Comments on: SQL SERVER – Unique Nonclustered Index Creation with IGNORE_DUP_KEY = ON

By: SQL SERVER – Weekly Series – Memory Lane – #010 « SQL Server Journey with SQL Authority

Sat, 05 Jan 2013 01:31:43 +0000

[...] Unique Nonclustered Index Creation with IGNORE_DUP_KEY = ON [...]

By: Ashok

Ashok — Fri, 28 Sep 2012 07:50:21 +0000

Hey man.

Yo are doing wonderful job.

Thanks for sharing,
Ashok.

By: Nagendra

Nagendra — Thu, 28 Jun 2012 09:29:21 +0000

Can I use IGNORE_DUP_KEY property on Clustered index

By: santhosh

santhosh — Thu, 07 Apr 2011 13:38:34 +0000

Hi Pinal,
Thanks for the article.
I’ve a question. Can i Insert duplicate values on a non clustered indexed column.
If true can you please expalin how.

Thanks in advance

By: Simon

Simon — Thu, 03 Mar 2011 19:44:44 +0000

Hi Dave, thanks for your excellent blogs – you always seem to answer questions simply and succinctly, without being patronising or drowning everything in endless complexity, or assuming that people already know almost everything – if tha was the case, they wouldn’t need to ask, would they?

My extension to this question is, if you have a duplicate key, is there a standard way to identify the duplicate? The “duplicate” message always drives me mad, since SQL Server must know what the duplicate is or it couldn’t say it was a duplicate. So why not add the key values to the duplicate message, to give you a fighting chance of doing something about it?

Not so hard on single inserts, but can be a nightmare for bulk inserts, esp where you may not have much control about the data contents.

By: Bharati

Bharati — Wed, 07 Jul 2010 02:30:40 +0000

Hi Pinal,

I have came across a problem in my inhouse website.

With Unique, Non-Clustered index and IGNORE_DUP_KEY=ON, duplicate value is inserted in a column, without throwing any error.

This happened only once. Otherwise, it is running as expected. When I tried to insert duplicate value, it does not allow, which is correct.

Only this one record has got duplicate value which is puzzling me. I am trying to figure out the reason.

Any help appreciated.

Thanks

By: Chavoux Luyt

Chavoux Luyt — Thu, 01 Jul 2010 12:23:15 +0000

Thanks Pinal

So do I understand correctly that even with IGNORE_DUP_KEY=ON, I still will not have duplicates in the table? If so, this is exactly what I wanted… then I don’t have to check my SELECT data for duplicates before inserting!

Cheers
Chavoux

By: samir

samir — Sat, 20 Mar 2010 15:08:27 +0000

thanks u r index query but i get error when i am trying to create UNIQUE NONCLUSTERED INDEX they give me

Error
—-Server: Msg 170, Level 15, State 1, Line 4
Line 4: Incorrect syntax near ‘(‘.

GO
CREATE TABLE DupIndex (ID INT,
FirstName VARCHAR(100),
LastName VARCHAR(100),
City VARCHAR(100))
GO

Go
CREATE UNIQUE NONCLUSTERED INDEX [IX_DupIndex_FirstName]
ON [dbo].[DupIndex]
(
[FirstName] ASC
) WITH (IGNORE_DUP_KEY = ON) ON [PRIMARY]
GO

Plz Help me ………..
and
How to data retrieval fast in server to client

By: Lewis Keller

Lewis Keller — Sat, 23 Jan 2010 23:57:36 +0000

Pinaldave,

Which version of SQL Server did you use? Did you use 2000, 2005, or 2008? I’m very curious about this.

By: Jay Patel

Jay Patel — Thu, 21 Jan 2010 19:03:16 +0000

In what scenarios would you want to turn the IGNORE_DUP_KEY setting ON?

From a data consistency standpoint, I would want it ‘OFF’, but perhaps there are other uses for it.

What do you guys use the ON setting for?

By: SQL SERVER – Unique Nonclustered Index Creation with IGNORE_DUP_KEY = ON – A Transactional Behavior Journey to SQL Authority with Pinal Dave

Wed, 06 Jan 2010 01:31:46 +0000

[...] 6, 2010 by pinaldave Earlier, I had written on SQL SERVER – Unique Nonclustered Index Creation with IGNORE_DUP_KEY = ON, and I received a comment regarding when this option can be useful. On the same day, I met Jacob [...]

By: Anand

Anand — Mon, 04 Jan 2010 21:22:50 +0000

Just a small update, it works with Unique Clustered Index too.

By: pinaldave

pinaldave — Mon, 04 Jan 2010 13:12:37 +0000

Marko,

I have follow up post demonstrating the usage of this particular feature.

Kind Regards,
Pinal

By: Pinal Dave

Pinal Dave — Mon, 04 Jan 2010 12:36:38 +0000

Hello Marko,

After implementing the uniqueness of a column if we take the below requirement:

When a duplicate value arrive in an INSERT statement, Should the INSERT statement failed without inserting a single record or just ignore the records of duplicate value?

The answer ‘yes’ or ‘no’ depends on business and SQL Server provides us an handy option to implement that is “IGNORE_DUP_KEY”.

Regards,
Pinal Dave

By: Ganesh

Ganesh — Mon, 04 Jan 2010 09:33:58 +0000

Thanks dave. Its very useful to me to take the unique records.

By: Marko Parkkola

Marko Parkkola — Mon, 04 Jan 2010 07:07:03 +0000

But if I’m creating UNIQUE index, why would I want it NOT to throw an error when trying to add duplicate value? I mean, I don’t see any situation where I wouldn’t want it to throw an error.