一.問題描述： 版本:mariadb 10.0.19啟動MySQL:/apps/svr/mariadb10/bin/mysqld_safe --defaults-file=/apps/dbdat/mariadb10_data3306/my.cnf  --socket=/tmp/mysql.sock &無法啟動，mysql進(jìn)程退出。

二.問題分析： 檢查error log，如下：160826 15:33:33 [Note] Server socket created on ip: '0.0.0.0'.160826 15:33:33 [ERROR] Can't start server : Bind on unix socket: Address already in use160826 15:33:33 [ERROR] Do you already have another mysqld server running on socket: /tmp/mysql3306.sock ?160826 15:33:33 [ERROR] Aborting

160826 15:33:33 [Note] InnoDB: FTS optimize thread exiting.160826 15:33:33 [Note] InnoDB: Starting shutdown...160826 15:33:36 [Note] InnoDB: Shutdown completed; log sequence number 452295626292160826 15:33:36 [Note] /apps/svr/mariadb10/bin/mysqld: Shutdown complete

160826 15:33:36 mysqld_safe mysqld from pid file /apps/dbdat/mariadb10_data3306/mysql.pid ended

上面提示unix socket /tmp/mysql3306.sock 被占用，用ps -ef|grep mysql檢查沒有其它mysql進(jìn)程.用netstat -an|grep 3306檢查端口也不存在。對比分析，正常關(guān)閉mysql時，/tmp/mysql3306.sock文件是不存在的，而現(xiàn)在mysql已關(guān)閉后，還存在/tmp/mysql3306.sock文件。猜測是由于上次mysql未正常關(guān)閉或關(guān)閉異常引起。 

三.問題解決：  在root用戶下，刪除/tmp/mysql.sock文件，再重啟mysql。