您的位置:首页 > 编程语言 > Python开发

基于binlog来分析mysql的行记录修改情况(python脚本分析)

2017-05-26 15:24 549 查看
最近写完mysql flashback,突然发现还有有这种使用场景:有些情况下,可能会统计在某个时间段内,MySQL修改了多少数据量?发生了多少事务?主要是哪些表格发生变动?变动的数量是怎么样的? 但是却不需要行记录的修改内容,只需要了解 行数据的 变动情况。故也整理了下。
昨晚写的脚本,因为个人python能力有限,本来想这不发这文,后来想想,没准会有哪位园友给出优化建议。

如果转载,请注明博文来源: www.cnblogs.com/xinysu/ ,版权归 博客园 苏家小萝卜 所有。望各位支持!

1 实现内容

有些情况下,可能会统计在某个时间段内,MySQL修改了多少数据量?发生了多少事务?主要是哪些表格发生变动?变动的数量是怎么样的? 但是却不需要行记录的修改内容,只需要了解 行数据的 变动情况。
这些情况部分可以通过监控来大致了解,但是也可以基于binlog来全盘分析,binlog的格式是row模式。
在写flashback的时候,顺带把这个也写了个脚步,使用python编写,都差不多原理,只是这个简单些,介于个人python弱的不行,性能可能还有很大的提升空间,也希望园友能协助优化下。
先贴python脚步的分析结果图如下,分为4个部分:事务耗时情况、事务影响行数情况、DML行数情况以及操作最频繁表格情况。



2 脚本简单描述

脚本依赖的模块中,pymysql需要自行安装。
创建类queryanalyse,其中有5个函数定义:_get_db、create_tab、rowrecord、binlogdesc跟closeconn。

2.1 _get_db

该函数用来解析输入参数值,参数值一共有7个,都是必须填写的。分别为host,user,password,port,table name for transaction,table name for records,对应的简写如下:

ALL options need to assign:

-h : host, the database host,which database will store the results after analysis
-u : user, the db user
-p : password, the db user's password
-P : port, the db port
-f : file path, the binlog file
-tr : table name for record , the table name to store the row record
-tt : table name for transaction, the table name to store transactions

比如,执行脚本:python queryanalyse.py -h=127.0.0.1 -P=3310 -u=root -p=password -f=/tmp/stock_binlog.log -tt=flashback.tbtran -tr=flashback.tbrow,该函数负责处理各个选项的参数值情况,并存储。

2.2 create_tab

创建两个表格,分别用来存储 binlog file文件的分析结果。一个用来存储事务的执行开始时间跟结束时间,由选项 -tt来赋值表名;一个是用来存储每一行记录的修改情况,由选项 -tr来赋值表名。
事务表记录内容:事务的开始时间及事务的结束时间。
行记录表的内容:库名,表名,DML类型以及事务对应事务表的编号。

root@localhost:mysql3310.sock  14:42:29 [flashback]>show create table tbrow \G
*************************** 1. row ***************************
Table: tbrow
Create Table: CREATE TABLE `tbrow` (
`auto_id` int(10) unsigned NOT NULL AUTO_INCREMENT,
`sqltype` int(11) NOT NULL COMMENT '1 is insert,2 is update,3 is delete',
`tran_num` int(11) NOT NULL COMMENT 'the transaction number',
`dbname` varchar(50) NOT NULL,
`tbname` varchar(50) NOT NULL,
PRIMARY KEY (`auto_id`),
KEY `sqltype` (`sqltype`),
KEY `dbname` (`dbname`),
KEY `tbname` (`tbname`)
) ENGINE=InnoDB AUTO_INCREMENT=295151 DEFAULT CHARSET=utf8
1 row in set (0.00 sec)

root@localhost:mysql3310.sock  14:42:31 [flashback]>SHOW CREATE TABLE TBTRAN \G
*************************** 1. row ***************************
Table: TBTRAN
Create Table: CREATE TABLE `tbtran` (
`auto_id` int(10) unsigned NOT NULL AUTO_INCREMENT,
`begin_time` datetime NOT NULL,
`end_time` datetime NOT NULL,
PRIMARY KEY (`auto_id`)
) ENGINE=InnoDB AUTO_INCREMENT=6390 DEFAULT CHARSET=utf8
1 row in set (0.00 sec)


2.3 rowrecord

重点函数,分析binlog文件内容。这里有几个规律:

每个事务的结束点,是以 'Xid = ' 来查找
事务的开始时间,是事务内的第一个 'Table_map' 行里边的时间

事务的结束时间,是以 'Xid = '所在行的 里边的时间

每个行数据是属于哪个表格,是以 'Table_map'来查找

DML的类型是按照 行记录开头的情况是否为:'### INSERT INTO' 、'### UPDATE' 、'### DELETE FROM'

注意,单个事务可以包含多个表格多种DML多行数据修改的情况。

2.4 binlogdesc

描述分析结果,简单4个SQL分析。

分析修改行数据的 事务耗时情况

分析修改行数据的 事务影响行数情况

分析DML分布情况

分析 最多DML操作的表格 ,取前十个分析

2.5 closeconn

关闭数据库连接。

3 使用说明

首先,确保python安装了pymysql模块,把python脚本拷贝到文件 queryanalyse.py。
然后,把要分析的binlog文件先用 mysqlbinlog 指令分析存储,具体binlog的文件说明,可以查看之前的博文:关于binary log那些事——认真码了好长一篇。mysqlbinlog的指令使用方法,可以详细查看文档:https://dev.mysql.com/doc/refman/5.7/en/mysqlbinlog.html
比较常用通过指定开始时间跟结束时间来分析 binlog文件。

mysqlbinlog --start-datetime='2017-04-23 00:00:03' --stop-datetime='2017-04-23 00:30:00' --base64-output=decode-rows -v /data/mysql/logs/mysql-bin.007335 > /tmp/binlog_test.log

分析后,可以把这个 binlog_test.log文件拷贝到其他空闲服务器执行分析,只需要有个空闲的DB来存储分析记录即可。
假设这个时候,拷贝 binlog_test.log到测试服务器上,测试服务器上的数据库可以用来存储分析内容,则可以执行python脚本了,注意要进入到python脚本的目录中,或者指定python脚本路径。

python queryanalyse.py -h=127.0.0.1 -P=3310 -u=root -p=password -f= /tmp/binlog_test.log -tt=flashback.tbtran -tr=flashback.tbrow

没了,就等待输出吧。
性能是硬伤,在虚拟机上测试,大概500M的binlog文件需要分析2-3min,有待提高!

4 python脚本

import pymysql
from pymysql.cursors import DictCursor
import re
import os
import sys
import datetime
import time
import logging
import importlib
importlib.reload(logging)
logging.basicConfig(level=logging.DEBUG,format='%(asctime)s %(levelname)s %(message)s ')

usage=''' usage: python [script's path] [option]
ALL options need to assign:

-h     : host, the database host,which database will store the results after analysis
-u     : user, the db user
-p     : password, the db user's password
-P     : port, the db port
-f     : file path, the binlog file
-tr    : table name for record , the table name to store the row record
-tt    : table name for transaction, the table name to store transactions
Example: python queryanalyse.py -h=127.0.0.1 -P=3310 -u=root -p=password -f=/tmp/stock_binlog.log -tt=flashback.tbtran -tr=flashback.tbrow

'''

class queryanalyse:
def __init__(self):
#初始化
self.host=''
self.user=''
self.password=''
self.port='3306'
self.fpath=''
self.tbrow=''
self.tbtran=''

self._get_db()
logging.info('assign values to parameters is done:host={},user={},password=***,port={},fpath={},tb_for_record={},tb_for_tran={}'.format(self.host,self.user,self.port,self.fpath,self.tbrow,self.tbtran))

self.mysqlconn = pymysql.connect(host=self.host, user=self.user, password=self.password, port=self.port,charset='utf8')
self.cur = self.mysqlconn.cursor(cursor=DictCursor)
logging.info('MySQL which userd to store binlog event connection is ok')

self.begin_time=''
self.end_time=''
self.db_name=''
self.tb_name=''

def _get_db(self):
#解析用户输入的选项参数值,这里对password的处理是明文输入,可以自行处理成是input格式,
#由于可以拷贝binlog文件到非线上环境分析,所以password这块,没有特殊处理
logging.info('begin to assign values to parameters')
if len(sys.argv) == 1:
print(usage)
sys.exit(1)
elif sys.argv[1] == '--help':
print(usage)
sys.exit()
elif len(sys.argv) > 2:
for i in sys.argv[1:]:
_argv = i.split('=')
if _argv[0] == '-h':
self.host = _argv[1]
elif _argv[0] == '-u':
self.user = _argv[1]
elif _argv[0] == '-P':
self.port = int(_argv[1])
elif _argv[0] == '-f':
self.fpath = _argv[1]
elif _argv[0] == '-tr':
self.tbrow = _argv[1]
elif _argv[0] == '-tt':
self.tbtran = _argv[1]
elif _argv[0] == '-p':
self.password = _argv[1]
else:
print(usage)

def create_tab(self):
#创建两个表格:一个用户存储事务情况,一个用户存储每一行数据修改的情况
#注意,一个事务可以存储多行数据修改的情况
logging.info('creating table ...')
create_tb_sql ='''CREATE TABLE IF NOT EXISTS  {} (
`auto_id` int(10) unsigned NOT NULL AUTO_INCREMENT,
`begin_time` datetime NOT NULL,
`end_time` datetime NOT NULL,
PRIMARY KEY (`auto_id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
CREATE TABLE IF NOT EXISTS  {} (
`auto_id` int(10) unsigned NOT NULL AUTO_INCREMENT,
`sqltype` int(11) NOT NULL COMMENT '1 is insert,2 is update,3 is delete',
`tran_num` int(11) NOT NULL COMMENT 'the transaction number',
`dbname` varchar(50) NOT NULL,
`tbname` varchar(50) NOT NULL,
PRIMARY KEY (`auto_id`),
KEY `sqltype` (`sqltype`),
KEY `dbname` (`dbname`),
KEY `tbname` (`tbname`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;
truncate table {};
truncate table {};
'''.format(self.tbtran,self.tbrow,self.tbtran,self.tbrow)

self.cur.execute(create_tb_sql)
logging.info('created table {} and {}'.format(self.tbrow,self.tbtran))

def rowrecord(self):
#处理每一行binlog
#事务的结束采用 'Xid =' 来划分
#分析结果,按照一个事务为单位存储提交一次到db
try:
tran_num=1    #事务数
record_sql='' #行记录的insert sql
tran_sql=''   #事务的insert sql

self.create_tab()

with open(self.fpath,'r') as binlog_file:
logging.info('begining to analyze the binlog file ,this may be take a long time !!!')
logging.info('analyzing...')

for bline in binlog_file:

if bline.find('Table_map:') != -1:
l = bline.index('server')
n = bline.index('Table_map')
begin_time = bline[:l:].rstrip(' ').replace('#', '20')

if record_sql=='':
self.begin_time = begin_time[0:4] + '-' + begin_time[4:6] + '-' + begin_time[6:]

self.db_name = bline[n::].split(' ')[1].replace('`', '').split('.')[0]
self.tb_name = bline[n::].split(' ')[1].replace('`', '').split('.')[1]
bline=''

elif bline.startswith('### INSERT INTO'):
record_sql=record_sql+"insert into {}(sqltype,tran_num,dbname,tbname) VALUES (1,{},'{}','{}');".format(self.tbrow,tran_num,self.db_name,self.tb_name)

elif bline.startswith('### UPDATE'):
record_sql=record_sql+"insert into {}(sqltype,tran_num,dbname,tbname) VALUES (2,{},'{}','{}');".format(self.tbrow,tran_num,self.db_name,self.tb_name)

elif bline.startswith('### DELETE FROM'):
record_sql=record_sql+"insert into {}(sqltype,tran_num,dbname,tbname) VALUES (3,{},'{}','{}');".format(self.tbrow,tran_num,self.db_name,self.tb_name)

elif bline.find('Xid =') != -1:

l = bline.index('server')
end_time = bline[:l:].rstrip(' ').replace('#', '20')
self.end_time = end_time[0:4] + '-' + end_time[4:6] + '-' + end_time[6:]
tran_sql=record_sql+"insert into {}(begin_time,end_time) VALUES ('{}','{}')".format(self.tbtran,self.begin_time,self.end_time)

self.cur.execute(tran_sql)
self.mysqlconn.commit()
record_sql = ''
tran_num += 1

except Exception:
return 'funtion rowrecord error'

def binlogdesc(self):
sql=''
t_num=0
r_num=0
logging.info('Analysed result printing...\n')
#分析总的事务数跟行修改数量
sql="select 'tbtran' name,count(*) nums from {}  union all select 'tbrow' name,count(*) nums from {};".format(self.tbtran,self.tbrow)
self.cur.execute(sql)
rows=self.cur.fetchall()
for row in rows:
if row['name']=='tbtran':
t_num = row['nums']
else:
r_num = row['nums']
print('This binlog file has {} transactions, {} rows are changed '.format(t_num,r_num))

# 计算 最耗时 的单个事务
# 分析每个事务的耗时情况,分为5个时间段来描述
# 这里正常应该是 以毫秒来分析的,但是binlog中,只精确时间到second
sql='''select
count(case when cost_sec between 0 and 1 then 1 end ) cos_1,
count(case when cost_sec between 1.1 and 5 then 1 end ) cos_5,
count(case when cost_sec between 5.1 and 10 then 1 end ) cos_10,
count(case when cost_sec between 10.1 and 30 then 1 end ) cos_30,
count(case when cost_sec >30.1 then 1 end ) cos_more,
max(cost_sec) cos_max
from
(
select
auto_id,timestampdiff(second,begin_time,end_time) cost_sec
from {}
) a;'''.format(self.tbtran)
self.cur.execute(sql)
rows=self.cur.fetchall()

for row in rows:
print('The most cost time : {} '.format(row['cos_max']))
print('The distribution map of each transaction costed time: ')
print('Cost time between    0 and  1 second : {} , {}%'.format(row['cos_1'],int(row['cos_1']*100/t_num)))
print('Cost time between  1.1 and  5 second : {} , {}%'.format(row['cos_5'], int(row['cos_5'] * 100 / t_num)))
print('Cost time between  5.1 and 10 second : {} , {}%'.format(row['cos_10'], int(row['cos_10'] * 100 / t_num)))
print('Cost time between 10.1 and 30 second : {} , {}%'.format(row['cos_30'], int(row['cos_30'] * 100 / t_num)))
print('Cost time                     > 30.1 : {} , {}%\n'.format(row['cos_more'], int(row['cos_more'] * 100 / t_num)))

# 计算 单个事务影响行数最多 的行数量
# 分析每个事务 影响行数 情况,分为5个梯度来描述
sql='''select
count(case when nums between 0 and 10 then 1 end ) row_1,
count(case when nums between 11 and 100 then 1 end ) row_2,
count(case when nums between 101 and 1000 then 1 end ) row_3,
count(case when nums between 1001 and 10000 then 1 end ) row_4,
count(case when nums >10001 then 1 end ) row_5,
max(nums) row_max
from
(
select
count(*) nums
from {} group by tran_num
) a;'''.format(self.tbrow)
self.cur.execute(sql)
rows=self.cur.fetchall()

for row in rows:
print('The most changed rows for each row: {} '.format(row['row_max']))
print('The distribution map of each transaction changed rows : ')
print('Changed rows between    1 and    10 second : {} , {}%'.format(row['row_1'],int(row['row_1']*100/t_num)))
print('Changed rows between   11 and   100 second : {} , {}%'.format(row['row_2'], int(row['row_2'] * 100 / t_num)))
print('Changed rows between  101 and  1000 second : {} , {}%'.format(row['row_3'], int(row['row_3'] * 100 / t_num)))
print('Changed rows between 1001 and 10000 second : {} , {}%'.format(row['row_4'], int(row['row_4'] * 100 / t_num)))
print('Changed rows                       > 10001 : {} , {}%\n'.format(row['row_5'], int(row['row_5'] * 100 / t_num)))

# 分析 各个行数 DML的类型情况
# 描述 delete,insert,update的分布情况
sql='select sqltype ,count(*) nums from {} group by sqltype ;'.format(self.tbrow)
self.cur.execute(sql)
rows=self.cur.fetchall()

print('The distribution map of the {} changed rows : '.format(r_num))
for row in rows:

if row['sqltype']==1:
print('INSERT rows :{} , {}% '.format(row['nums'],int(row['nums']*100/r_num)))
if row['sqltype']==2:
print('UPDATE rows :{} , {}% '.format(row['nums'],int(row['nums']*100/r_num)))
if row['sqltype']==3:
print('DELETE rows :{} , {}%\n '.format(row['nums'],int(row['nums']*100/r_num)))

# 描述 影响行数 最多的表格
# 可以分析是哪些表格频繁操作,这里显示前10个table name
sql = '''select
dbname,tbname ,
count(*) ALL_rows,
count(*)*100/{} per,
count(case when sqltype=1 then 1 end) INSERT_rows,
count(case when sqltype=2 then 1 end) UPDATE_rows,
count(case when sqltype=3 then 1 end) DELETE_rows
from {}
group by dbname,tbname
order by ALL_rows desc
limit 10;'''.format(r_num,self.tbrow)
self.cur.execute(sql)
rows = self.cur.fetchall()

print('The distribution map of the {} changed rows : '.format(r_num))
print('tablename'.ljust(50),
'|','changed_rows'.center(15),
'|','percent'.center(10),
'|','insert_rows'.center(18),
'|','update_rows'.center(18),
'|','delete_rows'.center(18)
)
print('-------------------------------------------------------------------------------------------------------------------------------------------------')
for row in rows:
print((row['dbname']+'.'+row['tbname']).ljust(50),
'|',str(row['ALL_rows']).rjust(15),
'|',(str(int(row['per']))+'%').rjust(10),
'|',str(row['INSERT_rows']).rjust(10)+' , '+(str(int(row['INSERT_rows']*100/row['ALL_rows']))+'%').ljust(5),
'|',str(row['UPDATE_rows']).rjust(10)+' , '+(str(int(row['UPDATE_rows']*100/row['ALL_rows']))+'%').ljust(5),
'|',str(row['DELETE_rows']).rjust(10)+' , '+(str(int(row['DELETE_rows']*100/row['ALL_rows']))+'%').ljust(5),
)
print('\n')

logging.info('Finished to analyse the binlog file !!!')

def closeconn(self):
self.cur.close()
logging.info('release db connections\n')

def main():
p = queryanalyse()
p.rowrecord()
p.binlogdesc()
p.closeconn()

if __name__ == "__main__":
main()
内容来自用户分享和网络整理,不保证内容的准确性,如有侵权内容,可联系管理员处理 点击这里给我发消息
标签: