17

Given the following

-- table ddl
create table dbo.f_word(
    sentence_id int NULL,
    sentence_word_id int NULL,
    word_id int NULL,
    lemma_id int NULL,
    source_id int NULL,
    part_of_speech_id int NULL,
    person_id int NULL,
    gender_id int NULL,
    number_id int NULL,
    tense_id int NULL,
    voice_id int NULL,
    mood_id int NULL,
    case_id int NULL,
    degree_id int NULL,
    citation nvarchar(100) NULL
);
-- create partition function
create partition function pf_f_word_source_id (int)
as range left for values 
(
    1,2,3,4,5,6,7,8,9,10,11,12,13,14,
    15,16,17,18,19,20,21,22,23
);

-- create the partition scheme
create partition scheme ps_f_word as partition pf_f_word_source_id to 
(
    [primary],[primary],[primary],[primary],[primary],[primary],[primary],[primary],[primary],
    [primary],[primary],[primary],[primary],[primary],[primary],[primary],[primary],[primary],
    [primary],[primary],[primary],[primary],[primary],[primary]
);

-- partition the index
create unique clustered index cix_fword on dbo.f_word 
(
    source_id,
    sentence_id,
    sentence_word_id,
    word_id,
    lemma_id,
    part_of_speech_id,
    person_id,
    gender_id,
    number_id,
    tense_id,
    voice_id,
    mood_id,
    case_id,
    degree_id 
)
on ps_f_word (source_id);

-- swapin table ddl

create table dbo.f_word_swapin(
    sentence_id int NULL,
    sentence_word_id int NULL,
    word_id int NULL,
    lemma_id int NULL,
    source_id int NULL,
    part_of_speech_id int NULL,
    person_id int NULL,
    gender_id int NULL,
    number_id int NULL,
    tense_id int NULL,
    voice_id int NULL,
    mood_id int NULL,
    case_id int NULL,
    degree_id int NULL,
    citation nvarchar(100) NULL
) on [primary];

-- create the same index on the swapin table
create unique clustered index cix_fword_swapin on dbo.f_word_swapin 
(
    source_id,
    sentence_id,
    sentence_word_id,
    word_id,
    lemma_id,
    part_of_speech_id,
    person_id,
    gender_id,
    number_id,
    tense_id,
    voice_id,
    mood_id,
    case_id,
    degree_id 
);

-- add check constraints WITH CHECK
ALTER TABLE dbo.f_word_swapin
WITH CHECK
ADD CONSTRAINT ck_f_word_swapin_lb
CHECK ( source_id > 12);

ALTER TABLE dbo.f_word_swapin
WITH CHECK
ADD CONSTRAINT ck_f_word_swapin_ub
CHECK ( source_id <= 13);

Then, move the data around:

-- switch data OUT of the partitioned table
ALTER TABLE dbo.f_word
SWITCH PARTITION 13 TO dbo.f_word_swapin;

-- attempt to switch data back IN 
ALTER TABLE dbo.f_word_swapin
SWITCH TO dbo.f_word PARTITION 13;

Below is the "Script Table As ... CREATE" DDL just to verify the same table structures.

/****** Object:  Table [dbo].[f_word_swapin]    Script Date: 9/10/2014 10:01:01 AM ******/
SET ANSI_NULLS ON
GO

SET QUOTED_IDENTIFIER ON
GO

CREATE TABLE [dbo].[f_word_swapin](
    [sentence_id] [int] NULL,
    [sentence_word_id] [int] NULL,
    [word_id] [int] NULL,
    [lemma_id] [int] NULL,
    [source_id] [int] NULL,
    [part_of_speech_id] [int] NULL,
    [person_id] [int] NULL,
    [gender_id] [int] NULL,
    [number_id] [int] NULL,
    [tense_id] [int] NULL,
    [voice_id] [int] NULL,
    [mood_id] [int] NULL,
    [case_id] [int] NULL,
    [degree_id] [int] NULL,
    [citation] [nvarchar](100) NULL
) ON [PRIMARY]

/****** Object:  Table [dbo].[f_word]    Script Date: 9/10/2014 10:09:43 AM ******/
SET ANSI_NULLS ON
GO

SET QUOTED_IDENTIFIER ON
GO

CREATE TABLE [dbo].[f_word](
    [sentence_id] [int] NULL,
    [sentence_word_id] [int] NULL,
    [word_id] [int] NULL,
    [lemma_id] [int] NULL,
    [source_id] [int] NULL,
    [part_of_speech_id] [int] NULL,
    [person_id] [int] NULL,
    [gender_id] [int] NULL,
    [number_id] [int] NULL,
    [tense_id] [int] NULL,
    [voice_id] [int] NULL,
    [mood_id] [int] NULL,
    [case_id] [int] NULL,
    [degree_id] [int] NULL,
    [citation] [nvarchar](100) NULL
)

GO

SWITCHing OUT works just fine. SWITCHing IN produces the following error:

Msg 4972, Level 16, State 1, Line 1 ALTER TABLE SWITCH statement failed. Check constraints or partition function of source table 'greek.dbo.f_word_swapin' allows values that are not allowed by check constraints or partition function on target table 'greek.dbo.f_word'.

Running:

select target_partition_id = $PARTITION.pf_f_word_source_id(source_id), 
    *
from dbo.f_word_swapin;

verifies that all data should go back into partition 13.

I'm really pretty new to partitioning so I'm sure that I'm doing things incorrectly, I just don't know what it is.

1
  • This also can happen if at least on one of the tables your check constraint was created WITH NOCHECK.
    – user62502
    Commented Mar 31, 2015 at 18:30

1 Answer 1

22

The thing about CHECK constraints is they only disallow rows for which the predicate returns FALSE. If the check returns UNKNOWN, that is not FALSE, so the row passes the check:

CREATE TABLE dbo.T1 (id int NULL CHECK (id = 1));

INSERT dbo.T1 VALUES (1); -- Ok
INSERT dbo.T1 VALUES (2); -- Error
INSERT dbo.T1 VALUES (NULL); -- Ok!

Your check constraint does not disallow NULL values, which is the out-of-range 'value' the SWITCH statement is objecting to. Your switch-in table might contain nulls, which do not belong in partition 2.

Add AND source_id IS NOT NULL to your CHECK constraint, when the destination partition is not partition 1 (where the nulls go).

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.