Why left join not returning nulls?


I am trying to get all products and category name if category is still active with below left join script. But I cannot get production information if category is not active. What’s wrong with this left join script?

select p.name, c.name
from products p
left join categories c on p.categoryid = c.categoryid
where c.isactive = 1 or c.isactive is null


To find out what’s the issue of this script have, we need to understand order of a SQL query. In the script above we used SELECT, FROM, LEFT JOIN and WHERE

  1. FROM and JOIN: The FROM clause, and subsequent JOINs are first executed to determine the total working set of data that is being queried
  2. WHERE: Once we have the total working set of data, the WHERE constraints are applied to the individual rows, and rows that do not satisfy the constraint are discarded.
  3. SELECT: Any expressions in the SELECT part of the query are finally computed.

The WHERE clause is execution after the ON clause. which means the predicates in the ON clause are applied to the table before the join, WHERE clause is semantically applied to the result of the join.

So below script will get all the data from product table and will keep categories data as null if cannot find category.

select p.name, c.name
from products p
left join categories c on p.categoryid = c.categoryid

WHERE clause is applied to the JOIN result which will remove products result if category missing.

where c.isactive = 1 or c.isactive is null


if we want to get all product information even if categories is missing or is not active, we need to changed the WHERE clause to ON clause as below:

select p.name, c.name
from products p
left join categories c on p.categoryid = c.categoryid
and (c.isactive = 1 or c.isactive is null )

SSIS Converting date to YYYYMMDD

SSIS Expression format date to YYYYMMDD

Mostly, when we create an auto file creation SSIS package we want to have timestamp in file name. To achieve this purpose, we can simply use below SSIS Expression to produce file name with current timestamp.

+ (DT_STR, 4, 1252) DATEPART(“yyyy” , GETDATE())
+ RIGHT(“0” + (DT_STR, 2, 1252) DATEPART(“mm” , GETDATE()), 2)
+ RIGHT(“0” + (DT_STR, 2, 1252) DATEPART(“dd” , GETDATE()), 2)
+ “.csv”

SSIS Expression format YYYYMMDD to Date

+ “-” + SUBSTRING(@[yourDateAsStr],5,2)
+ “-” + SUBSTRING(@[yourDateAsStr],7,2))

Set SSIS Expression to format File Connection Manager Property

Powershell copy or move files from multiple folder to another folder

In the daily data processing task, we always need to copy or move files from multiple source place to one destination folder. In this blog, I’ll show you how to do those task quickly and easily via PowerShell.

Sample Code

I am using below code to move all csv files under Archives folder to Source folder. Archives folder has more than thousands sub-folders.

get-childitempath e:\archives\*.csv -recurse | move-item -destination d:\source


Get-ChildItem by default only can get the items in one or more specified folder, to perform sub-folders copy and move , we need to user parameter -recurse .

Pip the result

To perform loop copy or move, we need to pipe (|) the recurse result first, then we can use move-item and copy-item cmdlets to do copy or move.

RDP unable to connect Azure server

This week I have been having trouble connecting to my Azure server via RDP. I keep getting below errors:

-The number of connections to this computer is limited and all connections are in use right now.

I can connect to server if I changed inbound security rule to allowed my IP address only. But I cannot use white IP list inbound rule as I don’t have static IP.

Then I did some deep investigation and try to tracking failed logon information from system event.

I checked Event Viewer -> Windows Logs -> Security there are lots of Audit Failure event looks like below:

But as you see, it is completely useless. I only can guess their are some attackers tried to logon with a username of Administrator and the Logon Type is set to 3 (generic network logon), and there is no Source Network Address recorded.

But in Event Viewer -> Applications and Services Logs -> Microsoft -> Windows -> RemoteDesktopServices-RDPCoreTS I found lots of below warnings:

In Windows Server 2012 and later version, if an attacker attempts to logon but fails to do so AND uses a username that DOES NOT EXIST on the targeted RDS host or domain that the host is a member of, Event ID 140 is logged, showing you the source IP of the attacker.

I added this IP to Azure blocked IP list then issue is fixed.

SQL server nchar and nvarchar size

This blog I will show you a very trick issue thing a bout nchar and nvarchar actual size and storage size.


When I use sp_help to check the table property. I got below result:

which shows length 100

but when I using SSMS to check the column size, it shows nvarchar(50)

so what’s the different between sp_help and SSMS design view?


I checked sp_help , it is get the max length field from sys.all_columns. In Microsoft document website, it mentioned maximum length (in bytes) of the column.

But we know nchar and nvarchar, the storage size is two times n bytes (n is the actual number of characters ). so this explained, sp_help is not real character size, it is the storage size in bytes. But SSMS design view is using real character size.

For details you can check Microsoft article nchar and nvarchar and sys.all_columns

SSIS : Script component error – Output0Buffer does not exist

Before we start to talk about the error, let me introduce a Script component editor property SynchronousInputID first. In Microsoft document, described SynchronousInputID as below:

The SynchronousInputID property has a non-zero value only in transformations with synchronous outputs. If the value of this property is zero, it means that the output is asynchronous. For a synchronous output, where rows are passed through to the selected output or outputs without adding any new rows, this property should contain the ID of the component’s input.

Which means if output data only one line, then the output columns are just part of the Input Columns, and there is no need for Output0Buffer. If your output data has many rows, then you need to create new Output0Buffer.

In SSIS, SynchronousInputID default value is zero. If you created a new Script component and have change this default value, but in C# script you are try to use Output0Buffer.AddRow(); to create multiple rows output, then you will get error message:

Output0Buffer does not exist

If you got this error, just follow below steps to change the SynchronousInputID value to non.

clicking on the Script Transformation Editor -> clicking the catagory Inputs and Outputs -> clicking on Output 0 -> changing SynchronousInputId to None in the right hand panel

SSIS: Extract and import JSON file to SQL Server via Script Component as Transformation

Currently there are lots of JSON format stored data. It’s important to be able to read the JSON data stored in files, load the data into SQL Server, and analyze it. This blog describes how to import JSON files into SQL Server via SSIS Script Component as Transformation.

I have JSON format file looks like below


need import this format json file to database table, table schema looks like below

CREATE TABLE import_raw_accounts(
id varchar(60) NULL,
account_alias varchar(60) NULL,
home_currency varchar(60) NULL,
division varchar(3) NULL,
userid varchar(60) NULL,
marginUsed varchar(60) NULL

Create Script Component as Source

Step 1 : Drag and drop Script Component in the SSIS toolbox to the data flow region. Once you drop the Script component, a new pop up window called Select Script Content Type opened. Then selecting the Source option

Step 2: Within the Input and Outputs tab, Go to Output Columns to add output columns.

Step 3 : Within the Script tab, please click on the Edit Script.. button to write the actual C# Script

Once you click on the Edit Script, it will open the main.cs class file to write the C# code. Please write your custom code inside the CreateNewOutputRows function

Step 4 : Before start extract Json data, we need to create a new class for imported JSON object, which will be the same JSON data structure.

public class Accounts
public string Id { get; set; }
public string accountAlias { get; set; }
public string homeCurrency { get; set; }
public string division { get; set; }
public string userid { get; set; }
public string marginUsed { get; set; }

Step 5: Add JSON deserialize code to method CreateNewOutputRows

using (StreamReader r = new StreamReader(Variables.JsonFileName))
while (!r.EndOfStream)
string json = r.ReadLine();
var item = JsonConvert.DeserializeObject(json);
if (item != null)
OutputAccountsBuffer.id = item.Id;
OutputAccountsBuffer.accountalias = item.accountAlias;
OutputAccountsBuffer.homecurrency = item.homeCurrency;
OutputAccountsBuffer.marginUsed = item.marginUsed;
var acc = item.Id.Split(‘-‘);
if (acc.Length >= 1)
OutputAccountsBuffer.division = acc[1];
OutputAccountsBuffer.userid = acc[2];

Final Step: add destination source and mapping output data to table import_raw_accounts.

SSIS Expression Task

According to Microsoft document

The Expression Task creates and evaluates expressions that set variable values at runtime, using the Expression Builder

It’s a pretty useful task which can help you avoid having to write any .net code. In this blog, I will you show you how to use Expression Task to do conditional check.

I have one SSIS package need to export data to flat file. But need to determine if the exported row count is grater than zero or not. If it’s grater than zero then I need to update the row count to database table. If not then need to delete empty file.

I am using Expression Task to do this. The expression in my Expression Task looks like this.

Using conditional logic to compare the row_count then determine which action need to run.

Left side Precedence Constraint Editor

Right side Precedence Constraint Editor

SSIS : Script Component as Transformation

The Script Component is one of the most powerful tools in SSIS. It can be used as a Source, Destination, and Transformation.

In this blog, I will show you how to use the SSIS Script Component as a Transformation with a example. Normally, we can use Script Component to do some task that is not possible for the built-in transformations, or we can use SSIS Script Component as a Transformation to combine the work of multiple transformations in one place.

Example Data

Before we start creating the SSIS Script Component as a Transformation package, Let us see what the data we are going to use. If you want to know what the example doing you can look into like SSIS : Compare two flat files

We have one of big Merge Full Outer Join result and need to compare ‘field'(come from left join) and ‘field(1)’ (come from right join). I am going to using Script Component to do this compare. Below the the merge output screenshot, there are more than hundred field in both left and right join table.

Create Script Component as transformation

To create Script Component as transaformation:

Step 1 : Drag and drop Script Component in the SSIS toolbox to the data flow region. Once you drop the Script component, a new pop up window called Select Script Content Type opened. Then selecting the Transformation option

Step 2 : Double click on the Script component will open the following editor to configure the properties.

Step 3: Within the SSIS Script Component as Transformation Input Columns tab, you can cross-check the input columns.

Step 4: Within the Input and Outputs tab, Go to Output Columns to add output columns.

Step 5 : Within the Script tab, please click on the Edit Script.. button to write the actual C# Script

Once you click on the Edit Script, it will open the main.cs class file to write the C# code. Please write your custom code inside the Input0_ProcessInputRow(Input0Buffer Row) function

Step 6 : Add your custom C# code here. For this example, I am using C# reflection to do comparison. In this way, I don’t need to compare field by field, even if column name changed we don’t need to change the comparison code.

Step 7 : Once you finished editing the Script, Please close the main.cs file. Next, drag the OLE DB Destinations, and join the script component Output Arrow to this new OLE DB Destination

Configure File & Folder Permissions for SQL Server

Since SQL Server 2012 permissions are assigned to the per-service SID for each of its instance. This blog will show how to configure file & folder permission for the SQL Server Database Engine. SQL Server must have permission of the Windows file system to access the file folder where database files are stored. The default SQL server data and backup folder permission are configured during setup. But if your database files saved in a different folder, you might need to follow below steps to setup SQL Server permission to that folder

Configure folder permission for SQL Server instance

  1. Using Windows Explorer, navigate to the file system location where the database files are stored. Right-click the file system folder, and then click Properties.
  2. On the Security tab, click Edit, and then Add.
  3. In the Select Users, Computer, Service Account, or Groups dialog box, click Locations, at the top of the location list, select your computer name, and then click OK.
  4. In the Enter the object names to select box, type the name of instance. If you database use default name then use NT SERVICE\MSSQLSERVER, or for named instance please use NT SERVICE\MSSQL$InstanceName
  5. Click Check Names to validate the entry
  6. In the Group or user names box, select the name added at step 4, and then in the Permissions for box, select the Allow check box for Full control.
  7. Click Apply, and then click OK twice to exit.