Anyone integrating Cloverleaf to Biztalk?

[Jim Kosloskey]

All,

Recently we have been exposed to trying to establish TCP/IP interfaces  with a Biztalk environment.

Is there anyone currently doing something like this who would like to exchange information?

Also if you are going to the User Conference October in Dallas, we would be interested in meeting with anyone who wants to discuss the pros/cons/etc. of interfacing with Biztalk.

email: jim.kosloskey@jim-kosloskey.com 29+ years Cloverleaf, 59 years IT - old fart.

Topic

[Charlie Bursell]

Jim:

What would be the difference between a TCP/IP interface to Biztalk and any others?  TCP/IP is goverened by an RFC which I would hope they have read.  If HL7, MLLP is also goverened by an HL7 standard.

Perhaps the internals of what they do is different from Cloverleaf but I can’t see any TCP/IP differences

Just a thought  🙂

[David Teh]

Hi Jim,

Affirm what Charlie has written: no difference as far as MLLP is concern.

Beyond MLLP….

It’s like any other systems with their unique implementation of queues etc. Usually boils down to the person configuring and maintaining the server to get things right.

Earlier versions of Biztalk seem to have some problem using ‘&’ as an encoding character.

[Rick Martin]

Hi Jim,

We’ve interfaced with a few BizTalk instances.  No real TCP/IP problems, but BizTalk does have it’s own quirks especially with how it handles acknowledgements and things like trailing pipes.

Rick

[Russ Ross]

Let me give some history as to why Jim Kosloskey is asking if BizTalk will handle doing a TCP/IP real-time dedicated HL7 interface.  We have a departmental silo that excluded those with interface experience like Jim and I from having any input.  Now we are starting to find out about point-to-point interfaces that have been done behind our back using BizTalk.  From what limited information we can gather, they have been using BizTalk to do web services interfaces.  Consequently, at this point Jim and I suspect we have likely lost the business of doing web services integrations.  Then one day this silo needed to do an allergy integration to pharmacy (Centricity).  Centricity requires all messages be delivered on one port only and in HL7 format, making another proprietary XML point-to-point web services interface not an option because cloverleaf is already sending ADT and Scheduling messages to that pharmacy (Centricity) port.  Now they call me up and I start to work with the in-house developers that will be my counterpart for doing a BizTalk real-time HL7 interface to cloverleaf.  After the developers pass the assignment off like a hot potato because none of them wanted the assignment, I finally ended up with a developer that proceeded to work with me.  Based on what that developer told me, it turned out to be quite a challenge for him to take their typical BizTalk web services setup and make it behave like a persistent TCP/IP interface.

The developer doing the BizTalk configuration was challenged by the following:

Russ Ross
RussRoss318@gmail.com

[Russ Ross]

My director has also asked that I construct an email explaining why BizTalk may not be a good tool for doing TCP/IP real-time dedicated HL7 interfaces and I

Russ Ross
RussRoss318@gmail.com

[Charlie Bursell]

Russ:

If what you say is true, and I don’t know a lot about Biztalk either, why fight it?  If they want a web-based iterface, why not do it that way.  It is certainly doable via Cloverleaf.  Cloverleaf, as-is, can act as a web client and with the addition of IHB , can act as as server.

I recently had to do multiple interfaces with a prescription service that only dealt with HTTPS.  The client side was easy but they wanted to send non-realtime responses back via a discrete web server.  The client did not feel they had enough work to justify IHB.  So I installed Apache on their AIX box and set up a CGI script to interface to Cloverleaf.  It wasn’t very hard except trying to get a secure version of Apache compiled and working on their AIX box

In todays modern world of computer science there are those of that have already done web interfaces and those of us that will soon be doing them or get left behind.  The web is here to stay and like it or not we will have to deal with it  🙂

Just a thought

[Russ Ross]

Charlie:

I had a similar thought to what you are suggesting.  I understand at a high level your suggestion and did some proof of concept along those lines using a multi-server interface test.  I

Russ Ross
RussRoss318@gmail.com

[Charlie Bursell]

I understand

[Author]

Replies