Azure Logic Apps are a great solution to write workflows of the business processes, and that too without writing any code. As a result, unit tests can not be written because there’s no code to write unit tests against. How do one go around testing a logic app then?

Latest version of Visual Studio Extension Azure Functions and Web tools helps you create pre-compiled C# Azure Functions, and lets you write functions as a C# library. Being a pre-compiled function it has better performance while starting up. Along with that, it lets you use Webjobs attributes for declaring bindings, thus not needing a functions.json file. And, it lets you debug and run azure functions locally.

If you’re developing a solution with micro services architecture with Azure, there are chances you’ve been using Azure Functions to develop a service. For example, If you’re solution can involving sending email alerts. One of the microservices here can be Email Service. This can further be broken down into two micro services. One to enqueue mail message to Azure storage queue. This can be a Http Triggered function. Second one can be a Queue trigerred function that prcesses these messages and actually sends the email, using a third party API.

Few days back I was thinking of creating an Image Gallery of the collection of photos I have. Although, there are multiple options available over the internet that you can download and get ready on the go; most of them involves saving the images on your own server. But what I was more concerned was to just have a display only image in my web site to showcase my photos to the word. And, I wanted to make use of one of my social network account – Facebook, Google+, Twitter or Flickr to host the images.

Azure VM Role involves creating a base image, uploading it to Azure using csupload, and then creating a servic model in Visual Studio to point to the uploaded base image.

I have been really frustrated in waiting for days to log in to my VM Role instance via RDP. I have encountered this issue multiple times. Once a VM Role is deployed or it is restarted, we can not login via RDP for a long period, and it takes a lot of time, sometimes 2-3 (random) days once it is back and allows us to connect via RDP.

Since the release of Windows Azure SDK 1.3 we can now have startup tasks defined in the *ServiceDefinition *file for web/worker roles. However, this is not allowed for a Virtual Machine role. So what to do in case you need to have some tasks/scripts to be run on your VM Role’s startup? For example, your VM Role runs some software applications that require some configuration information that can be available only at run time.

The latest Azure SDK 1.3 makes it possible to login to the VM of your web/worker role instances via Remote desktop. So you are not limited to use VM Role in case you need to establish a RDP connection to your VM.

We already have seen how Windows Azure connect can help us in connecting our windows azure role instances to our local computers in the previous post. We also saw in the use cases/scenarios for azure connect, that we can domain join our windows azure roles with our on-premise Active Directory, and this is possible by using connect plug-in.

Building applications for cloud and hosting them on cloud is one of the great things that happened in recent times. However, you might be having number of existing applications that you wish to migrate to cloud, but you do not want to move your database server to the cloud. Or you want to create a new application and host it in the cloud, but this new application needs to communicate with your existing on-premise applications hosted in your enterprise’s network. Other case might be that your new application that you wish to host in cloud will rely for its authentication on your enterprise’s Active Directory.

After the September release if you are configuring your service namespace as per old method you might get following error:

An application that accepts tokens from an
STS is called as a Relying Party (or RP). In modern scenarios, web applications use WIF and accept tokens from an STS to manage authentications process.

The security tokens generated by
STS contain various attributes based on which a grant/deny access is provided or based on which user experience is customized. These attributes are called as Claims.

Traditionally, access control was implemented within the main application by writing a code against user’s credentials to authenticate them and based on their attributes grant/deny access to various resources. This required application developers to be skilled in implementing security and writing a code which is hard to implement and maintain.

Well, i was just over with integrating AppFabric ACS with windows azure cloud app, using the August release, and now i hear that September release is out.

When i started working with ACS on Appfabric I kept on encountering various terminologies, one of them was Passive Federation and it was widely used. I could understand the term Federation but passive was not clear to me ever.

While working with your SSIS package, have not you ever tried debugging a script component transformation by putting a breakpoint in the VB code? Well, i did and found that, unfortunately, it does not work.

This article explains how to download a file that is stored in the database with the possibility of forcing the browser to open a “Save As” dialog box to save the file on to the client system. </p>

The content of the file is stored in a column of a table of image data type.</p>

Such a dialog-box for saving a file can be displayed by using HttpContext.Current.Response property. </p>

HttpContext is a class that encapsulates all HTTP-specific information about an individual HTTP request. The property HttpContext.Current gets the HttpContext object for the current HTTP request. </p>

Here is an example: </p>

Today when i was nearly at the verge of finishing my SSIS package i just found out that i really can not complete it at the moment. I am having lot of validations on the input data, lot of look ups to get the reference values and after that i need to insert the rows into a SQL server database. It is not simple insertion, there is a big logic behind inserting rows, checking if the rows containing individual numbers form a sequence so that those can be inserted into a range rather than individual entries. Not only that i need to check if that number is already in the table in some range, if yes i need to discard it, redirect that row to an error output file.

Often while working in a SSIS package you will require to temporary hold your data in a staging table in one of the Data Flow Tasks and then in another task you will require to fetch data from the staging table, perform transformations and load it and delete the staging table.

If you have used System.Web.Mail namespace in .NET 1.x for sending emails programmatically, expect a surprise. All classes within this namespace have been deprecated in favor of the new System.Net.Mail namespace. System.Net.Mail contains classes such as MailMessage, Attachment, MailAddress, SmtpClient, etc to help us send emails in the 2.0 world. The features provided by this namespace, in a nutshell, are given below.</p>

</p>

• MailMessage is the main class that represents an email message. </li>
• Use MailAddress class to represent the sender and each recipient. </li>
• Use SmtpClient class to connect to the SMTP server and send the email, both synchronously and asynchronously. </li>
• Use AlternateView class to create the email content in alternate formats, say one in HTML and other plain text, to support different recipient types. </li>
• Use LinkedResource class to associate an image with the email content. </li>
• SmtpPermission class and SmtpPermissionAttribute can be used for code access security. </li>

  </p>

  To know more about some of these classes, [read this article](http://aspnet.4guysfromrolla.com/articles/072606-1.aspx).</p>

If you are using VS.NET 2005 Team Suite, code analysis is built into the IDE itself. In older version of VS.NET, you might have used FxCop externally for comparing against pre-defined rules or would have integrated with the IDE by adding FxCop as an Add-In.