Steve Spencer's Blog

This is a follow on post from my introduction to Windows Azure Websites and shows you how you can synchronise your website in TFS with Windows Azure.

One of the biggest problems with the way you deploy applications to  Windows Azure is that minor changes (e.g .markup, content and styling) require a redeploy to publish the changes. Windows Azure Websites solves this problem by allowing you to synchronise your website with Team Foundation Server or GIT.

In this post I will show you how easy it is to manage your websites in version controlled environment using Team Foundation Service. Team Foundation Service is a cloud hosted version of Team Foundation Server.

This works by creating a continuous integration build with your source code that will automatically deploy your website after successful build each time code is checked in.

This is configured as follows:

Click the “+” button at the bottom of your portal screen and select Website –> Quick Create

Enter the url details and click Create Web Site

An Empty site has now been created.

This site now needs to be link to your Team Foundation Service. Click on the website in the dash board and then select “Setup TFS Publishing”. you will also note that you can use a GIT repository as well as TFS.

Enter your TFS url (or create a new one), then click Authorize Now.

this connects through to your TFS service and setup the CI build that will deploy your application to the cloud.

The TFS site will now be displayed asking you to authorize the connection

You now need to pick the website you want to deploy. If you haven’t create a site yet then you need to go to ~Visual Studio, create your site and check it in to TFS.

You have now linked your web site in TFS to the Azure Website. This will take a few moments to synchronise.

Your website has not been deployed yet. You need to make a change and then check the changes in

upon check-in the build is started

When the build is complete the new website is deployed

You can also revert back to older versions of the web site by clicking the desired version and then clicking redeploy:

This will start the redeploy of the older version:

A new build is kicked off using the same changeset details as the original deployment. Once the build is complete the  web site is reverted back. this whole cycle only took a few minutes so it is a lot faster than the redeploy mechanism you had previously.

TFS and Windows Azure provide a good mechanism for version controlling your website. Adding application life cycle management to any software development activity is a good thing.

If you have installed the Window Azure Toolkit For Windows 8 you may want to deploy it to a real live environment so that you can try out the notification service and start wiring useful Windows 8 applications. I assumed this was going to be a quick task, but it took me a little longer than expected. Firstly when you try and deploy the web application and service to Windows Azure it won’t just deploy out of the box. You need to sort out the certificates for SSL, Out of the box the certificates come as a cer file and Windows Azure only accepts pfx file so you will need to convert the file.  I set up the certificate and deployed to Azure but when I connected my windows 8 application to register for notifications I kept getting errors connecting to my registration service. After a number of attempts to connect I determined that it was an issue with the certificates. The Windows 8 application has an appmanifest file which contains the certificate information. I set this up as I thought was correct but I still could not get the application to talk to my Azure service. Running in the debugger didn’t seem to give me any error diagnostics. Eventually I found this article which provided me with a bit more detail as to what was required (I was doing most of what was suggested). A number of additional issues arose which slowed me down a bit further.

1. When creating a new certificate I needed to run the command prompt as administrator. On my computer my user account is not an administrator so when I created a new certificate. In order to export the certificate I needed to run the certmgr as administrator.

2. Selecting the certificate in Visual Studio to assign to the endpoint was also an issue as it is deployed as administrator so it didn’t seem to appear in the list. I found the certificate and then copied its thumbprint (converted it to uppercase letters) and pasted it into the thumbprint field in the certificate in the role properties.

The Azure application was then deployed to Azure and the new certificate added to the Windows 8 client as per the instructions in the article above.

You should now be able to login, upload images and send notifications.

Now that’s working I can start to build a proper notification service.

We recently moved a web application into Windows Azure that was using session state. As it was running on a single webserver the session state was set to InProc but this is not useful when in a multi-server environment as the session is stored on the specific machine and is therefore not accessible to other machines. There were a number of options:

1. Use the Windows AppFabric Caching service (http://msdn.microsoft.com/en-us/library/windowsazure/gg278339.aspx)
2. Use SQL Azure (http://blogs.msdn.com/b/sqlazure/archive/2010/08/04/10046103.aspx)
3. Use Windows Azure Storage

Windows Azure Storage seemed to be the more cost effective version as the site does not currently use SQL Azure and they have purchased a subscription for Azure which includes both transaction and storage costs.

There is a sample asp.net session provider that uses Windows Azure Table Storage as its backing store. The sample can be downloaded from MSDN at

http://code.msdn.microsoft.com/windowsazure/Windows-Azure-ASPNET-03d5dc14

How to use the Azure Storage Session State Provider

Add the following Session State provider config to the web.config file of the project

   1: <!-- SessionState Provider Configuration -->

   2: <sessionState mode="Custom"

   3:               customProvider="TableStorageSessionStateProvider">

   4:   <providers>

   5:     <clear/>

   6:     <add name="TableStorageSessionStateProvider" type="Microsoft.Samples.ServiceHosting.AspProviders.TableStorageSessionStateProvider"/>

   7:   </providers>

   8: </sessionState>

Add your windows azure storage connection string (DataConnectionString) to each web role that requires session state (Not setting this will result in an object reference not set to an instance of an object exception)

Add a reference to the ASPProviders.dll taken from the sample project and make sure that the Copy Local property is set to true (Not setting this will cause an unable to load exception)

We also added a reference to System.DataServices.Client and set copy local to true on this too.(Not sure if this is needed)

Once this is setup and running, add multiple instances to your role configuration and run in the debugger. Make sure you can navigate to the page that has the session data in. I put a break point onto the action of the page and added a watch for Microsoft.WindowsAzure.ServiceRuntime.RoleInstance.CurrentRoleInstance.Id and checked to see if it changed and if it did change checked to see if the session data was visible.

You may well get the following error when you are using session as all the objects that are put into the Azure Table Storage session object need to be serializable.

Unable to serialize the session state. In 'StateServer' and 'SQLServer' mode, ASP.NET will serialize the session state objects, and as a result non-serializable objects or MarshalByRef objects are not permitted. The same restriction applies if similar serialization is done by the custom session state store in 'Custom' mode.

You can check to see the session data in the Azure Storage Server Explorer.

We are going to run this for a while to se how well it works and also see what debris is left behind in the table and blob storage due to ended sessions. We might have to have a job running that tidies up the expired sessions later.

This week we had a very useful side effect of using the Window Azure Service bus. We have an Azure hosted website that connects to a CRM backend using the service bus in relay mode to communicate between the two systems. We had a test system that worked fine but when we moved to a Live system we had a configuration error in one of the systems but it was difficult to identify.

The way the service bus works means that the server can easily be moved (as long as the server has an outgoing internet connection).Our service bus at the server side is a Windows Service but we also have a console application to help us with debugging as all the traces are logged to the console window. We turned off the service on the Live system and started it on the test system. As the Azure hosted website is connecting to the service bus rather than a specific server the website will now connect to our test system. By running a successful connection test on the Azure hosted site we could prove that the Azure website configuration was correct.

The next change was to configure the test system to point to the Live CRM system. This would prove whether our data was correct or not. Running the same test as before proved that our data migration to the Live CRM system was fine.

This left us with the service bus and the business logic web service running on the test system, so we reconfigured the Live service bus service to point to the Test web service (which we had previously configured to connect to the live CRM system) and this also work. Thus proving we had an issue with the business logic service.

What we were able to do then was to move the service bus console application on to a developers machine and run it in Visual Studio so that we could debug and break on the calls to the business logic service which helped us to easily identify the problem. All this was done without needing to reconfigure or redeploy our Azure hosted website.

I wish I could say that this ease of debugging was one of the reasons we chose to use the service bus, but I would be lying. The fact that it has made our debugging so much easier will now have an influence on its future use.

A new release of the Windows Azure Tools for Visual Studio 2010 is available here

The release adds the following features:

• Profile applications running in Windows Azure.
• Create ASP.Net MVC3 Web Roles.
• Manage multiple service configurations in one cloud project.
• Improved validation of Windows Azure packages.

The Windows Azure Platform Training Kit has also be updated in line with this release. It can be downloaded from here. The training kit has the following changes:

• [Updated] Labs and Demos to leverage the August 2011 release of the Windows Azure Tools for Microsoft Visual Studio 2010
• [Updated] Windows Azure Deployment to use the latest version of the Azure Management Cmdlets
• [Updated] Exploring Windows Azure Storage to support deleting snapshots
• Applied several minor fixes in content

When trying to access the Azure Storage SDK in a non-Azure application I kept getting the following error:

“The type or namespace name 'WindowsAzure' does not exist in the namespace 'Microsoft' (are you missing an assembly reference?)    “

References to Microsoft.WindowsAzure.ServiceRuntime and Microsoft.WindowsAzure.StorageClient were both already included as references to the assembly that was trying to store data in a table. After a bit of investigation the reason for the error was that the assembly was set to have the target framework of “.Net Framework 4 Client Profile”. Changing to “.Net Framework 4” solved the problem.

Recently I was helping out at the Azure Bootcamp in London and during the labs a common theme kept occurring when the labs were deployed to a real Azure account. The roles failed to run and it appeared that the deployment was taking forever.  This is something I experienced first hand when I was starting out with Azure. There is a way to diagnosing these deployment errors and it is by using IntelliTrace. During deployment you can enable IntelliTrace as part of the publish dialog

The IntelliTrace option is only available if you have Visual Studio 2010 Ultimate. Once deployed to Azure the Roles will attempt to start and any errors during this phase will lead to the symptoms mentioned above. You can then connect to your Azure environment using the Server Explorer in Visual Studio to retrieve the IntelliTrace files which can be opened in Visual Studio and show any exceptions that may have been thrown. Further information can be found here. Once you have diagnosed your issue please ensure at you then disable the IntelliTrace by redeploying the fixed application as it will have a negative impact on performance if left enabled.

Getting back to the problem we have at the Bootcamp, the issues was that the deployed application was trying to writing information to Azure storage and the connection string was still pointing to Development storage. This was strange because none of the deployed applications had got to the Azure storage part of the lab so you would have thought that there was no need for a connection string. Luckily I had the exact same problem with one of my earlier deployments and it turns out that when a project is created the Diagnostic plug-in is automatically enabled. The diagnostic plug-in requires its own connection string to Azure storage so that the diagnostic information can be stored. Looking at the role configuration in Visual Studio you can see the Diagnostic plug-in configuration.

To fix the deployment issue click the button next to the connection string text box and enter the details of your Azure Storage account.

You will need to redeploy the application or upload the new ServiceConfiguration.cscfg to fix this issue. If this still does not resolve the issue then try disabling the Diagnostics plug-in and redeploy.

If you are an MSDN subscriber (Ultimate or Premium) you now get more Windows Azure resources. Professional subscribers are now also included in the promotion.

See here for more details. If you have already taken up this offer and you are an MSDN Premium subscriber then you will automatically be upgraded to the MSDN Ultimate offer.

In order to activate your benefits

Moving to Windows Azure is not as difficult as you might think. Using a basic MVC 2.0 web application created in Visual Studio 2010, the Azure tools within Visual Studio enable you to migrate the web side really easily.

Load your MVC 2.0 application into Visual Studio and right click on the solution in Solution Explorer and add a new project to this solution.

Select Windows Azure Project and enter a new project name.

When the next dialog appears, do not select any role and press OK.

In the Windows Azure project rigt click on the roles folder and select Add->Web Role Project in solution

Select the MVC project and press OK

You MVC web application should now be ready for deployment to azure. Hit F5 and make sure that the solution builds and runs in development fabric.

For most people this is not the end. There are a number of issues that you may need to resolve for example file access, database access, security and on premise integration. There are solutions for each of these and it depends upon what you are trying to achieve.

With data access, if you are currently using a SQL database and providing the database is not too complicated then it should just port across my running the same SQL scripts on your SQL azure database. You will then just need to change your connection string to point to SQL Azure rather than whatever database you are currently using.  To see the issues the TFS team had converting to the cloud see the PDC video

Information about setting up security using the Access Control Service can be seen in my earlier blog

File access can be done using Azure Drives if there is a lot of file access, but it may be easier to store the file as a single blob or migrate the data you are storing to able storage. I depends upon what you are using the files for.

Connectivity to on premise applications could be done a number of ways by using the service bus or Azure connect. Again it depends on what you are trying to connect to and how “real time” the connection needs to be. I will be putting together a comparison of the methods as soon as I get some free time.

I suddenly started to get a CommunicationObjectFaultedException after I checked my azure code in to TFS

I could get it working by editing the web.config file manually, but it didn’t seem to matter what I actually changed!! It was the act of editing the web.config file that made it writable and it could therefore be written to by the development fabric. When comparing the files it looks like the machine key section is changed. Further investigation pointed me in the direction of the changes made in the Azure SDK 1.3 to support full IIS. During deployment “automatic configuration [of the machine key] occurs at the site-level, overriding any user-supplied value”. When the file is read-only the error occurs. Making the file writable fixes the problem.

The following links explain:

http://blogs.msdn.com/b/windowsazure/archive/2010/12/08/specifying-machine-keys-with-windows-azure-sdk-1-3.aspx

http://msdn.microsoft.com/en-us/library/gg494981.aspx

10 March 2011 : Update – Issue now fixed in Azure SDK 1.4. See Azure SDK 1.4 Released