需要注意的是，如果所有的数据都是通过DELETE从表中删除的(即没有WHERE子句)，那么只要a)权限允许，b)没有fk引用表(这里似乎就是这种情况)，使用TRUNCATE table将是首选，因为它可以更有效地DELETE并同时重置IDENTITY种子。以下细节取自TRUNCATE TABLE的MSDN页面:

Compared to the DELETE statement, TRUNCATE TABLE has the following advantages: Less transaction log space is used. The DELETE statement removes rows one at a time and records an entry in the transaction log for each deleted row. TRUNCATE TABLE removes the data by deallocating the data pages used to store the table data and records only the page deallocations in the transaction log. Fewer locks are typically used. When the DELETE statement is executed using a row lock, each row in the table is locked for deletion. TRUNCATE TABLE always locks the table (including a schema (SCH-M) lock) and page but not each row. Without exception, zero pages are left in the table. After a DELETE statement is executed, the table can still contain empty pages. For example, empty pages in a heap cannot be deallocated without at least an exclusive (LCK_M_X) table lock. If the delete operation does not use a table lock, the table (heap) will contain many empty pages. For indexes, the delete operation can leave empty pages behind, although these pages will be deallocated quickly by a background cleanup process. If the table contains an identity column, the counter for that column is reset to the seed value defined for the column. If no seed was defined, the default value 1 is used. To retain the identity counter, use DELETE instead.

下面是:

DELETE FROM [MyTable];
DBCC CHECKIDENT ('[MyTable]', RESEED, 0);

变成:

TRUNCATE TABLE [MyTable];

请参阅TRUNCATE TABLE文档(上面有链接)以获得有关限制等的其他信息。

DBCC CHECKIDENT ('TestTable', RESEED, 0)
GO

其中0是标识起始值

用新id重置标识列

DECLARE @MAX INT
SELECT @MAX=ISNULL(MAX(Id),0) FROM [TestTable]

DBCC CHECKIDENT ('[TestTable]', RESEED,@MAX)

DBCC CHECKIDENT management命令用于重置标识计数器。命令格式为:

DBCC CHECKIDENT (table_name [, { NORESEED | { RESEED [, new_reseed_value ]}}])
[ WITH NO_INFOMSGS ]

例子:

DBCC CHECKIDENT ('[TestTable]', RESEED, 0);
GO

以前版本的Azure SQL数据库不支持，但现在支持了。

多亏了Solomon Rutzky，该命令的文档现在已经修复。

使用这个存储过程:

IF (object_id('[dbo].[pResetIdentityField]') IS NULL)
  BEGIN
    EXEC('CREATE PROCEDURE [dbo].[pResetIdentityField] AS SELECT 1 FROM DUMMY');
  END
GO

SET  ANSI_NULLS ON
GO
SET  QUOTED_IDENTIFIER ON
GO

ALTER PROCEDURE [dbo].[pResetIdentityField]
  @pSchemaName NVARCHAR(1000)
, @pTableName NVARCHAR(1000) AS
DECLARE @max   INT;
DECLARE @fullTableName   NVARCHAR(2000) = @pSchemaName + '.' + @pTableName;

DECLARE @identityColumn   NVARCHAR(1000);

SELECT @identityColumn = c.[name]
FROM sys.tables t
     INNER JOIN sys.schemas s ON t.[schema_id] = s.[schema_id]
     INNER JOIN sys.columns c ON c.[object_id] = t.[object_id]
WHERE     c.is_identity = 1
      AND t.name = @pTableName
      AND s.[name] = @pSchemaName

IF @identityColumn IS NULL
  BEGIN
    RAISERROR(
      'One of the following is true: 1. the table you specified doesn''t have an identity field, 2. you specified an invalid schema, 3. you specified an invalid table'
    , 16
    , 1);
    RETURN;
  END;

DECLARE @sqlString   NVARCHAR(MAX) = N'SELECT @maxOut = max(' + @identityColumn + ') FROM ' + @fullTableName;

EXECUTE sp_executesql @stmt = @sqlString, @params = N'@maxOut int OUTPUT', @maxOut = @max OUTPUT

IF @max IS NULL
  SET @max = 0

print(@max)

DBCC CHECKIDENT (@fullTableName, RESEED, @max)
go

--exec pResetIdentityField 'dbo', 'Table'

回顾一下我的答案。我在sql server 2008 r2中遇到了一个奇怪的行为，你应该知道。

drop table test01

create table test01 (Id int identity(1,1), descr nvarchar(10))

execute pResetIdentityField 'dbo', 'test01'

insert into test01 (descr) values('Item 1')

select * from test01

delete from test01

execute pResetIdentityField 'dbo', 'test01'

insert into test01 (descr) values('Item 1')

select * from test01

第一个选择生成0,Item 1。

第二个生成1，项目1。如果在表创建后立即执行重置，则下一个值为0。老实说，我并不惊讶微软不能把这些东西做好。我发现它是因为我有一个填充引用表的脚本文件，有时在重新创建表后运行，有时在已经创建表时运行。

这里的大多数回复似乎都假定表为空，Identity值需要重置。 然而，我如何阅读这个问题是@xorpower现在有一个记录1,2,3,5,6,7,12,13,14等的表…并需要一个方法将其返回到一个连续列表。(1、2、3、4、5、6、7、8、9等…)

但是由于我必须从表中删除一些记录， 恕我直言，这里有个神奇的词。

AFAIK在MSSQL中没有这样的东西;在最坏的情况下，您确实可以将现有的记录表转储到一个新表中，然后从那里开始。我的问题是:你为什么要这样做?IDENTITY列是最好的方法吗?

Anyway,the solutions provided here that do care about existing data are mostly about copying everything in a temp-table, TRUNCATEing the existing table; reseeding the table and then copying everything back again. I'm sure that works but if you have a lot of data then this is a pretty heavy operation. Personally I would rather go with creating an identical table, copying the data in that new table (maybe in batches?) and then finally SWITCHing the data to the original table and dropping the newly created table again. You're likely to need to do a CHECKIDENT after the SWITCH. This way you only need to move the data from one table to another once. To save space you could even DELETE the relevant records from the original table after a batch is copied.

PS:是的，我知道这是一个老问题，但考虑到它的高得分，它仍然出现在类似问题的顶部，因为没有人提到SWITCH，它似乎值得添加。