Zabbix 监控 tcp 连接状态的第二版(python+zabbix_sender)

前言

之前写过一篇文章 Zabbix设置自定义监控项之——监控tcp连接状态。
最近完善了一点东西，也算是对第一个版本的升级。
升级的内容：

1. 脚本通过 python 实现。
2. 监控项类型为 zabbix 采集器，使用 zabbix_sender 进行发送数据到服务端。
3. 增加了一个触发器，对 close_wait 状态的预警.

脚本

/var/lib/zabbix/ 这个目录要事先自己创建。

#!/usr/bin/env   python
# -*- coding: utf-8 -*-
# @Time    :  2020/3/12 9:29
# @Author  : dongjiaxiao
# @Email   : 
# @File    : tcp_connect_status_monitor.py
# @Desc    :
import sys
import os
import subprocess


def execute_cmd(cmd):
    """
    命令执行，并获取返回状态与执行结果
    :param cmd: 要执行的命令
    :return: 字典包含 执行的状态码和执行的正常和异常输出, 0为正常，1为异常
    """
    cmd_res = {'status': 1, 'stdout': '', 'stderr': ''}
    try:
        res = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
        res_stdout, res_stderr = res.communicate()
        cmd_res['status'] = res.returncode
        cmd_res['stdout'] = res_stdout.replace('\n', '').strip()
        cmd_res['stderr'] = res_stderr.replace('\n', '').strip()
    except Exception as e:
        cmd_res['stderr'] = e
    finally:
        return cmd_res


def tcp_connect_status_value(zabbix_conf_path,tcp_conn_status_cmd):
    """
    发送数据到 zabbix 服务端，监控项类型需为 zabbix 采集器
    :param zabbix_conf_path: zabbix conf 路径
    :param tcp_connect_status_cmd: 获取tcp连接信息的执行命令
    :return: send_data_flg['status'] 发送数据是否成功 ，1为成功，0为失败，send_data_flg['error_info'] 为错误信息。
    """
    tcp_conn_status_dic = {
        'CLOSED': 0,
        'CLOSE_WAIT': 0,
        'CLOSING': 0,
        'ESTABLISHED': 0,
        'FIN_WAIT1': 0,
        'FIN_WAIT2': 0,
        'LAST_ACK': 0,
        'LISTEN': 0,
        'SYN_RECV': 0,
        'SYN_SENT': 0,
        'TIME_WAIT': 0,
    }  # tcp 连接状态码
    tmp_file = "/var/lib/zabbix/zabbix_sender_tcp_connect_status_file.txt"  # 临时文件，用完会删除
    send_data_flg = {'status': 0, 'error_info': ''}  # 发送数据状态与信息记录
    sender_bin_ret = execute_cmd("which zabbix_sender")
    if sender_bin_ret['status'] == 0:  # 判断是否安装 zabbix_sender 是否存在
        zabbix_sender_bin = sender_bin_ret['stdout']
    else:
        send_data_flg['error_info'] = "zabbix sender 未安装" + sender_bin_ret['stderr']
        return send_data_flg
    tcp_connect_status_ret = execute_cmd(tcp_conn_status_cmd)  # 执行命令获取tcp连接信息
    if tcp_connect_status_ret['status'] == 0:
        tcp_connect_status_data = tcp_connect_status_ret['stdout'].split()
        front_data = 0  # 记录列表当前数据的前一个数据，因为 tcp 数量的值在状态的前一行。
        for tcp_status in list(tcp_conn_status_dic.keys()):  # 遍历tcp状态码然后进行对比
            for tcp_status_data in tcp_connect_status_data:
                if tcp_status == tcp_status_data:
                    tcp_conn_status_dic[tcp_status] = front_data
                front_data = tcp_status_data
        with open(tmp_file, mode='w') as f_zs:  # 将要发送的数据写入临时文件
            for status, number in tcp_conn_status_dic.items():
                f_zs.write("- custom.tcp.connect.status[{0}] {1}  \n".format(status.lower(), number))
        sender_data_ret = execute_cmd('{0} -c {1} -i {2}'.format(zabbix_sender_bin, zabbix_conf_path, tmp_file)) # 发送到 zabbix server
        if sender_data_ret['status'] == 0 and sender_data_ret['stdout'].find(
                "failed: 0") >= 0:  # 当命令执行成功，并且找到"failed: 0"
            send_data_flg['status'] = 1
            os.remove(tmp_file)  # 删除临时文件
        else:
            send_data_flg['error_info'] = sender_data_ret['stdout'] + " or  failed is not eq 0"
    else:
        send_data_flg['error_info'] = "{0} 执行失败 ，result:{1}".format(tcp_conn_status_cmd, tcp_connect_status_ret['stderr'])
    return send_data_flg


if __name__ == '__main__':
    tcp_conn_status_cmd = "netstat -ant|awk '{print $NF}' |sort |uniq -c"  # 获取连接状态的信息
    zabbix_conf_path = '/etc/zabbix/zabbix_agentd.conf '    # zabbix 配置文件
    if len(sys.argv) == 2 and sys.argv[1] == "sender_data":  # 发送监控数据到服务端，单独监控项判断是否发送成功
        sender_status = tcp_connect_status_value(zabbix_conf_path, tcp_conn_status_cmd)
        print(sender_status['status'])
        # print(sender_status['error_info'])  # 打印错误信息，便于调试。
    else:
        print("please  use parameter  sender_data")

配置文件

UserParameter=custom.tcp.connect.status,python /etc/zabbix/scripts/tcp_connect_status_monitor.py sender_data

模板文件

    4.0
    2020-03-13T10:24:49Z
    
        
            Templates
        
    
    
        
    
    
        
            {Template_Linux_TCP_Connect_Status_Monitor:custom.tcp.connect.status.count(#3,0,eq)}=3
            1
            {Template_Linux_TCP_Connect_Status_Monitor:custom.tcp.connect.status.last()}=1
            Linux TCP Connect Status Sender Data Failed
            0
            
            
            0
            3
            
            0
            1
            
            
        
        
            {Template_Linux_TCP_Connect_Status_Monitor:custom.tcp.connect.status[close_wait].avg(1h)}>=500
            0
            
            There are too many TCP CLOSE_WAIT status
            0
            
            
            0
            2
            
            0
            1
            
            
        
        
            {Template_Linux_TCP_Connect_Status_Monitor:custom.tcp.connect.status[time_wait].last()}>10000
            0
            
            There are too many TCP TIME_WAIT status
            0
            
            
            0
            4
            
            0
            0
            
            
        
    
    
        
            TCP Status
            900
            200
            0.0000
            100.0000
            1
            1
            0
            1
            0
            0.0000
            0.0000
            0
            0
            0
            0
            
                
                    0
                    0
                    C80000
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[closed]
                    
                
                
                    1
                    0
                    00C800
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[close_wait]
                    
                
                
                    2
                    0
                    0000C8
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[closing]
                    
                
                
                    3
                    0
                    C800C8
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[established]
                    
                
                
                    4
                    0
                    00C8C8
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[fin_wait1]
                    
                
                
                    5
                    0
                    C8C800
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[fin_wait2]
                    
                
                
                    6
                    0
                    C8C8C8
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[last_ack]
                    
                
                
                    7
                    0
                    960000
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[listen]
                    
                
                
                    8
                    0
                    009600
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[syn_recv]
                    
                
                
                    9
                    0
                    000096
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[syn_sent]
                    
                
                
                    10
                    0
                    960096
                    0
                    2
                    0
                    
                        Template_Linux_TCP_Connect_Status_Monitor
                        custom.tcp.connect.status[time_wait]