用Python连接SQLServer抓取分析数据、监控 （pymssql）

Python 环境：python3

服务器环境： centos6.5

数据库: Mysql

大概流程：在装有Python服务器，利用pymssql库连接MSSQL生产数据库取出数据然后写进mysql数据库里，再从mysql里取出需要的数据，制作对应的报表；

SQLServer 权限：master只读权限，grant view server state to user，或者限定具体系统视图只读权限

贴出部分代码：

后续可以写成类，可以大大提高执行速度，不过此脚本执行频率不高。

###############################   DB_SERVER_SERVICES  ###########################
#!/usr/bin/python
#coding=utf8

import  pymssql
import pymysql

#Define mssql connections Dic.
servers = {'server':'192.168.10.10','user':'sa','password':'******','database':'master'}
connmysql = pymysql.connect(host='localhost',port=3306,user='root',passwd='******',db='mssql',charset='utf8')
connmssql = pymssql.connect(**servers)

curmssql_services = connmssql.cursor()
curmysql_insert = connmysql.cursor()

sql01 = """select @@SERVERNAME as servername,servicename,startup_type_desc,status_desc,convert(varchar(100),last_startup_time,23) as last_startup_time,service_account,is_clustered,cluster_nodename
from sys.dm_server_services where servicename like 'SQL Server%' and servicename not like 'SQL Server 代理%'"""

curmssql_services.execute(sql01)

sql02 = "insert into db_server_services (servername,servicename,startup_type_desc,status_desc,last_startup_time,service_account,is_clustered,cluster_nodename) values(%s,%s,%s,%s,%s,%s,%s,%s)"
curmysql_insert.executemany(sql02,curmssql_services.fetchall())
connmysql.commit()

curmssql_services.close()
curmysql_insert.close()
connmssql.close()
connmysql.close()

############################   db_space_percent  ##############################
import  pymssql
import pymysql

#Define mssql connections Dic.
servers = {'server':'10.252.130.198','user':'sa','password':'abc123..','database':'master'}
connmysql = pymysql.connect(host='localhost',port=3306,user='root',passwd='abc123..',db='mssql',charset='utf8')
connmssql = pymssql.connect(**servers)

curmssql01 = connmssql.cursor()
curmysql01 = connmysql.cursor()

sql01 = """with tt as (
select distinct
@@servername as servername,
vs.volume_mount_point as drive_name ,
cast(vs.total_bytes / 1024.0 / 1024 / 1024 as numeric(18,2)) as total_space_gb ,
cast(vs.available_bytes / 1024.0 / 1024 / 1024  as numeric(18,2)) as free_space_gb,
convert(varchar(100), getdate(), 23)as check_time
from    sys.master_files as f
cross apply sys.dm_os_volume_stats(f.database_id, f.file_id) as vs
)
select
servername,
drive_name,
total_space_gb,
total_space_gb-free_space_gb as used_space_gb,
free_space_gb,
cast(cast((total_space_gb-free_space_gb)*100/total_space_gb as numeric(18,2))as varchar(10))+'%' as used_percent,
check_time
from tt
"""

curmssql01.execute(sql01)

sql02 = "insert into db_space_percent (servername,drivername,total_space_gb,used_space_gb,free_space_gb,used_Percent,check_time) values(%s,%s,%s,%s,%s,%s,%s)"
curmysql01.executemany(sql02,curmssql01.fetchall())
connmysql.commit()

curmssql01.close()
curmysql01.close()
connmssql.close()
connmysql.close()