Flat file interchanges with BizTalk are really useful for handling large files that contain messages that need to be individually processed. However one of their limitation is that they only support one flat file schema as the interchange’s document schema:
In some instances, one might want to process interchanges that contain more than one type of […]
Just to let you folks know –
you no longer have to create special pipeline components, cases etc. for handling
flat files.
R2 will resolve (dynamically) the schema from those that are deployed!!! (c.f.
to xml)
To set it up:
– create a custom pipeline + add a FFAsm component to it.
– leave the schemas properties blank.
– BizTalk will now use dynamic resolution for the incoming schemas.
This is great news!!!! I can now throw out my ‘dynamic FF schema resolver custom component’
Enjoy!
Management and monitoring
Eventually I got through with the installation procedure, and got the portal up and running.
Being a BizTalk developer you are pretty much neglected from any kind of user-friendly experience. -This looks really good, and it seems easy to manage. Exceptions are caught, and nicely presented in various graphs and lists. Alerts can be set up and later subscribed to. BizTalk receive locations and send ports can be published to the uddi server and much more.
In my opinion, there are two main advantages with this portal. First of all, in my experience, BizTalk server is usually installed down in the deepest and darkest of catacombs, behind several firewalls and as far away from any user as possible. This makes it somewhat challenging for the support personnel since they need to use the BizTalk Administration Console. The BizTalk Administration Console, as you may know, need to to connect to the management database directly, wish pretty much puts our support guys behind the same firewalls. However, using the ESB portal, you'll still need the BizTalk console, but you might at least delegate some of the operational tasks to colleagues on the other side of the wall.
The second benefit is all about exception handling. Usually, we need to handle exceptions differently depending whether it is caught in a mapping, pipeline or orchestration. ESB Guidance solves this by relying on failed message routing. Exceptions are routed to the All.Exceptions Send Port (SQL Adapter) which stores the data in the EsbExceptionDb database. From this source, all exceptions are presented in a unified way in the portal. – Sweet!
ESB Guidance part 3
Message routing is one of the most used and attractive features of Enterprise Service Buses (ESB). Unfortunately, the traditional view of message routing, most of the times, is reduced exclusively to service orchestration and choreography scenarios. In…(read more)
I love the Dallas/Fort Worth developer community, I’ve got friends who work for all
sorts of companies around the metroplex. Of course, as a consultant myself some
of the best and brightest in the area unfortunately work for other firms. Two
of the biggest hoarders of .NET Talent in D/FW are Improving
Enterprises and of course, Telligent.
As you all know I’m a little passionate about the We
Are Microsoft event, so I decided to take some time after the InstallFest the
other day and call out Improving Enterprises and Telligent on camera, this is what
ensued.
So you’ve heard it, both Improving Enterprises and Telligent will be there.
If your a Telligent employee and interested, I’d suggest contacting Jay
Leask and if you’re an Improving Enterprises employee I suggest Caleb
Jenkins.
Many thanks to Cory
Smith and Dave
O’Hara for working the camera for me so I could call people out properly on camera
myself.
Tim Rayburn is a consultant for Sogeti in the Dallas/Fort
Worth market.
Way to often I get a request to deliver a message from BizTalk without any XML namespace. Comments like the one below aren’t that rare.
Why do you put those “ns0″ I front of every tag!? We can’t read the XML when it’s written like that! All our XPath and XSLT seems to fail.
We don’t want <ns0:OutMessage xmlns:ns0=”Acme.Messages.OutMessage/1.0″> … […]
Rahul (MS BTS TS in our parts – great guy
and is hugely knowledgeable in both Java integration technologies as well as MS) has
been hard at work again weaving his magic… I just couldn’t go past his post without
sharing it with you guys.
Great step
by step instructions Rahul!
Well done.
Several people are starting to blog about the introduction of the Volta project on live.labs.com. This is a pretty interesting research project and points to an interesting development at Microsoft. I get a little frustrated at times b/c I hear developers complain that a) Microsoft isn’t agile enough and it takes them too long to release stuff, and b) there is too much stuff coming out of Microsoft for me to keep up. Well which is it!
The stuff on live labs is really early stuff that is only interesting if you want to see what Microsoft is thinking about long term and provide feedback. If you are going to get frustrated when you can’t use this stuff in 6 months in production, then it is not for you. That being said, looking at the list of projects out there, two really stick out for me: Volta and Deep Fish. Volta is a really intesting project to me in that it potentially allows me to write client code once and have the framework take advantage of the richest platform on the client. I think that is where it really shines, where I can create the UI and it will use Ajax on clients that don’t have .NET, and use something richer like Silverlight 2.0 when it is there. I think that is more compelling than figuring out architecture/deployment later on. I watched a Channel 9 video a this summer about Volta: check out part 1 and part 2 to hear more about the ideas from
Jeff was just complaining about the fact that IE mobile has not evolved. I look at the Deepfish project and it gives me hope that someday my Windows Mobile device will have a browser half as cool as the one on the IPhone.
A couple of weeks ago, I had the pleasure of participating in a roundtable event with some really smart folks from around the Minneapolis area. Jeff Brand, our local evangelist and .NET sex symbol (self-proclaimed) organized the whole thing and recorded it using a very cool device from Microsoft that came out of their research group. It is a panning video camera that can react to the voices and capture video of the person speaking. I think Jeff had some problems getting it working just how he wanted, but it is still pretty cool stuff and I’m sure we’ll get it ironed out for future chats.
Jeff has some posts with the links to download the conversation if you are interested. Unfortunately, I was in NC at the time teaching a class so I had to participate via phone and Live Meeting. While it gave me a remote user look at the experience with the Roundtable device, it was less than optimal way to be an active participant in the conversation. I’m looking forward to the next conversation when I can hopefully be in town.
As
I’ve mentioned before, there is a big event coming up in January that has developers
donating their time to benefit local charities. The event is called We
Are Microsoft and is put on by Microsoft (care of Chris
Koenig) and is being run by Toi Wright. I had a chance to catch up with
Toi at the Visual
Studio 2008 InstallFest last night, and ask her about the event. Here is
that chat.
Tim Rayburn is a consultant for Sogeti in the Dallas/Fort
Worth market.