There is a long list of new features getting introduced in SQL Server 2016. In this post, we would talk about one of the security feature called Dynamic Data Masking. How to Use it?

Whenever you access your account on your bank site, would you be comfortable in seeing your credit card or bank account number in clear text on the web page? There are multiple ways to do this at the application level, but as a human nature, it leaves room for error. One small mistake from the developer can leak sensitive data and can cost a huge loss. Wouldn’t it be great if a credit card number would be returned with only its last 4 digits visible – XXXX-XXXX-XXXX-1234 with no additional coding? Sounds interesting, read on!

Before experimenting this feature please remember that if you are using CTP2.0 then you need to turn on trace flags using below command.

DBCC TRACEON(209,219,-1)

If you don’t enable, then here is the error which you would receive while trying this sample script given later.

Msg 102, Level 15, State 1, Line 14
Incorrect syntax near ‘masked’.

Don't forget that this is SQL Server 2016 feature. Running the script on earlier version of SQL would cause below:

Msg 102, Level 15, State 1, Line 14
Incorrect syntax near ‘MASKED’.
Msg 319, Level 15, State 1, Line 14
Incorrect syntax near the keyword ‘with’. If this statement is a common table expression, an xmlnamespaces clause or a change tracking context clause, the previous statement must be terminated with a semicolon.

And here is the script to create objects (database, table, masked column):

SET NOCOUNT ON
GO

Drop database MaskingDemo, if already exists

USE [master]
GO
IF DB_ID('MaskingDemo') IS NOT NULL
BEGIN
ALTER DATABASE [MaskingDemo] SET SINGLE_USER WITH ROLLBACK IMMEDIATE
DROP DATABASE [MaskingDemo]
END

Create new database called MaskingDemo

CREATE DATABASE MaskingDemo
GO
USE MaskingDemo
GO

Create table with different data type columns

CREATE TABLE MyContacts (
ID INT IDENTITY(1, 1) PRIMARY KEY
,fName NVARCHAR(30) NOT NULL
,lName NVARCHAR(30) NOT NULL
,CreditCard VARCHAR(20) NULL
,SalaryINR INT NULL
,OfficeEmail NVARCHAR(60) NULL
,PersonalEmail NVARCHAR(60) NULL
,SomeDate DATETIME NULL
)

Insert a Row

INSERT INTO [dbo].[MyContacts]
([fName],[lName] ,[CreditCard],[SalaryINR],[OfficeEmail],[PersonalEmail], SomeDate)
VALUES('Rebecca','C','1234-5678-1234-5678',999999,'[email protected]','[email protected]', '31-March-2013')
GO

Apply Masking

ALTER TABLE MyContacts
ALTER COLUMN CreditCard ADD MASKED
WITH (FUNCTION = 'partial(2,"XX-XXXX-XXXX-XX",2)')
ALTER TABLE MyContacts
ALTER COLUMN SalaryINR ADD MASKED
WITH (FUNCTION = 'default()')      -- default on int
ALTER TABLE MyContacts
ALTER COLUMN SomeDate ADD MASKED
WITH (FUNCTION = 'default()')      -- default on date
ALTER TABLE MyContacts
ALTER COLUMN fname ADD MASKED
WITH (FUNCTION = 'default()')      -- default on varchar
ALTER TABLE MyContacts
ALTER COLUMN OfficeEmail ADD MASKED
WITH (FUNCTION = 'email()')
GO

Create a new user and grant select permissions

USE MaskingDemo
GO
CREATE USER WhoAmI WITHOUT LOGIN;
GRANT SELECT ON MyContacts TO WhoAmI;

As we can see above, those fields which are masked are showing obfuscated data based on masking rule.
For your information, versions after CTP2 release, the trace flag will not be needed. If you add trace flag, you would start getting “Incorrect syntax” error.

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.

In this article, I'm introducing you new feature in SQL Server 2016 CTP 2.0 named Query Store. This is a very useful feature for the DBA and developers from the performance point of view.

Query store feature allows to captures multiple query plan for a query and run time statistics. Query store can store multiple execution plans per query, it can force query processor to use a particular execution plan which is referred as plan forcing using USE PLAN query hint.

By default, Query Store is not active so you can enable it in two ways:

Step 1

First Using SSMS, Right Click on DatabaseName -> Go to properties -> Query Store options -> Enable -> True

Step 2

Second way to enable it by using ALTER Database script in this manner:

ALTER DATABSE  Database_name SET QUERY_STORE = ON;

Query store option is not enabled for master or tempdb database. If you try to enable it then you get below error:

Msg 12420, Level 16, State 1, Line 1

Cannot perform action because Query Store is not started up for this database.

Msg 5069, Level 16, State 1, Line 1

ALTER DATABASE statement failed.

Step 3

To determine the current options available for query store we can query the system view sys.database_query_store_options. Query stores contains two stores:

1. Plan store – Stores execution plan information
2. Running Stats store – Stores execution statistics information

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.

In this article, I will explain you how to parse XML or import XML to SQL Server Database Table.

Here are the XML File:

<?xml version="1.0"?>
<Orders>
<Order OrderNumber="99503" OrderDate="2013-10-20">
  <Address Type="Shipping">
    <Name>Ellen Adams</Name>
    <Street>123 Maple Street</Street>
    <City>Mill Valley</City>
    <State>CA</State>
    <Zip>10999</Zip>
    <Country>USA</Country>
  </Address>
  <Address Type="Billing">
    <Name>Tai Yee</Name>
    <Street>8 Oak Avenue</Street>
    <City>Old Town</City>
    <State>PA</State>
    <Zip>95819</Zip>
    <Country>USA</Country>
  </Address>
  <DeliveryNotes>Please leave packages in shed by driveway.</DeliveryNotes>
  <Items>
    <Item PartNumber="872-AA">
      <ProductName>Lawnmower</ProductName>
      <Quantity>1</Quantity>
      <USPrice>148.95</USPrice>
      <Comment>Confirm this is electric</Comment>
    </Item>
    <Item PartNumber="926-AA">
      <ProductName>Baby Monitor</ProductName>
      <Quantity>2</Quantity>
      <USPrice>39.98</USPrice>
      <ShipDate>2013-05-21</ShipDate>
    </Item>
  </Items>
</Order>
</Orders>

To Query XML File, you need to store in to xml variable @MyXML:

DECLARE @MyXML XML
SET @MyXML = '<?xml version="1.0"?>
<Orders>
<Order OrderNumber="99503" OrderDate="2013-10-20">
  <Address Type="Shipping">
    <Name>....</Name>
    <Street>123 Maple Street</Street>
    <City>Mill Valley</City>
    <State>CA</State>
    <Zip>10999</Zip>
    <Country>...</Country>
  </Address>
  <Address Type="Billing">
    <Name>Tai Yee</Name>
    <Street>8 Oak Avenue</Street>
    <City>Old Town</City>
    <State>PA</State>
    <Zip>95819</Zip>
    <Country>...</Country>
  </Address>
  <DeliveryNotes>Please leave packages in shed by driveway.</DeliveryNotes>
  <Items>
    <Item PartNumber="872-AA">
      <ProductName>Lawnmower</ProductName>
      <Quantity>1</Quantity>
      <USPrice>148.95</USPrice>
      <Comment>Confirm this is electric</Comment>
    </Item>
    <Item PartNumber="926-AA">
      <ProductName>Baby Monitor</ProductName>
      <Quantity>2</Quantity>
      <USPrice>39.98</USPrice>
      <ShipDate>2013-05-21</ShipDate>
    </Item>
  </Items>
</Order>
</Orders>

Now, you use SQL query to get Order header from @MyXML:

SELECT Col.value('@OrderNumber', 'int') AS 'Order',
Col.value('@OrderDate', 'date') AS 'OrderDate',
Col.value('(Address[@Type=''Shipping'']/Name/text())[1]', 'nvarchar(50)') AS 'Shipping_Name',
Col.value('(Address[@Type=''Shipping'']/Street/text())[1]', 'nvarchar(50)') AS 'Shipping_Street',
Col.value('(Address[@Type=''Shipping'']/City/text())[1]', 'nvarchar(50)') AS 'Shipping_City',
Col.value('(Address[@Type=''Shipping'']/State/text())[1]', 'nvarchar(50)') AS 'Shipping_State',
Col.value('(Address[@Type=''Shipping'']/Zip/text())[1]', 'nvarchar(50)') AS 'Shipping_Zip',
Col.value('(Address[@Type=''Shipping'']/Country/text())[1]', 'nvarchar(50)') AS 'Shipping_Country',
Col.value('(Address[@Type=''Billing'']/Name/text())[1]', 'nvarchar(50)') AS 'Billing_Name',
Col.value('(Address[@Type=''Billing'']/Street/text())[1]', 'nvarchar(50)') AS 'Billing_Street',
Col.value('(Address[@Type=''Billing'']/City/text())[1]', 'nvarchar(50)') AS 'Billing_City',
Col.value('(Address[@Type=''Billing'']/State/text())[1]', 'nvarchar(50)') AS 'Billing_State',
Col.value('(Address[@Type=''Billing'']/Zip/text())[1]', 'nvarchar(50)') AS 'Billing_Zip',
Col.value('(Address[@Type=''Billing'']/Country/text())[1]', 'nvarchar(50)') AS 'Billing_Country',
Col.value('(DeliveryNotes/text())[1]', 'nvarchar(250)') AS 'DeliveryNotes'
FROM @MyXML.nodes('/Orders/Order') AS T(Col)

Then, you can use SQL query to get Order Items from @MyXML:

SELECT
Col.value('(../../../Order/@OrderNumber)[1]', 'int') AS 'Order',
Col.value('(@PartNumber)[1]', 'nvarchar(50)') AS 'PartNumber' ,
Col.value('(ProductName/text())[1]', 'nvarchar(250)') AS 'ProductName' ,
Col.value('(Quantity/text())[1]', 'int') AS 'Quantity',
Col.value('(USPrice/text())[1]', 'float') AS 'USPrice',
Col.value('(Comment/text())[1]', 'nvarchar(250)') AS 'Comment',
Col.value('(ShipDate/text())[1]', 'date') AS 'ShipDate'
FROM @MyXML.nodes('/Orders/Order/Items/Item') AS T(Col)

To make it easier, here's the full code:

DECLARE @MyXML XML
SET @MyXML = '<?xml version="1.0"?>
<Orders>
<Order OrderNumber="99503" OrderDate="2013-10-20">
  <Address Type="Shipping">
    <Name>...</Name>
    <Street>123 Maple Street</Street>
    <City>Mill Valley</City>
    <State>CA</State>
    <Zip>10999</Zip>
    <Country>...</Country>
  </Address>
  <Address Type="Billing">
    <Name>Tai Yee</Name>
    <Street>8 Oak Avenue</Street>
    <City>Old Town</City>
    <State>PA</State>
    <Zip>95819</Zip>
    <Country>...</Country>
  </Address>
  <DeliveryNotes>Please leave packages in shed by driveway.</DeliveryNotes>
  <Items>
    <Item PartNumber="872-AA">
      <ProductName>Lawnmower</ProductName>
      <Quantity>1</Quantity>
      <USPrice>148.95</USPrice>
      <Comment>Confirm this is electric</Comment>
    </Item>
    <Item PartNumber="926-AA">
      <ProductName>Baby Monitor</ProductName>
      <Quantity>2</Quantity>
      <USPrice>39.98</USPrice>
      <ShipDate>2013-05-21</ShipDate>
    </Item>
  </Items>
</Order>
</Orders>'
   SELECT
    Col.value('@OrderNumber', 'int') AS 'Order',
    Col.value('@OrderDate', 'date') AS 'OrderDate',
    Col.value('(Address[@Type=''Shipping'']/Name/text())[1]', 'nvarchar(50)') AS 'Shipping_Name',
    Col.value('(Address[@Type=''Shipping'']/Street/text())[1]', 'nvarchar(50)') AS 'Shipping_Street',
    Col.value('(Address[@Type=''Shipping'']/City/text())[1]', 'nvarchar(50)') AS 'Shipping_City',
    Col.value('(Address[@Type=''Shipping'']/State/text())[1]', 'nvarchar(50)') AS 'Shipping_State',
    Col.value('(Address[@Type=''Shipping'']/Zip/text())[1]', 'nvarchar(50)') AS 'Shipping_Zip',
    Col.value('(Address[@Type=''Shipping'']/Country/text())[1]', 'nvarchar(50)') AS 'Shipping_Country',
    Col.value('(Address[@Type=''Billing'']/Name/text())[1]', 'nvarchar(50)') AS 'Billing_Name',
    Col.value('(Address[@Type=''Billing'']/Street/text())[1]', 'nvarchar(50)') AS 'Billing_Street',
    Col.value('(Address[@Type=''Billing'']/City/text())[1]', 'nvarchar(50)') AS 'Billing_City',
    Col.value('(Address[@Type=''Billing'']/State/text())[1]', 'nvarchar(50)') AS 'Billing_State',
    Col.value('(Address[@Type=''Billing'']/Zip/text())[1]', 'nvarchar(50)') AS 'Billing_Zip',
    Col.value('(Address[@Type=''Billing'']/Country/text())[1]', 'nvarchar(50)') AS 'Billing_Country', 
    Col.value('(DeliveryNotes/text())[1]', 'nvarchar(250)') AS 'DeliveryNotes'
  FROM  @MyXML.nodes('/Orders/Order') AS T(Col)
  
   SELECT
    Col.value('(../../../Order/@OrderNumber)[1]', 'int') AS 'Order',
    Col.value('(@PartNumber)[1]', 'nvarchar(50)') AS 'PartNumber'  ,
    Col.value('(ProductName/text())[1]', 'nvarchar(250)') AS 'ProductName' ,
    Col.value('(Quantity/text())[1]', 'int') AS 'Quantity',
    Col.value('(USPrice/text())[1]', 'float') AS 'USPrice',
    Col.value('(Comment/text())[1]', 'nvarchar(250)') AS 'Comment',
    Col.value('(ShipDate/text())[1]', 'date') AS 'ShipDate' 
  FROM  @MyXML.nodes('/Orders/Order/Items/Item') AS T(Col)

Here is the Output:

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.

In this article, you will play around with some SQL Script to identify blocking behaviour. The most basic script that I have been using and used by many DBA will include Activity Monitor, sp_who2, sysprocesses etc.

Recently, I saw a script written by one person:

SELECT * FROM sys.dm_os_waiting_tasks
WHERE blocking_session_id IS NOT NULL
GO

This will shows the rows like:

This is so cool because I know who is waiting for whom. In the above image 53 is waiting for 68. While 79 is waiting for 53. This was a cool way to look at things. Sure, you wanted to show the same data in slightly different way using T-SQL. So, I wrote a Blocking Tree TSQL script:

SET NOCOUNT ON
GO
SELECT SPID, BLOCKED, REPLACE (REPLACE (T.TEXT, CHAR(10), ' '), CHAR (13), ' ' ) AS BATCH
INTO #T
FROM sys.sysprocesses R CROSS APPLY sys.dm_exec_sql_text(R.SQL_HANDLE) T
GO
WITH BLOCKERS (SPID, BLOCKED, LEVEL, BATCH)
AS
(
SELECT SPID,
BLOCKED,
CAST (REPLICATE ('0', 4-LEN (CAST (SPID AS VARCHAR))) + CAST (SPID AS VARCHAR) AS VARCHAR (1000)) AS LEVEL,
BATCH FROM #T R
WHERE (BLOCKED = 0 OR BLOCKED = SPID)
AND EXISTS (SELECT * FROM #T R2 WHERE R2.BLOCKED = R.SPID AND R2.BLOCKED <> R2.SPID)
UNION ALL
SELECT R.SPID,
R.BLOCKED,
CAST (BLOCKERS.LEVEL + RIGHT (CAST ((1000 + R.SPID) AS VARCHAR (100)), 4) AS VARCHAR (1000)) AS LEVEL,
R.BATCH FROM #T AS R
INNER JOIN BLOCKERS ON R.BLOCKED = BLOCKERS.SPID WHERE R.BLOCKED > 0 AND R.BLOCKED <> R.SPID
)
SELECT N'    ' + REPLICATE (N'|         ', LEN (LEVEL)/4 - 1) +
CASE WHEN (LEN(LEVEL)/4 - 1) = 0
THEN 'HEAD -  '
ELSE '|------  ' END
+ CAST (SPID AS NVARCHAR (10)) + N' ' + BATCH AS BLOCKING_TREE
FROM BLOCKERS ORDER BY LEVEL ASC
GO
DROP TABLE #T
GO

The output would look like:

That's a simple way to look at the same Blocking data inside SSMS.

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.

SQL Server database files are organized in 8KB (8192 bytes) chunks, called pages. When we create the first row in a table, SQL Server allocates an 8KB page to store that row. Similarly every row in every table ends up being stored in a page.

Say one of the pages in your table is corrupt and while repairing the corrupt pages, you may eventually end up loosing some data. You may want to find out which records are on the page. To do so, use the following undocumented T-SQL %%physloc%% virtual column:

USE AdventureWorks2014
GO
SELECT *, %%physloc%% AS physloc
FROM Person.AddressType
ORDER BY physloc;

As you can see, the last column represents the record location. However the hexadecimal value is not in a human readable format. To read the physical record of each row in a human readable format, use the following query:

SELECT *
FROM Person.AddressType
CROSS APPLY sys.fn_PhysLocCracker(%%physloc%%)

The sys.fun_PhysLocCracker function takes the %%physloc%% and represents a human readable format fileid, pageid i.e. 880 and record number on the page 880.

If you are interested in knowing what’s inside the sys.fn_PhysLocCracker function, use sp_helptext as follows:

EXEC sp_helptext 'sys.fn_PhysLocCracker'
which display the definition of sys.fn_PhysLocCracker
-------------------------------------------------------------------------------
-- Name: sys.fn_PhysLocCracker
--
-- Description:
--    Cracks the output of %%physloc%% virtual column
--
-- Notes:
-------------------------------------------------------------------------------
create function sys.fn_PhysLocCracker (@physical_locator binary (8))
returns @dumploc_table table
(
    [file_id]    int not null,
    [page_id]    int not null,
    [slot_id]    int not null
)
as
begin
    declare @page_id    binary (4)
    declare @file_id    binary (2)
    declare @slot_id    binary (2)
    -- Page ID is the first four bytes, then 2 bytes of page ID, then 2 bytes of slot
    --
    select @page_id = convert (binary (4), reverse (substring (@physical_locator, 1, 4)))
    select @file_id = convert (binary (2), reverse (substring (@physical_locator, 5, 2)))
    select @slot_id = convert (binary (2), reverse (substring (@physical_locator, 7, 2)))
  
    insert into @dumploc_table values (@file_id, @page_id, @slot_id)
    return
end

The undocumented sys.fn_PhysLocCracker works on SQL Server 2008 and above.

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.

Today, I'm gonna talk about how to handle error in SQL Server using THROW. In versions prior to SQL Server 2012, you can use @@RAISE_ERROR to generate error messages dynamically or using the sys.messages catalog.

Consider the following example:

SELECT ROUND(800.0, -3)

On executing this statement, you get the following error:

This error is caused by the value does not fit into the decimal data type.

You can use @@RAISERROR to raise a message:

BEGIN TRY
SELECT ROUND(800.0, -3)
END TRY
BEGIN CATCH
DECLARE @ErrorMsg nvarchar(1000), @Severity int
SELECT @ErrorMsg = ERROR_MESSAGE(),
@Severity = ERROR_SEVERITY()
RAISERROR (@ErrorMsg, @Severity, 1)
END CATCH

Note: The old syntax of RAISERROR syntax specifying the error number and message number got deprecated (RAISEERROR 50005 ‘Exception Occurred’). Instead the new syntax RAISEERROR(50005, 10, 1) allowed you to specify the messageid, severity and state). For new applications use THROW.

However in SQL Server 2012 and above, there’s a better way to this without much efforts, using THROW. Consider the following code:

BEGIN TRY
SELECT ROUND(800.0, -3)
END TRY
BEGIN CATCH
THROW
END CATCH

As you can see, with just one word THROW, we were able to handle the error with grace and get a result too.

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.

If you want to know if a table has a primary key or foreign key, you can find out this in many ways. Just follow these steps on this tutorial.

Step 1

First, let us create these a table using code below:

CREATE TABLE product_master
(
prod_id INT PRIMARY KEY,
prod_name VARCHAR(100),
price DECIMAL(12,2)
)
GO
CREATE TABLE product_details
(
prod_id INT,
sales_date DATETIME,
sales_qty INT,
sales_amount DECIMAL(16,2)
)
GO
CREATE TABLE company_master
(
compnay_id INT,
company_name VARCHAR(100),
address VARCHAR(1000)
)
GO

Step 2

Now, let us create foreign key:

ALTER TABLE product_details ADD CONSTRAINT ck_item FOREIGN KEY(prod_id) REFERENCES product_master(prod_id)

Step 3

If you want to check if a table has a primary key, you can use the following methods:

1. Use sp_pkeys system stored procedure

The result will be:

2. Use Objectproperty function

SELECT
*
FROM
INFORMATION_SCHEMA.TABLES
WHERE
OBJECTPROPERTY(OBJECT_ID(TABLE_NAME),'TABLEHASPRIMARYKEY')=1 AND
TABLE_TYPE='BASE TABLE'
ORDER BY
TABLE_NAME

Step 4

If you want to check if a table has a foreign key, you can use the following method:

SELECT
*
FROM
INFORMATION_SCHEMA.TABLES
WHERE
OBJECTPROPERTY(OBJECT_ID(TABLE_NAME),'TABLEHASFOREIGNKEY')=1 AND
TABLE_TYPE='BASE TABLE'
ORDER BY
TABLE_NAME

And the result is:

Step 5

If you want to check for the tables that do not have primary key or foreign key, you can use the following method:

SELECT
*
FROM
INFORMATION_SCHEMA.TABLES
WHERE
OBJECTPROPERTY(OBJECT_ID(TABLE_NAME),'TABLEHASPRIMARYKEY')=0 AND
OBJECTPROPERTY(OBJECT_ID(TABLE_NAME),'TABLEHASFOREIGNKEY')=0 AND
TABLE_TYPE='BASE TABLE'
ORDER BY
TABLE_NAME

This is the result:

Easy right?

HostForLIFE.eu SQL Server 2016 Hosting
HostForLIFE.eu is European Windows Hosting Provider which focuses on Windows Platform only. We deliver on-demand hosting solutions including Shared hosting, Reseller Hosting, Cloud Hosting, Dedicated Servers, and IT as a Service for companies of all sizes.