Forum Replies Created
-
AuthorPosts
-
September 9, 2008 at 8:31 AM in reply to: Has anybody tried custom pipeline on WCF receive port? #20650
You can help diagnose the problems by attaching the Visual Studio debugger to the pipeline component. Take a look at this post for more info: http://www.biztalkgurus.com/forums/p/10659/21286.aspx#21286
That error is most likely occurring in the pipeline component. To further diagnose the error, you can attach the Visual Studio debugger to the BizTalk process and step through the code. In visual studio set a breakpoint in the pipeline component in the Execute method. Then click Debug -> Attach to process, and select BTSNTSVC.exe. Now send a file through your process. It should stop on the breakpoint. If it does not stop, then try putting the following line in the pipeline component code:
System.Diagnostics.Debugger.Break();
Now, when the pipeline componet runs, you should see a message on the screen asking if you would like to attach the debugger. Click Yes.
Sure. You just need to create 2 new messages in your orchestration, 1 for the header data and 1 for the body data. Then in a message assignment shape, you can use XPath to extract the data from the SOAP envelope and assign it to the messages. Ex:
msgHeaderData = xpath(soapEnvIn, “/*[local-name()=’Envelope’]/*[local-name()=’Header’]/*[local-name()=’HeaderData’]”);
msgBodyData = xpath(soapEnvIn, “/*[local-name()=’Envelope’]/*[local-name()=’Bodyr’]/*[local-name()=’BodyData’]”);
Note: This example assumes that the data that you want from the SOAP envelope is stored in nodes named “HeaderData” and “BodyData”. This is probably not the case, so your actual xpath will be a little different.
That should’ve been TransmitWorkID, not WorkTransmitId.
/Thomas Østergaard
Thanks for the link.
I don’t think I’ve come across that particular page before, probably because I’ve been searching mainly for some of the Guid context properties. Mainly ReceiveInstanceId and WorkTransmitId, which unfortunately are not included in that documentation.
I have a pipeline component used for some custom tracking that must only be executed once. By storing a Guid with my tracked data that uniquely identifies (indirectly) the contents of my message, I can assure that my tracking only executes once. In other words, I need to verify that ReceiveInstanceId is guaranteed to be the same across receive pipeline resumes and that there is one ReceiveInstanceId per message independent of the receive adapter. The same goes for WorkTransmitId on the send side.
I’m not at all sure that these context properties can be used for that purpose, since I haven’t been able to find any documentation on them.
/Thomas Østergaard
Hi Russel,
Thank you so much for the reply. I tried whatever you said and it works.
But, I have follow up question, How I pull the header and body out of the SOAP message as you explained. It seems to be the correct way for me.. Could you please explain..
September 8, 2008 at 12:49 PM in reply to: Sending Message to MSMQ which resides in extranet from Intranet. #20641Yes, I am checking using All Service Instances..There are no dehydrated messages. I am using MSMQ. I also looked into HAT but there also not much information which i can get. I will try with WCF-netMSMQ.
Thanks,
Baba
September 8, 2008 at 12:39 PM in reply to: Sending Message to MSMQ which resides in extranet from Intranet. #20640Did the message maybe get dehydrated? You can check by running a query for “All Service Instances”, rather than just for suspended instances. I find this is the easiest way to look for messages stuck in BizTalk. This way, in addition to suspended messages, you will be able to see the dehydrated messages, ready to run messages, etc. Also, you might want to view the message flow in HAT. That will show you the chain of receive ports, orchestrations, and send ports that processed the message. This could be a firewall issue, but there will be an error or a stuck message somewhere.
Also, are you using the MSMQ adapter? You might want to consider using the WCF-NetMSMQ adapter instead, so you can leverage the WCF tracing stack to help debug your issue.
EDI file processing in R1 is strange because it uses a special File Adapater, not the standard File adapter. This special adapter didn’t populate/promote the Recevied File Name or the %Filename% macro. Because of the changes to EDI processing in R2 I was hoping that this would now be possible.
I haven’t done any EDI work, but I think the most common method is to send the files over HTTP. It seems strange that the file adapter would behave differently for you depending on what type of files you are processing. It seems like the file adapter would set the same context properties no matter what kind of files it is processing. Does MSFT know about this issue?
September 8, 2008 at 10:30 AM in reply to: Exception while sending message thru WCF-BasicHTTP adapter #20635That’s a tough one… It’s hard to say whether it would be easier to troubleshoot it, or to start reinstalling things. I know this isn’t much help to you now, but I like to use a virtual machine for development. That way, I can create a fully configured development image, then mark it as read only and create a differencing disk on top of it for day-to-day work. When something gets messed up, I can just create a new differencing disk and have my environment working again in 10-15 minutes. If you are interested in this approach, I can elaborate…
Ya, we receive EDI files and unfortunately we need to know what the filename is. This has been a bit of a pain in R1 and I was hoping R2 with it’s changes to the way EDI files are process would allow me to get this value.
Are EDI files picked up using a file adapter in R2? I haven’t looked that deeply into R2 yet, but I have noticed that the EDI processing is drastically different.
Hi,
easyhl7 had some great samples of HL7 projects that were freely downloadable.
Most of them worked when I tested them.
Thanks,
Michael
September 8, 2008 at 8:25 AM in reply to: Exception while sending message thru WCF-BasicHTTP adapter #20632Yes, I reset them and other functionalities are working as it is. Only wcf-basic http giving the above exception but I am not sure it is because of this change.
-
AuthorPosts