Microsoft SQL Server provides a tool to connect to the SQL Server database to run SQL queries all from the command line. This is useful for database administrators who need to quickly connect to run a SQL command without having to start the SQL Server Management Studio (SSMS).

In this example I will concentrate on 3 simple examples to give a quick overview of what you can do with SQLCMD. The first example shows you how to connect to a SQL Server database using the Windows Authentication. The second example shows you how to connect to SQL Server database using the SQL Authentication and finally the third example shows you how to run a simple SELECT query. All examples are based on SQL Server 2005.

Connect to the master database using Windows Authentication

Using sqlcmd to connect to SQL Server using Windows Authentication

-E indicates Windows Authentication.
-S the SQL Server name instance. In this example it is Merovingian.
-d name of the database. In this example the database is master.

Connect to the master database using SQL Authentication

using sqlcmd to connect to sql server using sql authentication

-S the SQL Server name instance. In this example it is Merovingian.
-d name of the database. In this example the database is master.
-U the username used to login to SQL Server using SQL Authentication.
-P the password used to login to SQL Server using SQL Authentication.

SELECT query using SQLCMD

SELECT query using SQLCMD

In this example, we want to query the TOP 10 contacts displaying the first name and last name.

Some of you may have come across this error message in SQL Server 2005. SQL Server 2005: “Internal error. Buffer provided to read column value is too small. Run DBCC CHECKDB to check for any corruption”. If you run a DBCC CHECKDB, it will return without any errors.

When do you see the error in SQL Server 2005?

When you do a SELECT…INTO a temporary table. Example: SELECT * INTO #mytable FROM dbo.datatable

Note: where #mytable is the temporary table and dbo.datatable is the table in the database you are working on.

How do you resolve the error?

This was issued Microsoft:
This issue has been fixed in the Cumulative update package 2 for SQL Server 2005 Service Pack 2 (http://support.microsoft.com/kb/936305) as part of SQL Hotfix 50001299. Please install the more recent Cumulative update package 4 for SQL Server 2005 Service Pack 2 (http://support.microsoft.com/kb/941450) which contains this fix and more.

There was no easy way to produce a cross tabulation in SQL Server 2000 but in SQL Server 2005, this is achievable using the PIVOT command. Here is a example on how to produce a cross tabulation in SQL Server 2005.

In the example below, the first table, named EventResult contains the results for an athletic competition. The second table is what we will produce and it’s the cross tabulation of the EventResult table.

Cross tabulation in SQL Server 2005

The T-SQL example below shows you how to produce the cross tabulation mentioned earlier using the PIVOT command.

T-SQL example using PIVOT command to produce cross tabulation

Explanation: Let’s start with the sub-query within the T-SQL query.

What is being PIVOT are the values of the Event column, taking each value and making them into columns ([100m], [200m], [800m]) and each pivoted column will contain SUM(Time).

Sub-Query showing PIVOT in T-SQL

The SELECT command highlighted in grey shows the result of the pivoted table as Name, [100m], [200m], [800m].

SELECT command showing the PIVOT column headers

It may look a little confusing at first glance but I suggest that you create the example in SQL Server 2005 and try out the example and you will see the relationship between the EventResult table, the PIVOT table (cross tabulation) and the T-SQL.

SQL Server 2005

Here’s a quick guide on how to create indexed views in SQL Server 2005.

How do you create indexed views in SQL Server 2005?

1) Create a table in your database, e.g.

CREATE TABLE [dbo].[TestTableSC] (
[ID] [INT] NOT NULL,
[name] [varchar] (100) NULL,
[description] [varchar] (255) NULL
)

2) SET ANSI_NULLS, ANSI_WARNINGS ON….

SET NUMERIC_ROUNDABORT OFF;
SET ANSI_PADDING,
ANSI_WARNINGS,
CONCAT_NULL_YIELDS_NULL,
ARITHABORT,
QUOTED_IDENTIFIER,
ANSI_NULLS ON;
GO

3) In you database create a view WITH SCHEMABINDING, e.g.

CREATE VIEW [dbo].[TestTableSCView] WITH SCHEMABINDING AS
SELECT ID, name, description FROM dbo.TestTableSC

GO

4) Create a CLUSTERED INDEX

CREATE UNIQUE CLUSTERED INDEX TestTableSC_PKey
ON TestTableSCView(ID)
GO

Here are the Gotchas on creating indexed views in SQL Server 2005

You can only create a SCHEMABINDING view when the base table is in the same database. If the base table is in a different database or a linked server, you will get an error.

SQL Server does not allow SCHEMABINDING a view to a base table in a different database.

Example:

CREATE VIEW [dbo].[TestTableSCView2] WITH SCHEMABINDING AS
SELECT ID, name, description FROM database1.dbo.TestTableSC

GO

You’ll get the error:

Cannot schema bind view ‘dbo.TestTableSCView2′ because name ‘database1.dbo.TestTableSC’ is invalid for schema binding. Names must be in two-part format and an object cannot reference itself.

That concludes the quick guide to creating indexed views in SQL Server 2005.

SQLMail - Legacy in SQL Server 2005 64-bit

The legacy SQLMail support is no longer available in SQL Server 2005 64-bit. In lower version, i.e. SQL Server 2005 32-bit, SQL Mail is marked as a legacy functionality but you can still setup SQLMail and use it in your Stored Procedures. SQLMail has been superseded by DatabaseMail which does not require MAPI and the installation of Microsoft Outlook to run SQLMail.

Legacy SQLMail in SQL Server 2005

In SQL Server 2005 64-bit we have been advised to use DatabaseMail. This works perfectly for sending e-mails from SQL Server and it’s a whole lot easier when setting up DatabaseMail. The only drawback is you cannot read e-mails from an Outlook mailbox like how you could with SQLMail. Microsoft has not provided an alternative. The only way is to use a third party solution if you want to read e-mails from a mailbox.

In my next post, I will write about a third party solution that will allow you to read e-mails from a mailbox and process the e-mails within SQL Server.

SQL Server 2005

When maintaining a SQL Server database, you have the option to select if you want to restrict the growth of a database or transaction log or give SQL Server the liberty to grow automatically without setting a restriction. You select unrestricted file growth when you know you have enough disk space to cope with the growth of the database or transaction log over a period of time.

Recently, we had to make adjustments to cater for a larger transaction log but we were stumped when the “Unrestricted File Growth” option radio button kept reverting back to the “Restricted File Growth (MB)” option and the size of 2,097,152 MB. Every time I select the “Unrestricted File Growth” radio button and hit the save button, it reverts back to “Restricted File Growth (MB)”.

I discovered from Microsoft that it is a bug but Microsoft refused to call it a “bug”. Their explanation – there is a set limit of 2TB for “Unrestricted File Growth” for any database or transaction log. The Restricted File Growth (MB) size of 2,097,152 MB is 2TB. This means that the Restricted File Growth (MB) option is actually allowing auto growth until it reaches 2TB. This also means they have also confused database administrators by providing the “Unrestricted File Growth” option which is not selectable in SQL Server 2005.

So if you wish to allow your SQL Server database or transaction log to grow without a limit, select “Restricted File Growth (MB)” option and allow the database or transaction log to grow until it reaches the 2TB limit. That is the only work around for now.

See screenshot below.

SQL Server 2005 Autogrowth

How many times have you asked this question when you run a query. How do I copy the query result with the column names into Microsoft Excel?

In the SQL Server Management Studio, in the Tools menu, select Tools > Options > SQL Server > Results to Grid