Problem

So, I followed the instruction from the blog and uploaded the package and configuration file, waited for the staging application to get into a runnable state, and clicked on the staging link…

Windows Azure – 503 Service Unavailable

Clearly not a good thing! In addition to this error I occasionally got a random network error or a full-on connection error reported by the browser.

I tried googling for this error but there was nothing about this problem after deploying to the real Windows Azure hosting (the cloud). As far as I can tell there’s no way of getting any debug information or logs to work out what’s going wrong. Maybe this is something that Microsoft will add later on?

Solution

Then it struck me that I hadn’t actually set up any kind of TableStorage on my Windows Azure hosting so I removed all mentions of storage and database connections from the MVC application. This included editing Web.config and removing the following.

• The <section name=”authenticationService” type=”System.Web.Configuration.ScriptingAuthenticationServiceSection, System.Web.Extensions, Version=3.5.0.0, Culture=neutral, PublicKeyToken=31BF3856AD364E35″ requirePermission=”false” allowDefinition=”MachineToApplication” /> section element.
• The <connectionStrings> element
• The <membership>element
• The <authentication mode=”Forms”> element
• The <profile> element
• The <roleManager>element

Update: I’ve found that without the <authentication>element I started getting the error below so you may want to instead replace it with :

CCT: Role instances did not start within the time allowed. Please try again. If you continue to encounter this behavior please try shutting down the Development Fabric.

I’m guessing that not all of these elements actually need to be removed but doing so resulted in the MVC ASP.NET application working in Windows Azure.

Update 2: ServiceDefinition.csdef

If you happen to get network timeouts or 404 network connection issues it’s worth checking your ServiceDefinition.csdef to make sure that you’ve got port 80 configured. I noticed that mine had updated to port 8080 so obviously I couldn’t access the staging site on port 80 like I was trying. If you do have the port set to something else other than port 80 you can use that port to access your application but the ServiceDefinition.csdef does have a comment in there telling you your application should be on port 80 so you are probably best to stick with that.

<!– Must use port 80 for http and port 443 for https when running in the cloud –>
<InputEndpoint name=”HttpIn” protocol=”http” port=”80″ />

Related posts:

1. Response.Redirect in Windows Azure
2. System.Security.SecurityException: That assembly does not allow partially trusted callers
3. Linkstation NAS Folder share with Windows Vista

a small round-table event to discuss Bing and since we saw your recent tweet regarding comparative search results between Bing and Google, we thought that you might be interested in coming along. The idea would be to get your initial thoughts and you’ll get an update on the future steps for Bing which are being developed specifically for the UK. During the session you’ll have the chance to talk with Microsoft representatives and other people from social media like yourself, who have expressed some early opinion.

It would appear the event is being attended by a number of web 2.0, social media, marketing and internet types.

Phil Bradley

Phil is interested in Internet searching, Web 2.0 resources, search engines and their development. Phil has confirmed. I recommend following Phil Bradley on Twitter.

James Barlow

James Barlow is an independent consultant specialising in information technology and project/programme management.

Tom Warren

I don’t have any additional information on Tom.

Chris Hambly

Chris lists himself as a successful Digital Media Entrepreneur

Josh Russell

Josh has tweeted that he’s going. He provides micro-consulting on web application strategy

David Stuart

David is going. David blog provides thoughts of a Web 2.0 research fellow on all things in the technological sphere that capture his interest.

Karen Blakeman

Karen’s blogs about search tools, Internet technologies and resources for business information. Karen has confirmed.

Neil Robinson

Neil describes himself as passionate about using the right technology to drive business forward.

Kevin Harrington

Looks like Kevin had some initial trouble confirming his attendance but I think he’s coming now.

Mark Sharp

Mark designs and runs a number of websites for myself and for other people and organisations

Andrea Giammarchi

Andrea’s profile says JavaScript Ninja and both PHP/ECMAScript Certified Engineer

We are being encouraged to “live blog” the event.

Please come armed with all of your thoughts, opinions and questions for the Bing team. This meet-up is about everyone getting the chance to have their say. We’ll make sure you’re able to live blog during the event

So, I’ll be armed with my laptop (for tweeting) and HTC Touch Pro (for video, if appropriate) and will try and ask as many questions as possible and get as much information out to the real-time web as I can. My personal angle will be one of technology. For example:

• How Bing are pushing use of the Bing API and Bing developer centre.
• What are they doing to ensure information is available to the real-time web as quickly as possible.

Please let me know if you have any questions that you’d like asked and I’ll do my best to ask them. You can do this in a number of ways. During the event the hashtag in use will be #meetbing. Prior to the event I’d suggest people maybe use #askbing. Failing that you can contact me directly or leave a comment on this blog post.

Related posts:

1. Microsoft Bing.com round table thoughts
2. Wikipedia now has a definition for Real-Time Web
3. Caplin Systems Blog Post – What is the real-time web?

View the What is the real-time web? article here.

Related posts:

1. How to install Caplin Systems’ Liberator Free Edition on Windows
2. Wikipedia now has a definition for Real-Time Web
3. How real-time does real-time have to be?

No related posts.

 The silly thing is that I know exactly what mistakes I’ve made today that have hindered my progress and what’s worse, I’ve made the same mistakes before.

Don’t spend too long in front of the computer

This might sound a bit daft since you are supposed to be programming and you can’t do that unless you are in front of you computer. However, sometimes walking away from the computer for five minutes, and potentially taking your mind off your problem, can really help. Constantly thinking about the same thing can make you mind go stale. You need to freshen up, take some air, have a cup of tea or have a chat with somebody to take your mind off the problem just for a while. All of a sudden an alternative solution might pop into your head. This is just common sense.

My problem was one of threading, which can be tricky domain, and the mistake I made was that I thought the solution was so simple that I threw myself straight into code. After a couple of hours, and little progress, I took a step back and drew a simple diagram to make sure I understood the problem properly. Whilst the act of drawing the diagram didn’t shed any light on this occasion it took my eyes off the screen for a moment and changed the way I was thinking about things. In addition, the diagram will be handy when I speak to others about the problem…

Do ask for help

You can do this in a number of ways but the best way is to ask those around you. I’m lucky in that I work at a software house so there are loads of people that I can mosey on up to and chat with about a technical problem and even discuss some ins and outs of the technical detail. If you work on your own or don’t work with others that have a similar technical mindset you can see out help and advice from forums, Twitter, email and instant messenger to name but a few. Don’t wollow in the same problem for too long without looking for a fresh take on a problem. For me, this is one scenario where pair programming is really useful. By pairing on a problem you are less likely to run out of ideas because you will be constantly talking about the decisions you are making.

Don’t decide it needs a complete re-write at 6pm

This is exactly what I did today and I got myself into a right mood. What I generally mean is don’t make big decisions or changes right at the end of the day. I didn’t get out of work until around 8pm having only partially re-written the code I wanted to. Then, whilst sitting on the train having aired my brain a bit, I decided I didn’t actually need to re-write everything. There are still alternative solutions that I’ve not tried out yet. There are still other reasons why tests might be failing. I’ve not discussed the problem with others who may come up with a better solution.

Do learn from you mistakes

Today has been an example where it appears that I’ve not learnt from making similar mistakes in the past. But in actual fact I’ve probably corrected the mistakes a little quicker than I have previously so I have learnt, I just need to remind myself of these best practices more often. It’s not like putting your finger on the tail of a wasp – you may do it once, but never again. The problems I’ve described here sneak up on you and before you know it you have spent three hours staring at the same area of code without making any progress. Maybe if I then went out and harrassed a wasp I would learn faster?

One of the great things about working in software engineering is that you have the opportunity to be learning nearly all the time. You don’t just learn software engineering best practices but also common sense, problem solving best practices. Hopefully next time I’ll only waste a fraction of a day before I realise the error of my ways.

Related posts:

1. How I approach problem solving in code?
2. Live Mesh – my experience

I originally had a lot of web files being synchronised, around 700MB, but the synchronisation was taking ages and killing my CPU. CPU was sitting at 100% and there was no sign of things completing. I think it was having a problem with the volume of small files.

To summarise: it’s a great product with bags of potential but something needs to be done to reduce the CPU usage when dealing with the synchronisation of a large volume of small files.

Related posts:

1. Whinge of the Week: Elephant.co.uk – bad experience

Here’s the code. It’s obviously part of a full solution containing a Web Application and a SilverLight Application project. I’ve also uploaded it.

Useful Information:

• You can not do a cross domain request over HTTPS.
• You can do a cross domain request over HTTP.
• To be able to make the cross domain request the server that you are making the request to needs to have either (requires clarification)  crossdomain.xml and/or clientaccesspolicy.xml in the root of the webserver.

Resources/links:

The code:

private void RequestButton_Click(object sender, RoutedEventArgs e)
{
    try
    {
        // Create and being making the request/getting the response
        HttpWebRequest request = (HttpWebRequest)WebRequest.Create(new Uri("http://www.leggetter.co.uk/"));
        request.BeginGetResponse(HandleResponse, request);
    }
    catch (Exception ex)
    {
        Debug.WriteLine(ex);
    }
}

private void HandleResponse(IAsyncResult result)
{
    try
    {
        // Finish getting the response and then read the response in chunks
        HttpWebRequest request = (HttpWebRequest)result.AsyncState;
        HttpWebResponse response = (HttpWebResponse)request.EndGetResponse(result);

        StringBuilder readText = new StringBuilder();
        using (Stream responseStream = response.GetResponseStream())
        {
            byte[] buffer = new byte[1024];
            int read = responseStream.Read(buffer, 0, buffer.Length);
            while (read > 0)
            {
                readText.Append(Encoding.UTF8.GetString(buffer, 0, read));
                read = responseStream.Read(buffer, 0, buffer.Length);
            }
        }

        // Display the response text
        Dispatcher.BeginInvoke(delegate()
        {
            ResponseText.Text = readText.ToString();
        });
    }
    catch (Exception ex)
    {
        Debug.WriteLine(ex);
    }
}

Related posts:

1. Using Fiddler to trick Silverlight into allowing a crossdomain Web Request
2. Making cross domain JavaScript requests using XMLHttpRequest or XDomainRequest
3. Silverlight uses XCP tmp files for Web Requests

Here’s the contents of the list:

• http://www.remotesoft.com/salamander/obfuscator.html
• http://www.smartassembly.com/
• http://www.preemptive.com/products/dotfuscator/index.html
• http://www.9rays.net/products/Spices.Obfuscator/
• http://www.wiseowl.com/products/products.aspx
• http://www.eziriz.com/comparison.htm

Related posts:

1. A new look for Google

I then had to spend 30 minutes taking the whole set of drawers apart with a screwdriver to get this drawer open. I don’t know how many times I’ve come across this same problem but still the design of drawers, and more to the point their casing, stays the same. The drawer cavity always has space above the actual drawer to allow things within the drawer to move into this space and then catch when you try and open the drawer.

My point here is that this has proven time and time again to be bad design with flaws yet this design does not change. The same problem can be seen in other engineering designs including software.

I’ve recently been working on a product which all in all has been a big success but there is one particularly ticky piece of code that is continuing to be troublesome. The interface to the section of code is well defined but the implementation continues to cause problems. The implementation was designed and has organically changed throughout the project and at various stages we’ve said that it could probably do with a redesign. However, it generally works without problems so we keep putting of the redesign. This is the same as the drawer design; you can open the same drawer 1,000 times without the contents catching but when it catches on the 1001^st it’s still very annoying.

I did a quick google for “when to redesign” (ignoring “website” related posts) and came across a PDF on Law 3 from the 10 laws of design and development by Innovative Thermal Solutions.

It is usually difficult if not impossible to make the desired level of improvement without significant change to the original design. Everytime (yes, everytime!) I have been through this struggle, hindsight has shown that if we had bitten the bullet and made the significant design change called for at the first sign of a problem, the product cost and project schedule would have been better off.

The fact is that if something has a flaw which may require an overhaul in design then it should most probably be done. Something as simple as a drawer opening mechanism should work flawlessly. I’m not saying that software is as simple as this but the same design and redesign principles should be used across all forms of engineering.

If it works all the time and does everything you need then leave it alone. If it works most of the time then redesign and fix it – most of the time isn’t good enough.

Related posts:

1. Colonsay Holiday 2008

I work for Caplin Systems who have recently released Liberator FE which is the Free Edition of their real time Comet server product.

Liberator is the most performant and fully-featured server currently available for streaming real-time data to web pages. The technology behind many of the world’s leading financial trading portals, it is now available FREE for evaluation and non-commercial use.

In this post I will detail my experiences when installing Liberator FE on Windows with an aim to being able to create a RIA Client that consumes data from a Capin DataSource.

Liberator Free Edition comes with a comprehensive client library for integrating with Ajax or Flex, RIAs, a Java server library for connecting to sources of data and building application logic, and full documentation.

How to Install Liberator FE on Windows

Liberator FE will not run directly on Windows. The only way to acheive this in any way if you don’t have a Linux server handy is to run it on a Virtual Machine. The following sections explain how to get the pieces of software that you need and how to install Liberator FE on Windows.

VM Player

VMWare provide a free VMPlayer to allow you to run virtual machine images. Download it here. Alternatively you can use VMServer. Both are free to use but make sure you read through the licence agreement.

Installing the player is as simple as double clicking on the downloaded file and following the installation wizard.

Download Virtual Machine Image

Once you have the player downloaded and installed you will need a Linux Virtual Machine image so that you can install Liberator FE.

There are a variety of places to get a VMWare image. I found a Linux CentOS Image via the Virtual Appliance Marketplace.

Run the Virtual Machine Image

The VM Image will most probably be in a compressed archive so extract it. The image can the be opened using the VMPlayer by selecting the Open option and navigating to the directory with the VM Image and double clicking on the image file that will be visible.

There will most probably be some operating system setup steps to go through but using the Centos image that I chose this took less than 2 minutes.

At this point it’s probably worth taking a note of the IP Address of the VM is for the Liberator FE registration. To do this enter ifconfig eth0. If that doesn’t work you might need to enter the full path the the /sbin/ifconfig eth0.

Download Liberator FE

Open up a web browser in the VMWare Player (Guest OS) and go to http://www.freeliberator.com, fill in the details (remember to use the IP Address we got above) and download.

Install Liberator FE

Move the downloaded Liberator FE installer to an install location and run the script which will install to a new LiberatorFE folder:
./LiberatorFE-4.4.12-1-i686-pc-linux-gnu.sh

When prompted enter the licence information that you used when you registered on http://www.freeliberator.com. You will also be asked for port information. Since this is a clean OS install you should be able to use the default ports so just hit enter when prompted to choose ports. The default HTTP port is 8080 which we’ll be using in our next step.

Start the Liberator

Navigate to the /LiberatorFE/etc/ directory and run ./rttpd start.

To ensure that the Liberator has started up open up a web browser within the VM and navigate to http://localhost:8080 (substitute 8080 for your HTTP port if you didn’t choose the default. I chose 4040 just to be different). You should see the Liberator FE Status Page which means you have successfully installed Liberator FE on Windows.

References

• Liberator FE
• Liberator FE Getting Started Guide
• VMWare
• ThoughtPolice VMWare

Related posts:

1. Caplin Systems Blog Post – What is the real-time web?
2. Windows Azure – 503 Service Unavailable
3. Linkstation NAS Folder share with Windows Vista