#執行的sql
log=/tmp/logs/mysqld.log 
#記錄sql執行超過下面設定時間的sql
log-slow-queries = /tmp/mysqlslowquery.log
#執行時間大於等於1秒
long_query_time = 1 

然後你可以tail -f /tmp/logs/mysqld.log 監控所有執行的sql，同樣的方法可以監控mysqlslowquery.log 為執行時間超過long_query_time = 1（秒）的sql語句

比如通過第一步我們找到了某一個mysql 自定義函式執行慢func_getDevice(); 執行了15s,但並不知道這個方法裡面到底是那一條sql影響了效能，那麼就有了第二步。

第二步：進入mysql命令列，輸入

mysql> set profiling=1;
mysql> select func_getDevice(1);
mysql> show profiles;
+----------+------------+-----------------------+
| Query_ID | Duration   | Query                 |
+----------+------------+-----------------------+
|        1 | 0.00250400 | select * from TDevice |
+----------+------------+-----------------------+
1 row in set (0.00 sec)
 

這時候你就會看到一個詳細的sql執行列表，但預設只記錄15條sql，如果方法裡面的sql比較多，那麼可以通過設定

mysql> set profiling_history_size=20;
mysql> show variables like 'profiling%';
+------------------------+-------+
| Variable_name          | Value |
+------------------------+-------+
| profiling              | ON    |
| profiling_history_size | 15    |
+------------------------+-------+
2 rows in set (0.00 sec)
mysql> select func_getDevice(1);
mysql> show profiles;
 

mysql> show profile for query 1;詳細檢視執行一條sql的耗時情況
+--------------------------------+----------+
| Status                         | Duration |
+--------------------------------+----------+
| (initialization)               | 0.000003 | 
| checking query cache for query | 0.000042 | 
| Opening tables                 | 0.00001 | 
| System lock                    | 0.000004 | 
| Table lock                     | 0.000025 | 
| init                           | 0.000009 | 
| optimizing                     | 0.000003 | 

mysql> EXPLAIN select * from TDevice;
+----+-------------+---------+------+---------------+------+---------+------+------+-------+
| id | select_type | table   | type | possible_keys | key  | key_len | ref  | rows | Extra |
+----+-------------+---------+------+---------------+------+---------+------+------+-------+
|  1 | SIMPLE      | TDevice | ALL  | NULL          | NULL | NULL    | NULL |   70 |       |
+----+-------------+---------+------+---------------+------+---------+------+------+-------+
1 row in set (0.00 sec)