hmoban Pyhton常见问题 2024-03-29

用Python备份MYSQL 数据库

工作需要，对公司的MYSQL数据库进行备份，赶上刚刚开始学python，看了一套简单的python教学视频，简单的写了个备份脚本，个人表示对python 的class 、function、build-in function 、私有变量、全局变量等等，该怎么用，啥时候用等毫无概念，仅此记录一下吧，也欢迎路过的pythoner赐教。

个人已知的一些问题：

1、该脚本必须要求 mysql配置文件内的所有行为 key=value的格式，并且不能存在多余的注释，否则ConfigParser模块解析配置文件时会出错，由于没研究过ConfigParser是不是有容错的方法可以调用，也没时间写容错处理，而是通过整理my.ini 配置文件使其符合ConfigParser的要求解决的。后面会附上我用的mysql配置文件。

2、大量使用类私有成员变量，因为完全不知道python 变量、类方法、等等啥时候该私有化，以及有啥区别，只知道类私有成员变量在别的脚本中import 或者继承时，是不可见的。

3、比较多的进行文件操作，以及传值操作，目前只保证按正确格式传值没问题，没有做多余的容错处理。 4、大量的在进行字符串拼接，第一次写运维相关脚本，由于要调用系统命令，和传递很多参数，也不会subprocess模块，不知道别人写运维脚本都具体咋做，就直接拼接了。

5、其他未知的bug、未发现的逻辑错误等等。

环境：

– Server : Dell PowerEdge T110

– OS: CentOS 6.3_x86_64

– PythonVersion: 2.7.3

– MysqlVersion: 5.5.28 linux x86_64

MysqlBackupScript.py

#!/usr/bin/env python
# coding: utf8
# script MysqlBackupScript
# by Becareful
# version v1.0
 
"""
This scripts provides auto backup mysql(version == 5.5.x) database .
"""
 
import os
import sys
import datetime         #用于生成备份文件的日期
import linecache        #用于读取文件的指定行
import ConfigParser     #解析mysql配置文件
 
 
class DatabaseArgs(object):
    """
 
    """
 
    __MYSQL_BASE_DIR = r"/usr/local/mysql"            #mysql安装目录
    __MYSQL_BIN_DIR = __MYSQL_BASE_DIR + "/bin"       #mysql二进制目录
    __MYSQL_CONFIG_FILE = r"/usr/local/mysql/my.cnf"  #mysql配置文件
 
    __ONEDAY = datetime.timedelta(days=1)             #一天的时长，用于计算下面的前一天和后一天日期
    __TODAY = datetime.date.today()      #当天日期格式为 YYYY-MM-DD
    __YESTERDAY = __TODAY - __ONEDAY                  #计算昨天日期
    __TOMORROW = __TODAY + __ONEDAY                   #计算明天日期
    __WEEKDAY = __TODAY.strftime("%w")                #计算当天是一星期的星期几
 
    __MYSQL_DUMP_ARGS = {                             #用一个字典存储mysqldump 命令备份数据库的参数
        "MYISAM": " -v -E -e -R --triggers -F  -n --opt --master-data=2 --hex-blob -B ",
        "INNODB": " -v -E -e -R --triggers -F --single-transaction -n --opt --master-data=2 --hex-blob -B "
    }
 
    __DUMP_COMMAND = __MYSQL_BIN_DIR + "/mysqldump"   #mysqldump 命令的 路径 用于dump mysql数据
    __FLUSH_LOG_COMMAND = __MYSQL_BIN_DIR + "/mysqladmin"    #mysqladmin 命令的路径 ，用于执行 flush-logs 生成每天增量binlog
 
    __BACKUP_DIR = r"/backup/"                        # 指定备份文件存放的目录
 
    __PROJECTNAME = "example"                         # 指定需要备份的数据库对应的项目名，将来会生成 projectname-YYYY-MM-DD.sql 等文件
    __DATABASE_LIST = []                              # 指定需要备份的数据库名，可以是多个，使用列表
    __HOST = "localhost"
    __PORT = 3306
    __USERNAME = "root"
    __PASSWORD = ""
    __LOGINARGS = ""                                  # 如果在localhost登陆，需要密码，可以设定登陆的参数，具体在下面有说明
    __LOGFILE = __BACKUP_DIR + "/backup.logs"
 
    def __init__(self, baseDir=__MYSQL_BASE_DIR, backDir=__BACKUP_DIR, engine="MYISAM", projectName=__PROJECTNAME,
                 dbList=__DATABASE_LIST, host=__HOST, port=__PORT, user=__USERNAME, passwd=__PASSWORD):
 
        """
            实例化对象时传入的参数，如不传入默认使用类的私有成员变量作为默认值
    :param baseDir:    
    :param backDir:
    :param engine:
    :param projectName:
    :param dbList:
    :param host:
    :param port:
    :param user:
    :param passwd:
    """
        self.__MYSQL_BASE_DIR = baseDir
        self.__BACKUP_DIR = backDir
        self.__PROJECTNAME = projectName
        self.__DATABASE_LIST = dbList
        self.__HOST = host
        self.__PORT = port
        self.__USERNAME = user
        self.__PASSWORD = passwd
        self.__ENGINE = self.__MYSQL_DUMP_ARGS[engine]
        #下面定义了如需登陆时，参数 其实就是生成 这样的格式  “-hlocalhost -uroot --password=‘xxxx’”
        self.__LOGINARGS = " -h" + self.__HOST + " -P" + str(
            self.__PORT) + " -u" + self.__USERNAME + " --password="" + self.__PASSWORD + """
        self.checkDatabaseArgs()   #调用检查函数
 
    def __getconfig(self, cnf=__MYSQL_CONFIG_FILE, item=None):  # 解析mysql配置文件的小函数，简单封装了下，传入一个值作为my.cnf的key去查找对应的value
 
        __mycnf = ConfigParser.ConfigParser()
        __mycnf.read(cnf)
        try:
            return __mycnf.get("mysqld", item)
        except BaseException, e:
            sys.stderr.write(str(e))
            sys.exit(1)
 
    def __getBinlogPath(self): #  取每天需要增量备份的binlog日志的绝对路径，从mysql的binlog.index文件取倒数第二行
 
        __BINLOG_INDEX = self.__getconfig(item="log-bin") + ".index"
 
        if not os.path.isfile(__BINLOG_INDEX):
            sys.stderr.write("BINLOG INDEX FILE: [" + __BINLOG_INDEX + " ] NOT FOUND! 
")
            sys.exit(1)
        else:
            try:
                __BINLOG_PATH = linecache.getline(__BINLOG_INDEX, len(open(__BINLOG_INDEX, "r").readlines()) - 1)
                linecache.clearcache()
            except BaseException, e:
                sys.stderr.write(str(e))
                sys.exit(1)
            return __BINLOG_PATH.strip()
 
    def flushDatabaseBinlog(self):  # 调用此函数，将会执行  mysqladmin flush-logs ,刷新binlog日志
        return os.popen(self.__FLUSH_LOG_COMMAND + self.__LOGINARGS + " flush-logs")
 
    def dumpDatabaseSQL(self):  #|通过mysqladmin 对指定数据库进行全备
        if not os.path.isfile(self.__BACKUP_DIR + "/" + self.__PROJECTNAME + "/" +  str(self.__YESTERDAY) + "-" + self.__PROJECTNAME + ".sql"):
            return os.popen(self.__DUMP_COMMAND + self.__LOGINARGS + self.__ENGINE + " ".join(
                self.__DATABASE_LIST) + " >> " + self.__BACKUP_DIR + "/" + self.__PROJECTNAME + "/" + str(
                self.__YESTERDAY) + "-" + self.__PROJECTNAME + ".sql")
        else:
            sys.stderr.write("Backup File [" + str(self.__YESTERDAY) + "-" + self.__PROJECTNAME + ".sql]  already exists.
")
 
 
 
    def dumpDatabaseBinlog(self):#通过copy2() 将需要备份的binlog日志复制到指定备份目录
 
        if not os.path.isfile(self.__BACKUP_DIR + "/" + self.__PROJECTNAME + "/" + str(self.__YESTERDAY) + "-" + os.path.split(self.__getBinlogPath())[1]):
            from shutil import copy2
            try:
                copy2(self.__getBinlogPath(), self.__BACKUP_DIR + "/" + self.__PROJECTNAME + "/" + str(self.__YESTERDAY) + "-" + os.path.split(self.__getBinlogPath())[1])
            except BaseException, e:
                sys.stderr.write(str(e))
        else:
            sys.stderr.write("Binlog File [" + str(self.__YESTERDAY) + "-" + os.path.split(self.__getBinlogPath())[1] + "] already exists
" )
 
 
    def checkDatabaseArgs(self):  #对一些必要条件进行检查
        __rv = 0
 
        if not os.path.isdir(self.__MYSQL_BASE_DIR):  #检查指定的mysql安装目录是否存在
            sys.stderr.write("MYSQL BASE DIR: [ " + self.__MYSQL_BASE_DIR + " ] NOT FOUND
")
            __rv += 1
 
        if not os.path.isdir(self.__BACKUP_DIR):   #检查指定的备份目录是否存在，如不存在自动创建
            sys.stderr.write("BACKUP DIR: [ " + self.__BACKUP_DIR + "/" + self.__PROJECTNAME +  " ] NOT FOUND ,AUTO CREATED
")
            os.makedirs(self.__BACKUP_DIR + "/" + self.__PROJECTNAME)
 
        if not os.path.isfile(self.__MYSQL_CONFIG_FILE): #检查mysql配置文件是否存在
            sys.stderr.write("MYSQL CONFIG FILE: [" + self.__MYSQL_CONFIG_FILE + " ] NOT FOUND
")
            __rv += 1
 
        if not os.path.isfile(self.__DUMP_COMMAND):  #检查备份数据库时使用的mysqldump命令是否存在
            sys.stderr.write("MYSQL DUMP COMMAND: [" + self.__DUMP_COMMAND + " ] NOT FOUND
")
            __rv += 1
 
        if not os.path.isfile(self.__FLUSH_LOG_COMMAND): #检查刷新mysql binlog日志使用的mysqladmin命令是否存在
            sys.stderr.write("MYSQL FLUSH LOG COMMAND: [" + self.__DUMP_COMMAND + " ] NOT FOUND
")
            __rv += 1
 
        if not self.__DATABASE_LIST:  #检查需要备份的数据库列表是否存在
            sys.stderr.write("Database List is None 
")
            __rv += 1
 
        if __rv:   # 判断返回值，由于上述任何一步检查失败，都会导致 __rv 值 +1 ，只要最后__rv ！= 0就直接退出了。
            sys.exit(1)
 
 
def crontab():  # 使用字典，来进行相关参数传递，并实例化对象，调用相关方法进行操作
 
    zabbix = {
        "baseDir": "/usr/local/mysql/",
        "backDir": "/backup/",
        "projectName": "Monitor",
        "dbList": ["zabbix"],
        "host": "localhost",
        "port": 3306,
        "user": "root",
        "passwd": "xxxxxxx"
    }
 
    monitor = DatabaseArgs(**zabbix)
    monitor.dumpDatabaseSQL()
    monitor.dumpDatabaseBinlog()
    monitor.flushDatabaseBinlog()
 
if __name__ == "__main__":
    crontab()

my.cnf

[client]
port                            = 3306
socket                          = /mysql/var/db.socket
 
[mysqld]
socket                          = /mysql/var/db.socket
datadir                         = /mysql/db/
skip-external-locking           = 1
skip-innodb                     = 0 
key_buffer_size                 = 256M
max_allowed_packet              = 10M
table_open_cache                = 2048
sort_buffer_size                = 4M
read_buffer_size                = 4M
read_rnd_buffer_size            = 8M
myisam_sort_buffer_size         = 64M
myisam_max_sort_file_size       = 1G
myisam_repair_threads           = 1
myisam_recover                  = DEFAULT
thread_cache_size               = 32
query_cache_size                = 32M
query_cache_min_res_unit        = 2k
bulk_insert_buffer_size         = 64M
tmp_table_size                  = 128M
thread_stack                    = 192K
skip-name-resolve               = 1
max_connections                 = 65500
default-storage-engine          = myisam
federated                       = 0
server-id                       = 1
slave-skip-errors               = all
#log                            = /var/log/sql_query.log
slow-query-log                  = 1
slow-query-log-file             = /mysql/log/sql_query_slow.log
long-query-time                 = 5
log-queries-not-using-indexes   = 1
log-slow-admin-statements       = 1
log-bin                         = /mysql/var/log/binlog/bin-log
log-error                       = /mysql/var/log/mysql.err
master-info-file                = /mysql/var/log/master.info
relay-log                       = /mysql/var/log/relay-bin/relay-bin
relay-log-index                 = /mysql/var/log/relay-bin/relay-bin.index
relay-log-info-file             = /mysql/var/log/relay-bin/relay-bin.info
binlog_cache_size               = 8M
binlog_format                   = MIXED
max_binlog_cache_size           = 20M
max_binlog_size                 = 1G
binlog-ignore-db                = mysql
binlog-ignore-db                = performance_schema
binlog-ignore-db                = information_schema
replicate-ignore-db             = mysql
replicate-ignore-db             = performance_schema
replicate-ignore-db             = information_schema
 
innodb_data_home_dir            = /mysql/ibdata/
innodb_data_file_path           = ibdata:156M:autoextend
innodb_log_group_home_dir       = /mysql/ibdata/
log-slave-updates               = 0
back_log                        = 512
transaction_isolation           = READ-COMMITTED
max_heap_table_size             = 246M
interactive_timeout             = 120
wait_timeout                    = 120
innodb_additional_mem_pool_size = 16M
innodb_buffer_pool_size         = 512M
innodb_file_io_threads          = 4
innodb_thread_concurrency       = 8
innodb_flush_log_at_trx_commit  = 2
innodb_log_buffer_size          = 16M
innodb_log_file_size            = 128M
innodb_log_files_in_group       = 3
innodb_max_dirty_pages_pct      = 90
innodb_lock_wait_timeout        = 120
innodb_file_per_table           = 1
innodb_open_file                = 327500
open_files_limit                = 327500
 
[mysqldump]
quick                           = 1
max_allowed_packet              = 50M
 
[mysql]
auto-rehash                     = 1
socket                          = /mysql/var/db.socket
safe-updates                    = 0
 
[myisamchk]
key_buffer_size                 = 256M
sort_buffer_size                = 256M
read_buffer                     = 2M
write_buffer                    = 2M
 
[mysqlhotcopy]
interactive-timeout             = 100

最终生成的备份目录结构是这样的

[root@zabbix backup]# find ./
./
./Monitor
./Monitor/2013-03-16-bin-log.000008
./Monitor/2013-03-14-bin-log.000006
./Monitor/2013-03-16-Monitor.sql
./Monitor/2013-03-15-Monitor.sql
./Monitor/2013-03-15-bin-log.000007
./Monitor/2013-03-14-Monitor.sql
 
~END~

来源：PY学习网：原文地址：https://www.py.cn/article.html

hmoban主题是根据ripro二开的主题，极致后台体验，无插件，集成会员系统
自学咖网 » 用Python备份MYSQL 数据库

hmoban 普通

分享到：

MysqlBackupScript.py

my.cnf

最终生成的备份目录结构是这样的

hmoban 普通

相关推荐