Question

我正在开发一种管理nagios的自动化工具（nagiosql对我来说还不够）。

当我与nagios进行集成测试时，会发生一些奇怪的事情。

情况如下：

Nagios配置

# grep log_notification /usr/local/nagios/etc/nagios.cfg
log_notifications=1

#grep enable_notifications /usr/local/nagios/etc/nagios.cfg
enable_notifications=1

我创建了一个示例联系人，主持人，服务，所有这些都没有使用模板

define contact { 
    contact_name    jone.doe
    host_notifications_enabled  1
    service_notifications_enabled   1
    host_notification_period    7x24
    service_notification_period 7x24
    host_notification_options   d,u,r,f,s,n
    service_notification_options    w,u,c,r,f,s,n
    email   john.doe@xx.com
    host_notification_commands  notify_service_by_state
    service_notification_commands   notify_service_by_state
}

define host{
    host_name   localhost
    address localhost
    max_check_attempts  1
    check_period    7x24
    active_checks_enabled   1
    passive_checks_enabled  0
    check_command   check_host_alive
}

define service{
    host_name   localhost
    service_description check_the_process_count_of_the_local_machine
    check_command   check_local_procs!10!20!RSZDT
    max_check_attempts  3
    check_interval  3
    retry_interval  1
    notification_interval   3
    check_period    7x24
    notification_period 7x24
    notifications_enabled   1
    contacts   john.doe
}

以下是一些额外的配置文件：

define timeperiod{
    timeperiod_name 7x24
    alias   7x24
    monday  00:00-24:00
    tuesday 00:00-24:00
    wednesday   00:00-24:00
    thursday    00:00-24:00
    friday  00:00-24:00
    saturday    00:00-24:00
    sunday  00:00-24:00
}

define command { 
    command_name    check_local_procs
    command_line    $USER1$/check_procs -w $ARG1$ -c $ARG2$ -s $ARG3$
}

define command { 
    command_name    notify_service_by_state
    command_line    /usr/local/nagios/etc/scripts/notify.py "$SERVICESTATE$" "$CONTACTEMAIL$" "$CONTACTADDRESS1$" "$NOTIFICATIONTYPE$" "$HOSTALIAS$" "$SERVICEDESC$" "$HOSTADDRESS$" "$SERVICEOUTPUT$" "$LONGDATETIME$"
}

define command { 
    command_name    check_host_alive
    command_line    $USER1$/check_ping -H $HOSTADDRESS$ -w 3000.0,80% -c 5000.0,100% -p 5
}

这是notify.py源代码，安装了所有软件包。

#!/usr/bin/python
#coding:utf-8
import codecs
import sys
import urllib
import commands

import smtplib
from email.mime.text import MIMEText
from email.header import Header

import time

fi = codecs.open('/data/appdatas/logs/message.log', 'a' , encoding='utf-8')
fi.write("=======Begin Log\n");
fi.write("Notify\n");
fi.write(time.strftime('%Y-%m-%d %H:%M:%S') + "\n");
fi.write(sys.argv[1]+ "\n");
fi.write("End Log=========\n");
fi.close()

为了简化测试过程，我chmod +777已log_file。

chmod +x notify.py。我已经测试过在终端中执行python脚本，并且日志文件中有日志。

正如您可以看到check_local_proc命令。该命令将在执行时返回代码2.

在nagios.log文件中，我看到了这些日志：

[1449588806] SERVICE ALERT: localhost;check_the_process_count_of_the_local_machine;CRITICAL;SOFT;1;PROCS CRITICAL: 423 processes with STATE = RSZDT
[1449588866] SERVICE ALERT: localhost;check_the_process_count_of_the_local_machine;CRITICAL;SOFT;2;PROCS CRITICAL: 423 processes with STATE = RSZDT
[1449588926] SERVICE ALERT: localhost;check_the_process_count_of_the_local_machine;CRITICAL;HARD;3;PROCS CRITICAL: 423 processes with STATE = RSZDT

从日志中可以看到，此服务每1分钟检查一次。

但是没有登录message.log文件（没有执行notify.py）。

Answer 1

我知道为什么我不能发送任何通知。因为我错误地配置了联系人的配置。根据官方文件：Nagios Object Definitions，字段host_notification_options [d,u,r,f,s,n]，service_notification_options [w,u,c,r,f,s,n];选项n表示联系人不会收到任何通知。

Nagios无法发送通知

1 个答案: