
EMAIL_HOST = os.getenv('IZONE_EMAIL_HOST', 'smtp.163.com')
EMAIL_HOST_USER = os.getenv('IZONE_EMAIL_HOST_USER', 'your-email-address')
EMAIL_HOST_PASSWORD = os.getenv('IZONE_EMAIL_HOST_PASSWORD',
                                'your-email-password')  # 这个不是邮箱密码，而是授权码
EMAIL_PORT = os.getenv('IZONE_EMAIL_PORT', 465)  # 由于阿里云的25端口打不开，所以必须使用SSL然后改用465端口
EMAIL_TIMEOUT = 5
# 是否使用了SSL 或者TLS，为了用465端口，要使用这个
EMAIL_USE_SSL = os.getenv('IZONE_EMAIL_USE_SSL', 'True').upper() == 'TRUE'
# 默认发件人，不设置的话django默认使用的webmaster@localhost，所以要设置成自己可用的邮箱
DEFAULT_FROM_EMAIL = os.getenv('IZONE_DEFAULT_FROM_EMAIL', 'TendCode博客 <your-email-address>')
# *************************************** 邮箱配置结束 *******************************************

def action_check_host_status(recipient_list=None, times=None, ignore_hours=None):
    from django.conf import settings
    from django.core.mail import send_mail
    from .models import MonitorServer

    current_date = datetime.now()

    # 忽略的检查时段，这些时段不检查状态
    # 这个忽略的意义是因为运营商会定期断网更新IP，导致上报失败触发告警，比如电信是4点多断网一段时间
    ignore_hours = ignore_hours or []

    if current_date.hour in ignore_hours:
        return f'Ignore period for {ignore_hours}, do not check.'