One Blog for Biztalk Engineers

You cannot create an x64 host with the Standard edition

Manuel999 — Mon, 16 Nov 2009 15:01:00 GMT

A number of BizTalk users have been wondering lately that they are not able to create an x64 host on a Standard edition.

The comparison of the features of available BizTalk editions is not really clarifying this issue: http://www.microsoft.com/biztalk/en/us/editions.aspx

You find clearer information here:
http://msdn.microsoft.com/en-us/library/aa560166(BTS.10).aspx

where it says: "BizTalk Server 2009 Enterprise, Developer, Branch, and Evaluation Editions support native 64-bit execution on Windows Server 2008, Windows Vista, Windows Server 2003 x64 (including R2), and Windows XP Professional x64"
(The same is applicable for BizTalk 2006 R2).

The Standard edition is toned down in several areas. If you consider purchasing a BizTalk license, you should carefully define your requirements and check the available editions for the one

which meets your requirements.

Manuel Stern

Update: This is now well documented in KB 926628

BizTalk and Project References (continued)

Larry2 — Fri, 13 Nov 2009 17:08:33 GMT

In my last blog I mentioned issues with BizTalk 09 when creating references between projects. Here's some follow up thanks to Mandi on the BizTalk escalation team.

BizTalk solutions are often separated into various projects for the different artifacts, pipelines, schemas, orchestration, etc. When working with a BizTalk project with references to other BizTalk projects compilation fails.

A review the project shows an orchestration variable is missing and shows a red exclamation mark. When selecting the type, it shows the referenced BizTalk project but no defined types. This is a known issue and a fix is available (no link available at this time).

Maps are not immune. If there is a map using a schema from a referenced BizTalk project, the "Schema Type Picker" in the map doesn't show the referenced project and schemas. It only shows local schemas.

To resolve this problem check to see if the "Build Action" is set to "None" instead of "BtsCompile" for the schema. Here's the link with more information.

Issues with BizTalk 2009 on VS.NET 2008

http://blog.eliasen.dk/2009/07/21/IssuesWithBizTalk2009OnVSNET2008.aspx

There may be other related symptoms. The problems mentioned here are not from a single case so check with the BizTalk team regarding referencing issues and BizTalk 09.

Accelerator for RosettaNet Configuration Failure

evyang — Fri, 13 Nov 2009 16:17:52 GMT

I have seen quite a few issues since last year regarding to BTARN configuration wizard. The failures seemed to be in the wizard itself, not the typical SQL connectivity or permission related issues. In some cases the UI failed to load at all. Configuration log indicated access violation during form load. In other cases, configuration UI does load but configuration fail to complete. While the errors vary, the configuration log pointed to something UI related. These issues have been difficult to track down. In one case, we created a number of debug builds but the symptom continued to shift. So while we are still looking for the root cause, I thought it’d be helpful to post the workaround we have been using. The workaround is simply to use silent configuration. Logical since the problem seems to be specific to UI. For BTARN, there are only 2 features so the configuration XML is manageable. You can either export a configuration XML from a fully configured BTARN box and then modify the settings or you can copy configuration XMLs from a configuration log and combine the features manually. In case you don’t have access to a configuration XML for BTARN, this is a sample of what it should look like:

<SQL ID="RN_DB" DisplayName="Configuration Database" Description="This database stores configuration data for BizTalk Accelerator for RosettaNet.">
<Server>Insert Server Name</Server>
<Database>BTARNCONFIG</Database>
<WindowsSecurity Editable="no">yes</WindowsSecurity>
<UserName /><Password /></SQL>

<SQL ID="RN_DATADB" DisplayName="Storage Database" Description="This database stores runtime information for RosettaNet transactions.">
<Server>Insert Server Name</Server>
<Database>BTARNDATA</Database>
<WindowsSecurity Editable="no">yes</WindowsSecurity>
<UserName /><Password /></SQL>

<SQL ID="RN_ARCHIVEDB" DisplayName="Archive Database" Description="This database stores message content for archive and tracking purposes.">
<Server>Insert Server Name</Server>
<Database>BTARNARCHIVE</Database>
<WindowsSecurity Editable="no">yes</WindowsSecurity><UserName /><Password /></SQL>

</Answer>
</Question>

<Name ID="RN_IISSERVERNAME_ID" DisplayName="Web server name" Description="The name of the Web server where the RosettaNet applications are installed." Hidden="false">
<Value>Insert Server Name</Value>
</Name>

<WebSite ID="RN_HTTPRECEIVERWEBAPP" DisplayName="BizTalk HTTP Receive virtual folder " Description="Configure HTTP Receive virtual folder.">
<WebSiteName>Default Web Site</WebSiteName></WebSite>

<NTService ID="RN Service ID runtime" DisplayName="Application Pool service account for BTARN HTTP Receive location " Description="Service account application pool and database configuration.">
<UserName>insert user name</UserName>
<Domain>domain name</Domain><Password>password</Password>
</NTService>

</Feature>

<WebSiteName>Default Web Site</WebSiteName></WebSite>

<Name ID="RN_BTSSERVERNAME_ID" DisplayName="BizTalk Server name" Description="The name of the BizTalk Server where the BTARN pipeline component and the HTTP adapter are installed." Hidden="false">
<Value>Insert Server Name</Value>
</Name>

</Feature>

<InstalledFeature>Runtime</InstalledFeature>

<InstalledFeature>WebApps</InstalledFeature>

</Configuration>

From command prompt, browse to BTARN install folder and you can execute “configuration.exe /s <Configuration XML File Path>”. In some cases where UI partially configured BTARN before crashing, you can use “configuration.exe /u” to un-configure before running silent configuration again. Remember to remove the databases if they have been created previously. Certainly not an ideal solution at this point but hopefully this can help to unblock you in the time being.

New feature in SP1 for BizTalk 2006 R2: Override group certificate to sign outgoing messages

Manuel999 — Thu, 12 Nov 2009 14:49:00 GMT

Service Pack 1 for BizTalk 2006 R2 is now released as beta, and includes a number of hotfixes.

(See http://support.microsoft.com/kb/974563 )

SP1 also includes a new feature, namely the ability to configure different certificates for signing outgoing AS2 messages for different parties.

Prior to SP1 for R2, we could only set one certificate for signing outgoing AS2 messages per BizTalk group.

After installation of SP1, you will see in the AS2 party properties a new tab, where you can override the group certificate.

If you do not choose to override the group certificate, the group certificate will be used by default.

We are already describing how to use this in R2 with SP1. See the topic " To configure a certificate for signing outgoing AS2 messages for a specific party" in the R2 documentation.

http://msdn.microsoft.com/en-us/library/bb728096(BTS.20).aspx

We are currently working on porting this feature to BizTalk 2009.

Manuel Stern

BizTalk 09 and SQL R2 not Supported\BizTalk 09 and Project References

Larry2 — Mon, 09 Nov 2009 15:40:28 GMT

In a recent case the customer was attempting to install BizTalk 09 with SQL 2008 R2. It fails with the following error.

"Failed to create Management database 'BizTalkMgmtDB' on server 'XXXXX'. Error '0x80040E1d' occurred."

The server name is represented by the "XXXXX".

No follow up has been done to explain the problem. The BizTalk 09 product site does not mention support for this version of SQL.

Building Projects with BizTalk 09

Beware when building projects with BizTalk 09. There is a problem with references. There are various manifestations of the problem. Artifacts from referenced projects may not be available. Messages types requiring a reference to another project may not be available. Maps requiring schemas from referenced types are not available. The hot fix is under development and should be available in the near future. This could be a bigger problem as more symptoms are discovered. One fix may not fit all.

Unable to deploy the MSI Application that contains WCF bindings to BizTalk

Hari — Tue, 27 Oct 2009 09:33:00 GMT

I have seen few customers running into this issue. This forced me to write this blog.

Background: On a development environment, my customer had a BizTalk application. Application has few WCF bindings. They created MSI file that can be used to install the application onto production machines. When they try to deploy the application to target machines, following error is encountered

Validation failed for 1 resource(s).

Invalid URL "http://localhost:90/" provided as destination location. Valid format of URL is "http://localhost[:port]/path".

Environment: BizTalk 2006, Windows 2003

Cause: During the deployment of MSI application, For WCF bindings, BizTalk would verify whether URL is in <http://localhost[:port]/path> format or not. If it is not in the specified format BizTalk would display the error similar to the above.

In this specific case WCF service is hosted directly under root directory (under localhost) and hence it is referenced by http://localhost. When BizTalk checks the URL it finds that actual URL is not represented in http://localhost[:port]/path format (because it expects to see that WCF service is hosted under a Virtual directory, but not under root directly) and it errors out.

Resolution: You can work around this issue by following the below steps.

1) In the source machine host WCF service under virtual directory (i.e. For

example instead of hosting WCF service directly under localhost host it under localhost/WCFService)

2) Modify the receive location properties of WCF Receive port and specify the modified URI. It should be like /virtualdirectoryname/servicename.svc.

3) Export the package now

4) Make sure while exporting, under the “Specify IIS Hosts” window make sure Hostname is set to localhost and Path should contain /virtualdirectoryname (If we just have “/” here it would error out)

5) Import the MSI file in the target computer

Hope this helps!

Inserting parent child records with Identity column using WCF-SQL Adapter in one transaction

Atin Agarwal — Mon, 05 Oct 2009 14:40:00 GMT

There would be scenarios where you are inserting into a Parent table which have got an auto incrementing or an Identity field and you need to get this identity value back and use it to insert into child table records. And that too all, in a single transaction.

In Native SQL Adapter, this kind of composite operation can be achieved using updategram by capturing the value generated for the Identity field using the updg:at-identity attribute. This is documented at whitepaper: Working With BizTalk Adapter for SQL Server. But when it comes to WCF-SQL Adapter (Adapter Pack 2.0), since it would not work with updategram, only way I could figure out is to write a Stored Procedure which will do all the stuff in a single transaction and you can then call it from the Send Port. Of course, you can do the same in multiple operations without calling Stored Procedure where you first write to Parent table and get the identity value back as response and then write to child table; but this would not be a single transaction and any failures will writing to child table will not rollback the insertion into Parent table by itself.

Now, when it comes to writing Stored Procedure, if you just need to insert into one parent record and one child record, it would be simpler. But what about if there are multiple child records for a single parent record? For this, we would need to create an Xml message and pass this to the Stored Procedure as an Xml parameter. And then inside the stored procedure, we need to retrieve the values from the Xml message and insert it to our Parent and Child tables. Parsing the Xml message inside Procedure will be not an easy task if the message schema is complicated (especially for EDI Schemas :) ) and will involve good knowledge of writing Xpath Queries. But this becomes really simpler with use of Table Types and Table-Valued Parameters in SQL 2008. (Thanks to Mustansir Doctor for pointing me to this new feature of SQL 2008)

Here is an example of how we can achieve the same using Table Types and Table-Valued Parameters.

Lets say we have two tables Orders & Order Details in SQL 2008 with Primary Foreign relationship and OrderId is an auto incrementing field.

Now, first step will be to create Table Types for the above two tables:

We do not need to have OrderId field in the Table Type created as this is an incrementing field and we do not need to pass this in parameters passed to Stored Procedure.

Now, create a stored procedure to which we will pass this Table Type as parameter and inside it we will first insert into Orders table and read out the Identity value and use it to insert into OrderDetails table. Also, notice that reading out values from the Table Type objects is as simple as reading out form a Table.

Now, what all we have to do is to use Consume Adapter Service through Visual Studio to generate Schema.

It will generate two schemas TableType.dbo.xsd and TypedProcedure.dbo.xsd along with a binding file for the WCF-SQL Adapter.

TableType.dbo.xsd

TypedProcedure.dbo.xsd

TypedProcedure.dbo.xsd contains the request and response schema for WCF-SQL Adapter.

Now, only thing remaining is to construct a message of the request schema type inside the Orchestration and send it to WCF-SQL Adapter which will execute the Stored Procedure. I am not going into it more detail on how we do it inside Orchestration as it will be similar to any other operation we do using WCF-SQL Adapter.

You can see that using the Table Types simplify the things to a large extent and we do not need to write a single line of code to parse the XML.

MSDTC issue in BTS 2009 virtualization environment

evyang — Tue, 29 Sep 2009 19:28:31 GMT

This is a repost of Ahmed Irkakene’s mail to his fellow BizTalk colleagues.

======================================================================

I have been working with a customer last time to install several BTS 2009 systems. Customer was cloning his initial VMware system.

As a result he is facing problem with MSDTC: “MSDTC not configured correctly.”

MSDTC component, during its transactional operations, needs basically SID and NetBios name to be resolved.

As you imagine without MSDTC, BizTalk server can’t do much.

When cloning system, the customer should:

1. Use sysprep tool to get a new SID for each system ( DO NOT newsid tool as it’s NOT supported )

ð As a consequence, the system will be removed from its domain.

2. Re-install properly MSDTC

Before joining the domain again, please follow the steps below to make sure the MSDTC is correctly installed.

a. MSDTC –Uninstall

b. Reboot

c. MSDTC –Install

d. Reboot

Make sure the domain replication is occurred ( or force it to occur), otherwise you may get the following error even if the system joined the domain:

The security database on the server does not have a computer account for this workstation trust relationship

Also, one more important thing ( mainly when using static IP address but still valid for dynamic ones ):

As mentioned earlier, MSDTC needs name resolution and therefore “Enable NetBIOS over TCP/IP” should be enabled on all systems (adapters network) involved in the scenario.

Troubleshooting BTARN Web Pages and more on BTARN x64

Larry2 — Fri, 04 Sep 2009 17:08:09 GMT

Support for BTARN often involves issues connecting with the partner. Unfortunately the web pages deployed for this purpose are not the best when it comes to troubleshooting. It can be problematic to tell a customer they need to deploy new pages with increased tracking capabilities into their production environment. This is especially true if it is 3 AM. This was generally the situation prior to BTARN 3.3.

With .Net 2.0 some cut and paste changes to the web.config file often provide the same information without the need to redeploy pages to the production environment. The configuration changes provided here should be pasted in after the closing </system.web> tag. Then make sure the logging location exists and is available for output.

<system.diagnostics>

<trace autoflush="true" />

<sources>

<source name="System.Net">

<listeners>

<add name="System.Net"/>

</listeners>

</source>

<source name="System.Net.HttpListener">

<listeners>

<add name="System.Net"/>

</listeners>

</source>

<source name="System.Net.Sockets">

<listeners>

<add name="System.Net"/>

</listeners>

</source>

<source name="System.Net.Cache">

<listeners>

<add name="System.Net"/>

</listeners>

</source>

</sources>

<sharedListeners>

<add

name="System.Net"

type="System.Diagnostics.TextWriterTraceListener"

initializeData="c:\\trace\\System.Net.trace.log"

/>

</sharedListeners>

<switches>

<add name="System.Net" value="Verbose" />

<add name="System.Net.Sockets" value="Verbose" />

<add name="System.Net.Cache" value="Verbose" />

<add name="System.Net.HttpListener" value="Verbose" />

</switches>

</system.diagnostics>

A log will be created with great deal of detail. This includes the messages being passed. The log eliminates the previous issues with same-box transfers, SSL encryption, and sometimes limited access to network resources. Armed with this information it should be straight forward to understand the problem. The same information from BTARN 3.0 would require many lines of code added to the web pages to get the same information and as mentioned before redeployment of the enhanced pages.

BTARN and Windows 08 x64

In my previous blog I published help when using x64 with BTARN. This is particularly bothersome for Windows 03 x64 because BTARN is only available as a 32 bit application. Another little discovery has been detected with Windows 08 x64.

It turns out Windows 08 uses TLS protocol 1.0 as the default when attempting connections for BTARN (any HTTP). This was not the case with Windows 03. In some cases this turns out to be a problem. Close examination of the conversation shows the "Client Hello" being responded to with an "Encryption Error" or some other server drop. When things work properly the first attempt to use the TLS protocol should be followed with a second attempt to revert back to SSL 3.0 if the server cannot use TLS 1.0. Some servers don't allow the second attempt.

The fix for this situation is to change the code in the BTARN web pages. The following line of code should be added.

//this line for compatibility with Win08

ServicePointManager.SecurityProtocol = SecurityProtocolType.Ssl3;

//rest of the code

HttpWebRequest outerRequest = null;

ErrorLevel level = ErrorLevel.ProxyToOuterFailure;

This code sets a property. Location is not critical except I would keep it close to the HttpWebRequest object so it is easy to follow like listed above. This will prevent the use of the TLS protocol for this communication.

BizTalk DevCenter - check it out!

MandiAnez — Sat, 15 Aug 2009 00:08:00 GMT

The BizTalk support team is now adding content to the BizTalk Developer Center on MSDN, especially the Support page:

http://msdn.microsoft.com/en-us/biztalk/default.aspx

Our goal is to continually update these pages with the following information:

Common call generators, including Symptoms and potential Resolutions
Potential scenarios and their fixes
Contributing to the monthly theme
Providing learning materials, including some of the same BizTalk 2009 training materials that the BizTalk support team used:

BizTalk Server Training from Microsoft Customer Service and Support
http://www.microsoft.com/downloads/details.aspx?displaylang=en&FamilyID=459bd96f-0c52-49ac-bb78-e1868c2d3121

We encourage you to browse the BizTalk DevCenter. New information is being added on a regular basis.

Thanks!
Mandi

Error due to missing HIPPA Schema Trigger Field Annotations

Atin Agarwal — Thu, 30 Jul 2009 19:44:00 GMT

Recently, I was working on a support incident where the customer have modified 837I HIPPA schema by adding a child node under DTP segment. But, while we tried to validate an instance against the schema at both design time and run time, we will get error “Object reference not set to an instance of an object”.

Below is the error thrown by EDIReceive pipeline at run time.

Event Type: Error
Event Source: BizTalk Server 2009 EDI
Event Category: None
Event ID: 4097
Description:
Fatal error encountered in EDI Disassembler, error information is Object reference not set to an instance of an object.

The issue would not happen if we were using the default 837I HIPPA schema which comes out of the box with BizTalk. Now, it means that something going wrong with the node which we have added to schema and the error is also of not much help here. Also, if we remove the data from the input instance corresponding to the node added in the schema, it would get successfully parse.

After we did a thorough examination of schema (all thanks to Farida for this), we found that the child node added under DTP segment in the schema was missing the HIPPA Schema Trigger Field Annotations. First learn what are HIPPA Schema Trigger Field Annotations, for that click here. Here it was happening that, for DTP segment, we had other default child nodes which have got their Trigger Field Annotations specifying what their qualifier values (or trigger values) are. Now when the EDI Disassembler encounters DTP segment followed by the qualifier value, it should be able to resolve it to the corresponding XML child node under DTP segment. But when it encountered DTP segment for the child node added to the schema, it was not able to resolve the qualifier value to any node since there is no child node matching that qualifier value and hence the error.

Once we added the Trigger Field Annotations to the node added manually, it resolved the error. Note that, for adding the trigger info, we have to open the schema in a text editor.

EDI/AS2 in BizTalk 2009 : How to get the BatchId for a specific BatchName

Manuel999 — Thu, 16 Jul 2009 12:41:00 GMT

One of the new features in BizTalk 2009 EDI is the possibility to create multiple batch configurations for each party.
Each batch will have a specific name and specific batchID and can be managed individually.

For a specific batch name you can get the corresponding batch ID. For example you may need this for an external trigger message for that batch.

If you want to do that programmatically, you may expect you can just call Partner.GetBatchIdForBatchName method by passing the batch name.
http://msdn.microsoft.com/en-us/library/microsoft.biztalk.edi.partneragreementmanager.partner.getbatchidforbatchname.aspx

But looking at the signature of that method, it is probably not what you want. Instead of expecting just a string for the batch name, this method wants two int, the partyID and the messageType. But if you just want to pass in the batch name, you do not get very far with that method.

To get the batch ID for a batch name, we have a stored procedure you can call from code: edi_GetBatchIdForBatchName .
Calling
exec edi_GetBatchIdForBatchName 'MyBatch'
will return the ID of the batch name MyBatch.

Note: both calling GetBatchIdForBatchName and edi_GetBatchIdForBatchName directly is not supported.

Manuel Stern

Configuring BTARN on x64 boxes

Larry2 — Thu, 09 Jul 2009 21:27:34 GMT

As x64 O/S's start to become the norm documenting the additional steps required to configure BTARN to work with these may be helpful. Most of this information is documented in the troubleshooting or "known issues" section of the product documentation. The most important thing to know is BTARN is 32-bit.

The information provided here addresses Windows Server 03 x64 and 08. The changes are to IIS so these would apply to other operating systems depending on the version of IIS. BizTalk provides configuration guidance to enable x64 processing. This is not compatible with BTARN. Review these steps if 404's or 500's are showing up in the application log when attempting to operate BTARN.

Web Service Extensions

Windows 03 (IIS 6.0)

Open IIS manager and check the web extensions. Make sure the BizTalk externsion is not set to the x64 version. It should be "C:\Program Files (x86)\Microsoft BizTalk Server 2006\HttpReceive". The next configuration is a couple of command lines. The first sets processing to 32-bit and the second installs .Net.

cscript %SystemDrive%\inetpub\AdminScripts\adsutil.vbs set w3svc/AppPools/Enable32bitAppOnWin64 1

C:\WINDOWS\Microsoft.NET\Framework64\v2.0.50215>aspnet_regiis.exe -i

This may not be necessary if the ASP .Net 2.0 32-bit entry is visible and enabled under "Web Extensions". If this is not displayed the above code will make the extension available. Be sure to set the value to "allowed".

Windows 08 (IIS 7.0)

Again, BTARN has to run as a 32-bit application. The bigger issue with 08 is the relocation of several configuration options. Here are the instructions to install the BizTalk ISAPI filter from the BizTalk documentation:

Click Start, point to All Programs, point to Administrative Tools, and then click Internet Information Services (IIS) Manager.
Select the root Web server entry and in the Features View, double-click Handler Mappings and then in the Actions pane, click Add Script Map.
Note - Configuring the script mapping at the Web server level will cause this mapping to apply to all child Web sites. If you wish to restrict the mapping to a specific Web site or virtual folder, select the target site or folder instead of the Web server.
In the Add Script Map dialog box, enter BtsHttpReceive.dll in the Request path field.
In the Executable field, click the ellipsis (.) button and browse to drive:\Program Files\Microsoft BizTalk Server 2009\HttpReceive. Select BtsHttpReceive.dll, and then click OK.
Enter BizTalk HTTP Receive in the Name field, and then click Request Restrictions.
In the Request Restrictions dialog box, select the Verbs tab and then select One of the following verbs. Enter POST as the verb.
On the Access tab, select Script, and then click OK.
Click OK and when prompted to allow the ISAPI extension, click Yes.

This works up to a point. Note the location would implement the x64 version. Since BTARN only works as a 32-bit use "C:\Program Files (x86)\Microsoft BizTalk Server 2009\HttpReceive" instead.

Application Pools (app pool)

Make sure to use a custom app pool for each BTARN web virtual folder. Configure them to run under the same identity as the BizTalk service. This is the same for IIS 6.0 or 7.0. Move to the individual web folders and select the app pool for each.

That's it for IIS 6.0. IIS 7.0 requires additional configuration under each app pool. Select the "Advanced" configuration option and look for "Enable 32-bit Applications" setting and change it to "True". This should get BTARN working. Use the "LoopBack" tutorial to verify the configuration.

Client Authentication (ELEMICA)

Client authentication is not required to get BTARN up and running. However, in many cases this is required by partners. The existing procedures work for Windows 03 x64, but 08 is a different animal. The "ELEMICA" whitepaper (http://msdn.microsoft.com/en-us/library/cc563935.aspx) provides the bulk of configuration details. The process for 08 is easier but not unless instruction is provided. The client authentication certificate installation requires more explanation.

Open MMC and add the "Certificates" add-in. Select the machine account and open the "Personal" store. Import the private key used for client authentication.

Right click the new entry and select "All tasks". Select "Manage Keys" and give "Everyone" read access to the key.

The whitepaper includes a link to replacement web pages required to reference the certificate used for client authentication. The documentation is not so clear on the configuration of the web page. The additional parameter in the web.config must point to the public key of the client authentication certificate. This is the unusual situation where both the public and private keys are required.

Call orchestration with ref parameter in a loop

Atin Agarwal — Thu, 18 Jun 2009 18:25:00 GMT

Sometime back I worked on a case where the spool would go very high and I have been thinking to blog this for some time now and today I decided I will process this awaiting queued item in my mind.

I had a customer who was seeing a surge in the spool count whenever an orchestration instance started and spool would go back to its original count only once this instance get completed. As we all know, the high count of spool impact Biztalk processing, the customer was too worried about it.

Looking into the orchestration, they have got a loop shape which could loop up to 40K times or even more. In this loop, they were calling an orchestration with Biztalk message being passed as ref parameter.

We found that with REF we need to keep messages around until the orchestration completes, so that is why the spool count keeps growing till the loop completes and once the orchestration instance completes, Biztalk jobs would clean up messages from spool. This is by design.

To demonstrate this, we got a simple orchestration with a loop inside. This will loop for 50 times only.

Inside the loop, we are making a call to another orchestration and passing Biztalk messages as REF parameter.

If we test it, and monitor the spool count in perfmon, we will see that spool keeps growing until the loop completes and once the orchestration instance completes, Biztalk jobs are quick to clear the spool.

As a workaround for REF parameter, if we pass a message as IN parameter and have one more OUT parameter of the same type in the called orchestration to return the result, I see that spool growth is lower than while using REF parameter, but still for very large loops spool count can grow substantially.

Also, if possible we can try using variables instead - which can be reused as opposed to message.

So moral of the story is we have to avoid large loops with call orchestration shape passing ref parameter. We would suggest using smaller loop, or removing the call structure by using expression Shape or custom object etc.

How to join two schemas in a map when they contain namespaces

Atin Agarwal — Mon, 01 Jun 2009 23:03:00 GMT

I worked on an issue where we were receiving two different messages (Message1 and Message2) inside orchestration each with multiple records. Now, we have to join these messages in a map inside orchestration based on TRANID and PROCID. So basically, we have to get the value of TRANID from Message1 and look for the record inside Message2 where value of PROCID matches to value of TRANID and get the value of SUM element and populate it in the output message and this process have to be repeated for every TRAN record inside Message1.

Input Message1

Input Message2

Output Message

Though, there is already a solution for this which could be found at the blog <http://geekswithblogs.net/synboogaloo/archive/2005/04/22/37335.aspx>, but only thing different is that in this case, input schema contains namespace which would complicate the XPATH Query used in XSLT Call Template. Let us figure it out later, how to quickly build XPATH query using namespaces in this scenario.

1) First, create the schema for the three messages.

2) Now create orchestration like below. It will receive these two input messages, constructs the output message using map inside Transform shape and then sends the output to a file location. Since we have to receive two different messages using File adapter and we do not have any unique identifier for correlation purpose, I used BTS.ReceivePortName for correlation. Though, this may not be very practical, I just want to demonstrate joining of two schemas when they have got namespace. But yes, make sure that both the receive locations are in the same Receive Port.

3) In the transform shape, select the msgInput1 and msgInput2 as Inputs and msgOutput as Output. Open the map now.

4) In the map, put a looping functoid as in the figure below. Link TRANID and DESC from input to output. The main thing remaining is to retrieve value of SUM from INPUT2 when TRANID matches to PROCID. Now drop a scripting functoid to the map and connect it to TRANID as Input and SUM as Output. It should look something like below.

5) Now, for the scripting functoid, things would have been something like below if the schema does not have the namespaces. But here, this XPATH query inside the ‘Inline XSLT Call Template’ will give us only blank results.

6) So we need to use the namespaces prefix in out query to get the desired output. Now we need to find out the namespace prefix being used in map for INPUT 2 schema. To find it out, we need to view the XSLT of the map. Therefore, we have to validate the map file first. So, right click the map file and say validate map. Now in the output, we will get a link to the map XSLT. Open this XSL file. Now in this XSL file, we can find the prefix used for the namespace “http:\\Input2” and that is ‘s1’.

7) Now, use the prefix "s1" inside the XPATH Query as below. There could be other ways to use XPATH Query but this was the best I could figure out.

8) Deploy the solution and test it. That’s all.

I believe there would be several other solutions for this and if you know one, please share it. Would really be interested in knowing that.