SqlServer Possible Errors

1.The scenario was:
   - A stored procedure that uses Linked Server to pull record-set from another SQL Server database.
   - Pull records and load this result set on a temp-table.
   - For implementing the business logic and to do some calculations the query uses some aggregate functions
     like SUM(), MIN(), MAX() on a few columns in that temp table.

Here if any of these aggregated column contain NULL values, it results in an obvious Warning message (not Error), i.e.: Warning: Null value is eliminated by an aggregate or other SET operation.
We could ignore this warning message, but when used in SSIS packages it causes the package to fail. So to avoid this Warning message, “SET ANSI_WARNINGS OFF” option was set at the beginning of the stored procedure.

The stored procedure compiled fine, but when we executed it, it resulted into this strange error: “Heterogeneous queries require the ANSI_NULLS and ANSI_WARNINGS…”.

We observed that this error is due to setting ANSI_WARNINGS to OFF just before the linked server call.
So the workaround was to shift the “SET ANSI_WARNINGS OFF” statement just below the linked server call.
And to be on the safer side applied the “SET ANSI_WARNINGS ON” statement at the end of the stored procedure.


2.UPDATE statement conflicted with the REFERENCE constraint

When tables are referenced by Foreign Key constraints it gets difficult to UPDATE the key columns.

-- Now lets UPDATE the FK column values
   UPDATE emp SET d_fk=3 WHERE d_fk=2

Error Message:
Msg 547, Level 16, State 0, Line 1
The UPDATE statement conflicted with the REFERENCE constraint "fk_dept_d_fk".
The conflict occurred in database "tempdb", table "dbo.emp", column 'd_fk'.
The statement has been terminated.

We can disable the FK constraints by ALTER statement before the UPDATE and enable them after the UPDATE, like:

-- Disable FK Constraint
   ALTER TABLE emp NOCHECK CONSTRAINT fk_dept_d_fk

-- Perform UPDATE
   UPDATE emp SET d_fk=3 WHERE d_fk=2

-- Enable FK Constraint
   ALTER TABLE emp WITH CHECK CHECK CONSTRAINT fk_dept_d_fk

String Function

Create FUNCTION [dbo].[UF_CSVToTable]
(
@psCSString VARCHAR(8000)
)
RETURNS @otTemp TABLE(sID VARCHAR(20))
AS
BEGIN
DECLARE @sTemp VARCHAR(10)
WHILE LEN(@psCSString) > 0
BEGIN
SET @sTemp = LEFT(@psCSString, ISNULL(NULLIF(CHARINDEX(',', @psCSString) - 1, -1),
LEN(@psCSString)))
SET @psCSString = SUBSTRING(@psCSString,ISNULL(NULLIF(CHARINDEX(',', @psCSString), 0),
LEN(@psCSString)) + 1, LEN(@psCSString))
INSERT INTO @otTemp VALUES (@sTemp)
END
RETURN
END


Input : Select * from UF_CSVToTable('1,2,3')

Output:: SID
             1
             2
             3

Split Example

DECLARE @xml xml,@str varchar(100),@delimiter varchar(10)
SET @str= 'AAA 65785 89'
SET @delimiter =' '
SET @xml = cast(('<X>'+replace(@str,@delimiter ,'</X><X>')+'</X>') as xml)
select @xml
SELECT a.value('.','varchar(10)') as value FROM @xml.nodes('X') as X(a)

Triggers Article

INTRODUCTION

TRIGGERS IN SQL SERVER

BACKGROUND

This article gives a brief introduction about Triggers in Sql Server 2000/2005.

What is a Trigger

A trigger is a special kind of a store procedure that executes in response to certain action on the table like insertion, deletion or updation of data. It is a database object which is bound to a table and is executed automatically. You can’t explicitly invoke triggers. The only way to do this is by performing the required action no the table that they are assigned to.

Types Of Triggers

There are three action query types that you use in SQL which are INSERT, UPDATE and DELETE. So, there are three types of triggers and hybrids that come from mixing and matching the events and timings that fire them.

Basically, triggers are classified into two main types:- 

(i) After Triggers (For Triggers) 
(ii) Instead Of Triggers 

(i) After Triggers

These triggers run after an insert, update or delete on a table. They are not supported for views. 
AFTER TRIGGERS can be classified further into three types as: 

(a) AFTER INSERT Trigger. 
(b) AFTER UPDATE Trigger. 
(c) AFTER DELETE Trigger. 

Let’s create After triggers. First of all, let’s create a table and insert some sample data. Then, on this table, I will be attaching several triggers. 

 Collapse | Copy Code

CREATE TABLE Employee_Test
(
Emp_ID INT Identity,
Emp_name Varchar(100),
Emp_Sal Decimal (10,2)
)

INSERT INTO Employee_Test VALUES ('Anees',1000);
INSERT INTO Employee_Test VALUES ('Rick',1200);
INSERT INTO Employee_Test VALUES ('John',1100);
INSERT INTO Employee_Test VALUES ('Stephen',1300);
INSERT INTO Employee_Test VALUES ('Maria',1400);

I will be creating an AFTER INSERT TRIGGER which will insert the rows inserted into the table into another audit table. The main purpose of this audit table is to record the changes in the main table. This can be thought of as a generic audit trigger. 

Now, create the audit table as:-

 Collapse | Copy Code

CREATE TABLE Employee_Test_Audit
(
Emp_ID int,
Emp_name varchar(100),
Emp_Sal decimal (10,2),
Audit_Action varchar(100),
Audit_Timestamp datetime
)

(a) AFTRE INSERT Trigger

This trigger is fired after an INSERT on the table. Let’s create the trigger as:-

 Collapse | Copy Code

CREATE TRIGGER trgAfterInsert ON [dbo].[Employee_Test] 
FOR INSERT
AS
	declare @empid int;
	declare @empname varchar(100);
	declare @empsal decimal(10,2);
	declare @audit_action varchar(100);

	select @empid=i.Emp_ID from inserted i;	
	select @empname=i.Emp_Name from inserted i;	
	select @empsal=i.Emp_Sal from inserted i;	
	set @audit_action='Inserted Record -- After Insert Trigger.';

	insert into Employee_Test_Audit
           (Emp_ID,Emp_Name,Emp_Sal,Audit_Action,Audit_Timestamp) 
	values(@empid,@empname,@empsal,@audit_action,getdate());

	PRINT 'AFTER INSERT trigger fired.'
GO

The CREATE TRIGGER statement is used to create the trigger. THE ON clause specifies the table name on which the trigger is to be attached. The FOR INSERT specifies that this is an AFTER INSERT trigger. In place of FOR INSERT, AFTER INSERT can be used. Both of them mean the same. 
In the trigger body, table named inserted has been used. This table is a logical table and contains the row that has been inserted. I have selected the fields from the logical inserted table from the row that has been inserted into different variables, and finally inserted those values into the Audit table. 
To see the newly created trigger in action, lets insert a row into the main table as : 

 Collapse | Copy Code

 insert into Employee_Test values('Chris',1500);

Now, a record has been inserted into the Employee_Test table. The AFTER INSERT trigger attached to this table has inserted the record into the Employee_Test_Audit as:-

 Collapse | Copy Code

6   Chris  1500.00   Inserted Record -- After Insert Trigger.	2008-04-26 12:00:55.700

(b) AFTER UPDATE Trigger

This trigger is fired after an update on the table. Let’s create the trigger as:-

 Collapse | Copy Code

CREATE TRIGGER trgAfterUpdate ON [dbo].[Employee_Test] 
FOR UPDATE
AS
	declare @empid int;
	declare @empname varchar(100);
	declare @empsal decimal(10,2);
	declare @audit_action varchar(100);

	select @empid=i.Emp_ID from inserted i;	
	select @empname=i.Emp_Name from inserted i;	
	select @empsal=i.Emp_Sal from inserted i;	
	
	if update(Emp_Name)
		set @audit_action='Updated Record -- After Update Trigger.';
	if update(Emp_Sal)
		set @audit_action='Updated Record -- After Update Trigger.';

	insert into Employee_Test_Audit(Emp_ID,Emp_Name,Emp_Sal,Audit_Action,Audit_Timestamp) 
	values(@empid,@empname,@empsal,@audit_action,getdate());

	PRINT 'AFTER UPDATE Trigger fired.'
GO

The AFTER UPDATE Trigger is created in which the updated record is inserted into the audit table. There is no logical table updated like the logical table inserted. We can obtain the updated value of a field from theupdate(column_name) function. In our trigger, we have used, if update(Emp_Name) to check if the column Emp_Name has been updated. We have similarly checked the column Emp_Sal for an update. 
Let’s update a record column and see what happens. 

 Collapse | Copy Code

 update Employee_Test set Emp_Sal=1550 where Emp_ID=6

This inserts the row into the audit table as:- 

 Collapse | Copy Code

6  Chris  1550.00  Updated Record -- After Update Trigger.	  2008-04-26 12:38:11.843 

(c) AFTER DELETE Trigger

This trigger is fired after a delete on the table. Let’s create the trigger as:- 

 Collapse | Copy Code

CREATE TRIGGER trgAfterDelete ON [dbo].[Employee_Test] 
AFTER DELETE
AS
	declare @empid int;
	declare @empname varchar(100);
	declare @empsal decimal(10,2);
	declare @audit_action varchar(100);

	select @empid=d.Emp_ID from deleted d;	
	select @empname=d.Emp_Name from deleted d;	
	select @empsal=d.Emp_Sal from deleted d;	
	set @audit_action='Deleted -- After Delete Trigger.';

	insert into Employee_Test_Audit
(Emp_ID,Emp_Name,Emp_Sal,Audit_Action,Audit_Timestamp) 
	values(@empid,@empname,@empsal,@audit_action,getdate());

	PRINT 'AFTER DELETE TRIGGER fired.'
GO

In this trigger, the deleted record’s data is picked from the logical deleted table and inserted into the audit table. 
Let’s fire a delete on the main table. 
A record has been inserted into the audit table as:- 

 Collapse | Copy Code

 6  Chris	1550.00  Deleted -- After Delete Trigger.  2008-04-26 12:52:13.867 

All the triggers can be enabled/disabled on the table using the statement 

 Collapse | Copy Code

ALTER TABLE Employee_Test {ENABLE|DISBALE} TRIGGER ALL 

Specific Triggers can be enabled or disabled as :- 

 Collapse | Copy Code

ALTER TABLE Employee_Test DISABLE TRIGGER trgAfterDelete

This disables the After Delete Trigger named trgAfterDelete on the specified table. 

(ii) Instead Of Triggers

These can be used as an interceptor for anything that anyonr tried to do on our table or view. If you define anInstead Of trigger on a table for the Delete operation, they try to delete rows, and they will not actually get deleted (unless you issue another delete instruction from within the trigger)
INSTEAD OF TRIGGERS can be classified further into three types as:- 

(a) INSTEAD OF INSERT Trigger. 
(b) INSTEAD OF UPDATE Trigger. 
(c) INSTEAD OF DELETE Trigger. 

(a) Let’s create an Instead Of Delete Trigger as:-

 Collapse | Copy Code

CREATE TRIGGER trgInsteadOfDelete ON [dbo].[Employee_Test] 
INSTEAD OF DELETE
AS
	declare @emp_id int;
	declare @emp_name varchar(100);
	declare @emp_sal int;
	
	select @emp_id=d.Emp_ID from deleted d;
	select @emp_name=d.Emp_Name from deleted d;
	select @emp_sal=d.Emp_Sal from deleted d;

	BEGIN
		if(@emp_sal>1200)
		begin
			RAISERROR('Cannot delete where salary > 1200',16,1);
			ROLLBACK;
		end
		else
		begin
			delete from Employee_Test where Emp_ID=@emp_id;
			COMMIT;
			insert into Employee_Test_Audit(Emp_ID,Emp_Name,Emp_Sal,Audit_Action,Audit_Timestamp)
			values(@emp_id,@emp_name,@emp_sal,'Deleted -- Instead Of Delete Trigger.',getdate());
			PRINT 'Record Deleted -- Instead Of Delete Trigger.'
		end
	END
GO

This trigger will prevent the deletion of records from the table where Emp_Sal > 1200. If such a record is deleted, the Instead Of Trigger will rollback the transaction, otherwise the transaction will be committed. 
Now, let’s try to delete a record with the Emp_Sal >1200 as:- 

 Collapse | Copy Code

delete from Employee_Test where Emp_ID=4

This will print an error message as defined in the RAISE ERROR statement as:- 

 Collapse | Copy Code

Server: Msg 50000, Level 16, State 1, Procedure trgInsteadOfDelete, Line 15
Cannot delete where salary > 1200 

And this record will not be deleted. 
In a similar way, you can code Instead of Insert and Instead Of Update triggers on your tables. 

CONCLUSION

In this article, I took a brief introduction of triggers, explained the various kinds of triggers – After Triggers and Instead Of Triggers along with their variants and explained how each of them works. I hope you will get a clear understanding about the Triggers in Sql Server and their usage.

Easy way to find the salary of an Employee

To find 5th Highest salary from the table
   
       SELECT TOP 1 SALARY FROM (SELECT DISTINCT TOP 5 SALARY FROM TblSalary ORDER BY SALARY DESC) a ORDER BY SALARY

To find 5th Lowest salary from the table
   
       SELECT TOP 1 Salary FROM (SELECT DISTINCT TOP 5 Salary FROM TblSalary ORDER BY Salary ASC) a ORDER BY Salary desc

To find Minimum salary from the table
    
      SELECT TOP 1 Salary FROM (SELECT DISTINCT TOP 1 Salary FROM TblSalary ORDER BY Salary ASC) a ORDER BY Salary

To find maximum salary from the table
    
       SELECT TOP 1 Salary FROM (SELECT DISTINCT TOP 1 Salary FROM TblSalary ORDER BY Salary desc) a ORDER BY Salary

With(NOLOCK) Advantages and Disadvantages

Usage of with(NOLOCK)

When data in a database is read or modified, the database engine uses special types of controls, called locks, to maintain integrity in the database. Locks basically work by making sure database records involved in a transaction cannot be modified by other transactions until the first transaction has committed, ensuring database consistency.

When designing database applications, you should keep in mind the different types of locks that will be issued, and the different levels of isolation your transactions will occur. Typically, the SQL Server defaults work fine for what you are trying to accomplish. However, there will be times when it is advantageous to manually make hints to how locks are issued on your tables in your SQL statements.

This article focuses on table : NOLOCK. I'll set up a table to use for our example queries. Execute the script in  to create the Orders table and populate the table with data.

IF EXISTS(SELECT TABLE_NAME FROM INFORMATION_SCHEMA.VIEWS WHERE TABLE_NAME = 'Orders')

DROP TABLE Orders

GO

CREATE TABLE Orders ( OrderId INT, ProductName VARCHAR (30), OrderDare DATE, PriceMONEY, )

GO

--Insert values to the table

INSERT INTO Orders VALUES (1,'Bag',GETDATE(),1500.00)

INSERT INTO Orders VALUES (1,'Book',GETDATE(),200.00)

GO

NOLOCK

This table hint, also known as READUNCOMMITTED, is applicable to SELECT statements only. NOLOCK indicates that no shared locks are issued against the table that would prohibit other transactions from modifying the data in the table.

The benefit of the statement is that it allows you to keep the database engine from issuing locks against the tables in your queries; this increases concurrency and performance. The downside is that, because the statement does not issue any locks against the tables being read, some "dirty," uncommitted data could potentially be read. A "dirty" read is one in which the data being read is involved in a transaction from another connection. If that transaction rolls back its work, the data read from the connection using NOLOCK will have read uncommitted data. This type of read makes processing inconsistent and can lead to problems. The trick is being able to know when you should use NOLOCK.

The following example shows how NOLOCK works and how dirty reads can occur. In the script below, I begin a transaction and insert a record in the SalesHistory table.

BEGIN TRANSACTION

INSERT INTO Orders

    (OrderId, ProductName (30), OrderDare, Price)

     VALUES

     (3,'Shoes',GETDATE(),2000.00)

         
                  The transaction is still open, which means that the record that was inserted into the table still has locks issued against it. In a new query window, run the following script,

SELECT * FROM Orders

The above query is in Executiong stage only and resultwon’t display.

Since the table Orders is locked by the insert statement.

With  using  the NOLOCK table hint in returning the  records in the Orders table.

SELECT * FROM Orders WITH(NOLOCK)

The number of records returned is 3. Since the transaction that entered the record into theOrders table has not been committed, I can undo it. I'll roll back the transaction by issuing the following statement:

ROLLBACK TRANSACTION

This statement removes the record from the SalesHistory table that I previously inserted. Now I run the same SELECT statement that I ran earlier:

SELECT COUNT(*) FROM Orders WITH(NOLOCK)

This time the record count returned is 2. My first query read a record that was not yet committed -- this is a dirty read.

When to use  WITH (NOLOCK)

WITH (NOLOCK) is the equivalent of using READ UNCOMMITED as a transaction isolation level. So, you stand the risk of reading an uncommitted row that is subsequently rolled back, i.e. data never made it into the database. So, while it can prevent reads being deadlocked by other operations, it comes with a risk. In a banking application with high transaction rates, it's probably not going to be the right solution to whatever problem you're trying to solve with it IMHO.

Most banking applications can safely use nolock because they are transactional in the business sense. You only write new rows, you never update them

Most of the Real time scenario’s when we used to read the date we need to use theWITH(NOLOCK)