Forum Replies Created
-
Posts
-
I have solved the problem!!!! When I was asked for account to connect to the SharePoint site I have entered the local admin account and configuration eneded succesfully.
I am trying to consume the tourico webservices in a biztalk orchestration
the webservices are available here –[url]http://h2hdemo.touricoholidays.com/tws-h2h/Hote lsService.asmx[url]
I am using Biztalk 2006 and still unable to add a web reference to this
webservice. When I use the WSE 3 adapter to generate the proxy/schemas and get the orchestration up and running – i get a login failurewhen i looked at the tcp traffic in ethreal – i noticed that WSE added its own set of SOAP headers and didnt add the soap headers i specified in the
orchestration ie user name, password and culture. I am wondering whether its really possible to call a plain vanilla web service with the WSE 3 without adding all sorts of WSE 3 items in the soap header and how to add my user name and password in the header. The approach of creating a property schema – and then having a property the name of the main element of the header tag etc doesnt workI cant understand why i cant add a web reference to the service in the first place…i mean this was supposed to hav been sorted out in biztalk 2006
You can write a custom pipeline component.
Is the jpg the only data you wish to keep from the message. If not you will have to create a multi-part message in the pipeline.
i.e. keep the data part and add a binary jpg image part.
If you only wish to pass the jpg image, then you can just replace the received data stream with a jpg data stream.You could also do this in an orchestration using a custom .Net class
Hi guys
I am building a BTS app that needs to use a solicit-response mechanism to send HTTP commands out to a server and capture its XML response.
Unfortunately I am a complete HTTP n00b and am not sure quite what to do. I had built some stuff already and had confidently assumed it would work but it didn’t (I got some 405 method not allowed message). Could you help?
What I essentially need to do is send an http://myserver.com?app=abc&cmd=xyz (value-pair) type command out to a server that may be over the intra- or internet and receive a response in the form of an XML-format ADO recordset. If I issue this command over IE I see the response I expect to capture in BTS. The actual server URI is variable so I am defining the solicit-response port as Dynamic and setting the server address at runtime.
The way I attempted to approach the problem was as follows:
On the send side, I defined a schema to represent the various bits I need (uri, app, cmd as in example above) in XML form and then define a flat file extension for it to turn it into the appropriately-delimited string. The send port runs over a pipeline that uses the Flat File assembler component. At runtime, the BTS orchestration builds an XML message that conforms to the schema I defined; I strip the uri out of the message and assign it to the send port address, and then post the message remnant to the send port: the Flat File component turns it into the appropriate string and the string should then get posted to the desired address. I have tested this by sending to a file rather than http port and the http string is what I would expect to see. Do I need to implement URI encoding myself using a custom pipeline that I’d slot in after the Flat File one has run, or should the http send port manage this my itself? I had assumed the latter…
On the receive side, I have built a pipeline component to capture the returned XML ADO Recordset and turn it into a more recognisable schema that I have declared in BTS. The http receive port runs over a pipeline that uses this component. I have tested this by using the pipeline on a file-based receive port and it seems to work ok.
However, I strongly suspect this is not the way I am supposed to be doing it!
It could be that this method is ok but that something else is causing the 405 error: I have set up the BTS http receive extension (though it was already there for hws anyway), but I reckon the problem is on the send rather than the receive side, and my knowledge of IIS and SharePoint is so shallow that I do not know what to do next.
Any help would be GREATLY appreciated!!
Thanks and best regards
Bob
Hi,
I am new to BizTalk 2004. I created small project for Xml file trasfer. I
have build and deployed the project. I have following issues:
1) The schema doent not registered in the HAT – When i go to Find Message View in HAT I couldnt see any schema names there. Nothing there. Just empty.2) I am getting following Error message in the Event Viewer
There was a failure executing the receive pipeline:
\”Microsoft.BizTalk.DefaultPipelines.XMLReceive\” Source:
\”Microsoft.BizTalk.Pipeline.Components\” Receive Location:
\”file path\” Reason: There was an authentication failure. \”The party corresponding to the inbound message cannot be identified\”.Any idea?
Thank you,
I.m trying to goal it, with exceptions in the orchestration, checking each row… if you found something about this, please tell me..
Thanks again, Stephen…
Just to sum all this up – an orchestration \”receive port\” is not a \”read file\”. A receive is a subscription to a message from the message box.
(Is that the right wording?).Also, the way Biztalk physical \”Receive Ports/Locations\” work, if the file was not deleted, how would it know not to keep reading (and publishing) the same file over and over again. [This goes back to an issue of which I have heard conflicting stories – does Biztlak poll the directory or is it NTFS event driven. ]
It might be cool feature to have a \”read\” file \”port\”.
The point would be to configure the filename etc… externally using the Admin Hub. But then would it do pipelines, and mapping and all that cool stuff???If you go back to the .NET method, maybe you just parameterize the name of the file you want to read (you could do this with an app/config variable in the BTSNTSvc.exe.config file – or there is a now a sample in 2006 of how to store parms in SSO).
Yes, it does. Here’s a problem I used to have.
I was on my machine, then I would deploy to remote MachineX.
Then I would undeploy. But later I would try to run some orch’s on my machine and they would fail for missing load modules.So they point is, that the Deploy/Undeploy does GAC/Un-GAC, but only on your machine. If you are doing a remove deploy/undeploy, be careful, especially of the undeploy. To deploy, you must manually GAC the .DLLs.
Currently, I’m using some NANT scripts to deploy to my remote MachineX. The basically copy the .DLLs, GAC, and restart Biztalk, all with the click of a single button. However, I typically just move the .DLLs, and don’t do a full deploy/undeploy, which usually works as long as we don’t change the schema layouts. It can cause issues in the HAT Orch debugger (shapes missing), but it’s so much easier!
Neal
In that case, if your file was not too large you might be able to pass the whole document to an Orchestration, map it (use the map to apply your validation logic and set a flag on the record), xpath out the healer and trailers, and then xpath out all success / failed records. Then, build your two response messages inside the Orchestration.
That would probably be an easier way to go but it probably wouldn’t perform as well. Plus you’ll have more control over the validation part.
I would have to play around with it for a bit since it’s been a long time since I did something like this
Hope this helps.
I was recently faced with a similar problem, except using FTP.
The only options I could think of were a .Net class or writing my own FTP request/response adapter.
The .Net class took about 10 lines of code.
I started investigating the FTP request/response adapter. After a day I decided the .Net class while not ideal was good enough.
I may get back to the FTP request/response adapter one day.Thank you very much, Stephen !!!
I have to evaluate the format of the fields, and some contain of that (if the telephone, email is correct, etc..)in the body of each line.
but I have some initials doubts… Have I do three new schemas? (by the moment, I have a schema, with three roots, header, body and end file, build with the flat file wizard..)..
Thanks again!!!Well, you can’t have an activateable receive on a called Orchestration because when it’s called it will always be active. Make sense?
BizTalk is a pub sub system. So, if you use a receive port to read in your file it will just be published to the message box as soon as there is a file in that location. In this case, you will need some type of correlation to get that file into your already running Orchestration and it will need to arrive at the correct time in your business process. Plus, the file will be deleted.
If you know the file name and location at run time, I’d just use .net code to read the file and not delete it. I think it’s as simple as Xml.Load(“filename”). Not really ideal – but straight forward and simple.
Hope this helps.
Hi
The scenario I have is to integrate 2 systems. The 1st system sends a csv file to the 2nd system. My first question is this: I know that BizTalk can accept a csv file as input, then convert to XML etc for processing within BizTalk, but can it then output the same data as a CSV file?
System 2 currently has a manual process, which basically involves running an exe or shell script of some sort. My second question is: Can BizTalk instruct the second system to run this exe/shell script?
Cheers
