如何通过Python收集MySQL MHA 部署及运行状态信息的功能

[server default]
manager_workdir=/var/log/masterha/app1.log              //设置manager的工作目录
manager_log=/var/log/masterha/app1/manager.log          //设置manager的日志
master_binlog_dir=/data/mysql                         //设置master 保存binlog的位置，以便MHA可以找到master的日志，我这里的也就是mysql的数据目录
master_ip_failover_script= /usr/local/bin/appxxx_master_ip_failover    //设置自动failover时候的切换脚本
master_ip_online_change_script= /usr/local/bin/appxxx_master_ip_online_change  //设置手动切换时候的切换脚本
passWord=用户密码         //设置mysql中root用户的密码，这个密码是前文中创建监控用户的那个密码
user=root               设置监控用户root
ping_interval=1         //设置监控主库，发送ping包的时间间隔，默认是3秒，尝试三次没有回应的时候自动进行railover
remote_workdir=/tmp     //设置远端mysql在发生切换时binlog的保存位置
repl_password=用户密码    //设置复制用户的密码
repl_user=repl          //设置复制环境中的复制用户名
report_script=/usr/local/send_report    //设置发生切换后发送的报警的脚本          
shutdown_script=""      //设置故障发生后关闭故障主机脚本（该脚本的主要作用是关闭主机放在发生脑裂,这里没有使用）
ssh_user=root           //设置ssh的登录用户名

[server1]
hostname=110.110.110.50
port=3306

[server2]
hostname=110.110.110.60
port=3306
candidate_master=1   //设置为候选master，如果设置该参数以后，发生主从切换以后将会将此从库提升为主库，即使这个主库不是集群中事件最新的slave
check_repl_delay=0   //默认情况下如果一个slave落后master 100M的relay logs的话，MHA将不会选择该slave作为一个新的master，因为对于这个slave的恢复需要花费很长时间，通过设置check_repl_delay=0,MHA触发切换在选择一个新的master的时候将会忽略复制延时，这个参数对于设置了candidate_master=1的主机非常有用，因为这个候选主在切换的过程中一定是新的master

[server3]
hostname=110.110.110.70
port=3306

create table `mysqldb_mha_info` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
   mha_manager_ip varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA管理节点所在集群的IP',
   mha_name varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA的名字，类似于副本集',
   mha_file_name varchar(250) NOT NULL DEFAULT '' COMMENT 'MHA .cnf 配置文件名字',
   mha_name_path varchar(250) NOT NULL DEFAULT '' COMMENT 'MHA .cnf 配置文件路径和名字',
  `cnf_server1_ip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA cnf 配置文件中的节点1',
  `cnf_server2_ip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA cnf 配置文件中的节点2',
  `cnf_server3_ip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA cnf 配置文件中的节点3',
   failover_script varchar(250) NOT NULL DEFAULT '' COMMENT 'MHA failover scripts的文件',
   failover_script_vip varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA failover scripts 中定义的VIP',
   online_script varchar(250) NOT NULL DEFAULT '' COMMENT 'MHA online change scripts的文件',
   online_script_vip varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA online change scripts 中定义的VIP',
   script_remark varchar(1500) NOT NULL DEFAULT '' COMMENT 'MHA scripts VIP 检查结果',
   masterha_status varchar(10) NOT NULL DEFAULT '' COMMENT 'MHA 检查是否开启,来自于 masterha_check_status 检查结果',
   master_serverip varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA 检查是否开启,来自于 masterha_check_status 检查结果',
  `current_master_ip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA 当前主节点,来自check_repl',
  `mha_current_vip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA 当前VIP ,来自check_repl',
 `slave1_ip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA 当前从节点1,来自check_repl',
  `slave2_ip` varchar(50) NOT NULL DEFAULT '' COMMENT 'MHA 当前从节点2 ,来自check_repl',
  mha_cnf_remark varchar(1500) NOT NULL DEFAULT '' COMMENT 'MHA check conf/cnf 检查结果',
  check_repl_remark varchar(1500) NOT NULL DEFAULT '' COMMENT 'MHA check repl检查结果',
  remark varchar(1500) NOT NULL DEFAULT '' COMMENT 'MHA 检查结果',
  `creator` varchar(50) NOT NULL DEFAULT '',
  `create_time` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
  `operator` varchar(50) NOT NULL DEFAULT '',
  `modify_time` datetime NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
 PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8mb4

#!/usr/bin/python3
# -*- coding: UTF-8 -*-

##import MySQLdb 安装模块麻烦
import mysql.connector
db = mysql.connector.connect(user='nideuid', password='nidepwd',host='nideseverip',database='DBname',port=XXXX)

#!/usr/bin/Python
# -*- coding: UTF-8 -*-

import os
import io
import re
import ConfigParser
import Socket

import db_conn
mysqldb = db_conn.db
cursor = mysqldb.cursor()

## 第1部分 获取本机IP
try:
  s=socket.socket(socket.AF_INET,socket.SOCK_DGRAM)
  s.connect(('8.8.8.8',80))
  mha_manager_ip=s.getsockname()[0]
  print('mha manager 所在主机的IP如下:')
  print(mha_manager_ip)
finally:
  s.close()
###

##第2部分: 循环遍历mha cnf 所在的文件夹，取出 cnf 进行判断和检查
Path='/date/funcation/python/mha_conf'
#fout=open('输出文件名','w')
for Name in os.listdir(Path) :
  Pathname= os.path.join(Path,Name)
 ## print(Pathname)
 ## print(Name)
  mha_name = Name.replace(".cnf", "").replace(".conf", "") ###为MHA集群启个名字
  ##print(mha_name)
  ##注意此处为r,不能为w,否则报错:IOError: File not open for reading
  with open(Pathname,'r') as f:
    filecontent=f.read()
    #print(filecontent)
    remark = ''
    ####调整为ConfigParser,注意python2 和 python 的模块名字是不一样的.ConfigParser与configparser
    config =ConfigParser.ConfigParser()
    try:
      config.read(Pathname)
      server_item = config.sections()
      ##print(server_item)
      ### start 获取 MHA 切换时的scripts 文件名字
      mha_failover_script = ''
      mha_online_change_script =''
      mha_cnf_remark =''
      if 'server default' in server_item:
        mha_failover_script = config.get('server default','master_ip_failover_script')
        ###
        mha_failover_script=mha_failover_script.replace(" --ssh_user=root", "")
        ##print(mha_failover_script)
      else:
        mha_cnf_remark = mha_cnf_remark + 'mha_failover_script 未配置;'
      if 'server default' in server_item:
        mha_online_change_script = config.get('server default','master_ip_online_change_script')
        ##print(mha_online_change_script)
      else:
        mha_cnf_remark = mha_cnf_remark + 'mha_online_change_script 未配置;'
      ###1.1 end  获取结束
      ##1.2 start 获取MHA配置文件中的节点信息
      server1_host = ''  ##MHA cnf 配置文件中的节点1
      server2_host = ''  ##MHA cnf 配置文件中的节点2
      server3_host = ''  ##MHA cnf 配置文件中的节点3
      if 'server1' in server_item:
        server1_host = config.get('server1','hostname')
        print(server1_host)
      else:
         server1_host = ''
         mha_cnf_remark = mha_cnf_remark + 'Server1未配置;'
         print(server1_host)
      if 'server2' in server_item:
        server2_host = config.get('server2','hostname')
        print(server2_host)
      else:
        server2_host = ''
        mha_cnf_remark = mha_cnf_remark + 'Server2未配置;'
        print(server2_host)
      if 'server3' in server_item:
        server3_host = config.get('server3','hostname')
        print(server3_host)
      ##else:
        ##server3_host = ''
        ##mha_cnf_remark = mha_cnf_remark + 'Server3未配置;'
        ##print(server3_host)
      ##1.2 获取server节点信息结束
      print(mha_cnf_remark)
    except Exception as e:
      print(e)

    #####第3部分 start 从 mha scripts 中提取 配置的VIP
    mha_remark = ''
    mha_failover_my_vip = ''
    mha_failover_flush_vip = ''
    mha_onlinechange_my_vip = ''
    mha_onlinechange_flush_vip =''
    if len(mha_failover_script) <> 0 and len(mha_online_change_script) <> 0 :
      ##3.1 先来处置 failover_script，解析其中的VIP
      with open(mha_failover_script,'r') as f:
        failscript_lines=f.readlines()
        for failscript_line in failscript_lines:
          failscript_ip=re.findall(r"\b(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\b", failscript_line)
          if failscript_ip:
            if 'my $vip =' in failscript_line:
              mha_failover_my_vip = failscript_ip[0]
              print('解析出mha_failover_my_vip:')
              print(mha_failover_my_vip)
            if  'my $ssh_flush_vip' in failscript_line:
              mha_failover_flush_vip = failscript_ip[0]
              print('解析出mha_failover_flush_vip:')
              print(mha_failover_flush_vip)

        ##文件读取完毕，对读取结果进行判断，判断两种情况
        ## 一种是否未定义
        if mha_failover_my_vip =='':
          mha_remark = mha_remark + 'MHA failover  未提取到VIP的设置，请检查；'
        ## 另外一种，，定义了，但是值不相等
        if mha_failover_my_vip <> mha_failover_flush_vip:
          mha_remark = mha_remark + 'MHA failover scripts文件中设置的两处VIP不一致，请检查；'

      ## 3.2 处理online change scripts ,解析提取其中的VIP信息
      with open(mha_online_change_script,'r') as f:
        onlinescript_lines=f.readlines()
        for onlinescript_line in onlinescript_lines:
          onlinescript_ip=re.findall(r"\b(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\b", onlinescript_line)
          if onlinescript_ip:
            if 'my $vip =' in onlinescript_line:
              mha_onlinechange_my_vip = onlinescript_ip[0]
              print('解析出mha_onlinechange_my_vip:')
              print(mha_onlinechange_my_vip)
            if  'my $ssh_flush_vip' in onlinescript_line:
              mha_onlinechange_flush_vip = onlinescript_ip[0]
              print('解析出mha_onlinechange_flush_vip:')
              print(mha_onlinechange_flush_vip)
        #### online change 文件读完了，判断定义的VIP是否符合要求
        if mha_onlinechange_my_vip =='':
          mha_remark = mha_remark + 'MHA online change scripts未提取到VIP的设置，请检查；'
        if mha_onlinechange_my_vip <> mha_onlinechange_flush_vip:
          mha_remark = mha_remark + 'MHA online change scripts文件中设置的两处VIP不一致，请检查；'

      #### 两个文件都读取了，判断两个文件中定义的VIP是否一致
      if mha_onlinechange_my_vip <> mha_failover_my_vip:
        mha_remark = mha_remark + 'MHA online change script  和 failover script 中的VIP不一致,请检查。'

    else:
      mha_remark = mha_remark + 'MHA init 的配置文件未正确定义切换的scripts,请检查。'
      #print('MHA init 的配置文件未正确定义 切换的scripts,请检查。')
      print(mha_remark)
    #####第2部分 end 从 mha scripts 中提取 配置的VIP

    #### 第4部分，从masterha_check_status执行结果中判断mha进程状态
    ## 从 执行masterha_check_status结果中解析出的  masterha_status 和 master_serverip 的数据
    masterha_status =''
    master_serverip =''
    ## 从 执行masterha_check_repl结果中解析出的 current_master 、current_slave1、current_slave2 和 mha_current_vip 的数据
    current_master = ''
    current_slave1 = ''
    current_slave2 = ''
    mha_current_vip =''
    ##判断下文件是否是MHA的配置文件，判断方式就是文件中 必须有 server default\server1的sections
    if 'server default' in server_item and 'server1' in server_item :
      ##cmd_mha_status ='/usr/local/bin/masterha_check_status --conf=/etc/mha/opszabbix.cnf'
      cmd_mha_status ='/usr/local/bin/masterha_check_status --conf='+Pathname
      try:
        mha_status=os.popen(cmd_mha_status)
        mha_status_result = mha_status.read()
        print(mha_status_result)  ##返回样式为 XXXX (pid:------) is running(0:PING_OK), master:XXX.XXX.XXX.XXX
        ### 判断状态是否为运行中
        if 'running(0:PING_OK)' in mha_status_result:
          masterha_status='Running'
          ##抓取MHA的Master 节点
          ##master_serverip = mha_status_result[mha_status_result.rfind('master:'):]
          master_serverip = mha_status_result.split('master:')[1]
          print(master_serverip)
          print('MHA启动运行正常')
        elif 'stopped(2:NOT_RUNNING)' in mha_status_result:
          masterha_status='stopped'
          print('MHA未启动!!!')
      finally:
         if mha_status:
            mha_status.close()
      #### 第5部分，从masterha_check_repl的执行结果中，判断解析 主、从节点、VIP节点
      ##  判断 副本集 的状况
      cmd_repl_status ='/usr/local/bin/masterha_check_repl --conf='+Pathname
      try:
        ##### 添加 2> error 参数,不需要打印出调试信息。
        cmd_repl_status_result = cmd_repl_status + '     2> checkrepl.log'
        repl_status=os.popen(cmd_repl_status_result)
        repl_status_result = repl_status.read()
        ##print(repl_status_result)
        if 'MySQL Replication Health is OK' in repl_status_result:
          print('MHA集群的主从正常')
          ###获取ServerIP
          #######调试信息是输出到2号流中的，所以一定 添加 2>&1，否则抓取不到节点信息，只能抓到一个VIP。
          cmd_repl_status_info = cmd_repl_status + '     2>&1'
          with os.popen(cmd_repl_status_info,'r') as repl_status_check2:
            #repl_status_lines=repl_status_check2.readlines()
            repl_status_lines=repl_status_check2.readlines()
            ##print(len(repl_status_lines))  ####打印出list的元素个数
            for repl_status_line in repl_status_lines:
              ##print('##################  start   ###########################')
              ##print(str(repl_status_line).replace("\n", "").replace("\t", ""))
              ##repl_status_line ='Current Alive Master: 10.200.58.63(10.200.58.63:55988)'
              serverip_result=re.findall(r"\b(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\b", repl_status_line)
              if serverip_result:
                if 'Current Alive Master:' in repl_status_line:
                  current_master = serverip_result[0]
                  print('已解析到主节点IP')
                  print(current_master)
                elif 'Checking replication health on' in repl_status_line and current_slave1 == '':   ###有可能有2个从节点，此处为第1个从节点
                  current_slave1 = serverip_result[0]
                  print('已解析到从节点1')
                  print(current_slave1)
                elif 'Checking replication health on' in repl_status_line and current_slave1 <> '':  ###有可能有2个从节点，此处为第2个从节点
                  current_slave2 = serverip_result[0]
                  print('已解析到从节点2')
                  print(current_slave2)
                elif 'Checking replication health on' in repl_status_line and current_slave1 <> '':  ###有可能有2个从节点，此处为第2个从节点
                  print('集群有3个或更多的从节点，请确认。')
                if 'down==/sbin/ifconfig ' in repl_status_line:
                  mha_current_vip = serverip_result[0]
                  print('已解析到MHA集群的VIP')
                  print(mha_current_vip)

                ##print('包含serverip')
                ##print(serverip_result)
              #else:
                #print('不包含ServerIP')
            ##else:
              ##print(repl_status_line)
              ##print('##################  end   ###########################')
          ####获取IP部分结束
        else:
          print('MHA集群的主从异常,请及时检查')

      finally:
         if repl_status:
            repl_status.close()

    else:
       remark = Pathname + '...... 不是MHA的配置文件,请检查!'
       print(remark)


    ##### 第6部分，将数据保存到表中
    sql_insert = "insert into mysqldb_mha_info(mha_manager_ip,mha_name,mha_file_name,mha_name_path,cnf_server1_ip,cnf_server2_ip,cnf_server3_ip,failover_script,failover_script_vip,online_script,online_script_vip,masterha_status,master_serverip,current_master_ip,mha_current_vip,slave1_ip,slave2_ip,mha_cnf_remark,script_remark,remark) " \
                      "values('%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s','%s')" % \
                      (mha_manager_ip,mha_name,Name,Pathname,server1_host,server2_host,server3_host,mha_failover_script,mha_failover_my_vip,mha_online_change_script,mha_onlinechange_my_vip,masterha_status,current_master,master_serverip,mha_current_vip,current_slave1,current_slave2,mha_cnf_remark,mha_remark,remark)
    ##print(sql_insert)
    cursor.execute(sql_insert)
    mysqldb.commit()

    #####


# 关闭游标
cursor.close()
# 关闭数据库连接
mysqldb.close()

python /data/XXXX路径/collect_mysqldbmha_info.py