Chinaunix首页 | 论坛 | 博客
  • 博客访问: 442223
  • 博文数量: 56
  • 博客积分: 2705
  • 博客等级: 少校
  • 技术积分: 740
  • 用 户 组: 普通用户
  • 注册时间: 2008-08-21 13:52
文章分类

全部博文(56)

文章存档

2016年(1)

2014年(1)

2013年(5)

2012年(4)

2011年(8)

2010年(10)

2009年(15)

2008年(12)

分类: Mysql/postgreSQL

2011-07-28 10:47:50

MYSQL数据库安装完成后,默认最大连接数是100,一般流量稍微大一点的论坛或网站这个连接数是远远不够的,虽然通过连接池可以将连接数设置为固定的100,可是在下那BT同事将所有的连接池都设置成了100,话说我们小组有5个人……再话说,个人调试有必要设置那么高么……不过牢骚归牢骚,问题依然需要解决。

方法一:进入MYSQL安装目录 打开MYSQL配置文件 my.ini 或 my.cnf查找 max_connections=100   修改为 max_connections=1000 服务里重起MYSQL即可。此方法用于Windows下的mysql设置,简单方便。如果是Linux的话,请看下面。

方法二:MySQL的最大连接数默认是100

客户端登录:mysql -uroot -p

设置新的最大连接数为1000:mysql> set GLOBAL max_connections=1000

显示当前运行的Query:mysql> show processlist

显示当前状态:mysql> show status

退出客户端:mysql> exit

查看当前最大连接数:mysqladmin -uroot -p variables

这个办法治标不治本,重启服务以后最大连接数还是100。这个方法用于紧急扩充最大连接数用,不是长久之计。

方法三:以centos 5.4 下面的mysql 5.0.83 rpm版本为例说明:

找到/usr/bin/mysqld_safe编辑它,找到mysqld启动的那两行,在后面加上参数 :

-O max_connections=1000

用红字特别说明:

if test -z "$args"
  then
    $NOHUP_NICENESS $ledir/$MYSQLD $defaults --basedir=$MY_BASEDIR_VERSION --datadir=$DATADIR $USER_OPTION --pid-file=$pid_file --skip-external-locking
-O max_connections=1000
>> $err_log 2>&1
  else
    ev
al "$NOHUP_NICENESS $ledir/$MYSQLD $defaults --basedir=$MY_BASEDIR_VERSION --datadir=$DATADIR $USER_OPTION --pid-file=$pid_file --skip-external-locking $args -O max_connections=1000
>> $err_log 2>&1"
  fi

这段代码很不好找,在下也无法说清楚它具体在哪个位置。不过建议从最后往前找会比较快。

重启mysql服务:# service mysql restart

查看当前最大连接数:# /usr/local/mysql/bin/mysqladmin -uroot -p variables

输入root数据库账号的密码后可看到

max_connections 1000 即新改动已经生效。

第四种方法:修改mysqld的启动文件/etc/init.d/mysqld 在start()函数末尾添加代码

mysql -e 'set GLOBAL max_connections=1000;'如

start(){
    [ -x $exec ] || exit 5
    # check to see if it's already running
    RESPONSE=`/usr/bin/mysqladmin --socket="$socketfile" --user=UNKNOWN_MYSQL_USER ping 2>&1`
    if [ $? = 0 ]; then
        # already running, do nothing
        action $"Starting $prog: " /bin/true
        ret=0
    elif echo "$RESPONSE" | grep -q "Access denied for user"
    then
        # already running, do nothing
        action $"Starting $prog: " /bin/true
        ret=0
    else
        # prepare for start
        touch "$errlogfile"
        chown mysql:mysql "$errlogfile"
        chmod 0640 "$errlogfile"
        [ -x /sbin/restorecon ] && /sbin/restorecon "$errlogfile"
        if [ ! -d "$datadir/mysql" ] ; then
            # First, make sure $datadir is there with correct permissions
            if [ ! -e "$datadir" -a ! -h "$datadir" ]
            then
                mkdir -p "$datadir" || exit 1
            fi
            chown mysql:mysql "$datadir"
            chmod 0755 "$datadir"
            [ -x /sbin/restorecon ] && /sbin/restorecon "$datadir"
            # Now create the database
            action $"Initializing MySQL database: " /usr/bin/mysql_install_db --datadir="$datadir" --user=mysql
            ret=$?
            chown -R mysql:mysql "$datadir"
            if [ $ret -ne 0 ] ; then
                return $ret
            fi
        fi
        chown mysql:mysql "$datadir"
        chmod 0755 "$datadir"
        # Pass all the options determined above, to ensure consistent behavior.
        # In many cases mysqld_safe would arrive at the same conclusions anyway
        # but we need to be sure.  (An exception is that we don't force the
        # log-error setting, since this script doesn't really depend on that,
        # and some users might prefer to configure logging to syslog.)
        # Note: set --basedir to prevent probes that might trigger SELinux
        # alarms, per bug #547485
        $exec   --datadir="$datadir" --socket="$socketfile" \
                --pid-file="$mypidfile" \
                --basedir=/usr --user=mysql >/dev/null 2>&1 &
        safe_pid=$!
        # Spin for a maximum of N seconds waiting for the server to come up;
        # exit the loop immediately if mysqld_safe process disappears.
        # Rather than assuming we know a valid username, accept an "access
        # denied" response as meaning the server is functioning.
        ret=0
        STARTTIMEOUT=120
        while [ $STARTTIMEOUT -gt 0 ]; do
            RESPONSE=`/usr/bin/mysqladmin --socket="$socketfile" --user=UNKNOWN_MYSQL_USER ping 2>&1` && break
            echo "$RESPONSE" | grep -q "Access denied for user" && break
            if ! /bin/kill -0 $safe_pid 2>/dev/null; then
                echo "MySQL Daemon failed to start."
                ret=1
                break
            fi
            sleep 1
            let STARTTIMEOUT=${STARTTIMEOUT}-1
        done
        if [ $STARTTIMEOUT -eq 0 ]; then
            echo "Timeout error occurred trying to start MySQL Daemon."
            ret=1
        fi
        if [ $ret -eq 0 ]; then
            action $"Starting $prog: " /bin/true
            touch $lockfile
        else
            action $"Starting $prog: " /bin/false
        fi
    fi
    mysql -e 'set GLOBAL max_connections=1000;'
    return $ret
}

阅读(9091) | 评论(0) | 转发(1) |
给主人留下些什么吧!~~