MySQL按月自动设置表分区

开始检查

首先，确保 ticket_history_info表是一个分区表。如果未设置分区，需要修改表结构以支持分区，将 ticket_history_info 表设置为按日期范围进行分区

ALTER TABLE ticket_history_info
PARTITION BY RANGE (TO_DAYS(CALL_DATE)) (
    PARTITION p0 VALUES LESS THAN (TO_DAYS('2022-09-01'))
);

1.创建分区函数

-- 创建分区函数
DELIMITER //
CREATE FUNCTION get_partition_name(p_date DATE)
RETURNS VARCHAR(20)
DETERMINISTIC
BEGIN
    DECLARE p_month VARCHAR(2);
    DECLARE p_year VARCHAR(4);
    SET p_month = LPAD(MONTH(p_date), 2, '0');
    SET p_year = YEAR(p_date);
    RETURN CONCAT('p', p_year, p_month); 
END//
DELIMITER ;

检查是否成功创建函数

SHOW FUNCTION STATUS LIKE 'get_partition_name';

2.创建存储过程，用于自动生成分区

-- 存储过程
DELIMITER //
CREATE PROCEDURE create_monthly_partition()
BEGIN
    DECLARE next_month VARCHAR(20);
    DECLARE next_month_first_day DATE;
    
    -- 计算下一个月份的名称和下个月的第一天
    SET next_month = CONCAT('p', DATE_FORMAT(DATE_ADD(CURDATE(), INTERVAL 1 MONTH), '%Y%m'));
    SET next_month_first_day = LAST_DAY(DATE_ADD(CURDATE(), INTERVAL 1 MONTH)) + INTERVAL 1 DAY;
    
    -- 检查分区是否已存在
    IF NOT EXISTS (
        SELECT NULL
        FROM information_schema.PARTITIONS
        WHERE TABLE_NAME = 'ticket_history_info' AND TABLE_SCHEMA = 'ccm' AND PARTITION_NAME = next_month
    ) THEN
        -- 创建新的分区
        SET @sql = CONCAT('ALTER TABLE ticket_history_info ADD PARTITION (PARTITION ', next_month, ' VALUES LESS THAN (TO_DAYS(\'', next_month_first_day, '\')))');

        PREPARE stmt FROM @sql;
        EXECUTE stmt;
        DEALLOCATE PREPARE stmt;
    END IF;
END//
DELIMITER ;

** 进行测试**

检查是否成功创建函数

SHOW PROCEDURE STATUS LIKE 'create_monthly_partition';

执行函数

CALL create_monthly_partition();

查询表分区，查看是否成功分区。

SELECT PARTITION_NAME FROM information_schema.PARTITIONS WHERE TABLE_NAME = '表名' AND TABLE_SCHEMA = '数据库名';

可通过修改本地系统时间，来进行反复测试是否可按照月份进行分区。

返回成功，显示已创建表分区。

3.创建自动删除半年以前的表空间函数

DELIMITER //

CREATE PROCEDURE delete_old_partitions()
BEGIN
    DECLARE done INT DEFAULT FALSE;
    DECLARE drop_partition_name VARCHAR(64);
    DECLARE cur CURSOR FOR 
        SELECT PARTITION_NAME 
        FROM information_schema.PARTITIONS 
        WHERE TABLE_SCHEMA = '数据库名' 
          AND TABLE_NAME = 'ticket_history_info' 
          AND PARTITION_DESCRIPTION < TO_DAYS(NOW() - INTERVAL 6 MONTH);
    DECLARE CONTINUE HANDLER FOR NOT FOUND SET done = TRUE;

    OPEN cur;

    read_loop: LOOP
        FETCH cur INTO drop_partition_name;
        IF done THEN
            LEAVE read_loop;
        END IF;
        SET @sql = CONCAT('ALTER TABLE ticket_history_info DROP PARTITION ', drop_partition_name);
        PREPARE stmt FROM @sql;
        EXECUTE stmt;
        DEALLOCATE PREPARE stmt;
    END LOOP;

    CLOSE cur;

   -- 删除超过半年的数据
    DELETE FROM ticket_history_info_copy1
    WHERE CALL_DATE < CURDATE() - INTERVAL 6 MONTH;

END//

DELIMITER ;

4.创建调度任务，修改到每月最后一天执行执行存储过程

CREATE EVENT IF NOT EXISTS monthly_partition_event
ON SCHEDULE EVERY 1 MONTH
STARTS CONCAT(DATE_FORMAT(LAST_DAY(CURRENT_DATE), '%Y-%m-'), DAY(LAST_DAY(CURRENT_DATE)), ' 23:00:00')
DO
BEGIN
    CALL create_monthly_partition();
    CALL delete_old_partitions();
END;

查看事件调度器是否启动

SET GLOBAL event_scheduler = ON;

查看事件

SHOW EVENTS;

或

SHOW EVENTS FROM `数据库名`;

创建成功

5.进行测试

插入测试数据进行测试，根据CALL_DATE字段进行数据分区，

INSERT INTO ticket_history_info (CALL_DATE, SRC_ADD, CALLING_NUM, DURATION) VALUES
('2024-05-24 00:00:00', 'Address1', '1234567890', 60),
('2024-06-20 00:00:00', 'Address2', '0987654321', 120),
('2024-05-10 00:00:00', 'Address3', '1122334455', 180),
('2024-10-01 00:00:00', 'Address4', '5566778899', 240);

-查询每个分区的行数

SELECT PARTITION_NAME, TABLE_ROWS
FROM information_schema.PARTITIONS
WHERE TABLE_SCHEMA = 'ccm' AND TABLE_NAME = 'ticket_history_info';

或者单独查询某一个分区的数据

SELECT *
FROM ticket_history_info PARTITION (p202406);