2. 技术
    3. mysql全量备份、增量备份实现方法

    mysql全量备份、增量备份实现方法

    技术2025-08-17  12
    <p>mysql全量备份、增量备份。开启mysql的logbin日志功能。在/etc/my.cnf文件中加入以下代码：</p> [mysqld] log-bin = "/home/mysql/logbin.log" binlog-format = ROW log-bin-index = "/home/mysql/logindex" binlog_cache_size=32m max_binlog_cache_size=512m max_binlog_size=512m

    重启mysql即可。其中路径 /home/mysql的用户和组要改成mysql。

     

    2、增量备份 在/home/mysql/目录下建立以下目录：

     

    增量备份脚本

    cd /home/mysql

    vi binlogbak.sh

    #!/bin/bash export LANG=en_US.UTF-8

    BakDir=/home/mysql/backup/daily

    BinDir=/home/mysql

    LogFile=/home/mysql/backup/binlog.log

    BinFile=/home/mysql/logindex.index

    mysqladmin -uroot -proot123 flush-logs

    #这个是用于产生新的mysql-bin.00000*文件

    Counter=wc -l <span class="hljs-variable">$BinFile</span> |awk <span class="hljs-string">'{print $1}'</span>

    NextNum=0

    #这个for循环用于比对 C o u n t e r , Counter, Counter,NextNum这两个值来确定文件是不是存在或最新的。

    for file in cat <span class="hljs-variable">$BinFile</span>

    do

        base=basename <span class="hljs-variable">$file</span>

        #basename用于截取mysql-bin.00000*文件名，去掉./mysql-bin.000005前面的./

        NextNum=expr <span class="hljs-variable">$NextNum</span> + 1

        if [ N e x t N u m < / s p a n > − e q < s p a n c l a s s = " h l j s − v a r i a b l e " > NextNum</span> -eq <span class="hljs-variable"> NextNum</span>−eq<spanclass="hljs−variable">Counter ]

        then

            echo KaTeX parse error: Expected 'EOF', got '&' at position 19: …e</span> skip! &̲gt;&gt; <span c…LogFile

        else

            dest= B a k D i r < / s p a n > / < s p a n c l a s s = " h l j s − v a r i a b l e " > BakDir</span>/<span class="hljs-variable"> BakDir</span>/<spanclass="hljs−variable">base

            if(test -e $dest)

            #test -e用于检测目标文件是否存在，存在就写exist!到$LogFile去。

            then

                echo KaTeX parse error: Expected 'EOF', got '&' at position 20: …</span> exist! &̲gt;&gt; <span c…LogFile

            else

                cp B i n D i r < / s p a n > / < s p a n c l a s s = " h l j s − v a r i a b l e " > BinDir</span>/<span class="hljs-variable"> BinDir</span>/<spanclass="hljs−variable">base $BakDir

                echo KaTeX parse error: Expected 'EOF', got '&' at position 21: …/span> copying &̲gt;&gt; <span c…LogFile

            fi

        fi

    done

    echo date +<span class="hljs-string">"%Y年%m月%d日 %H:%M:%S"</span> Bakup succ! >> $LogFile

    ?赋予binlogbak.sh执行权限

    chmod a+x /home/mysql/binlogbak.sh

    3、全量备份

     

    vi databak.sh

    #!/bin/bash export LANG=en_US.UTF-8

    BakDir=/home/mysql/backup

    LogFile=/home/mysql/backup/bak.log

    Date=date +%Y%m%d

    Begin=date +<span class="hljs-string">"%Y年%m月%d日 %H:%M:%S"</span>

    cd $BakDir

    DumpFile=$Date.sql

    GZDumpFile=$Date.sql.tgz

    mysqldump -uroot -proot123 --all-databases --flush-logs --delete-master-logs --single-transaction > $DumpFile

    tar -czvf G Z D u m p F i l e < / s p a n > < s p a n c l a s s = " h l j s − v a r i a b l e " > GZDumpFile</span> <span class="hljs-variable"> GZDumpFile</span><spanclass="hljs−variable">DumpFile

    rm $DumpFile

    count=$(ls -l *.tgz |wc -l)

    if [ $count -ge 5 ]

    then

    file=$(ls -l *.tgz |awk ‘{print $9}’|awk ‘NR==1’)

    rm -f $file

    fi

    #只保留过去四周的数据库内容

    Last=date +<span class="hljs-string">"%Y年%m月%d日 %H:%M:%S"</span>

    echo 开始: B e g i n < / s p a n > 结 束 : < s p a n c l a s s = " h l j s − v a r i a b l e " > Begin</span> 结束:<span class="hljs-variable"> Begin</span>结束:<spanclass="hljs−variable">Last KaTeX parse error: Expected 'EOF', got '&' at position 24: …le</span> succ &̲gt;&gt; <span c…LogFile

    cd $BakDir/daily

    rm -f *

     

    赋予databak.sh 执行权限

    chmod a+x /home/mysql/databak.sh

    ?4、开启定时任务

    vi /etc/crontab

    #每个星期日凌晨3:00执行完全备份脚本

    0 3 * * 0 /home/mysql/databak.sh >/dev/null 2>&1

    #周一到周六凌晨3:00做增量备份

    0 3 * * 1-6 /home/mysql/binlogbak.sh >/dev/null 2>&1

     

     

    使上述定时任务生效

    crontab /etc/crontab

    查看定时任务

    crontab -l

    完成。

     

     

    Processed: 0.011, SQL: 9