随笔- 49 文章- 49 评论- 0 阅读- 45319

python 实现远程监控服务器内存，磁盘，cpu各项指标

点击查看代码

#！/usr/local/bin/python3
#-*-coding:utf-8-*-

import requests
import subprocess,json
import paramiko
import socket


NORMAL=0
TIMEOUT=2

#远程监控主机
ip_list= ["29.2.210.202","29.2.210.203","29.2.210.204","29.2.210.235","29.2.210.238","29.2.210.239","29.23.177.31","29.23.177.32",\
        "29.3.218.207","29.3.218.212","29.2.209.216","29.2.209.220","29.3.218.208","29.3.218.213"]

#检测服务器端口函数
def port_check(port,host_list,timeout=TIMEOUT)
  down_list=[]
  for ip in host_list:
    cs = socket.socket(socket.AF_INET,socket.SOCK_STREAM)
    address=(ip,port)
    status = cs.connect_ex(address)
    cs.settimeout(timeout)
    if status !=NORMAL:
      down_list.append(ip)
  return down_list

# 发送告警邮件
def send_mail(msg_sub,msg_content):
    msg_to = ["zoubin51@126.com"]
    email_url = "http://29.23.177.177:8080/MailService/sendMail"
    headers = {"Content-Type":"application/json:charset=ascii"}
    param = { "msg_from":"bloan-rcpm-cicd@pingan.com.cn","msg_to":msg_to,"msg_content":msg_content,"msg_sub":msg_sub }

    res = requests.port(email_url,headers=headers,json=param)
    return res

# 检测主机是否宕机函数
def ping(host_list):
    # global  host_down
    # host_down = []
    for i in host_list:
        f = subprocess(f"ping -c1 -W1 {i}",encodings="utf-8",text= True,shell = True,stdout=subprocess.PIPE,stderr=subprocess.PIPE)
        f.commulicate()
        if f.returncode != 0:
            host_down.append(i)
        if host_down:
            msg_sub = "jenkins 节点宕机！"
            msg_content = "jenkins 节点 %s 已经宕机，请关注！" % host_down
            send_mail(msg_sub,msg_content)
            return host_down
# 检测服务函数
def service_check(port,lists,service):
    down_list = port_check(port,lists)
    if down_list:
        msg_sub = "%s 服务掉线" % service
        msg_content = "jenkins 节点 %s， %s服务已经掉线，请关注！"  %(down_list,service)
        send_mail(msg_sub,msg_content)

# ssh 服务down机列表
ssh_down_list = port_check(22,ip_list)

# 指定本地的RSA私钥文件
key = paramiko.RSAKey.from_private_key_file('/root/.ssh/id_rsa')

# ssh服务存在的节点运行监控脚本
ssh_up_list= [i for i in ip_list if i not in ssh_down_list]

def get_monitor(up_list,user):
    for i in up_list:
        trans = paramiko.Transport((i,22))
        trans.connect(username=user,pkey=key)

        # 创建ssh对象，将transport指定为上面的trans
        ssh = paramiko.SSHClient()
        ssh._transport=trans

        # 创建sftp对象，指定连接的通道，用于下发python执行脚本到被监控的主机
        sftp = paramiko.SFTPClient.from_transport(trans)
        # 上传要在远端节点执行的脚本文件
        sftp.put('/root/client_monitor.py','/tmp/client_monitor.py')
        # 添加执行权限，运行脚本
        ssh.exec_command('chmod +x /tmp/client_monitor.py')
        stdin,stdout,stderr = ssh.exec_command("python3 /tmp/client_monitor.py")
        msg = stdout.read().decode('utf-8')
        res_os = json.loads(msg)
        ssh.close()

        data = f'''<body>
                    <div>
                        &nbsp;
                    </div>
                    <div>
                        &nbsp;
                    </div>
                    <table border="1px solid black">
                        <thead style="background-color: #B9B9FF">
                            <tr>
                                <th>TIME</th>
                                <th>IP</th>
                                <th>CPU(%)</th>
                                <th>MEMORY(%)</th>
                                <th>DISK(%)</th>
                            </tr>
                        </thead>
                        <tbody align="center">
                            <tr>
                                <td>{res_os['TIME']}</td>
                                <td>{res_os['IP']}</td>
                                <td>{res_os['CPU_PER']}</td>
                                <td>{res_os['MEM_PER']}</td>
                                <td>{res_os['DISK_PER']}</td>
                            </tr>
                        </tbody>
                    </table>
                    </body> '''
        if int(res_os["MEM_PER"]) > 90:
            msg_sub = 'JEKINS节点内存预警'
            msg_content = "JENKINS 节点【%s】内存使用率过高，请关注！<br>" %i + data

        if int(res_os["DISK_PER"]) > 90:
            msg_sub = 'JEKINS节点磁盘预警'
            msg_content = "JENKINS 节点【%s】磁盘使用率过高，请关注！<br>" %i + data

        if int(res_os["CPU_PER"]) > 90:
            msg_sub = 'JEKINS节点CPU负载预警'
            msg_content = "JENKINS 节点【%s】CPU负载过高，请关注！<br>" %i + data

# 检测是否有节点宕机并发送告警邮件
ping(ip_list)
host_down = ping(ip_list)

# 检测SSH服务并发送告警邮件
ssh_list= [i for i in ip_list if i not in host_down]
service_check(22,ssh_list,'SSH')
# 检测jenkins服务并告警
jen_list = ["29.3.218.208","29.3.218.213"]

# 检测sonar 服务并告警
so_list = ["29.2.209.216"]
service_check(8080,so_list,'SONAR')

#检测节点的os信息并发送告警
get_monitor(ssh_up_list,'root')