The PK blog by "Cloud Boy"

Running JBoss 7 on Windows Azure — Part II

pkefal — Tue, 06 Dec 2011 22:44:01 +0000

Continuing on where I left it on my previous post, I’m going to explain how the Announcement service works and why we choose that approach.

The way JBoss and mod_proxy work now is that every time something changes in the topology, either a new proxy is added or removed or a JBoss node, then the proxy list has to be updated and both the node and the proxy have to be aware of their existence.

Mod_proxy is using multicast to announce itself to the cluster but as this is not supported on Windows Azure, we created our own service that runs on the proxy and on the node also. Each time a new proxy or a node is added/removed, the service notifies the rest of the instances that something changed in the topology and they should update their lists with the new record.

The service is not running under a dedicated WorkerRole but it’s part of the same deployment as the proxy and the JBoss node. It’s a WCF service hosted inside a Windows NT Service listening on a dedicated port. That approach gives us greater flexibility as we keep a clear separation of concern between the services on the deployment and we don’t mix code and logic that has to do with the proxy, with the Announcement service. Originally the approach of using an NT Service caused some concerns as how this service is going to be installed on the machines and how can we keep one single code base for that service, running on both scenarios.

First of all, you should be aware that any port you open through your configuration is only available to the host process of the Role. That means if the port is not explicitly open again on the firewall, your service won’t be able to communicate as the port it’s blocked. After we realized that, we fix it by adding an extra line to our Startup Task which was installing the service on the machines. The command looks like this:

which is part of the installer startup task

To make our service even more robust and secure we introduced a couple of NetworkRules that they only allow communication between Proxies and Jboss nodes:

Any kind of communication between the services is secured by certificate based authentication and message level encryption. The service it’s a vital component in our approach and we want it to be as secure as possible.

The service is monitoring a couple of things that helps us also collect telemetry data from the Jboss nodes, but it’s also wired to a couple of RoleEnvironment events like OnStopping and OnChanged. Everytime there is an OnStopping, we send messages out to all of the other service instances to de-register that proxy from their list because it’s going down. Also, the service itself is checking on specific intervals if the others nodes are alive. If they don’t respond after 3 times, they are removed. The reason we do this, is to handle possible crashes of the proxy as fast as possible. Lastly, everytime there is an OnChanged event fired, we verify that everything is as we know they should be (nodes available etc).

Next post in the series, the cluster setup.

Running JBoss 7 on Windows Azure

pkefal — Thu, 01 Dec 2011 10:58:00 +0000

I’m going to start a series of posts to explain how we made JBoss run on Windows Azure, not just on standalone mode but with full cluster support.

Let me start with one simple definition, I’m NOT a Java guy, but I work with some very talented people under the same roof and under the same practice at Devoteam.

So everything started when I posted on our internal collaboration platform a post about the new Windows Azure Starter Kit CTP being released for Java. Since my very beginning at Devoteam I’ve been brainwashing them to try it out, so that post was the kick-off they needed. Knowing our customer base, which is highly mixed and hybrid environments with Java and .NET and SAP systems etc etc, we want to create this:

“An integrated, enterprise-grade demo of a Windows Azure cloud setup, containing a .NET front-end, a JEE application in the cloud and a local mainframe instance; completely integrated.”

We used:

mod_cluster
jGroups
mod_proxy

The basic reason behind choosing JBoss, besides the fact that our customers use it also, was that it’s open and free and open means that we can change whatever we want to make it work and fit in our environment.

One of those talented guys Francois Hertay, modified the code for cluster discovery (jGroups) already provided to make it work in a more robust way and more important, make it work with JBoss 7 because it currenctly works only with version 6. We still use the BLOB approach but we changed it a little bit to make it more robust. In a typical enterprise scenario we have the proxies in front of the JBoss cluster and mod_proxy is also the one achieving the much needed state consistency as you might already know that Windows Azure is using a non-sticky load balancer. Also, based on the dynamic nature of how a Windows Azure instance behaves, it was impossible to have static IPs for the proxies and the instances and it was obvious we needed a little something for:

Discovering the proxies and announce them to the JBoss cluster
Make sure that this is removed when a proxy goes down or when a new node joins the cluster, it finds the proxies and registers them

We needed something different as mod_proxy uses multicast to announce itself to the cluster and this is not supported in Windows Azure. The solution was to create our own home-brewed announcer service and will take care of this.

Our final setup was 1 WorkerRole for the Proxy and 1 WorkerRole for the JBoss node.

We choose this setup so we can independently scale either the proxies or the JBoss nodes, which is pretty typical in an Enterprise environment.

On the next post, I will explain how the Announcement Service works and what can be improved in the future in the service.

‘Till then,

PK.

Reviewing Neil Mackenzie’s Windows Azure Development Cookbook

pkefal — Sun, 18 Sep 2011 18:35:22 +0000

In the next couple of days I’ll be reading and reviewing Neil’s book. Neil is a fellow Windows Azure MVP and judging from what I’ve heard and read up to this point, this is probably going to be a very nice book not that I expected anything less from him.

You can find the book here -> http://www.packtpub.com/microsoft-windows-azure-development-cookbook/book and stay tuned from my review

Picture of the cover:

PK.

Windows Azure SDK 1.5 and Windows Azure AppFabric SDK 1.5 released

pkefal — Wed, 14 Sep 2011 22:34:00 +0000

Today Windows Azure SDK 1.5 and Windows Azure AppFabric SDK 1.5 were released, fixing issues and bugs detected during the beta. There are also some new enhancements to it:

Re-architected emulator, which enables higher fidelity between local and cloud developments.
Support for uploading service certificates in csupload.exe.
A new csencrypt.exe tool to manage remote desktop encryption passwords.
Enhancements in the Windows Azure Tools for Visual Studio for developing and deploying cloud applications.
The ability to create ASP.NET MVC3 Web Roles and manage multiple service configurations in one cloud project.
Improved validation of Windows Azure packages to catch common errors like missing .NET assemblies and invalid connection strings.

and also some changes and new features on AppFabric:

Asynchronous Cloud Eventing – Distribute event notifications to occasionally connected clients (for example, phones, remote workers, kiosks, and so on)
Event-driven Service Oriented Architecture (SOA) – Building loosely coupled systems that can easily evolve over time
Advanced Intra-App Messaging – Load leveling and load balancing for building highly scalable and resilient applications

Best of all, is that the new features on Windows Azure AppFabric Service Bus, are free, you still only pay for the number of connections/relays although, you will see some new meters on your monthly bill, called “Entity Hours” and “Message Operations” but as I said, you’re not going to be billed for those.

I’ll be going into details in the next few posts, so stay tuned

PK.

Windows Azure Wins Best Cloud Service at Cloud Computing World Series Awards

pkefal — Fri, 24 Jun 2011 21:28:22 +0000

Earlier this week the Windows Azure platform was named Best Cloud Service at the Cloud Computing World Forum in London. Now in its third year, the Cloud Computing World Series Awards celebrate outstanding achievements in the IT market. This year’s winners were selected by an independent panel of industry experts.

“It’s fantastic for us to see this type of recognition for the Windows Azure platform. We’re seeing companies creating business solutions in record times, reinforcing the new possibilities created by the cloud,” said Michael Newberry, Windows Azure lead, Microsoft UK.

Click here to read the press release about this award.

Source: Windows Azure Blog

Advanced Installer

pkefal — Wed, 22 Jun 2011 11:21:32 +0000

So, this is not a direct Windows Azure post, but I use it also on my roles, so here it goes:

I’ve been looking for a decent alternative to what I was using in the past for my installations, that is InstallAware. Sometimes things were breaking (support for .NET, SQL Server installations failing for nothing etc) all because of the re-packaging habbit of the InstallAware engineers. Then I found Advanced Installer.

I use it on my Windows Azure roles whenever I want to do a chain install (MSI 4.5) and instead of doing startup tasks one by one, I just create an MSI package using it and do everything in there. It’s fast, it’s reliable and it works.

I also use it on installation solutions for the software we get out through my company. Updates support is a really cool thing that I was looking to just work out of the box, instead of having to necessarily wrap it around some clusmy scripting code.

It might get a little time to find out how you might be able to do something, some polishing details are missing but they listen to feedback and they update it regurarly so sooner or later what you ask, we’ll make it to a version.

All in all, I’ll give a 8.5/10 to the product, not because I can find something better that combines usability/price etc, but just because those details are missing and to tease the Advanced Installer team to get even better

PK.

April Update of Windows Azure Platform training kit (WAPTK April v2.9)

pkefal — Tue, 19 Apr 2011 21:27:54 +0000

April’s Update (v2.9) of Windows Azure Platform Training Kit is available now and you can get it here.

“The Windows Azure Platform Training Kit includes a comprehensive set of technical content including hands-on labs, presentations, and demos that are designed to help you learn how to use the Windows Azure platform, including: Windows Azure, SQL Azure and the Windows Azure AppFabric.

The April 2011 update of the Windows Azure Platform Training Kit has been updated for Windows Azure SDK 1.4, Visual Studio 2010 SP1, includes three new HOLs, and updated HOLs and demos for the new Windows Azure AppFabric portal.

Some of the specific changes with the April update of the training kit includes:

[New] Authenticating Users in a Windows Phone 7 App via ACS, OData Services and Windows Azure lab
[New] Windows Azure Traffic Manager lab
[New] Introduction to SQL Azure Reporting Services lab
[Updated] Connecting Apps with Windows Azure Connect lab updated for Connect refresh
[Updated] Windows Azure CDN lab updated for CDN refresh
[Updated] Introduction to the AppFabric ACS 2.0 lab updated to the production release of ACS 2.0
[Updated] Use ACS to Federate with Multiple Business Identity Providers lab updated to the production release of ACS 2.0
[Updated] Introduction to Service Bus lab updated to latest AppFabric portal experience
[Updated] Eventing on the Service Bus lab updated to latest AppFabric portal experience
[Updated] Service Remoting lab updated to latest AppFabric portal experience
[Updated] Rafiki demo updated to latest AppFabric portal experience
[Updated] Service Bus demos updated to latest AppFabric portal “

Enjoy!

New Windows Azure CDN features and SDK 1.4

pkefal — Thu, 10 Mar 2011 21:27:14 +0000

Windows Azure SDK 1.4 was released yesterday with no breaking changes and a lot of stability fixes. You can get the bits from here. Among an important fix, specially if you use a source control system, where the web.config was being locked but Windows Azure Tools still needed write access to update the machine key, there are new features for the CDN (quoting from the release announcement):

Windows Azure CDN for Hosted Services
Developers can use the Windows Azure Web and VM roles as “origin” for objects to be delivered at scale via the Windows Azure Content Delivery Network. Static content in your website can be automatically edge-cached at locations throughout the United States, Europe, Asia, Australia and South America to provide maximum bandwidth and lower latency delivery of website content to users.

Serve secure content from the Windows Azure CDN
A new checkbox option in the Windows Azure management portal to enable delivery of secure content via HTTPS through any existing Windows Azure CDN account.

Get the bits and enjoy, I’m already updated

PK.

Using Windows Azure AppFabric Cache (CTP) on CodeProject

pkefal — Thu, 17 Feb 2011 21:26:40 +0000

I posted an article at CodeProject explaining how you can use Windows Azure AppFabric Cache (CTP) in your applications.

You can find the article here -> http://www.codeproject.com/KB/azure/WA-AppFabric-cache.aspx

Please let me know if you liked or not and of course any comments are more than welcome!

Thank you,

PK.

CDN — Do you get corrupt data while an update on the content is happening?

pkefal — Fri, 28 Jan 2011 21:26:00 +0000

A lot of interesting things have been going on lately on the Windows Azure MVP list and I’ll be try to pick the best and the ones I can share and make some posts.

During an Azure bootcamp another fellow Windows Azure MVP, had a very interesting question “What happens if someone is updating the BLOB and a request come in for that BLOB to serve it?”

The answer came from Steve Marx pretty quickly and I’m just quoting his email:

“The bottom line is that a client should never receive corrupt data due to changing content. This is true both from blob storage directly and from the CDN.

The way this works is:
·         Changes to block blobs (put blob, put block list) are atomic, in that there’s never a blob that has only partial new content.
·         Reading a blob all at once is atomic, in that we don’t respond with data that’s a mix of new and old content.
·         When reading a blob with range requests, each request is atomic, but you could always end up with corrupt data if you request different ranges at different times and stitch them together. Using ETags (or If-Unmodified-Since) should protect you from this. (Requests after the content changed would fail with “condition not met,” and you’d know to start over.)

Only the last point is particularly relevant for the CDN, and it reads from blob storage and sends to clients in ways that obey the same HTTP semantics (so ETags and If-Unmodified-Since work).

For a client to end up with corrupt data, it would have to be behaving badly… i.e., requesting data in chunks but not using HTTP headers to guarantee it’s still reading the same blob. I think this would be a rare situation. (Browsers, media players, etc. should all do this properly.)

Of course, updates to a blob don’t mean the content is immediately changed in the CDN, so it’s certainly possible to get old data due to caching. It should just never be corrupt data due to mixing old and new content.”

So, as you see from Steve’s reply, there is no chance to get corrupt data, unlike other vendors, only old data.

PK.