0

我正在关注这篇文章:

http://www.chrismoos.com/2010/01/31/mysql-partitioning-tables-with-millions-of-rows

但是,当我运行查询对产品表(包含 500,000 行)进行分区时,出现错误:

#1503 - A UNIQUE INDEX must include all columns in the table's partitioning function

我的查询是:

ALTER TABLE parts_library PARTITION by HASH(manufacturerId) PARTITIONS 200

我的主键是id和manufacturerId的复合键,和文章中的一样,所以不明白为什么会出现这个错误。

这是我的表的创建语句:

CREATE TABLE IF NOT EXISTS `parts_library` (
  `id` int(11) NOT NULL,
  `dateAdded` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
  `typeId` int(3) NOT NULL COMMENT 'Reference to part_types',
  `manufacturerId` int(11) NOT NULL DEFAULT '0',
  `familyId` int(11) NOT NULL DEFAULT '454',
  `partNumber` varchar(255) COLLATE utf8_unicode_ci NOT NULL COMMENT 'e.g. 6ES5123B62',
  `idealForm` varchar(255) COLLATE utf8_unicode_ci NOT NULL COMMENT 'E.g. 6ES5-123-B6/2',
  `comCodeId` int(11) DEFAULT NULL,
  `countryOriginId` int(3) NOT NULL DEFAULT '258',
  `minStockLevel` int(11) DEFAULT NULL,
  `weight` decimal(11,2) DEFAULT NULL,
  `width` decimal(11,2) DEFAULT NULL,
  `height` decimal(11,2) DEFAULT NULL,
  `depth` decimal(11,2) DEFAULT NULL,
  `validated` tinyint(1) NOT NULL DEFAULT '0',
  `onWeb` tinyint(1) DEFAULT '0',
  `indexed` tinyint(1) NOT NULL DEFAULT '1',
  `averageMargin` decimal(11,2) NOT NULL,
  PRIMARY KEY (`id`,`manufacturerId`),
  UNIQUE KEY `partNumber` (`partNumber`),
  KEY `fk_parts_library_parts_categories1` (`typeId`),
  KEY `fk_parts_library_manufacturers1` (`manufacturerId`),
  KEY `fk_parts_library_geo_countries1` (`countryOriginId`),
  KEY `fk_parts_library_parts_families1` (`familyId`),
  KEY `indexed` (`indexed`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;

如何对我的表进行分区?

4

1 回答 1

1

你的代码和arcticle之间的区别 - 你有一个

UNIQUE KEY 'partNumber' ('partNumber')

文章中没有提到。DB抱怨您需要将用于分区的列包含在键定义中,例如:

UNIQUE KEY 'partNumber' ('manufacturerId','partNumber')

于 2013-05-30T10:01:07.120 回答