Known Issues for Oracle SOA Products and Oracle AIA Foundation Pack for 11g Release 1 (11.1.1.7)

 

This document lists issues and workarounds for Oracle SOA and Oracle AIA Foundation Pack 11g Release 1 (11.1.1.7).

For a list of issues and workarounds pertaining to Oracle SOA Installation, Upgrade, High Availability, Enterprise Deployment, Performance Tuning, and Web Services, as well as SOA on IBM Websphere, see the Oracle Fusion Middleware Release Notes.

For known issues and workarounds for other Oracle SOA and AIA Foundation Pack releases, see Known Issues for Oracle SOA Products and AIA Foundation Pack.

On this page: (Last updated: 20-September-2013)

AIA Foundation Pack

SOA and BPM Suite Common Functionality

Technology Adapters

Adapter for Oracle Applications (Oracle E-Business Suite Adapter)

Application Adapters for WebLogic Server

Application Server Legacy Adapters

BPEL Process Manager

Business Activity Monitoring

Business Process Analysis Suite

Healthcare Integration

Business Process Management

Business Rules

Business-to-Business Integration (B2B)

Event Processing

Enterprise Repository

Human Workflow

Mediator

Service Bus



AIA Foundation Pack


Installation Fails with <AIA_Home>\Infrastructure\Install\AID\AIAPLWImportData.xml:48: exec returned: 1 error message


Bug: 12806332
Added: 19-September-2011
Platform: All

This error occurs because there is an issue with your environment and database set up.

Workaround: You need to configure timezone related startup parameters before you start the servers and install again.

To set up a timezone perform the following tasks:

  1. Login to WebLogic console.
  2. Navigate to Environment, Servers, soa_server1 (or your SOA Server).
  3. Click the Server Start tab.
  4. In the Arguments field enter -Duser.timezone=XYZ.

Where XYZ should be a valid timezone value in the database. Before you enter the value, check the timezone value in the AIA database.

To check the timezone value, connect to AIA database and run ALTER SESSION SET TIME_ZONE = '<Your Timezone>';

For example ALTER SESSION SET TIME_ZONE = 'America/New_York'. If it does not throw any error, use the value. If it throws an error, apply DST patches for the database. For example Note 975450.1 - DSTv13 update for the Oracle Database.


AIA Lookup Values in Project Lifecycle Workbench and CAVS Are Displayed in English Even When Your Locale Is Non-English Admin Language


Bug: 12840782
Added: 19-September-2011
Platform: All

When you access Project Lifecycle workbench UI and CAVS, the table gets populated with values in English language even if your locale is any of the nine other admin languages that include German, French, Spanish, Italian, Brazilian Portuguese, Japanese, Korean, Simplified Chinese, Traditional Chinese. This happens because values in AIA lookup table are stored in English language.


Warning in Installation Log


Bug: 12985128
Added: 19-September-2011
Platform: All

When you install, install log shows the following warning "WARNING: UNRECOGNIZED OPTION STAGE IS BEING IGNORED".

Workaround: Ignore this warning as it does not affect any functionality. No workaround required.


Configuration Fails as RAC Does Not Support SID


Bug: 13434421
Added: 27-November-2011
Platform: All

Deployment Configuration Wizard looks for System Identifier (SID) for Database details. However, the configured RAC does not support SID and you will not be able to proceed with the configuration.

Workaround: Silent mode of configuration with the following workaround:

  1. Open the aiaconfig.sh/cmd file.
  2. Search for ./runInstaller -invPtrLoc $AIA_HOME/oraInst.loc -jreLoc $JAVA_HOME -weblogicConfig and add -novalidation to it.
  3. Run through the Deployment Configuration Wizard.
    Note: The values of the Databases you configure are not validated. Ensure they are right.
  4. Open the AIAInstallProperties.xml from the AIA_INSTANCE/config directory
  5. Change the participatingapplication\<participatingapplicationname\db\jdbc-url property to jdbc:oracle:thin:@<Database Host>:<Database Port>/<Database service name>.

Failed Configuration of Pre-built Integration Shows as Configured When You Rerun Deployment Configuration Wizard


Bug: 13434355
Added: 27-November-2011
Platform: All

When configuration of a pre-built integration fails, Deployment Configuration Wizard still makes changes to AIAInstallProperties.xml. Because of this, when you rerun the Deployment Configuration Wizard, it assumes that pre-built integration is completely configured.

Workaround: Re-launch Deployment Configuration Wizard and reconfigure the failed pre-built integration. There is no loss of functionality.


Cannot Specify Participating Application Information in Deployment Configuration Wizard If the Participating Application Is Down


Bug: 13434385
Added: 27-November-2011
Platform: All

AIA Foundation Pack 11.1.1.6.0, Deployment Configuration Wizard does not let you suppress validation of participating application and configure.

Workaround: You can suppress validation of participating application using the steps described in “Performing Silent Installation and Deployment of Foundation Pack” of Oracle Fusion Middleware Installation and Upgrade Guide for Oracle Application Integration Architecture Foundation Pack 11g Release 1 (11.1.1.6.0).

If you are configuring the participating application for the first time, and you need to suppress validation of participating application, take the silent installation template from your pre-built integration team.


Installer Does Not Tell You Why a Managed Server Is Not Added to the Managed Server List in the SOA Server Details Screen


Bug: 13434393
Added: 27-November-2011
Platform: All

If any of the managed servers are down, the drop down for Managed Server does not auto-populate cluster names. This happens because the installer checks whether the managed servers are up and running to ensure that installation does not fail because the managed servers are down.

Workaround: If you do not see a managed server, start the server and select it when it gets populated.


AIADemo Does Not Update aiainstallproperties.xml in Metadata Service Repository


Bug: 13434397
Added: 27-November-2011
Platform: All

AIADemo does not update aiainstallproperties.xml in Metadata Service Repository. As a result AIADemo details do not get reflected in product information in AboutAIA.

Workaround: No workaround required as it does not affect functionality.


Prerequisites Required for Opening Demo Files and Composites in Oracle JDeveloper


Bug: 13434400
Added: 27-November-2011
Platform: All

When you try to open demo files in JDeveloper, you get an error relating to adf-config.xml. The mentions that the credential will not connect to the MDS so this file needs to be changed.

Workaround: Replace this file with adf-config.xml present in $AIA_INSTANCES/config/MDSConfig/.adf/META_INF.

Open the file and add the line <property value="<MDS_PASSWORD>" name="jdbc-password"/> at two occurrences beneath <property value="<MDS_USER>" name="jdbc-userid"/>


Exception: Internal Error While Activating AIAJMSSERVER (JMS Server)


Bug: 13434407
Added: 27-November-2011
Platform: All

When the domain is set up in development mode but if you start the installation in production mode, AIADataStore does not get created under the domain and the exception is logged in the log files.

Workaround: Create a folder under the domain home with the name AIADataStore before you start the installation.


Back to top


SOA Suite and BPM Suite Common Functionality


Disabling Automatic Failover Behavior on Endpoint Addresses


Bug: 18615120
Added: 10-July-2014
Platform: All

For scenarios in which you do not want a failover attempted on the next configured endpoint address, you can set the oracle.webservices.endpoint.failover property to false on the impacted reference binding component in the composite.xml file. This setting overrides the default attempt to fail over to the next configured endpoint address. When set to false, there is only one service invocation using the address configuration. If the invocation fails, an invocation exception is raised and propagated.

Note: Contact Oracle Support Services about the patch related to bug 14165813. You must apply this patch for this property to work.

<reference name="my-reference" ui:wsdlLocation="
my-server/soa-infra/services/default/WSBPELFAProj/
wsbpelfaprocess_client_ep?WSDL"><interface.wsdl
interface="http://xmlns.oracle.com/WSMEDWSBPELFA_jws/WSBPELFAProj
/WSBPELFAProcess#wsdl.interface(WSBPELFAProcess)"/>
    <binding.ws port="http://xmlns.oracle.com/WSMEDWSBPELFA_jws/WSBPELFAProj
/WSBPELFAProcess#wsdl.endpoint(wsbpelfaprocess_client_ep/WSBPELFAProcess_pt)"
location="my-server/soa-infra/services/default/WSBPELFAProj/wsbpelfaprocess_cl
ient_ep?WSDL">
      <property name="oracle.webservices.endpoint.failover">false
  </binding.ws>
</reference>

 


Flow Page is Not Correctly Displayed When Using Internet Explorer 11


Bug: 18766838
Added: 9-June-2014
Platform: All

If you use Internet Explorer 11 with Oracle Enterprise Manager Fusion Middleware Control, click an instance in the Trace table of the Flow Trace page, and then click the Flow tab on the Instance page, the flow is not correctly displayed. Instead, a Loading rendering engine window is displayed.


Resolving remoteFault Errors When the Oracle Exalogic Virtual Machine Template Environment is Rewired to an Oracle Database


Bug: 18152232
Added: 9-June-2014
Platform: All

If you are using BPEL message aggregation, and receive the following remoteFault error in an Oracle Exalogic Virtual Machine template environment that is rewired to an Oracle database, increase the sleep time between subsequent invokes to 10 seconds. For information on increasing the sleep time, see the documentation included with your third-party tool.

[2014-01-28T22:19:07.573-08:00] [SOAServer_1] [ERROR] []
[oracle.soa.bpel.engine] [tid: orabpel.engine.pool-6. thread-2]
[userId:
<anonymous>] [ecid:
e35347cf94a50218:64440355:143dbb5a500:-8000-
000000000000804f,1:17227]
[APP:
soa-infra] [composite_instance_id: 10003] [component_instance_id:
10004]
[composite_name: AggregationMats20_true!1.0] component_name:
AggregationMats20_trueBPELProcess1] This error contained the
exceptions
thrown by the underlying routing system. Root cause : [[
com.oracle.bpel.client.BPELFault: faultName:
{{http://schemas.oracle.com/bpel/extension}remoteFault}
messageType: {{http://schemas.oracle.com/bpel/extension}
RuntimeFaultMessage}
parts: {{
summary=<summary>Waiting for response has timed out. The conversation
id is urn:2717FC9088AD11E3BFBB33A3C357C75A. Please check the process
instance for detail.</summary>
,detail=<detail>Waiting for response has timed out. The conversation id
is urn:2717FC9088AD11E3BFBB33A3C357C75A. Please check the process
instance for detail.<detail> ,code=<code>0</code>}

For information about BPEL message aggregation, see Oracle Fusion Middleware Developer's Guide for Oracle SOA Suite.


Differing Behaviors of getComponentInstances()


Bug: 14681788
Added: 5-November-2012
Platform: All

You can use the Facade API to migrate instances between different SOA composite application revisions, as described in Chapter "Managing SOA Composite Application Instances" of the Oracle Fusion Middleware Administrator's Guide for Oracle SOA Suite and Oracle Business Process Management Suite.

When instance migration fails, the following Java code is used to retrieve the list of failed components:

private static void validateMigrationResult(MigrationResult mr)
{
if (mr != null)
{
List failed = mr.getFailedInstances();

for (CompositeInstanceMigrationResult f: failed)
{
List failedComponents =
f.getComponentInstances();
System.out.println("Failed components list size: " +
failedComponents.size() + " Failed components list: " +
failedComponents);
}
}
}

Note the following differing behaviors for getComponentInstances():

  • If a composite instance state is marked as completed faulted and migration of the composite instance to another revision fails, f.getComponentInstances() returns a list of ComponentInstanceMigrationResult objects that can be used to retrieve details about the component instances. This is the expected behavior.
  • However, if a composite instance state is marked as completed and migration of the composite instance to another revision fails, f.getComponentInstances() returns an empty list, even though it should have returned a list of failed component instances.

Dequeuer Returns the Same Message in High Concurrency Environments with Oracle Database 11.2


Bug: 13729601
Added: 20-February-2012
Platform: All

The dequeuer returns the same message in multiple threads in high concurrency environments when Oracle database 11.2 is used. This means that some messages are dequeued more than once.

For example, in Oracle SOA Suite, if Service 1 suddenly raises a large number of business events that are subscribed to by Service 2, duplicate instances of Service 2 triggered by the same event may be seen in an intermittent fashion.

The same behavior is not observed with a 10.2.0.5 database or in an 11.2 database with event 10852 level 16384 set to disable the 11.2 dequeue optimizations.

Workaround: Perform the following steps:

  1. Log in to the 11.2 database:

    CONNECT /AS SYSDBA

  2.  

  3. Specify the following SQL command in SQL*Plus to disable the 11.2 dequeue optimizations:

    SQL> alter system set event='10852 trace name context forever,
    level 16384'scope=spfile;

     

  4. Restart the database.

Composite Instance Creation in Oracle Enterprise Manager is Disabled in an SSL-Enabled Environment


Bugs: 13387571
Added: 09-December-2011
Platform: All

If you enable SSL ports for the Administration Server and all Managed Servers, you cannot create a test instance of a SOA composite application in Oracle Enterprise Manager Fusion Middleware Control.

For example, perform the following steps:

  1. From the Oracle WebLogic Server Administration Console, disable the non-SSL ports and enable the SSL ports for the Administration Server and all Managed Servers.
  2. Deploy a composite from Oracle JDeveloper.
  3. On the Configuration page during deployment, select the Always use SSL checkbox.
  4. Attempt to create an instance of the deployed composite in Oracle Enterprise Manager Fusion Middleware Control and note that the Test button at the top of the SOA composite application home page is disabled.
  5. Click the disabled Test button, and note that an error message similar to the following is displayed:

    Unable to retrieve composite details.
    Error retrieving Composite E2ETestComposite (1.0) details from
    soa-infra runtime. This could happen due to the errors in soa-infra initialization.
    Please view the log files for details.
    EJB Exception: ; nested exception is:
    java.lang.NullPointerException


Default/Last Active Revision of a Composite Is Not Identified When Retiring All Composites in a Partition


Bugs: 13368213
Added: 09-December-2011
Platform: All

Be aware that when you select Retire All from the Composite Control list on the home page of a SOA partition in Oracle Enterprise Manager Fusion Middleware Control, all composites in that partition are retired with no warning message to indicate that the default/last active composite is being retired.

This is the expected behavior when performing a bulk retirement of all composites in a partition.


Same Deployment JAR File Name is Displayed for Projects in the SOA Bundle


Bugs: 13396758
Added: 09-December-2011
Platform: All

The names of the deployment JAR files created for a project in an application are displayed as the same in Oracle JDeveloper, when they should not be the same.

For example, perform the following steps:

  1. Create an application named mdstest.
  2. Create BPEL processes named mdsproj1 and mdsproj2 in the application mdstest.
  3. Create a deployment JAR named archive1 for mdsproj1 and mdsproj2.
  4. Create a SOA bundle by including the archive1 JAR of mdsproj1 and mdsproj2.
  5. When the SOA bundle is deployed, the archive1 of mdsproj1 is displayed under the MDS connection > applications tree. The archive1 of mdsproj2 is not displayed.

The deployment JAR file created for the two projects in the application should not have the same name.


Duration Period Details Are Not Displayed on the Partition Home Page


Bug: 13328823
Added: 09-December-2011
Platform: All

On the partition home page in Oracle Enterprise Manager Fusion Middleware Control, the instance and faulted instance count messages do not contain the duration period in the title.

For example, perform the following steps:

  1. Enable the following properties on the SOA Infrastructure Common Properties page:

    • Select the Restrict display of instances and faults to the last check box and specify a duration period (for example, the last 30 minutes).

    • Enable the Disable fetching of instance and fault count metrics check box.

  2. Go to the partition home page.

  3. Click the link of an instance or faulted instance count. Because you enabled the properties described in Step 2, links are displayed instead of actual instance and faulted instance counts.

    The message that appears is entitled Number of Instances. It should be entitled Number of Instances (For the last 30 minutes).


Fusion Order Demo JMS Adapter Destination Value in Deployment Plan is Hard-Coded


Bug: 12316777
Added: 6-May-2011
Platform: All

The deployment plan shipped with the Fusion Order Demo ignores the JMS adapter destination value and replaces it with a hardcoded value:

<wsdlAndSchema name="FulfillmentBatch_jms.jca">
   <jca:property name="DestinationName">
      <replace>jms/DemoSupplierTopicUdd</replace>
   </jca:property>
</wsdlAndSchema>

Therefore, if you change the destination value used by the JMS adapter, the JCA file included in the composite's JAR file reflects the changed value. However, when you deploy the new version of the composite and select the corresponding reference in Oracle Enterprise Manager Fusion Middleware Control, the old destination value in the deployment plan is still displayed.

Workaround: Change the destination value in the deployment plan or in the build.properties file.


ant Compilation Error with SOA Composite Applications on Linux


Bug: 10255512
Added: 06-May-2011
Platform: Linux

You can receive the following compilation error when using ant to compile a SOA composite application.

Error(12,61): Parse of component type files failed, check the adf-config.xml
file : "java.lang.NoSuchMethodError:
org.apache.xerces.impl.xs.XMLSchemaLoader.loadGrammar([Lorg/apache/xerces/xni/
parser/XMLInputSource;)V:
org.apache.xerces.impl.xs.XMLSchemaLoader.loadGrammar([Lorg/apache/xerces/xni/
parser/XMLInputSource;)V"

This can occur if you have a .ant directory with older versions of JAR files that gets added to the classpath. The ${user.home}/.ant/lib directory is a standard location used by ant that is always added to the classpath when running ant.

Workaround: Perform one of the following steps:

  • Rename the .ant directory (for example, to .ant_old).
  • Specify the -nouserlib option when running ant from the Linux command prompt. In Oracle JDeveloper, this option is accessible by right-clicking your project, selecting Project Properties > Ant > Classpath, and deselecting the Include Jar Files from User Home checkbox.

The composite then compiles successfully.


Mediator Instances Can Display as Running After Transaction Completion


Bug: 8420049
Added: 06-May-2011
Platform: All

An Oracle Mediator service component relies on a successful commit of a local transaction to know whether an instance is recoverable. Even if a graceful shutdown is correctly implemented, and all transactions are allowed to complete (within the given time out window), there can still be a local transaction that is denied because it fails to occur within the time out window. For example, assume the following:

  1. A database adapter posts a message.

  2. An Oracle Mediator service component receives the message, sets its state to running, and invokes a BPEL process service component, whose state is also set to running.

  3. A JMS adapter is invoked, and encounters and throws an exception. The reference fault tables are updated.

  4. The BPEL process receives the exception, marks its state as completed with a fault, and rethrows the exception to the Oracle Mediator.

  5. The BPEL process triggers an asynchronous persistence of its audit information as the global transaction is rolled back.

  6. Oracle Mediator receives the exception and persists the fault information in its error tables (in a separate local transaction).

  7. Oracle Mediator sets its state to faulted.

If the server is shut down before the transaction can complete (for example, between Steps 5 through 7), the Oracle Mediator instance can remain in a running state. The corresponding composite instance whose state is inferred from the components also displays as running. As the global transaction is rolled back, the message can be retried and reprocessed by the inbound adapter. Note that in this scenario, no mediator instances are lost.


Importing Shared Schema from MDS Does Not Copy the Schema Imports


Bug: 9587562
Added: 06-May-2011
Platform: All

If you import shared schema from MDS (with schema imports within), only the base schema is copied and the schema imports are ignored.

Workaround: Manually localize the schema.


National Language Support Date Format and the Looped Purge Script


Bug: 12358938
Updated: 06-May-2011
Platform: All

You may encounter error ORA-01861 while running the looped purge script due to a mismatch in the date format. If your NLS_LANG setting represents a non-English language (for example, Japanese_Japan.AL32UTF8), set NLS_DATE_FORMAT to "YY-MON-DD" temporarily for the session in which you are running the purge script.

NLS_DATE_FORMAT = "YY-MON-DD"

For information about the looped purge script, see Chapter "Managing Database Growth" of the Fusion Middleware Administrator's Guide for Oracle SOA Suite and Oracle BPM Suite.


PermGen Space Out-of-Memory Error when Using the Sun JDK


Bug: 8589284
Added: 06-May-2011
Platform: All

When the Sun JDK is used as the JVM for the SOA managed server, Oracle recommends that the following memory settings be used. If proper memory settings are not used, repeated operations on task detail applications (human workflow) can result in PermGen space out-of-memory errors.

  1. For UNIX operating systems, open the $DOMAIN_HOME/bin/setSOADomainEnv.sh file.

  2. For Windows operating systems, open the DOMAIN_HOME\bin\setSOADomainEnv.cmd file.

  3. Increase the following values:

    if [ "${JAVA_VENDOR}" != "Oracle" ] ; then
      DEFAULT_MEM_ARGS="${DEFAULT_MEM_ARGS} -XX:PermSize=128m
    -XX:MaxPermSize=256m"
      PORT_MEM_ARGS="${PORT_MEM_ARGS} -XX:PermSize=256m -XX:MaxPermSize=512m"
    fi
    

    to these values:

    if [ "${JAVA_VENDOR}" != "Oracle" ] ; then
      DEFAULT_MEM_ARGS="${DEFAULT_MEM_ARGS} -XX:PermSize=512m
    -XX:MaxPermSize=1024m"
      PORT_MEM_ARGS="${PORT_MEM_ARGS} -XX:PermSize=256m -XX:MaxPermSize=512m"
    fi
    

Increasing the XA Transaction Timeout Value


Bug: 8576274
Added: 29-April-2011
Platform: All

If you send notifications for parallel task assignments in human workflow, you can see the transaction close, SQL exceptions appear, and notification processing take a long time at runtime. If this occurs, increase the XA transaction timeout for an XA data source in Oracle WebLogic Server Administration Console. When set, this value overrides the value of the global transaction timeout.

  1. Click Lock and Edit.

  2. Choose Services > Data Sources.

  3. Click the specific data source.

  4. Click the Transaction tab.

  5. Select Set XA Transaction Timeout.

  6. In the XA Transaction Timeout field, specify the value of the transaction timeout.

  7. Click Save.

  8. Log into SQL*Plus.

  9. Set the distributed_lock_timeout value.

    SQL> alter system set distributed_lock_timeout=value scope=spfile;
    

    Oracle recommends that you set distributed_lock_timeout to a value greater than or equal to that set in Step 6.

  10. Shut down and restart your database after running this command.

  11. Verify that the setting has changed.

    SQL> show parameter distributed_lock_timeout;
    
  12. Set this parameter to the same value for all instances in an Oracle Real Application Cluster.


Handling Business and Remote Faults in the Calling BPEL Processes


Bug: 8578316
Added: 06-May-2011
Platform: All

Oracle recommends that you handle both business and remote faults in BPEL processes during design time. If remote exceptions are not handled in the calling BPEL process, you can see discrepancies in the overall SOA composite application instance state and the server component instance states associated with the composite. For example, assume the composite instance is in a faulted state, while the service component instances are in a completed state. If the caller handles the remote exceptions, then the states of composite and component instances are accurate.


Storing Artifacts in Source Control and Deploying Them with the ant Utility


Bug: 8599824
Added: 06-May-2011
Platform: All

If you store application artifacts in a source control system that you later want to reuse for deployment with ant, the following changes are required if the application has metadata service (MDS)-based references.

Assume the original Oracle JDeveloper application has file-based MDS references, such as the following:

<metadata-store-usage id="mstore-usage_1">
            <metadata-store
class-name="oracle.mds.persistence.stores.file.FileMetadataStore">
              <property value="${oracle.home}/integration"
                        name="metadata-path"/>
              <property value="seed" name="partition-name"/>
            </metadata-store>

Modify this code to use database-based MDS references in the Application_Directory/.adf/META-INF/adf-config.xml file, as shown below:

<metadata-store-usage id="mstore-usage_1">
            <metadata-store
class-name="oracle.mds.persistence.stores.db.DBMetadataStore">
              <property name="jdbc-userid" value="${mdsuser}"/>
 <property name="jdbc-password" value="${mdspassword}"/>
              <property name="jdbc-url"
value="jdbc:oracle:thin:@${db.host.name}:${db.host.port}:${db.sid}"/>
               <property name="partition-name" value="soa-infra"/>
            </metadata-store>

Check the modified application with these adf-config.xml entries into a source control system. The following concrete example of a modified adf-config.xml file is provided.

 <metadata-store-usage id="mstore-usage_1">
            <metadata-store
class-name="oracle.mds.persistence.stores.db.DBMetadataStore">
              <property name="jdbc-userid" value="soainfra_mds"/>
              <property name="jdbc-password" value="soainfra_mds"/>
              <property name="jdbc-url"
value="jdbc:oracle:thin:@myhost56.us.oracle.com:15739:db2825"/>
              <property name="partition-name" value="soa-infra"/>
            </metadata-store>

You can use a variable for the user ID and password and replace the values prior to ant deployment in the application.

If the MDS database has a JNDI name, then use the following entries in adf-config.xml:

<metadata-store
  class-name="oracle.mds.persistence.stores.db.DBMetadataStore">
     <property name="jndi-datasource" value="${db.jndi.name}"/>
     <property name="partition-name" value="soa-infra"/>
</metadata-store>


SOA Composites Do Not Deploy for an SSL-enabled Database on Windows


Bug: 11060696
Added: 06-May-2011
Platform: Windows

When SSL is configured between the database and Oracle Fusion Middleware on Windows, SOA composite applications do not deploy to the SOA Server. This issue only occurs on Windows; composites deploy correctly on Linux.


Google Chrome Browser Limitation on Displaying XML


Bug: 11897467
Added: 06-May-2011
Platform: All

Consider the following:

  1. At the top of the home page for a SOA composite application, click the Show WSDL and endpoint URI icon. The Service Endpoint and WSDL dialog displays with links for the endpoint and WSDL.
  2. Click the WSDL link. Instead of the WSDL file contents being displayed, only a link is displayed.

Workaround: Right-click and select View Page Source to display the WSDL. Alternatively you can install a plug-in to see XML in the Google Chrome browser.


Cannot Undeploy All Composites at the Partition Level with the Undeploy SOA Composite Wizard in SSO-enabled Environments


Bug: 11825391
Added: 06-May-2011
Platform: All

You cannot undeploy all composites at the partition level with the Undeploy SOA Composite wizard in Oracle Enterprise Manager Fusion Middleware Control in SSO-enabled environments. You can only undeploy a single composite at a time with this wizard.

Workaround: Perform the following steps:

  1. Delete the partition and create a new one.
  2. Go to the SOA composite application home pages for each composite
  3. and undeploy each composite one-by-one.


Message Schema in UTF-16 Encoding Can Cause a Runtime Exception


Bug: 10178726
Added: 06-May-2011
Platform: All

You may encounter the following runtime parsing error if a message schema is encoded in UTF-16.

Unable to parse schema <name>.xsd

To resolve this error, change the schema encoding to UTF-8. In Source view in Oracle JDeveloper, set the encoding declaration in the XML prologue to encoding='UTF-8' for the message schema.


Maximum Heap Memory Size Recommendations on Windows 2008 Server


Bug: 12377414
Added: 29-April-2011
Platform: Windows 2008 Server

If you are running Oracle SOA Suite on a 64-bit Windows Server 2008 host, setting the heap memory size to 2G in the DOMAIN_HOME\bin\setSOADomainEnv.cmd file causes the server to crash.

Workaround: Set the heap memory size to 1.5G. For example:

-Xmx:1536M –Xms:1536M –Xgc:genpar

where -Xmx is the maximum heap memory size.


Removing Extra Set Text Dialog Characters


Bug: 9062901
Added: 29-April-2011
Platform: All

Adding &amp; to the Text field of the Edit Text dialog in Oracle JDeveloper adds extra characters to the XSL file. For example, perform the following steps:

  1. In the XSLT Mapper, right-click a string element in the target pane.

  2. Select Set Text > Enter Text.

  3. In the Edit Text dialog, enter &amp; in the Text field, and click OK.

  4. Go to the Source view of the XSL file, and see that an extra amp; is appended.

    <xsl:text disable-output-escaping="no">&amp;amp;</xsl:text>
    

Workaround: In Source view of the XSL file, remove the extra character amp; that was appended in the previous steps. The source view now looks as follows:

<xsl:text disable-output-escaping="no">&amp;</xsl:text>


Extra Business Event with Oracle Database 11.2.0.1 and One and Only One Level


Bug: 9558493
Added: 29-April-2011
Platform: All

If you subscribe to a business event with Oracle Mediator, select a consistency level of one and only one, and use a release 11.2.0.1 Oracle database for the SOA Infrastructure schema, the first event published produces two business events.

For example, if 20 messages are published, the subscriber receives 21 messages.

This issue occurs only with release 11.2.0.1 of the Oracle database and only one time after a restart of both the SOA Infrastructure database and the SOA Server. After that, the correct number of business events are inserted.


Inconsistent States for SOA Composite Application Instances in Recovery


Bug: 9574804
Added: 29-April-2011
Platform: All

When a SOA composite application instance is being recovered, callback messages and activities are displayed as Running, but invocation messages are displayed as Faulted in the State column of the Dashboard tab of a SOA composite application.


SOAP 1.2 <stackTrace> Fault Element Is Not Returned to the Caller


Bug: 9528496
Added: 29-April-2011
Platform: All

A fault from a SOAP 1.2 Web service contains a <stackTrace> element as part of the <exception> element. The <stackTrace> element provides debug information and assistance in understanding a fault. If you invoke a Web service directly from a client and a fault occurs, the <stackTrace> element is provided.

However, if a SOA composite application invokes a Web service (for example, through a BPEL process or Oracle Mediator) and a fault occurs, the <stackTrace> element is not passed back to the caller.

This is the expected behavior.


Composite State Not Updated After BPEL Recovery of Faulted Instance


Bug: 9035478
Added: 29-April-2011
Platform: All

The state of a failed SOA composite application instance is not updated after you successfully perform a recovery from the Recovery page of the BPEL service engine. For example, assume you perform the following steps.

  1. Enable payload validation on the SOA Infrastructure Common Properties page in Oracle Enterprise Manager Fusion Middleware Control Console.

  2. Invoke an instance of a SOA composite application. If payload validation fails (for example, because required elements in the schema are missing), a failure occurs at the BPEL service component level during execution of an invoke activity.

    The instance is displayed as faulted in the Dashboard and Instances pages of the composite.

  3. Go to the Recovery page of the BPEL service engine and note that the same instance is pending recovery.

  4. Disable payload validation on the SOA Infrastructure Common Properties page.

  5. From the Recovery page of the BPEL service engine, select the instance and click Recover. This recovers the instance.

  6. View the completed instance in the flow trace. However, note that the instance state is not updated and is still shown as faulted, whereas the instance has actually recovered and completed successfully.

    This is the expected behavior. BPEL recovery created a new BPEL instance, which completed successfully. However, the previous BPEL instance (the one that failed schema validation) is still shown as faulted. The overall composite instance is counted as faulted, since one of its two component instances (the previous BPEL instance) is faulted.


WSDL Interface and EJB Version 2 Are Not Supported During Runtime


Bug: 10266818
Added: 29-April-2011
Platform: All

The Create EJB Service dialog enables you to select the following combination:

  • The WSDL radio button selection in the Interface section.

  • The EJB2 selection from the Version dropdown list.

This combination is not supported during runtime.


Composite Instance Is Flagged as Faulted Even After Fault Is Caught


Bug: 9941332
Added: 29-April-2011
Platform: All

After manually recovering a fault, a composite instance state in Oracle Enterprise Manager Fusion Middleware Control Console can still display as faulted. For example, assume you perform the following steps:

  1. Deploy a SOA composite application.

  2. Test the web service from the Test Page in Oracle Enterprise Manager Fusion Middleware Control Console by deliberately passing a float value for an input string.

    A fault is correctly generated and the fault policy is applied. The composite instance is in a state requiring recovery.

  3. Recover the fault by modifying the payload and passing an integer value for input.

    The composite instance state changes to faulted.

  4. Check the flow trace and it shows that the fault has actually recovered.

The instance tracking infrastructure does not maintain the relationship of upstream and downstream components. When an upstream component recovers from the recovery required state and a downstream component is in the faulted state, the faulted state cannot be reset; only the recovery required state is reset by the component of the recovery required instance. The downstream component of the faulted instance does not know that the message is replayed; therefore, the composite instance is not reset from the faulted state.


Faulted Instances Waiting for Manual Recovery Are Shown as Running


Bug: 9474574
Added: 29-April-2011
Platform: All

Faulted instances that are waiting for manual recovery are shown when you search for running instances. For example, if you go to the Instances page of the SOA composite application and specify a search criteria to find running instances, the faulted instances waiting for manual recovery are shown.


Composites Calling Other Composites with Concrete WSDL Files


Bug: 9585631
Added: 29-April-2011
Platform: All

If you use concrete WSDL files, be aware that the order of startup for SOA composite applications is not guaranteed.

For example, if one SOA composite application calls a second SOA composite application with a concrete WSDL file, and the first SOA composite application gets started before the second SOA composite application, an error occurs. This is because the first SOA composite application cannot load the second SOA composite application's WSDL file.


Transactional Adapter Status Displays as Complete Even If Rollback Occurs


Bug: 9298953
Added: 29-April-2011
Platform: All

If a transaction rollback occurs because of a business fault or exception, the instance trial for a transactional adapter does not indicate whether or not there was a rollback.

For example, assume you have an asynchronous BPEL process invoking a database adapter service. An assert expression is defined in the invoke activity. If the assert expression returns false, then all transactional invocations should roll back. However, the flow trace shows a status of completed for the database adapter service.


Accessing Components from SSO-Enabled Oracle Enterprise Manager


Bug: 8498628
Added: 29-April-2011
Platform: All

If you log in to an SSO-enabled Oracle Enterprise Manager Fusion Middleware Control Console and click the links to the following components, you are directed to non-SSO URLs for these components, and prompted to log in again.

  • Oracle BPM Worklist

  • Oracle B2B

  • Oracle BAM

  • SOA Composer


Failure to Add and Update WSDL Binding After Migration


Bug: 9012064
Added: 29-April-2011
Platform: All

When you open Oracle JDeveloper and attempt to migrate a 10.1.3 application that includes WSIF binding information to version 11g, migration creates a web service reference with no WSDL binding information (binding.wsif) in the composite.xml file. When you then attempt to update the reference with the same WSDL in Oracle JDeveloper, the binding information is not saved, and you receive a compilation error.

This is because Oracle JDeveloper is handling the typical migration use case in which the reference is updated with a new WSDL location (same WSDL, but in a different location). Therefore, when you do not change the WSDL location, nothing happens.

Workaround: Perform either of the following steps:

  1. Delete the reference and create a new one using the same WSDL.

  2. (Easiest) Copy the WSDL to a different file name, but do not change its contents. Then, update the reference using this WSDL copy. This changes the location of the WSDL and correctly creates the binding.ws.


Understanding BPEL Global Transaction and Adapter Local Transaction Issues


Bug: 8476752
Added: 29-April-2011
Platform: All

BPEL processes run in global transactions, while other Oracle SOA Suite components, such as adapters, run in local transactions. This can cause inconsistencies with the information displayed in Oracle Enterprise Manager Fusion Middleware Control Console.

For example, assume you have a main synchronous BPEL process that invokes an asynchronous BPEL process. There is one database adapter insert in the main synchronous BPEL process before the invocation of the asynchronous BPEL process. There is a second database adapter insert in the asynchronous BPEL process, at which point a fault is expected to be thrown.

The expected behavior is as follows:

  • An initial insert into the database is performed from the main process that completes successfully as expected.

  • The second insert also executes in the asynchronous process, but after encountering the fault, the BPEL process rolls back the initial insert.

However, in Oracle Enterprise Manager Fusion Middleware Control Console, the second database insert also shows as having completed successfully, instead of the having faulted.

This is because the database adapter invocation is logged (instance tracked) in a local transaction that is not part of the global BPEL JTA transaction. Therefore, the database adapter invocation completes successfully, but the subsequent BPEL process incurs a global rollback.

<invoke name="Invoke_1" inputVariable="Invoke_1_insert_InputVariable"
   partnerLink="db" portType="ns1:db_ptt" operation="insert"/>
. . .
. . .
   <throw name="Throw_2" faultName="bpelx:rollback"/>

At that point, instance tracking has already recorded the outbound invocation as successful (in a local transaction). If the adapter throws an exception, instance tracking logs this as a fault. Also note that some partner links may invoke a file adapter, which cannot be rolled back, even if the BPEL process later throws a bpelx:rollback.


Descriptions for Data Source JNDI Fields Are Interchanged


Bug: 8399320
Added: 29-April-2011
Platform: All

In the SOA Infrastructure Common Properties page of Oracle Enterprise Manager Fusion Middleware Control Console, the descriptions for Server Data Source JNDI and Server Transaction Data Source JNDI are interchanged. However, the Configure links go to the correct locations.


Granting the Same Role to the Same User Twice Causes Unclear WLST Message


Bug: 8454184
Added: 29-April-2011
Platform: All

  1. Create a domain with Oracle WebCache and extend it with Oracle SOA Suite.

  2. Associate the Oracle WebCache application with LDAP and grant the admin role to a user (for example, user name test) in Oracle Internet Directory using the WebLogic Scripting Tool (WLST).

  3. Grant the same role (admin role) to the same user again (test).

    You expect to see a WLST error message such as the following:

    This particular user already has admin role to access your application
    

    Instead, you receive the following message, which is inaccurate:

    Command FAILED, Reason: Cannot add principal to application role
    

Flow Trace Fails to Load with Numerous Correlated Composite Instances


Bug: 10077813
Added: 29-April-2011
Platform: All

The flow trace in Oracle Enterprise Manager Fusion Middleware Control Console can fail to load in scenarios in which you have numerous correlated composite instances (for example, 100). This is because multiple midprocess receive activities result in separate ECIDs. Creating a consolidated audit trail involves sending separate requests for each ECID (for example, 100 separate queries for audit trails).

Workaround: Perform the following steps to increase the transaction timeout parameter.

  1. Open the fabric-ejb.jar file.

  2. Within this JAR file, open the weblogic-ejb-jar.xml file.

  3. Set the transaction timeout parameter to a higher value.


Back to top


Technology Adapters


There is no truncate support for multibyte streaming in PS6


Bug:14507570
Added: 10-Sept-2013
Platform: All

 

Because using truncate in MultiByte Streaming Support can lead to corrupt data, truncate functionality has been removed from the Native Format Builder.

 

UMS adapter cluster failover may lose messages


Bug:16439192
Added: 11-June-2013
Platform: All

 

Since XA is not supported for UMS Adapter Outbound in this Release, the UMS adapter cluster
failover can lose messages.

Outbound UMS adapter should use utf-8, not server encoding if not specified


Bug:16474430
Added: 13-March-2013
Platform: All

 

For the outbound UMS adapter the property JCA.UMS.MSG.CONTENT-TYPE can be used for specifying encoding. But if you do not set the value, the outbound UMS adapter will use server encoding for the email by default. When the SOA server is started in native encoding, for example iso8859-1, it cannot handle ccjk characters--the content will be garbled. Consequently, UTF-8 should be default encoding, not server encoding.

Workaround: Set the jca property for the outbound UMS adapter as: JCA.UMS.MSG.CONTENT-TYPE = 'text/plain; charset=utf-8'


Support for JMS Adapter to Connect with IBM Websphere MQ JMS for Multi Instance


Bug:15919519
Added: 28-January-2013
Platform: All

You can configure the JMS Adapter to connect to IBM WebSphere MQ Multi-Instance Queue Managers. To do so, use a custom property called connectionNameList. You can use this property to specify the names and ports of the different instances.

To make sure of the new custom property, you can set the FactoryProperties property in the WebLogic Server console for the used JNDI entry with the following settings:

QueueManager=<QUEUEMANAGER_NAME>TransportType=1;ConnectionNameList=<HOST1>,(<PORT1>)<HOST2>,(<PORT2>)Channel=<CHANNEL_NAME>;
ThirdPartyJMSProvider=true;ClientReconnectOptions=0

JMS Adapter(11.1.1.7) Never Reconects to TIBCO EMS 4.2.0 Due to Failing XA Recovery


Bug:15939498
Added: 7-December-2012
Platform: All

The JMS Adapter does not reconnect to TIBCO EMS 4.2.0 because of failing XA recovery

Workaround:All servers where the TIBCO resource is deployed must be shutdown and restarted, if the resource is deployed to a cluster then the entire cluster must be shutdown and restarted. When you are shutting down servers,ensure it is done gracefully to give a chance for prepared transactions to be written to the TLOG and recovered with commit after the server is restarted.


DBADAPTER Coherence Feature Not Working On WLS 10.3.5


Bug: 14233285
Added: 13-Dec-2012
Platform: All

The Database Adapter Coherence Feature introduced in 11.1.1.6.0 works out of the box against WebLogic Server 10.3.6, but not WebLogic Server 10.3.5. This is because it must bundle a dependent jar, but no version of the jar is compatible with both 10.3.5 and 10.3.6.

Workaround: To ensure the Database Adapter Coherence feature works against WebLogic Server 10.3.5 you must follow these steps:

  1. Find DbAdapter.rar in your SOA install.
  2. Remove the bundled 10.3.6 version of toplink-grid.jar from DbAdapter.rar.
  3. Rebuild DbAdapter.rar with its existing manifest which looks for the shared library toplink-grid.
  4. Deploy toplink-grid.jar as a shared library named 'toplink-grid'.
  5. Restart the WebLogic Server.

Adding User Patch JAR File Under /soa/modules --Adapter Diagnosibility


Bug: 14117683
Added: 11-October-2012
Platform: All

Adapter diagnosibility does not work in 11.1.1.7.0

Workaround: For adapter diagnosibilty to work in 11.1.1.7.0 you must apply patch 1036.jar to $fmwhome/AS11gR1SOA/soa/modules/user-patch.jar.


Some Specific Scenario Like Bug13803044 Needs a WLS Patch


Bug: 14335927
Added: 11-Oct-2012
Platform: All

If results occur such as that documented in Bug 13803044, when a sequence table, multi-threading, distributed polling and USESSKIPLOCK are used together, it can be an indication that a WebLogic Server patch is required.

Workaround: Apply the appropriate WebLogic Server patch.


NLS:MBCS Named Table Cannot Be Used Due to Progress Database Limitation


Bug: 14259650
Added: 15-Oct-2012
Platform: All

Due to a Progress DB limitation, you cannot use MBCS as database object name, such as table name and column name because Progress database can't support MBCS named table/fields/indexes/sequences..


Progress DB Stored Procedures Are Not Supported in the PS6 Release of the Database Adapter


Bug: 14150479
Added: 22-October-2012
Platform: All

Progress DB Stored Procedures are not supported in the PS6 release of the SOA Database Adapter.


UMS Adapter Does Not Support XA Transactions On Outbound Side in the PS6 Release of the Database Adapter


Bug: 14586707
Added: 15-Oct-2012
Platform: All

The UMS Adapter does not support XA Transactions on the Outbound message side of transactions.


IllegalStateException when WebSphere cleans up JCA connection pool


Bug: 14275766
Added: 11-Oct-2012
Platform: All

Following error is observed (at times) when running AQ adapter samples.

The error in question is:.

00000136 MCWrapper EJ2CA0079E: Method cleanup has detected an internal
illegal state and is throwing an IllegalStateException. The exception is:
java.lang.IllegalStateException: cleanup: illegal state exception. State =
STATE_INACTIVE MCW = 8210821.

WAS Uses Different JCA Connection Factory for JMS Async Request Reply Scenario


Bug: 14754024
Added: 11-Oct-2012
Platform: All

In the JMS asynch request reply scenario if the same jca connection factory is used for any other JMS receive operation, as well as for jms asych req-reply operation (in jca file), then the below error is encountered in the WebSphere server environment. This is only applicable when the JMS adapter is deployed on the WebSphere Server.

See the following trace.

Caused by: com.ibm.websphere.sib.exception.SIIncorrectCallException: 
CWSIP0777E: The transaction used for receive on the destination: Destination 
(1712612884) demoQueueCorr2: E4143EA44C604CC7E255F98C has already been 
completed.
at 
com.ibm.ws.sib.processor.
impl.JSLocalConsumerPoint.receive(JSLocalConsumerPoint.java:1590)
 
at 
com.ibm.ws.sib.processor.impl.ConsumerSessionImpl.receiveWithWait
(ConsumerSessionImpl.java:448)
at 
com.ibm.ws.sib.api.jms.impl.JmsMsgConsumerImpl.
receiveInboundMessage(JmsMsgConsumerImpl.java:1114)

Workaround: Use different JCA connection factory for JMS operations when using the WebSphere server. Please refer to bug14730484 for more information.


Retry Interval, Maxinterval, and Backoff Parameters Will Not Take Effect for UMS Adapter


Bug: 14735717
Added: 12-Oct-2012
Platform: All

Retry Interval,MaxInterval &Backoff parameters don’t take effect for UMSAdapter Inbound

The inbound retry interval, max interval and back off parameters cannot be controlled by UMS Adapter as it is not a local retry.So they may not take effect even if we specify some values. The UMS Adapter sends exception back to the UMS and UMS will redeliver the message again. As we never know when and how UMS redelivers the message once we do not acknowledge the message in poller mode.


16464175 - NLS:JCA UMS ADAPTER MAIL GARBLE ON NON UTF SERVER ENCODING


Bug: 16464175
Added: 15-March 2013
Platform: All

UMS Adapter composes outbound email using SOA server encoding. If the payload contains some characters not supported by server encoding, it can cause data corruption.

Workaround:

Set JCA.UMS.MSG.CONTENT-TYPE = 'text/plain; character=utf8'.This is a normalized message property set on the outbound invoke for the UMS Adapter for BPEL. If in a Mediator, use "Assign values" to specify the value.

 

UMS Adapter has issues parsing non-ascii characters in inbound email if the SOA server is running in non-UTF8 encoding.

Workaround:

Start SOA server in UTF-8 encoding.

 


Application <durable_subscriber_name> Is Already a Subscriber For Queue


Bug: 12903224
Added: 15-September-2011
Platform: All

When using the Oracle JMS adapter with an AQ/JMS provider and when the configuration involves multiple dequeuer threads in a multi-server cluster, the Oracle JMS Adapter logs the following exception.

Caused by: oracle.jms.AQjmsException: ORA-24034: application <durable-subscriber-name> is already
a subscriber for queue <queue-name>

ORA-06512: at "SYS.DBMS_AQADM_SYS", line 6270
ORA-06512: at line 1
ORA-06512: at "SYS.DBMS_AQJMS", line 129
ORA-06512: at line 1

at
oracle.jms.AQjmsSession.addDurableSubscriber(AQjmsSession.java:3993)
at
oracle.jms.AQjmsSession.createSubInCompliantMode(AQjmsSession.java:2961)
at
oracle.jms.AQjmsSession.createDurableSubscriber(AQjmsSession.java:2624)
at
oracle.jms.AQjmsSession.createDurableSubscriber(AQjmsSession.java:2559)
at
oracle.jms.AQjmsSession.createDurableSubscriber(AQjmsSession.java:2404)
at

Workaround: The exception can be ignored, as on subsequent retry the subscription is created successfully.


Application <subscriber-name> is Already a Subscriber For Queue


Bug: 12375483
Added: 26-October-2011
Platform: All

The Oracle AQ adapter has been changed. It now creates a new subscriber as specifiedby the Consumer property when you are modeling an adapter service. Previously, the subscriber was not created by the adapter but it was expected that the subscriber is created before the adapter service was executed. Due to the change, if the subscriber for the Oracle AQ queue already exists, you see the following warning in the log files. Oracle Application Adapter scenarios that rely on Oracle AQ adapter for interaction with EBS queues will be impacted by this as well.

<Subscriber <subscriber-name> already exists in the database. If the existing subscriber does not represent the correct subscriber, please undeploy the business process, remove the subscriber, and redeploy.>

ORA-24034: application <subscriber-name> is already a subscriber for queue <queue-name>

ORA-06512: at "SYS.DBMS_AQADM_SYS", line 5682

ORA-06512: at "SYS.DBMS_AQADM", line 314

ORA-06512: at line 1....

Workaround: As the warning states, if the subscriber is the correct one, no action is required.


Updating Required JAR List When Accessing IBM MQ SERVER 7.0 via SOA JMS Adapter


Bug: 12349381
Added: 20-April-2011
Platform: All

You must use the following required jar list when you use JMSadapter with MQ version 7.x (See the "Configuring Oracle JMS Adapter with IBM WebSphere MQ JMS" section in the Oracle Adapters User's Guide for more information.)

Non-XA Data Sources

  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mq.jar
  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mq.jmqi.jar
  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mqjms.jar
  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/dhbcore.jar

XA Data Sources

  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mq.jar
  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mq.jmqi.jar
  • <YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mqjms.jar
  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/dhbcore.jar
  • /<YOUR-MQSERIES-INSTALL-LOCATION>/java/lib/com.ibm.mqetclient.jar


Managed Connection Factory Name Has to Be Modified in Case of Migrated AQ Adapter Composites


Bug: 12369977
Added: 20-April-2011
Platform: All

If you migrated AQ Adapter projects from Release 10.1.3.x to Release 11g, you need to manually update the
managed connection factory from

<non-managed-connection
managedConnectionFactoryClassName="
oracle.tip.adapter.aq.AQManagedConnectionFactory">

to

<non-managed-connection
managedConnectionFactoryClassName="
oracle.tip.adapter.aq.v2.jca.AQManagedConnectionFactory">

so you can use AQadapter in non-managed mode . Note that non-managed mode is not recommended and not supported in a production environment.To run the adapter in managed mode, you need to configure the connection factory JNDI in the WebLogic Server console.


AQJMS Dequeue Failed Error When Redeployed By Adding Durable Subscription


Bug: 10434614
Added: 28-April-2011
Platform: All

When an AQJMS Topic inbound process that is without a durable subscription  is already deployed to the server,   and  is redeployed, adding a durable subscription
to the Inbound Topic, the following error occurs when the message is dequeued. (Note this is fixed when you use Oracle Database 11.2.0.2.0 or any later version of the Oracle Database.)


Caused by: oracle.jms.AQjmsException: JMS-120: Dequeue failed
at oracle.jms.AQjmsError.throwEx(AQjmsError.java:315)
at oracle.jms.AQjmsConsumer.jdbcDequeue(AQjmsConsumer.java:1620)
at oracle.jms.AQjmsConsumer.receiveFromAQ(AQjmsConsumer.java:1035)
at oracle.jms.AQjmsConsumer.receiveFromAQ(AQjmsConsumer.java:960)
at oracle.jms.AQjmsConsumer.receiveFromAQ(AQjmsConsumer.java:938)
at oracle.jms.AQjmsConsumer.receive(AQjmsConsumer.java:790)


DBADAPTER W.R.T. Sybase DB Gives Problem With BIGDATETIME & BIGTIME Data Type


Date is Always Set To One Month Higher Than What is Set For the Message


Bug: 13450180
Added: 01-December-2011
Platform: All

Refer to Bug-13435861.

The date that is set through MQMD header is based on Gregorian Calendar. So, it is always set one month higher than that specified in the header.

For example: If I set, jca.mq.MQMD.PutDateTime to "12:24:2011" - the actual date that is set for the message is: January 24th, 2012. Similarly, if I set jca.mq.MQMD.PutDateTime to "10:24:2011" - the actual date that is set for the message is: November 24th, 2011.


Limitations of Dynamic Queue While Using MQ Adapter


Bug: 13464470
Added: 05-December-2011
Platform: All

MQ Adapter supports Dynamic Queue with these limitations:

  • Dynamic Queue is supported only for Inbound operations and not for Outbound
  • The property InboundThreadCount has to be set to 1 in the Inbound JCA file.

Configuring 11g to Access Queues/Topics Present in 10.1.3.X OC4J


Bug: 9544466
Added: 17-April-2012
Platform: All

The following steps provide guidance in accessing queues/topics present in 10.1.3x Oracle Containers for J2EE (OC4J)

Workaround:

  1. Copy the jar files under the /domains/<DOMAIN_NAME>/lib folder of the WebLogicServer:
    • $J2EE_HOME/lib/jms.jar
    • $J2EE_HOME/lib/jta.jar
    • $J2EE_HOME/oc4jclient.jar
    • $AS_HOME/opmn/lib/optic.jar
     
  2. Add the Connector factory in the weblogic-ra.xml file.

<connection-instance>
<jndi-name>eis/oc4jjms/Queue</jndi-name>
<connection-properties>
<properties>
<property>
<name>ConnectionFactoryLocation</name>
<value>jms/XAQueueConnectionFactory</value>
</property>
<property>
<name>FactoryProperties</name>
<value>java.naming.factory.initial=com.evermind.server.rmi.
RMIInitialContextFactory;>
java.naming.provider.url=<PROVIDER_URL>;
java.naming.security.principal=<USER>java.naming.security.credentials=&ltPASSWORD></value>
</property>
<property>
<name>AcknowledgeMode</name>
<value>AUTO_ACKNOWLEDGE</value>
</property>
<property>
<name>IsTopic</name>
<value>false</value>
</property>
<property>
<name>IsTransacted</name>
<value>false</value>
</property>
<property>
<name>Username</name>
<value><USER></value>;
</property>
<property>
<name>Password</name>
<value><PASSWORD></value>
</property>
</properties>
</connection-properties>
</connection-instance>

where <PROVIDER_URL> is opmn://localhost:6003 or ormi://localhost:12401 to use with a specific node, or opmn:ormi://localhost:6003:oc4j_soa to use
with the oc4j_soa instance
 


Back to top


Adapter for Oracle Applications (Oracle E-Business Suite Adapter)


Sequence of Elements in XML and XSD Files Should Match


Bug: 14400869
Updated: 01-August-2012
Platform: All

From this release (11.1.1.7.0), order of elements passed as input to an API using Adapter for Oracle Applications should strictly be the same as the order in which those elements appear under the sequence tag of the corresponding schema.

Please note that in previous releases, Adapter for Oracle Applications was non-restrictive and these elements could be passed in any order. However, the sequence of elements should now match the sequence in the corresponding schema.


Support for Business Events through Existing Partner Links


Bug: 13263084
Added: 27-October-2011
Platform: All

Important: This workaround is applicable only if Adapter for Oracle Applications is being used for listening to Event Groups. It is NOT needed if the Business Event is not part of any Event Group that is being listened to.

With the additional support for business event groups, perform the following steps to modify all existing partner links for outbound business events contained in an event group that is being listened to:

  1. Updating the JCA property "MessageSelectorRule" for the existing Business Event partner links. This can be achieved by performing either one of the following steps:
    • Rerunning the Adapter Configuration (Recommended): Rerunning or editing the configuration wizard for an existing partner link of an outbound business event (on Oracle JDeveloper 11.1.1.6.0) would update the artifacts with the appropriate MessageSelectorRule.
       
    • Manually Updating the JCA Properties: Alternatively, modify the JCA property "MessageSelectorRule" for the business event partner link, located in the file XXX_apps.jca, as stated below:

      Change existing property from:

      <property name="MessageSelectorRule" value="tab.user_data.event_name = '<EVENT_NAME>'"/>

      To the following:

      <property name="MessageSelectorRule" value="tab.user_data.event_name = '<EVENT_NAME>' AND tab.user_data.getvalueforparameter('GROUP') IS NULL"/>

      <EVENT_NAME> is to be replaced with the name of the Business Event.

  2. Delete the Existing Subscriber on WF_BPEL_Q of the corresponding Oracle Applications.
    1. Undeploy the SOA Composite application containing the Business Event partner link.
    2. Delete the subscriber using the following snippet:begin subscriber := sys.aq$_agent('<CONSUMER_NAME>', NULL, NULL);
      dbms_aqadm.remove_subscriber(queue_name => 'WF_BPEL_Q', subscriber => subscriber);
      end;

       

      <CONSUMER_NAME> is to be replaced with the name of the subscriber. It can be found in the XXX_apps.jca file as the value for the JCA property 'Consumer'.
  3. Redeploy the SOA Composite application.

"DataSecurityCheck" Requires "IRepOverloadSeq" to Be Passed for Overloaded Functions


Bug: 13263451
Added: 27-October-2011
Platform: All

Adapter for Oracle Applications uses the following two new JCA properties to handle access control for overloaded PL/SQL APIs. This feature works in conjunction with the grants created from the Integration Repository user interface, and it does not depend on any profile options.

  • DataSecurityCheck
  • IRepOverloadSeq

If a user wants the Function Security Authorization check to be performed, then the following property information needs to be added in the WSDL file (XX_apps.jca):
<property name="DataSecurityCheck" value="yes"/>

However, the other property IRepOverloadSeq is derived automatically by Adapter for Oracle Applications at the design time during creation of the partner link. Based on these two properties, the function security check would be performed for the username, which is passed as a header property.

Please note that this security support would work with all interfaces which are available in the Integration Repository. For other interfaces, you need to enable the function security through the profile option "EBS Adapter for BPEL, Function Security Enabled" (EBS_ADAPTER_FUNCTION_SEC_ENABLED).


Back to top


Application Adapters for WebLogic Server


Missing Libraries in setDomainEnv Causes Application Adapters AE to Fail

Bug: 13929048
Added: 08-June-2012
Platform: All

This problem affects every release the adapters are certified for: 11.1.1.4.0, 11.1.1.5.0, and 11.1.1.6.0.

The following two libraries are missing when starting the starting the design tool for the Application Adapters WLS (Application Explorer) after installation:

  1. $MWH/Oracle_SOA1/soa/modules/oracle.soa.adapter_11.1.1/jca-binding-api.jar
  2. $MWH/Oracle_SOA1/soa/modules/oracle.soa.adapter_11.1.1/adapter_xbeans.jar

In the Oracle Fusion Middleware Application Adapters Installation Guide for Oracle WebLogic Server, section 2.3.1 Starting Application Explorer, the guide states that users must run the setDomainEnv.sh to set the environment accordingly. The two libraries need to be added to the setDomainEnv scripts to avoid class loading exceptions.

The following exceptions may occur when using the Application Explorer to generate EIS adapter artifacts:

20120405-11:51:04 INFO com.ibi.bse.gui.BseFrame:read manifest for iwae.jar
20120405-12:12:30 ERROR
com.ibi.bse.ConfigWorker:java.lang.NoClassDefFoundError:
oracle/tip/adapter/api/exception/PCResourceException
at
com.ibi.afjca.cci.IWAFConnection.createInteraction(IWAFConnection.java:81)
at
com.iwaysoftware.iwae.common.JCATransport.initJCA(JCATransport.java:100)
at
com.iwaysoftware.iwae.common.JCATransport.<init>(JCATransport.java:87)
at
com.iwaysoftware.iwae.common.AdapterClient.<init>(AdapterClient.java:85)
at com.ibi.bse.ConfigWorker.run(ConfigWorker.java:41)
at java.lang.Thread.run(Thread.java:662)
Caused by: java.lang.ClassNotFoundException:
oracle.tip.adapter.api.exception.PCResourceException
at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
at java.lang.ClassLoader.loadClass(ClassLoader.java:306)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
at java.lang.ClassLoader.loadClass(ClassLoader.java:247)
... 6 more

Exception occurred during event dispatching:

java.lang.NoClassDefFoundError:
oracle/integration/platform/blocks/adapter/fw/metadata/AdapterConfigDocument$Factory
at
com.iwaysoftware.af.container.tools.wsdl.oracle.InboundWSDLGenerator.get11g_jca
at
com.iwaysoftware.af.container.tools.wsdl.oracle.IWayWSILBrowser.writeFiles
(IWayWSILBrowser.java:137)
at
com.iwaysoftware.af.container.tools.wsdl.oracle.IWayWSILBrowser.generateIWayWSDL
(IWayWSILBrowser.java:132)
at
com.iwaysoftware.af.container.OracleAEManager.generateWSDL
(OracleAEManager.java:83)
at
com.iwaysoftware.af.container.OracleAEManager.generateWSDL
(OracleAEManager.java:64)
at
com.iwaysoftware.af.container.ae.AEComponentMessage.buildEXPORTWSDL
(AEComponentMessage.java:185)
at
com.iwaysoftware.af.container.ae.AEComponentMessage.build
(AEComponentMessage.java:89)
at
com.iwaysoftware.af.container.ae.AEComponentMessage.<init>(AEComponentMessage.java:61)
at
com.iwaysoftware.af.container.AEManager.dispatch(AEManager.java:134)
at
com.iwaysoftware.af.container.IWAFContainer.dispatchAERequest
(IWAFContainer.java:876)
at
com.ibi.afjca.cci.IWAFInteraction.execIWAE(IWAFInteraction.java:286)
at
com.ibi.afjca.cci.IWAFInteraction.exec(IWAFInteraction.java:182)
at
com.ibi.afjca.cci.IWAFInteraction.execute(IWAFInteraction.java:107)
at
com.iwaysoftware.iwae.common.JCATransport.execute(JCATransport.java:166)
at
com.iwaysoftware.iwae.common.AdapterClient.generateWSDL(AdapterClient.java:437)
at
com.ibi.bse.gui.ExportWSDLDialog.saveWSDL(ExportWSDLDialog.java:590)
at
com.ibi.bse.gui.ExportWSDLDialog.okButtonActionPerformed
(ExportWSDLDialog.java:476)
at
com.ibi.bse.gui.ExportWSDLDialog.access$000(ExportWSDLDialog.java:36)
at
com.ibi.bse.gui.ExportWSDLDialog$2.actionPerformed(ExportWSDLDialog.java:417)
at
javax.swing.AbstractButton.fireActionPerformed(AbstractButton.java:1995)
at
javax.swing.AbstractButton$Handler.actionPerformed(AbstractButton.java:2318)
at
javax.swing.DefaultButtonModel.fireActionPerformed(DefaultButtonModel.java:387)
at
javax.swing.DefaultButtonModel.setPressed(DefaultButtonModel.java:242)
at
javax.swing.plaf.basic.BasicButtonListener.mouseReleased
(BasicButtonListener.java:236)
at java.awt.Component.processMouseEvent(Component.java:6267)
at javax.swing.JComponent.processMouseEvent(JComponent.java:3267)
at java.awt.Component.processEvent(Component.java:6032)
at java.awt.Container.processEvent(Container.java:2041)
at java.awt.Component.dispatchEventImpl(Component.java:4630)
at java.awt.Container.dispatchEventImpl(Container.java:2099)
at java.awt.Component.dispatchEvent(Component.java:4460)
at
java.awt.LightweightDispatcher.retargetMouseEvent(Container.java:4577)
at
java.awt.LightweightDispatcher.processMouseEvent(Container.java:4238)
at
java.awt.LightweightDispatcher.dispatchEvent(Container.java:4168)
at java.awt.Container.dispatchEventImpl(Container.java:2085)
at java.awt.Window.dispatchEventImpl(Window.java:2478)
at java.awt.Component.dispatchEvent(Component.java:4460)
at java.awt.EventQueue.dispatchEvent(EventQueue.java:599)
at
java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:269)
at
java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:184)
at
java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:178)
at java.awt.Dialog$1.run(Dialog.java:1046)
at java.awt.Dialog$3.run(Dialog.java:1098)
at java.security.AccessController.doPrivileged(Native Method)
at java.awt.Dialog.show(Dialog.java:1096)
at
com.ibi.bse.ComponentNodeData.actionPerformed(ComponentNodeData.java:406)
at
javax.swing.AbstractButton.fireActionPerformed(AbstractButton.java:1995)
at
javax.swing.AbstractButton$Handler.actionPerformed(AbstractButton.java:2318)
at
javax.swing.DefaultButtonModel.fireActionPerformed(DefaultButtonModel.java:387)
at
javax.swing.DefaultButtonModel.setPressed(DefaultButtonModel.java:242)
at javax.swing.AbstractButton.doClick(AbstractButton.java:357)
at
javax.swing.plaf.basic.BasicMenuItemUI.doClick(BasicMenuItemUI.java:1223)
at
javax.swing.plaf.basic.BasicMenuItemUI$Handler.mouseReleased
(BasicMenuItemUI.java:1264)
at java.awt.Component.processMouseEvent(Component.java:6267)
at javax.swing.JComponent.processMouseEvent(JComponent.java:3267)
at java.awt.Component.processEvent(Component.java:6032)
at java.awt.Container.processEvent(Container.java:2041)
at java.awt.Component.dispatchEventImpl(Component.java:4630)
at java.awt.Container.dispatchEventImpl(Container.java:2099)
at java.awt.Component.dispatchEvent(Component.java:4460)
at
java.awt.LightweightDispatcher.retargetMouseEvent(Container.java:4577)
at
java.awt.LightweightDispatcher.processMouseEvent(Container.java:4238)
at
java.awt.LightweightDispatcher.dispatchEvent(Container.java:4168)
at java.awt.Container.dispatchEventImpl(Container.java:2085)
at java.awt.Window.dispatchEventImpl(Window.java:2478)
at java.awt.Component.dispatchEvent(Component.java:4460)
at java.awt.EventQueue.dispatchEvent(EventQueue.java:599)
at
java.awt.EventDispatchThread.pumpOneEventForFilters(EventDispatchThread.java:269)
at
java.awt.EventDispatchThread.pumpEventsForFilter(EventDispatchThread.java:184)
at
java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:174)
at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:169)
at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:161)
at java.awt.EventDispatchThread.run(EventDispatchThread.java:122)
Caused by: java.lang.ClassNotFoundException:
oracle.integration.platform.blocks.adapter.fw.metadata.
AdapterConfigDocument$Factory
at java.net.URLClassLoader$1.run(URLClassLoader.java:202)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:190)
at java.lang.ClassLoader.loadClass(ClassLoader.java:307)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:301)
at java.lang.ClassLoader.loadClass(ClassLoader.java:248)
... 73 more

The exceptions are related to missing libraries in the classpath when starting the Application Explorer.

The following libraries must be added to the classpath via the setDomainEnv script prior starting the Application Explorer:

  • • jca-binding-api.jar
  • • adapter_xbeans.jar

Workaround: The installation guide will be updated in a future release.


Back to top


Application Server Legacy Adapters


No issue has been discovered at the time of this release. As additional issues and workarounds are discovered after the release, they will be listed here.

For known issues and workarounds for earlier releases, see Known Issues for Oracle SOA Products and AIA Foundation Pack.

Back to top


BPEL Process Manager


The bpel.config.oneWayDeliveryPolicy Property is Added When Creating a Synchronous BPEL Process


Bug: 16248658
Added: 1-March-2013
Platform: All

When you create a BPEL process service component in the Create BPEL Process dialog and select Synchronous BPEL Process from the Templatelist, the bpel.config.oneWayDeliveryPolicy property is added with a value of async.persist to the composite.xml file. This property should not be added for a synchronous BPEL process.

As a workaround, perform the following steps.

  1. In the Application Navigator in Oracle JDeveloper, click the composite_name (composite.xml) file.
  2. Click Source view.
  3. Remove the bpel.config.oneWayDeliveryPolicy property.
  4. <component name="BPELProcess1" version="1.1">
    <implementation.bpel src="BPELProcess1.bpel"/>
    <property name="bpel.config.oneWayDeliveryPolicy" type="xs:string"
    many="false">async.persist</property>
    </component>

    When complete, the syntax looks as follows.

    <component name="BPELProcess1" version="1.1">
    <implementation.bpel src="BPELProcess1.bpel"/>
    </component>

Accessing and Setting a Unit of Order Property for the JMS Adapter


Bug: 13726051
Added: 08-March-2012
Platform: All

To access and set a unit of order property for the JMS adapter, you must set two different properties in Oracle BPEL Designer.

For example, assume you have the following SOA composite application:

demoQueue -> Inbound JMS adapter > BPEL process service component -> Outbound JMS adapter -> demoQueue2

For demoQueue2 to successfully receive the value, set the following properties:

  • jca.jms.JMSProperty.JMS_BEA_UnitOfOrder
    1. Click the Property tab in the receive activity.
    2. Select jca.jms.JMSProperty.
    3. In the Value column for this property, double-click the row.
    4. Click the dots that are displayed to access the Adapter Property Value dialog.
    5. Enter the unit of order value, and click OK.
    6. In the receive activity, click OK.
    7. Click Source in Oracle BPEL Designer to access the BPEL process in source mode.
    8. Rename jca.jms.JMSProperty to jca.jms.JMSProperty.JMS_BEA_UnitOfOrder.
      The incoming unit of order value is available in this property.

       

    9. Return to Design mode in Oracle BPEL Designer.
    10. Assign this property to a variable.
      In the receive activity, the incoming unit of order value is obtained using property jca.jms.JMSProperty.JMS_BEA_UnitOfOrder.

       

  • jca.jms.WeblogicUnitOfOrder
    1. Click the Property tab in the invoke activity.
    2. Select jca.jms.WeblogicUnitOfOrder.
    3. In the Type column for this property, double-click the row.
    4. From the dropdown list, select output.
    5. In the Value column, click the dots that are displayed to access the Adapter Property Value dialog.
    6. Enter the unit of order value, and click OK.
    7. In the invoke activity, click OK.
    8. Assign the variable to this property.
      In the invoke activity, this property carries the unit of order value to the destination (outbound) queue (demoQueue2).

       


Oracle BPEL Process Manager 2.0 Limitations


Bug: 9735542
Added: 29-April-2011
Platform: All

The following BPEL 2.0 features are not implemented in Oracle BPEL Process Manager:

  • Termination handlers

  • Message exchanges

  • Fault handlers and termination handlers in an invoke activity. (This is the same as the BPEL 1.1 implementation of Oracle BPEL Process Manager.)

    Workaround: Include the invoke activity in a scope activity.

  • Atomic assign activities. If an error occurs in the middle of an assign activity, all the changes roll back. This is the same as the BPEL 1.1 implementation of Oracle BPEL Process Manager.


Checkpoint Activity Wrongly Appears in Oracle BPEL Designer Context Menu


Bug: 10229577
Added: 29-April-2011
Platform: All

If you right-click inside your BPEL process in Oracle BPEL Designer and select Insert Into > Activities, note that checkpoint is incorrectly displayed in the list of activities. This activity is not supported, and selecting it does not insert it into the BPEL process.

The checkpoint activity has been replaced by the dehydrate activity, which is also available for selection in the same context menu.


Second XSD Created for API-Created, PL/SQL Partner Links with Complex Parameters


Bug: 10246232
Added: 29-April-2011
Platform: All

If you create a PL/SQL partner link with complex parameters from the API, a new XSD file is generated. If you then open the partner link in Oracle JDeveloper and make no modifications other than completing the flow and clicking the Apply button in the Edit Partner Link dialog, a second XSD file is also created.

Workaround: You must use the second XSD file.


Audit Trail Details Display in Server Language Instead of Browser Language


Bug: 9298375
Added: 29-April-2011
Platform: All

Audit trail information for a BPEL instance displays in the language of your server instead of the language of your web browser.


Do Not Specify the wsa and bpel Property Prefixes with the bpelx:property


Bug: 9124369
Added: 29-April-2011
Platform: All

Tables H-1 and H-2 of Appendix H, "Normalized Message Properties" of the Oracle Fusion Middleware Developer's Guide for Oracle SOA Suite describe the Oracle BPEL Process Manager and Oracle Web Services Addressing properties.To use these properties correctly with the bpelx:property in activities such as a receive activity, ensure that you do not include the wsa. or bpel. prefixes. For example:

<receive name="receiveInput" partnerLink="bpelprocessb_client"
portType="client:BPELProcessB" operation="process" variable="inputVariable"
 createInstance="yes">
   <bpelx:property name="action" variable="v1"/>
   <bpelx:property name="messageId" variable="v2"/>
   <bpelx:property name="relatesTo" variable="v3"/>
   <bpelx:property name="replyToAddress" variable="v4"/>
   <bpelx:property name="replyToPortType" variable="v5"/>
   <bpelx:property name="replyToService" variable="v6"/>
</receive>


Selecting the Key for an Entity Variable in the Browse Entity Variable Dialog


Bug: 9066090
Added: 29-April-2011
Platform: All

When selecting the local part of an entity key by browsing entity variables, entity icons display in the expanded element tree. These icons are identified by a blue x inside parentheses, and indicate which elements include entity keys. Selecting these elements does not update the Key Local Part or Key Namespace URI fields in the Specify Key dialog. Instead, you must expand these elements to select the actual key to get the key's local part and namespace. For example:

  1. In the Bind Entity dialog, click the Add icon to invoke the Specify Key dialog.
  2. To the right of the Key Local Part field, click the Browse Entity Variable icon.
  3. Note that a blue x inside parentheses indicates which elements include entity keys. Do not select these elements.
  4. Expand these elements to display the entity keys available for selection.
  5. Select an entity key, and click OK.
    The Key Local Part and Key Namespace URI fields of the Specify Key dialog are populated with information about your selection.

MIME Type for Voice Notification in BPEL Voice Activity Requires Modification


Bug: 9023004
Added: 29-April-2011
Platform: All

In a BPEL voice activity, the MIME type for message content generated in the source code is in text/vxml format. However, the message sent from the composite to the voice XML driver must be in text/plain format. This is because the voice XML driver adds the voice XML wrapper to the message before sending it to the destination point.

If the message content sent to the driver is in text/vxml format for the MIME type, a message delivery failure occurs.

Workaround: Manually change the MIME type for the voice payload content to be text/plain in Source mode in Oracle JDeveloper.

<scope name="VoiceApproval">
. . .
. . .
   <copy>
  <from expression="string('text/plain')"/>
  <to variable="varNotificationReq" part="VoicePayload"
  query="/VoicePayload/ns1:Content/ns1:MimeType"/>
   </copy>
. . .
. . .
</scope>

Client Java API Is Not Backward Compatible


Bug: 8554297
Added: 29-April-2011
Platform: All

Custom applications coded with the Oracle BPEL Process Manager 10.1.3 Client Java API cannot currently be upgraded to the 11g Release 1 (11.1.1.3, 11.1.1.4, 11.1.1.5, 11.1.1.6, and 11.1.1.7) Client Java API. Programs coded to the Oracle BPEL Process Manager 10.1.3 API must be rewritten. In addition, the current API is not compatible with the versions of the 11g Release 1 11.1.1.1 and 11.1.1.2 API, although the differences are minimal.

The version of the API in 11g Release 1 11.1.1.3 has been finalized. It has been fully supported since that release and is backward compatible from that release onwards.

Visit the following URL for details:

http://www.oracle.com/technology/products/soa/bpel/collateral/documentation.html

Back to top


Business Activity Monitoring


List Prompts Do Not Work at Runtime in IE9 (Report Level and Surface Prompts)


Bug: 16389540
Added: 18-March-2013
Platform: Internet Explorer 9

Expected Behavior: List prompts allow user to select a value, then apply to report.

Actual Behavior: Report launches fine, but list prompts either appearing as a pop-up at the report level, or in the report as a surface prompt, and do not highlight new value upon selection.

Pre-existing, reproduced with PS5. Not an issue with IE 8 or IE 10. Typed-in prompts work fine. Does not matter if default value is selected or not.

Oracle BAM 11.1.x Does Not Support Oracle Database 12


Bug: 16364606
Added: 11-March-2013
Platform: All

Due to an issue with Oracle BAM 11.1.x installed against the schema on Oracle Database 12, this version of Oracle BAM does not support Oracle Database 12.

WebSphere: WSWS7054E WSDL 'file could not be generated' Exception Can Be Ignored


Bug: 16340442
Added: 14-March-2013
Platform: IBM WebSphere Application Server environment

When you start the Deployment Manager (Dmgr process) in an IBM WebSphere Application Server environment, you can receive the following error message in the WAS_HOME/profiles/DmgrXX/logs/dmgr/SystemOut.log file. This message can be safely ignored.

"WSWS7054E: The Web Services Description Language (WSDL) file could not be generated for the oracle.bam.webservices.dataobjectdefinition.DataObjectDefinition Web service implementation class because of the following error: java.lang.Exception."

Oracle BAM Monitor Express Setup File


Bug: 14065933
Added: 18-March-2013
Platform: All

Currently the setup.sh/bat script will import both the dataobjects and the dashboard, and sometimes if you import/reimport the same thing multiple times it causes a constraint violation error.

Recommendation:

1) Rename current setup.sh /bat scripts to Demosetup.sh/bat * this will import both the dataobjects and the dashboard.

2) Create a new script setup.sh which has only the Dashboard import commands in it.

3) If you use this as a demo with no integration, you can use DemoSetup.

4) For Production usage, you should first deploy the Processes and then import the dashboard using the newly created setup.sh.

When Using Computers With Multiple Network Cards, Change Localhost to Either Server Name or IP Address


Bug: 14065906
Added: 03-January-2013
Platform: All

If you are using computers with multiple network cards, and you do not change the setting for either the localhost or the port in the BAMIcommandConfig.xml file, then ICommand does not respond. Moreover, if you then press Ctrl-X, you then receive a message saying that the import was successful--even though the import failed.

To prevent this, in the BAMIcommandConfig.xml file, change the setting for localhost to either the server name or the IP address as a best practice. Also, verify that the port number is the same as the one used in installation.

BAM PS6 + IE11 Not Supported


Bug: 18801612
Added: 05-June-2014
Platform: All

BAM PS6 (11.1.1.7) is not supported with the Internet Explorer 11 browser. An error message stating that it is not a supported browser version appears if you attempt to open the BAM URL using IE 11.

 

 

 

Back to top


Business Process Analysis Suite


No issue has been discovered at the time of this release. As additional issues and workarounds are discovered after the release, they will be listed here.

For known issues and workarounds for earlier releases, see Known Issues for Oracle SOA Products and AIA Foundation Pack.

Back to top


Business Process Management


Accessing Task With Task Details Throws Warning Dialog in Wrksp


Bug: 18640395
Added: 07-Jun-2014
Platform: All

When you click a Task to view its details, a warning dialog is displayed.

Workaround: Close the warning dialog to view the Task Details.


Admin Home: Task Menu Not Displayed Correctly


Bug: 18640360
Added: 07-Jun-2014
Platform: All

Some of the menu items are not displayed correctly and some of them are not aligned properly in the Workspace in IE11. This is a known issue.

Workaround: There is no workaround.


Non-ASCII Characters Cannot be Displayed In Web Form Print PDF File


Bug: 16434235
Added: 05-March-2013
Platform: All

When attempting to map a dropdown web form control to a string value in the human task data association, Business Process Composer generates an error when validating the BPM Project. The error reports that the data assocaition for the human task assigment is invalid.

Workaround: There is no workaround.


Generate simulation from runtime data not supported in JDeveloper


Bug: 16341922
Updated 11-March-2013
Platform: All

Generating a simulation from runtime data is not supported in the JDeveloper environment.

 

Workaround: There is no workaround.


Non-ASCII Characters Cannot be Displayed In Web Form Print PDF File


Bug: 16287282
Added: 05-February-2013
Platform: All

Web form controls do not support non-ASCII characters as label names. If a web form control contains non-ASCII character, they do not appear correctly in a generated PDF.


Task Payload Name Should Contain Only ASCII Characters When Using Web Forms


Bug: 16284318
Added: 05-February-2013
Platform: All

If the task payload element contains non-ASCII characters, Business Process Composer will attempt to generate a web form which also contains non-ASCII characters. However, these are not supported. You must edit the name part of the fields and change as follows:

  • The field names should only contain letters (a-z, A-Z), numbers (0-9) and the underscore ("_") character.
  • The field names should not begin with a number.

Must Access Business Process Composer Using SOA Server URL When Using Web Forms


Bug: 16272773
Added: 05-February-2013
Platform: All

Web Forms are not rendered inside Business Process Composer or Process Workspace unless those applications are accessed using fully qualified domain name. You must use the full URL of the SOA server.


Web Forms Editor Does Not Support Non-ASCII Characters in Graphic Filenames

Bug: 16220642
Updated: 01-February-2013
Platform: All

When adding a graphic file to a web form, you must ensure that the filename contains only ASCII characters.


Web Forms Functionality Not Supported on in Web Center Modeling Space


Bug: 16102020
Updated: 01-February-2013
Platform: All

Web forms functionality is not supported in Web Center's Modeling Space.


No Support for Deleting Case Management Stakeholders


Bug: 15951289
Updated 31-December-2012
Platform: All

The case management API does not support support deleting stakeholders.

 

Workaround: There is no workaround.


Setting the Case Management Permission Tag Using the setPermissionTag Function is Not Supported in Alfresco CMIS


Bug: 15951238
Updated 31-December-2012
Platform: All

Setting the case management permission tag using the function setPermissionTag() is not supported if the selected document content storage system is Alfresco CMIS.

 

Workaround: Set the permission tag when you upload the document using the uploadDocument() function.


Transaction Error When Calling a Subprocess with a Receive Task


Bug: 15940185
Updated: 01-February-2013
Platform: All

When calling a subprocess from a receive task, the following exception may be thrown:

java.sql.SQLException: ORA-12899: value too large for column "<schema>"."WORK_ITEM"."CUSTOM_ID"

 

This error occurs because the size of the transaction is too large. If you encounter this error, increase the size of the custom_id column to 512.


Web Forms Functionality Not Supported on Some Database Platforms


Bug: 15920624
Updated: 12-December-2012
Platform: All

Web Forms functionality is not supported on the following databases:

  • Microsoft SQL databases
  • IBM DB2 databases

Web Forms Functionality Not Supported on Websphere Application Server (WAS)


Bug: 14315026
Updated: 12-December-2012
Platform: Websphere Application Server

Web Forms functionality is not supported on IBM Websphere Application Server (WAS).


Case Management Functionality Not Supported on Websphere Application Server (WAS)

Bug: 14315026
Updated: 12-December-2012
Platform: Websphere Application Server

Case Management Functionality is not supported on Websphere Application Server (WAS).


Max Limit on Redeployment in BPM Workspace


Bug: 14840457
Added: 2-November-2012
Platform: All

If the composite being redeployed has more than 100 existing BPM process instances, only 100 of them are migrated during redeployment and the component is left suspended. This limit is used to minimize the deployment time. To migrate the remaining instances, use Process Management APIs.".


Expiration Date Field In Case Input Payload Is Not Supported


Bug: 14768646
Updated 29-October-2012
Platform: All

The expiration date field in the case input payload is not supported. The attribute exists for possible use in future versions.

 

Workaround: There is no workaround.


Restrictions in Case Management Design and Runtime


Bug: 14767324
Updated 11-March-2013
Platform: All

The following restrictions apply to case management design and runtime:

  • There is no schema validation for the case data payload. You can assign the payload value any value, even invalid random Strings, regardless of the type defined in the case file.
  • The Human Task case activity does not support task outcome and execData in their output arguments.

Workaround: There is no workaround.


Auto Update not available in BPM Workspace


Bug: 14671243
Added: 12-October-2012
Platform: All

If you have an assigned Initiator Task, when you invoke instances by clicking on the Process Link in the BPM Workspace Administration Tab, the Initiated Task count does not update automatically or after refreshing the webpage.

Workaround: After clicking the Initiated Task link, the count is updated accordingly.


Repeatable Option Not Support for Non-Conditional Automatic Case Activities


Bug: 14745503
Updated 29-October-2012
Platform: All

Non-conditional automatic case activities cannot be repeatable. The UI enables you to select this option but the BPM Runtime does not take it into account.

Workaround: There is no workaround.


Audit Trail not Maintained for Previous Process Version


Bug: 13466122
Updated: 06-December-2011
Platform: All

Consider this scenario:

  • One version of a process is deployed.
  • An instance of that version is completed.
  • Later, a second version of that process is deployed.

In this scenario, the first process instance shows the audit trail for the second version of the process rather than the first.


No Support for Deleting Case Management Comments


Bug: 14570446
Updated 29-October-2012
Platform: All

The case management API does not support support deleting comments.

 

Workaround: There is no workaround.


Logging in to BPM Workspace Gets Out of Memory Error


Bug: 13387403
Updated: 05-December-2011
Platform: All

At times, when you log into BPM Workspace and select a task for viewing the details, the taskform keeps loading endlessly. If you check the server console, then you note that it throws OutOfMemoryError.

The workaround is:
  1. Make sure that JRockit R28.2.0 is used for Oracle Weblogic Server.
  2. In the file setSOADomainEn.sh, increase the Perm Gen space to -XX:MaxPermSize=768m.

Characters in Comments Added to Flow-Altered or Migrated Instances Become "?" When Using Native Server Encoding


Bug: 13369711
Updated: 05-December-2011
Platform: All

If the server encoding is native encoding, and if you are doing either "Alter Flow" for a process instance or "Migrate" for a pending process instance, then strings added in comments that are not recognized by server encoding become question marks (?).


Audit Is Available When Turned Off


Bug: 13369711
Updated: 05-December-2011
Platform: All

Audit records are shown for all activities even when auditing is turned off in EM at both the BPMN engine and common soa-infra levels. Auditing is also shown even when auditing is turned off, and there is no measurement artifact in the process, and measurement is enabled.


Cannot Add An Attachment Having the Same Name as Another

Bug: 13404430
Updated: 05-December-2011
Platform: All

Adding an attachment having the same name as another without deleting the previous one can cause unexpected results. If an attachment with the same name that you need to attach exists, then:

  • Delete the attachment.
  • Save the task.
  • Add the new attachment.

Manual Refresh Needed After Resuming a Pending Component


Bug: 13425069
Updated: 05-December-2011
Platform: All

After you resume a component from the pending components panel in the Process Tracking tab page, you must manual refresh BPM Workspace manually to show that the resumed component has been removed from the Pending Components panel. The Refresh icon is located on top of the Process Tracking tab page.


Boundary Event Seen as an X Icon in Audit Trail

Bug: 13263092
Updated: 10-November-2011
Platform: All

During migration of an instance, an activity removed from the new process displays an X icon in the process instance audit trail. This can be confusing because a boundary event displays the same icon even though it is not removed. Note that this is a user interface issue only and causes no loss of functionality.


Issues with Input/Output Message in Audit Trail During Alter Flow


Bug: 13361923
Updated: 10-November-2011
Platform: All

Issue 1: Viewing comments after alterflow

Comments applied during grab/migrate are not visible unless you select the tree view in the audit trail list. Within that tree view, expand the activity where comment is applied. The comment column in audit trail is then populated with the comment.

Issue 2: Input output message shows the content of a previously selected record

If, in the audit trail of an instance, the input/output message is for a previously selected record, then do the following:

  1. Refresh the process tracking table.
  2. Select any other record.
  3. Select once again the record whose message you want to view. This time the input/output content is correct.

Use Only a Valid Combination When Changing the Variables ownerType and owner


Bug: 13355820
Updated: 10-November-2011
Platform: All

You can change the values of the process instance variables ownerType and owner during runtime while performing a grab/migrate action in the BPM Workspace Process Tracking page. When you do this, you must use the valid combination of ownerType and owner -- for example:

Valid combination: owner=jstein and ownerType=USER
Invalid combination: owner=jstein and ownerType=ROLE

Cannot Simultaneously Move a Token from both a User Task Within a Subprocess and from the Subprocess Itself


Bug: 13336279
Updated: 05-December-2011
Platform: All

Moving a token simultaneously from both a user task within a subprocess and the subprocess itself may result in an error. You should move the token from either the subprocess, or from the user task, but not from both.


You Can Neither Alter the Flow of an Errored Instance Nor Resume It


Bug: 13254650
Updated: 05-December-2011
Platform: All

You cannot alter the flow of an errored instance, nor can you resume it.


Custom Page Name and Layout Changes While Navigating Between Pages in Edit Mode

Bug: 13253389
Updated: 07-November-2011
Platform: All

You can edit only one custom page at a time. You can create a custom page, edit it, add panels to it, and then save it. Then, after you have saved the previous page, you can select another page to edit or create a new page. However, if you do not save the previous page that you were editing, then the settings of the older page-- for example, the page name--are copied on to the new page.


Unable to View Complete Error Message in Grab Panel

Bug: 13098620
Updated: 07-November-2011
Platform: All

When you attempt to alter the flow of a running instance, a Grab Confirmation dialog box tells you the result. If an exception is thrown, the error message may not be completely visible in the Grab results panel. To view the complete message, hover over the cell and read the tool tip that appears.


The weblogic.xml file Is Incorrect when Task Form Application Is Migrated from an Older Version in Oracle JDeveloper


Bug: 13051963
Updated: 07-November-2011
Platform: All

When task-details projects are either moved or deployed on the server based on Releases PS3 to PS5 onwards, then you must change the weblogic-application.xml file in the task details project to include the following:

<library-ref>
      <library-name>oracle.soa.worklist.webapp</library-name>
      <specification-version>11.1.1</specification-version>
  </library-ref>
reference 

No Auto-Migration While Redeploying Incompatible Process with "Force Deploy" Property Set to TRUE


Bug: 12918436
Updated: 05-December-2011
Platform: All

If a process is incompatible with a new implementation, and if, during design time, the process is implemented with both "keep running instances" and "force deploy" set to true, then, on deployment, instances of that process go into pending migration and the component is suspended. The "force deploy" property is set in the composite.xml file in Oracle JDeveloper. The property is oracle.bpm.bpmn.force.deploy.

In this scenario, a pending migration instance may not migrate successfully. For example, if the token is running inside a parallel gateway pair and the gateway pair is removed from the modified process, then the token will not have a valid target to move to and migrate will not succeed.


BPM Workspace Page Takes Only One Entry in BPM_USERAPPLICATIONDATA; Limits Test-to-Production (T2P) page migration


Bug: 12876023
Updated: 07-November-2011
Platform: All

In BPM Workspace, no matter how many workspace pages are created, all the info is recorded in one row of BPM_USERAPPLICATIONDATA with:

APPLICATIONDATANAME = WAM
APPLICATIONDATATYPE = PART_PROPS

This limits the Test-to-Production (T2P) page migration to either all or nothing.

It also limits the T2P page migration to replace all the existing pages.

For example:

Test env has following defined in page: "Widget1" , "Widget2" , "Widget3"
Prod env has following defined in page: "WidgetA" , "WidgetB"

In this example, T2P cannot just export "Widget1" and append to production which results to "WidgetA", "WidgetB", "Widget1". Only option after the export and import is that the production environment will have "Widget1", "Widget2", "Widget" defined, and "WidgetA", "WidgetB" removed.


Unable to Obtain Task ID to use Update Task Activity


Bug: 12813850
Updated 29-October-2012
Platform: All

The Update Task Activity enables you to update a task based on the task ID but currently there is no predefined dialog or XPATH function to obtain the taskId. Using the TaskExecutionData is not an option because the value is available after the task is completed.

Workaround: Assign the taskId variable to an output parameter in the Update Task data association. The BPM Engine evaluates and assigns a value to this variable after initiating the task, so the variable is assigned the task ID before completing the task.


Pending Migration Instance Details Show Activity ID Instead of Activity Name in Open Activities


Bug: 12681352
Updated: 12-September-2011
Platform: All

After installing the OPatch and performing the BPM-specific post-installation steps, but before starting the soa_server1, delete the WLS tmp directory for Oracle BPM Workspace.

 

Case Management Does Not Support SQL and DB2 Databases


Bug: 15920616
Updated 18-June-2013
Platform: All

Currently the case management feature does not support the SQL and DB2 databases.

Workaround: There is no workaround.


Mandatory Patches for Case Management


Bug: 14808494
Updated 18-June-2013
Platform: All

Install the following mandatory patches for the case management feature:

  • TopLink fix for Bug 14642695 - unable to use and reference alias inside subquery
  • TopLink fix for Bug 14306399 - stress bpm casem:get npe at casedecisionservicehandler.createdecisionpointfactli
  • Sun JDK fix for bug 14698557 stress bpm casem:get classcastexception at oracle.bpm.casemgmt.event. Multithreaded deserialization of Calendar leads to ClassCastException as at http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=7017458

 

Workaround: There is no workaround


Back to top


Business Rules


Creating a Dictionary Link Between Two Projects Does Not Work at Run Time


Bug: 13413704
Added: 13-February-2012
Platform: All

At design time, Oracle Business Rules allows you to link to a dictionary in the same application or in another application using the Links navigation tab in Rules Designer.

However, this feature is not supported at run time (by using Oracle SOA Composer).


Parent Taskflow of a Rules Dictionary Taskflow Needs to Be Notified When Handles a Save Operation


Bug: 11787788
Added: 13-February-2012
Platform: All

When a Rules Dictionary Taskflow performs a Save operation, the parent tasklow (of the Rules Dictionary taskflow) needs to be notified.

In the code, you need to add parent-action for both save with and without validation as the following:

	<parent-action id="updateMetadataDoneAct">
	   <parent-outcome>handleUpdateMetadataDone</parent-outcome>
	   </parent-action>
	   <parent-action id="updateMetadataNoValidateDoneAct">
		<parent-outcome>handleUpdateMetadataNoValidateDone>
		</parent-outcome>
	</parent-action>
	

You can also add handlers as the following in adfc-config.xml to achieve this:

	<method-call id="afterSave">
	   <method>#{MyBean.afterSaveDone}</method>
	   <outcome id="__22">
		 <fixed-outcome>return</fixed-outcome>
	   </outcome>
	 </method-call>
	 <method-call id="afterSaveNoValidate">
	   <method>#{MyBean.afterSaveNoValidate}</method>
	   <outcome id="__23">
		 <fixed-outcome>return</fixed-outcome>
	   </outcome>
	 </method-call>
	 <control-flow-rule id="__6">
	   <from-activity-id id="__7">*</from-activity-id>
	   <control-flow-case id="__10">
	<from-outcome id="__8">handleUpdateMetadataDone>
	</from-outcome>
	<to-activity-id id="__9">afterSave</to-activity-id>
	   </control-flow-case>
	   <control-flow-case id="__13">
		 <from-outcome id="__11">handleUpdateMetadataNoValidateDone<
		 <from-outcome>
		 <to-activity-id id="__12">afterSaveNoValidate</to-activity-id>
	   </control-flow-case>
	 </control-flow-rule>
	


Error When Trying to Locate a Schema in Projects Migrated from PS1 to PS3


Bug: 12868234
Added: 06-December-2011
Platform: All

When you open a Oracle SOA Suite (11.1.1.2.0) project in Oracle SOA Suite (11.1.1.3.0) and click Reload All Facts, an error is displayed when you try to locate the schema in the Project directory and not in the XSD directory (where it is available). This is because the schemaLocation has not been updated to reflect the correct location.

This error also occurs when you open Oracle SOA Suite (11.1.1.2.0) project and create a testsuite.

Workaround: You need to delete the facts pertaining to Oracle SOA Suite (11.1.1.2.0) projects and regenerate them again from their source schemas.


notserializableexception Is Thrown on Opening Dictionary in Oracle SOA Composer


Bug: 12415315
Added: 06-December-2011
Platform: All

In a clustered environment, when you open a dictionary in SOA Composer by clicking a ruleset name in the Select a dictionary to open dialog box, a NotSerializableException is thrown.

There is no loss of functionality because of the exception; the dictionary opens, however the selected ruleset is not displayed by default.

Workaround: Select the dictionary by clicking the row and click the Open button.


Difficult to Add a Variable Inside a Nested Case in Oracle SOA Composer


Bug: 12604367
Added: 11-September-2011
Platform: All

You cannot add a variable definition directly inside a nested block.

Workaround: The following are the workarounds for this issue:

  • Workaround 1: Add a variable definition first, and then surround it with parenthesis using the Surround condition(s) with parentheses (()+) icon at the top. However, this works only in the case of (...) but not with not(...).
  • Workaround 2: Add a nested (...) or not(...) block, add a variable definition inside the nested block, and then move the variable definition up by clicking the Move up (up arrow icon) icon at the top.

Refresh Needed to View Updated Rules Test Variable Names in SOA Composer Decision Tables


Bug: 12894814
Added: 11-September-2011
Platform: All

In Oracle SOA Composer, when using Oracle Business Rules test variables or inline aliases in rule and decision table conditions and actions, after you make changes to the variable names, the updated names do not appear in the decision table.

Workaround: In the decision table, do one of the following to force the decision table to redraw and display the updated variable names:

  • Click the Switch Rows to Columns icon on the decision table toolbar.
  • Click the Show Conflicts icon on the decision table toolbar.

Adding or deleting a rule, condition, or action also causes the decision table to redraw and hence display the updated variable names.


The Copy to Project Checkbox is Not Selected by Default While Using URL Location for a Schema


Bug: 11675232
Updated: 03-February-2011
Platform: All

While creating business rules, in the Import schema file dialog box, when you point to the URL location for a schema, the Copy to Project checkbox is not selected by default.

Workaround: In the Import Schema File" dialog box, manually select the Copy to Project checkbox. This is the default selection for schemas imported from local file system (file:/......).

RL.contains is Incorrectly Displayed in the List of Operators in Oracle SOA Composer


Bug: 11062184
Updated: 03-February-2011
Platform: All

In Oracle SOA Composer, when writing the "if" part of a if-then rule, RL.contains is incorrectly displayed as one of the relational operators in the operator list.


Too Many Open Files Message When Deploying a Business Rules Project


Bug: 9102568, 7418258
Added: 16-November-2009
Platform: IBM AIX on POWER Systems (64-Bit), HP-UX Itanium, HP-UX PA-RISC (64-Bit), Oracle Solaris

If you see the error "Too Many Open Files" while deploying an Oracle Business Rules Decision Component, you might need to increase the file descriptor limit.

The "Too many open files" issue is due to JDK6 bugs. These bugs could occur at runtime or at compile time, depending on the number of JAR files used and a few other variations around the use of file descriptors by the JDK/JRE.

Workaround: Increase the limit of file descriptors. Per-process, per-user file descriptor limit can be preconfigured by default to the value 1024. If you increase this limit to 4096 or larger, the new value should resolve this issue.

These are the potentially relevant JDK6 bugs:


And a blog entry describes this issue at:

http://coldfused.blogspot.com/2007/02/mystery-of-too-many-open-files.html

The command to increase per-process, per-user file descriptor differs across Operating Systems and shells. It is typically done with ulimit on UNIX platforms so consult the man pages for ulimit for your Operating System and shell.

For example:

$ ulimit -n 4096

Back to top


Business-to-Business Integration (B2B)


Resubmit Based on DUNS ID


Bug: 16166290
Added: 28-January-2013
Platform: All

In case of ebMS, resubmission of messages based on DUNS ID is not supported in Oracle SOA Suite 11.1.1.7.


XEngine Issue in HP-UX (64 bit)


Bug: 16057361
Added: 28-January-2013
Platform: HP-UX 64-bit

In the 64-bit HP-UX platform, to avoid server crash when working with Oracle B2B, you need to set the LD_PRELOAD environment variable in the setSOADomainEnv.sh file by adding the following:

export LD_PRELOAD=$XERoot/bin/libxejni.so


Configuring a Proxy Server for Multiple SFTP Channels in Oracle B2B


Bug: 15943006
Added: 21-December-2012
Platform: All

Oracle B2B enables you to configure proxy settings per SFTP channel. If you check the Use proxy check box in the Transport Protocol Parameters tab, and do not specify the proxy host or port, then Oracle B2B usea the Weblogic server proxy settings. But if you specify the proxy host and port values, then the proxy host, port, user, and password values are be used to send out messages for that SFTP channel.

This feature enables you to configure different proxies for different SFTP channels within the same Weblogic instance.


MLLP 2.0 and HLLP Are Not Supported in Oracle B2B


Bug: 15921028
Added: 21-December-2012
Platform: All

MLLP 2.0 and HLLP, which are currently available as options in the Transport protocols list in the Oracle B2B console, are actually not supported in Oracle B2B.

This is a known issue.


Message_Status Column Missing from the B2B_INSTANCEMESSAGE View


Bug: 15904985
Added: 21-December-2012
Platform: All

The Message_Status column is not available as a part of B2B_INSTANCEMESSAGE view.


Windows Server 2008 R2 Requirement for XEngine


Bug: 14851040
Added: 21-December-2012
Platform: Windows Server 2008 R2

On Windows Server 2008 R2, Oracle B2B requires Microsoft Visual C++ 2008 redistribute package to handle EDI/HL7 (XEngine) related messages. So, you need to install the corresponding Microsoft Visual C++ 2008 redistribute package (x86 or x64) if it is not already existing in the environment.

The package can be downloaded from:
http://www.microsoft.com/en-us/download/details.aspx?id=2092

You need to ensure that msvcm90.dll is present in the value of the PATH variable.


Remote Monitor User Is Incorrectly Provided More Privileges


Bug: 14799540
Added: 21-December-2012
Platform: All

Although the Remote Monitor user is not allowed to save any changes, some fields and buttons are still enabled that allows the Remote Monitor user to click and enter text. For example, under Agreements, this user is able to validate and export agreements using the Validate and Export buttons.

This is a known issue.


Resubmit Fails with the filepath Option when Using File Containing msgID Followed by Spaces


Bug: 14799092
Added: 21-December-2012
Platform: All

Resubmitting messages used with the DfilePath option through commandline is unable to handle file containing MsgIds followed by trailing spaces. It fails with StringIndexOutOfBoundException.


Unable to Delete Channels Using Command-Line Utilities


Bug: 14279743
Added: 21-December-2012
Platform: All

You cannot delete channels, including internal delivery channels, by using any command-line utility.


Negative Acknowledgement Not Sent for Duplicate RosettaNet Messages


Bug: 14246938
Added: 21-December-2012
Platform: All

While processing duplicate RosettaNet messages, Oracle B2B does not send any negative acknowledgment.


Negative MDN Sent for Duplicated Message IDs


Bug: 14165151
Added: 21-December-2012
Platform: All

In the case of the AS2 exchange protocol, a negative MDN is sent back if a message is duplicated. From the code, no additional flag or setting is necessary to implement this feature.


Change in Sync ACK URL in the Wire Message Table


Bug: 14156439
Added: 21-December-2012
Platform: All

When a response is received in sync mode, the URL displayed in Wire Message table is similar to:

http://<host>:8001//b2b/httpReceiver

However this is incorrect for sync case. So, it has been rectified to display the URL of the response to be the same as the URL of the request.


Backend Application Is Notified About ebMs Message Validation Failures


Bug: 14150417
Added: 21-December-2012
Platform: All

In case an ebMs inbound message fails in validation, such as an inbound ebMS message failing due to identifier present in the message is not present in the agreement then:

  • The error message sent to the backend contains the details of the identifier type and identifier value.
  • A negative ACK with the error information is sent back to the trading partner.

Web Service Validation Based on payload Namespace


Bug: 14146010
Added: 21-December-2012
Platform: All

In case of Generic Webservices in Oracle B2B/Oracle Healthcare, the XSD validation is entirely based on the namespace present in the payload. So for validating an incoming message, the namespace should match with the WSDL, otherwise validation will be disabled.


The Large Payload Directory Field Should Not Be Left Blank


Bug: 14000978
Added: 21-December-2012
Platform: All

The Large Payload Directory field in the Configuration tab under the Administration link contains a default value that you can configure. However, you need to ensure that the field should not be left blank.


Server Restart Required for Oracle B2B Server Property to Take Effect


Bug: 13644558
Added: 13-February-2012
Platform: All

You need to restart the mamaged server (SOA Server) for Oracle B2B Server property oracle.tip.b2b.edi.binaryDocType to take effect.


B2B Users Should Not Invoke Healthcare Postinstall Script


Bug: 13550031
Added: 19-January-2012
Platform: All
Release: 11.1.1.6.1

B2B users who do not want to use Oracle SOA Suite for healthcare integration should not invoke the postinstall script that is specifically for users of Oracle SOA Suite for healthcare integration 11.1.1.6.1.


Oracle B2B Does Not Initialize when Server Starts in WAS Platform


Bug: 13258868
Added: 21-December-2012
Platform: All

Oracle B2B does not get initialized when the SOA server starts in the Websphere Application Server platform. This is done to eradicate the occurence of the following exception:

 

[9/13/11 0:54:35:661 PDT] 00000022 FfdcProvider  W
com.ibm.ws.ffdc.impl.FfdcProvider logIncident FFDC1003I: FFDC Incident
emitted on
/scratch/aime1/work/was3613/profiles/Custom01/logs/ffdc/soa_server1_218e218e_1
1.09.13_00.54.35.6604783173679133928417.txt
com.ibm.ws.naming.util.Helpers.logJndiContextException 1411
[9/13/11 0:54:35:662 PDT] 00000022 engine        E   Unable to bind
SOAPProvider to B2B_SOAP_PROVIDER_IMPL in Directory
[9/13/11 0:54:35:664 PDT] 00000022 engine        E  
com.ibm.ws.naming.util.CannotBindObjectException: Object is not of any type
which can be bound.
        at
com.ibm.ws.naming.util.Helpers.processJavaObjectForBinding(Helpers.java:652)
        at
com.ibm.ws.naming.jndicos.CNContextImpl.doBind(CNContextImpl.java:1916)
        at
com.ibm.ws.naming.jndicos.CNContextImpl.bind(CNContextImpl.java:683)
        at com.ibm.ws.naming.util.WsnInitCtx.bind(WsnInitCtx.java:205)
        at com.ibm.ws.naming.util.WsnInitCtx.bind(WsnInitCtx.java:217)
        at javax.naming.InitialContext.bind(InitialContext.java:444)
        at
oracle.tip.b2b.transport.basic.soap.SOAPProvider.init(SOAPProvider.java:94)
        at
oracle.j2ee.ws.server.provider.ProviderProcessor.initProvider(ProviderProcessor. 

 


Change in Names of Configuration Parameters


Bug: 12918690
Added: 06-December-2011
Platform: All

The configuration parameter names FA Handled By B2B and Functional Knowledgement by B2B have been changed to FA Handled Automatically and Functional Knowledgement Automatically. There is no change in functionality.


Oracle B2B Commandline Utility b2bvalidate Uses -dmdsreference as an Option


Bug: 12794661
Added: 06-December-2011
Platform: All

The option -Dargs is not supported in the b2bvalidate commandline utility. Use -dmdsreference instead.


Setting the b2b.syncappdelivery Property


Bug: 10429627
Added: 06-December-2011
Platform: All

When using callouts, to enable delivery of messages to the back-end message queue, set this property to true. By default, this property is enabled (set to >code>true). To disable the message delivery, set this property to false.


RosettaNet Headers Cannot Be Seen if Message Security Is Enabled


Bug: 12976817
Added: 06-December-2011
Platform: All

You cannot see RosettaNet headers (Preamble, Delivery and Service) for any inbound or outbound message if the message security (encryption) is enabled.

The packed message does not show any content in the Wire Message report because data is encrypted, and the payload link in the Wire Message report shows only the payload part and not the complete RosettaNet MIME message.


Supported Document Types Are Not Refreshed After Switching Trading Partners


Bug: 13350675
Added: 06-December-2011
Platform: All

If you add a document type under the Supported Document Types section in the Oracle B2B console, the content of Supported Document Types does not get refreshed when you switch from one Trading Partner (Host) to another Trading Partner (Remote).


LDAP Cleanup Is Required to Avoid Monitor User to Have Administrator Privileges


Bug: 13342300
Added: 06-December-2011
Platform: All

In some cases, as the following, in Oracle B2B, a Host Monitor user can perform an Administrator's task, such as creating new users and change own user roles:

  1. In a clean environment (LDAP, B2B install), log on to the Oracle B2B console as weblogic.
  2. Import any given schema.
  3. In the Partner tab, under Host, create a new user called myUser1 with the Administrator role.
  4. Save and login as myUser1 to ensure that the user has administrator privilegesn and log out.
  5. Log on as weblogic.
  6. Under Administration -> Purge, purge the Design Metadata.
  7. Re-create the user, myUser1 (under Host), but this time as a Monitor user.
  8. Save and log out.
  9. Log on as myUser1. Notice that the user has Administrator privileges (instead of monitor).
This is an undesired situation.

 

Workaround: On the LDAP server, ensure that you clean up the entry for the user (myUser1 that was created as Administrator) before creating a new user with the same name (but with different role.)


Host Administrator Cannot Create User After Adding Document Type Support


Bug: 13255134
Added: 06-December-2011
Platform: All

When a Host Admin user has a new document added to Supported Document Types, the Host Admin cannot add a new user.


Signing AS2 Sync-ACK Needs a Trading Partner Delivery Channel


Bug: 12999861
Added: 06-December-2011
Platform: All

If a signed AS2 sync-ACK is needed, the Trading Partner delivery channel should contain security details:

  • ACK signed flag should be checked in the channel
  • Signing certificate should be selected in the channel

If the above configuration is not done, a runtime error is reported.


Adding the b2bmonitor Role to a User Does Not Work for DB-Based Policy Store


Bug: 13026205
Added: 06-December-2011
Platform: All

In Oracle B2B, a user needs to be granted the B2BMonitor role to be able to obtain the privileges of B2BMonitor. You can grant the B2BMonitor role to a user from the Oracle B2B console, and not from the Oracle Fusion Middleware Enterprise Management console.


No Support for Propagating JMS MessageID to JMS CorrelationID


Bug: 13430043
Added: 06-December-2011
Platform: All

In the case of default JMS queues, Oracle B2B does not support the propagation of JMS MessageID from an outbound MLLP request to the JMS CorrelationID field in corresponding Inbound Acknowledgment.


Oracle B2B Does Not Generate MessageID for JMS Internal Listening Outbound Channel


Bug: 13439585
Added: 06-December-2011
Platform: All

When back-end applications communicate with Oracle B2B by using JMS queues, if the Use JMS ID flag is not set, then Oracle B2B should not use JMS Message ID for outbound message and should use B2B-generated Message ID for outbound message. If the flag is set, then Oracle B2B uses the JMS ID as B2BMessageID. However, currently, Oracle B2B does not generate Message IDs when the flag is not set.


Oracle B2B Console Displays Empty Error Dialog for Remote Admin User


Bug: 13377596
Added: 06-December-2011
Platform: All

In Oracle B2B console, for a Remote Admin user, when you click any available agreements under Partners/Agreements, and then click any of the Document Types pertaining to the agreement under the Agreement section, an empty error dialog is displayed.


Oracle B2B User Interface Does Not Have Provision for the 'readonly' Parameter


Bug: 9669427
Added: 06-December-2011
Platform: All

While creating a delivery or listening channel for a Trading Partner in the Oracle B2B UI, currently, you do not have any provision for specifying a ReadOnly parameter. This feature was present in Oracle SOA Suite 10g.

Workaround: You can achieve the same functionality by setting the File Archival directory parameter available in the Oracle B2B UI.


Clicking Different Tabs under Oracle B2B Displays Spurious Warnings in the Managed Server Console Log


Bug: 12648474
Added: 17-November-2011
Platform: All

After you upgrade from Oracle SOA Suite 11.1.1.2.0 to Oracle SOA Suite 11.1.1.6.0, clicking the different tabs in the Oracle B2B homepage results in several instances of the following warning in the Managed Server Console log:

Could not find selected item matching value "" ...


WLST Command reassociatesecuritystore Works Differently in Oracle B2B


Bug: 12396118
Added: 01-May-2011
Platform: All

The WLST command reassociatesecuritystore that is used to reassociate a policy store with an LDAP Server works differently than what is described in Reassociation of Credentials and Policies in "Oracle Fusion Middleware Enterprise Deployment Guide for Oracle SOA Suite".

For the command to work properly in Oracle B2B:

Instead of using cn="jpsroot" or cn=<cn>, use jpsroot="dc=us,dc=oracle,dc=com".

For example:

reassociateSecurityStore(domain="soainfra",admin="cn=orcladmin",
password="welcome1",ldapurl="ldap://adc2100613.us.oracle.com:3060",
servertype="OID", jpsroot="dc=us,dc=oracle,dc=com");


Incorrect Date Format in Command-line Purge Destroys Repository


Bug: 11701338
Added: 13-February-2011
Platform: All

A user can destroy the entire Oracle B2B repository (both the design-time and run-time data) if an incorrect fromdate value is entered while executing a command-line purge.

Incorrect:

ant -f ant-b2b-util.xml b2bpurge -Dfromdate="sddsdsd"

Correct:

ant -f ant-b2b-util.xml b2bpurge -Dfromdate="01/02/2009 12:00 AM"


Oracle B2B Uses JMS as the Default Queue


Bug: 8935332
Added: 13-February-2012
Platform: All

Oracle B2B uses JMS (B2B_IN_QUEUE and B2B_OUT_QUEUE) as the default queue. However, AQ (IP_IN_QUEUE and IP_OUT_QUEUE) is fully supported. If you want to use the default AQ queues, you can change the default queue setting in the Configuration tab in the Oracle B2B console.


Back to top


Event Processing


The Sum And Avg Functions on a Float Type Produce Incorrect Values

Bug: 14134093
Updated: 19-December-2012
Platform: All

When you use the CQL Sum or Avg function on a floating point number, you might get an incorrect result. This is a known problem with Java floating point arithmetic.

Workaround: If you need a result that is accurate up to a specific number of decimal points, use a decimal number instead of a floating point number.

Java Event Properties with CQL Keywords for Property Names Cause Errors

Bug: 12426096
Updated: 19-December-2012

The Java event property timestamp is a native data type that clashes with the CQL keyword, timestamp. When you use the timestamp event property and CQL keyword in the same application, Oracle Event Processing generates an error similar to this:

The application context could not be started. Could not initialize component "MarketEventprocessor": Event property [cTimeStamp] defined in query [ExampleQuery] must exist in event type [MarketEvent]. Consider using the expression 'cTimeStamp AS ...' in the query.

Workaround: None.

No Support for Multi-Byte Character Sets

Bug:17277407
Updated: 6-August-2013
Platform: All

Oracle Event Processing does not support Chinese and other character sets that need more than a single byte to be represented.

Workaround: None.

CQL Processor Cannot be Named lifecycleprocessor

Bug: 13848408
Updated: 15-Mar-2012
Platform: All

In an Oracle Event Processing application, do not name a processor lifecycleprocessor. If you name a processor lifecycleprocessor, you get the The application context "lifecycleapp" could not be started error when you deploy the application.

Workaround: None.

Failed To Load The Bus Tracking Figure Of The OEP Spatial Sample On Oracle Enterprise Linux 6.1 Running FireFox 3.6.17

Bug: 16088260
Updated: 2-February-2013
Platform: All

The Flash file used in the spatial application for PS6 is incompatible with Oracle Enterprise Linux 6.1 running FireFox 3.6.17.

Workaround: Run earlier or later versions of the software.

Domain Creation Using Configuration Wizard Fails on AIX

Bug: 13397310
Updated: 10-April-2013 br /> Platform: AIX

Domain creation with Configuration Wizard fails on AIX with IBM VM. There is no workaround.

Browser Hangs When an Application Subscribes to too Many Channels

Bug: 10416835
Updated: 21-May-2013
Platform: All

When you start an event trace on a channel in Visualizer and then subscribe to the trace channel in the ViewStream panel, the browser becomes unresponsive after about 10 minutes.

Web Application Installed with Spatial Sample is Not Removed When Sample is Uninstalled


Bug: 16052630
Updated: 29-January-2013
Platform: All

Installing the spatial sample also installs a web application that can be accessed at http://localhost:9002/bus/main.html. When you uninstall the sample, the web application remains. As a result, if you reinstall the sample, it may fail to redeploy.

Workaround: Separately remove the included web app when you uninstall the sample.


Installation is Not Supported on 32-bit Linux


Bug: 14836138
Updated: 30-January-2013
Platform: Linux

Due a limitation with JDK 7 on 32-bit Linux, installation is not supported on that platform.

Workaround: None.


Operating System Locale Must Be Set to English for Installation


Bug: 14627283
Updated: 30-January-2013
Platform: All

Before installing on a non-English platform, the OS locale must be set to English in order for the installer's user interface to display properly. After installation, an English locale is not required for product user interfaces to display properly.

Workaround: Set operating system locale to English for installing.


A Cache Declared as a Foreign Stage Cannot Be Used as a Channel's Listener


Bug: 16026765
Updated: 30-January-2013
Platform: All

A cache declared in one OEP application can't be used as the listener for a channel in another OEP application. Declaring the cache as a foreign stage in the second application will generate an error when starting up the second application.

Workaround: None.


JMS Adapter Loses Event If Exception Occurs in JMS Message Conversion


Bug: 12919448
Updated: 25-August-2011
Platform: All

The JMS adapter is designed to respect transacted JMS session behavior (recovering input when the transaction is rolled back) when you set the adapter's session-transacted element to true. Due to an issue in this release, when an event provokes an exception during conversion between JMS message and OEP event, that event is not recovered.

Workaround: No workaround.


Limited Browser Support for Uploading an Application JAR Over HTTPS


Bug: 12793925
Updated: 10-November-2011
Platform: All

Uploading an application JAR file over HTTPS via Oracle CEP Visualizer is not supported in the following browsers: Mozilla Firefox, Google Chrome 15, Apple Safari 5.1. For uploading over HTTPS, use Internet Explorer.


bundler.sh Fails Due to Incorrect Classpath


Bug: 13522497
Updated: 21-December-2011
Platform: All

The bundler.sh file included with this release has an incorrect classpath that causes the tool to fail. The script's classpath contains a reference to org.osgi.eclipse.equinox_3.5.1.jar, but should refer to org.osgi.eclipse.equinox_3.6.2.jar.

Ordinarily, with bundler.sh you can create an OSGi bundle wrapper, a bundle JAR you can deploy to an OSGi container.

Workaround: You can open the bundler.sh file and edit the classpath to include the correct JAR file. You can also use the Oracle CEP IDE for Eclipse or the bnd tool to create the bundle JAR file.


Event Processing Language (EPL) is Deprecated


Bug: 13350828
Updated: 10-November-2011
Platform: All

Oracle Event Processing Language (EPL) is deprecated and may be removed in a future release. Use Oracle Continuous Query Language (CQL) for new development.


Creating a Two-Way SSL Connection with WebLogic Server


Bug: 12771217
Updated: 10-November-2011
Platform: All

While making a two-way SSL connection with WebLogic Server, use the following steps.

  1. Copy the DemoTrust.jks and DemoIdentity.jks from weblogic server to the CEP domain and run the following command. The command is displayed here on multiple lines for readability, but it should be run on a single line.
    keytool -importkeystore
        -srckeystore <domain_home>\defaultserver\ssl\DefaultIdentity.jks
        -srcstorepass DemoIdentityKeyStorePassPhrase
        -srcalias demoidentity
        -srckeypass DemoIdentityPassPhrase
        -destkeystore <domain_home>\defaultserver\ssl\server.jks
        -deststorepass welcome
        -destalias demoidentity
        -destkeypass welcome
  2. Using the java command, pass the following properties to the startwlevs.sh script:
    "%JAVA_HOME%\bin\java" %JVM_ARGS% %DGC% %DEBUG%
        -Dwlevs.home="%USER_INSTALL_DIR%"
        -Dbea.home="%BEA_HOME%"
        -Djavax.net.ssl.trustStore=ssl\DemoTrust.jks
        -Djavax.net.ssl.trustStoreType=JKS
        -Djavax.net.ssl.trustStorePassword=DemoTrustKeyStorePassPhrase
        -Djavax.net.ssl.keyStore=ssl\server.jks
        -Djavax.net.ssl.keyStoreType=JKS
        -Djavax.net.ssl.keyStorePassword=welcome
        -Dweblogic.StdoutDebugEnabled=true
        -Dssl.debug=true
        -jar "%USER_INSTALL_DIR%\bin\wlevs.jar" %ARGS% 
  3. In the WebLogic Server console, enable two-way SSL by setting “Two Way Client Cert Behavior” to “Client Certs Requested and Enforced”.
  4. In the WebLogic Server console, if your client certificate doesn’t specify the hostname, set “Hostname Verification” to “None”.
  5. If you are using your certificate instead of the certificate from “demoIdentity.jks”, you must import the certificate into the server’s trust store using the JDK keytool.

Back to top


Enterprise Repository


Enterprise Repository Access Becomes Unavailable

Bug: 13618692
Added: February, 2013
Platform: All

You cannot submit an OSB project and launch Enterprise Repository Access during the same Eclipse/OER session.

Workaround: Restart Eclipse and use one feature of the other. For example:

  1. Launch Eclipse and connect to OER.
  2. Submit an OSB project.
  3. Attempt to launch Enterprise Repository Access.  FAIL
  4. Clean OER connection, restart Eclipse.
  5. Launch Enterprise Repository Access.
  6. Attempt to submit OSB project.  FAIL
  7. Clean OER connection, restart Eclipse.
  8. Submit an OSB project.

Error Message Improved

Bug: 9973199
Added: February, 2013
Platform: All

In the SOA Suite Harvester, if you enter incorrect credentials or server URL during setup, it will fail but the error message provides details.


Password Error Details Added

Bug: 9157972
Added: February, 2013
Platform: All

In Harvester, if the encrypted password is incorrect, the error message provides details.


Harvesting a Composite Fails

Bug: 13684775
Added: February, 2013
Platform: All

When harvesting a composite that has an OWSM policy applied to it, the Harvester fails.

This issue is resolved in this release.


Harvester Error Message

Bug: 13332138
Added: February, 2013
Platform: All

In an Ant implementation for the Harvester, if the Harvester Solution Pack is not imported, the error message does not provide details.

This issue is resolved in this release.


Email Sent from Asset Editor Displays Subject Line Improperly

Bug: 13681198
Added: February, 2013
Platform: All

Email Sent from the Asset Editor displays the email's subject line improperly if the asset name contains multi-national characters (Japanese).

This issue is resolved in this release.


Update to Harvester Supports BPMN Projects

Bug: 13830060
Added: February, 2013
Platform: All

The Harvester has been updated to support both the older namespace as well as the newer one, when harvesting BPMN projects. The old namespace was http://www.omg.org/bpmn20. The new namespace is http://www.omg.org/spec/BPMN/20100524/MODEL.


Deleting Elements from the Harvester Configuration File Causes Harvester to Fail

Bug: 13723989
Added: February, 2013
Platform: All

The Harvester contains a configuration file called HarvesterSettings.xml. If you delete the <query> or <repository> elements, the Harvester fails.

This issue is resolved in this release.


Interface Tool Tip Updated

Bug: 9889339
Added: February, 2013
Platform: All

In the Asset Basic Search, there are two menus: Categorization Types and Asset Types. The tool tip was updated to include the full name. This only displays in browsers that support HTML 5.


Export to Excel Displays Some Characters Improperly

Bug: 12863887
Added: February, 2013
Platform: All

The Export to Excel function uses the Asset Type name for the Excel sheet name. If the Asset Type name has multi-national characters (Japanese), the sheet name in Excel does not display the name properly.

This issue is resolved in this release.


Encryption Tool Corrupts Already Encrypted Passwords

Bug: 9721356
Added: February, 2013
Platform: All

If the workflow.xml configuration file has a password that is encrypted and is then run through the encryption tool, the password will become corrupted and unusable.

This issue is resolved in this release.


Harvester Looks for src Directory

Bug: 13841298
Added: February, 2013
Platform: All

When designing a BPMN project in Jdeveloper, you can reference Java classes. If you store Java classes outside of the "src" directory, the Harvester continues looking in the "src" directory.

This issue is resolved in this release.


Asset Search

Bug: 13927795
Added: February, 2013
Platform: All

On the OER home page, the left-side frame shows the Asset Search utility.The frame is resizable to expand with the frame and increase viewing area.


Harvester Does Not Fail When a Project or Partition is Unspecified

Bug: 13971038
Added: February, 2013
Platform: All

When harvesting composites you can specify a project name. If no name is provided, the harvester will harvest every project within the provided partition. If no partition is provided , the Harvester harvests everything.


Import/Export Utility Fixed to Provide Multi-National Character Support

Bug: 12862348
Added: February, 2013
Platform: All

In the Import/Export utility, when exporting a large number of assets with multi-national characters (Japanese), the log at the end of the export would fail to display the exported entities properly. The characters were garbled and unreadable.


Retired Assets Hidden

Bug: 14081050
Added: February, 2013
Platform: All

The Asset Detail view no longer displays Retired related assets. They are now hidden.


Assets in Progress Disabled Causes Error

Bug: 14154070
Added: February, 2013
Platform: All

When Assets In Progress is disabled, the asset search in the Asset Editor displays an error and does not return any results.

This issue is resolved in this release.


Rex API Improved

Bug: 14563300
Added: February, 2013
Platform: All

In the REX API, you can read all of the metadata of a given asset based on a tab name.


Asset Navigator

Bug: 14666742
Added: February, 2013
Platform: All

In Linux, when you launch the Asset Navigator for an asset that had a policy assertion that passed, the expected green line connecting the asset to the policy did not appear and an exception was thrown.

This issue is resolved in this release.


Composite Harvest Ignoring WSDL Imports

Bug: 12921813
Added: November, 2011
Platform: All

In SOA Suite Harvester, if a Composite XML includes WSDL imports, the Harvester ignores them. Now, the SOA Suite Harvester does not ignore the imports; the WSDL imports are introspected.

This issue is resolved in this release.


Filter by Service Categorization in Orrxu.xml Broken For OER

Bug: 12921823
Added: November, 2011
Platform: All

When configuring the Exchange Utility, you can specify certain categorizations. Validations have been added to check that the categorization string entered exists in Oracle Enterprise Repository.If it does not, an error message displays. If no categorization value is entered, a warning displays but processing continues.


Browse Tree Does Not Work

Bug: 12921837
Added: November, 2011
Platform: All

When you log into Oracle Enterprise Repository and do not select Enable Automatic Login, the Browse Tree does not display on the homepage.

This issue is resolved in this release.


Composites with Multiple Revisions Causes Conflict

Bug: 12921856
Added: November, 2011
Platform: All

In the SOA Suite Harvester, a composite can have multiple versions, such as Composite (1.0) and Composite (2.0). When a composite is harvested, it creates child assets that are a part of that composite. When both of the composites are harvested, the child assets are related to both versions (Composite 1.0 and 2.0) in Oracle Enterprise Repository. This is incorrect, as each composite should have its own children.

This issue is resolved in this release.


Unicode Characters in Mailto Link

Bug: 12937813
Added: November, 2011
Platform: All

In the Asset Editor, an envelope icon displays next to each user’s name. If you select the icon, an email opens, addressed to the user. If the email address contains Unicode characters, the address does not display properly.

This issue is resolved in this release.


Unicode Characters in Asset Type Schema


Bug: 12937820
Added: November, 2011
Platform: All

If Unicode characters exist in an element of an asset type, the characters do not display properly when viewed from the Export Schema screen.

The Export Schema option is available from the Type Manager, which is accessible from the Asset Editor. To access this option, choose File->Export Schema.

This issue is resolved in this release.


Unicode Characters in Categorization Configuration


Bug: 12937834
Added: November, 2011
Platform: Al

When entering Unicode characters in the tModel Key v3 field, an error occurs. The tModel Key v3 field is available from the Configure Categorizations screen in the Asset Editor. To access this screen, choose Actions->Configure Categorizations from the Asset Editor.

This issue is resolved in this release.


Asset Editor 4000 Char Limitation on Asset Description


Bug: 11850937
Added: December, 2011
Platform: All

In the Asset Editor, the Asset description attribute has a known limitation of 4,000 characters. When you harvest a WSDL with a description of more than 4,000 characters and then return to the Asset Editor to edit anything on this asset, clicking Save leads to re-writing the description of only up to 4,000 characters.


Cannot Use the UDDI Registry to Search for Services


Bug: 12368396
Added: December, 2011
Platform: All

The "Working With UDDI Registries” section in the OER 11g Help documentation incorrectly states that one can "use the UDDI Registry to search for services and then import them into Oracle Enterprise Repository. (The UDDI Registry option must be enabled.)This procedure is performed in the Asset Editor."

This feature is not supported.


OER .Net Plugin Installer Fails Referencing vsusercontrollhost.dll


Bug: 12567801
Added: December, 2011
Platform: All

The installation of the plug in failed, generating an error indicating a problem with VSUserControlHost.dll and a dependency on msvcr71.dll could not be resolved.

Workaround: Copy the newest version of msvcr71.dll from the location where the VS.NET is installed. Paste into the plug in directory and then rerun the installer. Select the repair function on the plug-in installer and reboot.


Plug-In Login Module Class Name Causes Classnotfoundexception


Bug: 12641147
Added: December, 2011
Platform: All

In step 5 of section 3.1.2 “Enable LDAP Integration System Properties” in the Oracle Fusion Middleware Configuration Guide for Oracle Enterprise Repository 11g Release 1 (11.1.1.6.0), change the Plugin login module class name from com.oer.enterprise.authentication.server.loginmodule.LDAPLogin to com.flashline.enterprise.authentication.server.loginmodule.LDAPLogin.


NLS: Asset File Upload - File Name Garbled


Bug: 12802884
Added: December, 2011
Platform: All

When you create an asset using Submit an Asset on the OER Console, you can upload a file with the asset while creating the asset. The name of file gets garbled when uploaded.


Error "The Source XML Mapping Specified Cannot Be Empty or Null


Bug: 12814405
Added: December, 2011
Platform: All

OER provides an XML mapping file (customDataMigration.xml) to migrate the custom data and some of the 10g asset types to 11g types (that is, Endpoint: Web Service/ Interface: Web Service to Endpoint/Interface). If you run the tool without any custom data migration, just 10g to 11g types migration, you will get the following error:

4969 INFO Successfully previewed 31 asset(s) out of 31

4969 ERROR Artifact harvest failed due to: The source XML Mapping specified cannot be empty or null.

java.lang.RuntimeException: The source XML Mapping specified cannot be empty or null. at com.oracle.oer.sync.plugin.reader.migrate.hdp.CustomDataStep.call(CustomDataSt ep.java:167)

Workaround: Comment out the "<customDataFields>" elements in the customDataMigration.xml.


Usage Data Is Not Migrated


Bug: 12822142
Added: December, 2011
Platform: All

Usage data such as Asset Consumption /Asset download/Consumed projects are NOT migrated.


Unable to Access the Operational Data URL to Link Back to Enterprise Manager


Bug: 9694669
Added: 01-May-2011
Platform: All

In the EM Integration configuration file, there is an XML element called URLBase. If you have an EM Console server that is set up for HTTPS with the element set to HTTP, then the link that is created in Oracle Enterprise Repository, which is used to point back to the EM Console, does not work. This is a known issue in 11g Release 1 (11.1.1.5.0).


DTMConfiguration Exception When Running Oracle Enterprise Repository with WebLogic Server 10.3.3 on IBM JDK for AIX


Bug: 9729380
Added: 24-May-2010
Platform: IBM AIX on POWER Systems (64-Bit)

When you run Oracle Enterprise Repository server with WebLogic Server version 10.3.3 on IBM JDK for AIX, then the following exception message is displayed:

org.apache.xml.dtm.DTMConfigurationException: No default implementation found
at org.apache.xml.dtm.DTMManager.newInstance(DTMManager.java:177)
at org.apache.xpath.XPathContext.<init>(XPathContext.java:125)
at org.apache.xalan.transformer.TransformerImpl.<init>(TransformerImpl.java:398)
at org.apache.xalan.templates.StylesheetRoot.newTransformer(StylesheetRoot.java:197)

Workaround: Before starting the Oracle Enterprise Repository server set the JAVA_OPTIONS parameter in setOERDomainEnv.sh as follows:

JAVA_OPTIONS="-Dorg.apache.xml.dtm.DTMManager=
org.apache.xml.dtm.ref.DTMManagerDefault
$JAVA_OPTIONS"
export JAVA_OPTIONS


Back to top


Human Workflow


Cannot Select the Tasktype from Tasktype Browser when Creating Vacation Rule


Bug: 14090553
Added: 25-July-2012
Platform: All

When creating Task Namespace,Task Definition Name,and Task Display URL conditions for a vacation rule in Oracle BPM Worklist, you do not have the option to select the tasktype from tasktype browser.

 

Deployment of Heavy and Complex Human Workflow Task Forms Fail on BPM Server


Bug: 13465029
Added: 09-January-2012
Platform: All 64-bit computers

On 64-bit systems, Oracle Business Process Management (BPM) Server shows java.lang.OutOfMemoryError: PermGen space exception while deploying Heavy and Complex Human Workflow task forms. This is due to insufficient Max Permgen Space provided to SOA Server for deploying complex workflow task forms.

Workaround: Edit DOMAIN_HOME/bin/setSOADomainEnv.(sh|cmd) file to increase the MaxPermSize by 256m.

On Microsoft Windows x64 (64-Bit), change the PORT_MEM_ARGS value in setSOADomainEnv.cmd file as follows:

set PORT_MEM_ARGS=%PORT_MEM_ARGS% -XX:PermSize=256m -XX:MaxPermSize=768m

On HP-UX PA-RISC (64-Bit), IBM AIX on POWER Systems (64-Bit), Oracle Solaris, and Linux x86-64, change the PORT_MEM_ARGS value in setSOADomainEnv.sh file as follows:

PORT_MEM_ARGS="${PORT_MEM_ARGS} -XX:PermSize=256m -XX:MaxPermSize=768m"


Cannot Use Oracle JDeveloper to Search for Approle in an Environment with only SSL Enabled


Bug: 13372192
Added: 09-January-2012
Platform: All

In an environment in which SSL ports are enabled and non-SSL ports are disabled, you cannot search for approle by using Oracle JDeveloper. Attempting to do so prompts a SOAP exception.


Two Parameters Not Relevant in Federated Worklist Environment


Bug: 13111612
Added: 06-December-2011
Platform: All

The following parameters are irrelevant in a federated Worklist environment:

  • showCreateTODOTaskAction
  • showAssignmentFilter

If these are set in a federated environment, they will not be reflected in the Action list in Worklist.


FYI Stage Disappears from Task Sequence When It Follows a Parallel Participant


Bug: 13366984
Added: 10-November-2011
Platform: All

The FYI stage disappears from the task sequence when it follows a parallel participant and the parallel participant completes the task. This is apparent in the following scenario:

  1. Create an application with Human Workflow.
  2. Add stage1 with parallel users jcooper, jlondon, fkafka.
  3. After stage1, add stage2, sequential with stage1. Stage2 has only FYI user jstein.
  4. Deploy the application and invoke an instance.
  5. Login to BPM Worklist as jcooper, and click history in Task Details. The stage 2 user jstein is visible.
  6. After all of the parallel users in stage 1 complete the task, log in to BPM Worklist as stage2 FYI user jstein, and click the history.

Note that stage2 has disappeared from the task sequence.


Completed Task Disappears When There Are Both an FYI USER and a Non-FYI User After the Task


Bug: 13366978
Added: 10-November-2011
Platform: All

Consider the following scenario in which a task disappears from the list of completed tasks:

  1. Create an application with Oracle Human Workflow.
  2. Add stage1 with single user jcooper
  3. After stage1, add parallel stages: stage2, stage3, and stage4. These are sequential with stage1.
    • In stage2, there is only an FYI user, jstein
    • In stage3, there is only an FYI user, wfaulk
    • In stage4, there is a single user, cdickens
  4. Deploy the app and invoke an instance.
  5. In BPM Worklist, log in as stage1 user jcooper and approve the task.
  6. Change the "State" to "Completed" to list the completed task.

Note that the task has disappeared.


A Parallel Stage with Only an FYI User Shows as Sequential Stages in History


Bug: 13366516
Added: 10-November-2011
Platform: All

When parallel stages have only one or more FYI users in each of stages, the stages show up as sequential in the history graph.


Reusable Worklist Task Flow Regions Need Entry in the web.xml File


Bug: 13354559
Added: 08-November-2011
Platform: All

Newer versions of Oracle JDeveloper configure applications to use ojsp.next mode. To override that, include the following servlet and servlet-mapping entries in the web.xml file.

<servlet>
        <servlet-name>jsp</servlet-name>
        <servlet-class>oracle.jsp.runtimev2.JspServlet</servlet-class>
         <init-param>
             <param-name>static_text_in_chars</param-name>
             <param-value>true</param-value>
        </init-param>
         <init-param>
             <param-name>generic_servlet_mode</param-name>
             <param-value>single</param-value>
         </init-param>
         <init-param>
             <param-name>page_provider</param-name>

 <param-value>oracle.adf.library.webapp.ADFJspResourceProvider</param-value>
         </init-param>
         <load-on-startup>2</load-on-startup>
     </servlet>
     <servlet-mapping>
         <servlet-name>jspvservlet-namev
         <url-pattern>*.jspx</url-pattern>
     </servlet-mapping>
     <servlet-mapping>
         <servlet-name>jsp</servlet-name>
         <url-pattern>*.jsff</url-pattern>
     </servlet-mapping>


Incorrect Error Message When a Non-Existent UCM User Tries to Add UCM Attachments


Bug: 12917983
Added: 12-September-2011
Platform: All

If you create a task with a single assignee that does not exist in UCM server, and you then enable UCM in the task editor without adding any extra properties, then, when you log into Worklist and try to add a UCM attachment, the following incorrect error message appears:

Error occurred while inserting a task attachment record.

Error occurred while inserting attachment liu10.txt for task a6ad8d42-f588-4e10-aacd-8f4542d2ca6d.

Check the underlying exception and the database connection information. If the error persists, contact Oracle Support Services.


Exceptions While Loading Configurations From Workflow Identity Configuration


Bug: 16399934
Updated 18-June-2013
Platform: All

The exceptions that appear while loading configurations from the workflow identity configuration, do not provoke any problems. They are just warnings.

Workaround There is no workaround.

Back to top


Mediator


Publishing a Domain Value Map in SOA Composer Does Not Work Without Saving the Domain Value Map


Bug: 16306670
Added: 12-February-2013
Platform: All

If you edit a Domain Value Map (DVM) in a SOA Composer sandbox and publish it, the changes are not written to runtime data unless the DVM is saved before the publish operation.

The following steps reproduce the error:

1. Deploy a DVM file.

2. Log in to SOA Composer and open the DVM file.

3. Create a sandbox. Modify and publish the DVM.

The workaround is to first save, and then publish the DVM file.


Cannot Skip a Pending Mediator Resequencer Message in Enterprise Manager When Audit Is Off


Bug: 14253699
Added: 11-October-2012
Platform: All

If auditing is off (Audit = OFF) for the Mediator service engine, then instance tracking information is not collected, and you cannot monitor resequenced messages from the Mediator Service Component Home Page. As the resequenced messages are not monitored, you cannot skip a sequence ID in a timed-out group using the Mediator Resequencing Group Dialog.

Workaround: You can make use of Facade APIs to unlock a faulted or timed-out group. For example, the MediatorFacadeService.executeEngineMethod method can be used with the unlockGroup parameter.


populateLookupXRefRow Throws an Exception in Custom Mode


Bug: 13358019
Updated: 09-November-2011
Platform: All

When you use the populateLookupXRefRow function in LINK mode and both the reference value and the value to be added exist in the database, the unique constraint exception is caught and the value to be added is returned by the function. In custom mode, the value to be added is not returned and an exception is thrown.


Duplicate Message Processing for Standard Resequencer Does not Complete After Purging


Bug: 13043897
Updated: 04-November-2011
Platform: All

When you run the purge scripts and remove Oracle Mediator sequenced messages, group information is retained for standard resequencers. If at a later time, a duplicate message is received with the same group ID and sequence ID as a message that was processed prior to the purge, that message is shown as Running in Oracle Enterprise Manager and the duplicate message is not processed. This only occurs when using the resequencer and does not cause any runtime failure.


Faults Continue to Appear on Faults Tab After Recovery


Bug: 11894501
Updated: 15-April-2011
Platform: All

If you recover a faulted resequencer instance for Oracle Mediator and the first recovery fails, two recoverable faults appear in the instance flow trace of the Faults tab. If you then retry and succeed in recovering the instance, that instance still has the Recovery Required status on the Instances tab and the two recoverable faults still appear in the flow trace of the Faults tab.

This only occurs when using the resequencer, and it does not affect runtime performance.


Recoverable Case Instances for the Resequencer Are Tracked Incorrectly


Bug: 9574644
Updated: 15-April-2011
Platform: All

In Enterprise Manager, the number of recoverable case instances shown for the Oracle Mediator resequencer equals the number of retries made to that instance rather than the number of actual recoverable instances. This information is for tracking purposes only, and does not result in duplicate messages being sent.


Multi-lingual Support Depends on Database Character Set


Bug:8369798
Updated: 15-April-2011
Platform: All

To avoid any unexpected results when using the multi-language support feature for some functions in Oracle Mediator, ensure that these characters are supported by the character set of the database you are using. For example, for the Unit Test function, if the initial message payload in the unit test includes characters that are not supported by the database character set, the characters fail to display correctly in the unit test section of the Oracle Enterprise Manager Fusion Middleware Control Console. This issue does not occur for a database that supports Unicode characters.


Oracle Mediator - BPEL Process Manager Callback Processing


Bug: 8472652
Updated: 15-April-2011
Platform: All

If Oracle Mediator calls a BPEL process with callback processing and timeout, the best practice is to explicitly handle the callback exceptions returned from Oracle Mediator in the BPEL process. This ensures that the global transaction associated with callback processing gets committed and is not rolled back. This way, the timeout handler configured in Oracle Mediator will not kick in even if the callback handling in Oracle Mediator fails.


Oracle Mediator Components May Cause SOA Suite Server to Run Out of Memory


Bug: 9067255
Updated: 15-April-2011
Platform: All

Oracle Mediator applications may run out of memory under heavy loads. For example, if the rate of incoming messages to Oracle Mediator is faster than the rate of callback messages, memory accumulation occurs in the Oracle Mediator cache.

Workaround: To fix this issue, change the cache size through the Oracle Enterprise Manager Fusion Middleware Control Console by adding the following property in the Parameters attribute of the Oracle Mediator configuration properties:

mediator.runtime_cache.limit=value

 

For example, if you set the value of the mediator.runtime_cache.limit property to 100, the number of objects that can be stored in the cache is limited to 100. You can switch off caching completely by setting this property to zero.


Mapper Test Functionality Does Not Work for Mediator Advanced Functions in Design Time


Bug: 9072259
Updated: 15-April-2011
Platform: All

Oracle Mediator advanced functions, such as mhdr:getCompositeName() and mhdr:getHeader, require a runtime context to execute. If these functions are run using the Mapper Test feature during design time, the target XML file is not generated and the following error is thrown:

XML-22044: (Error) Extension function error: Error invoking 'getComponentName':'

java.lang.IndexOutOfBoundsException: Index: 0, Size: 0'

These functions work correctly in the runtime environment.



Back to top


Service Bus

Bugs Fixed in Oracle Service Bus 11.1.1.7

The following bugs were fixed in Oracle Service Bus 11.1.1.7:

  • 14731268 - FTP proxy services throw an exception when processing files with spaces in the names
  • 14702242 - Handling time-to-live for expired messages and messages with no expiration
  • 14667914 - Handling time-to-live for expired messages and messages with no expiration for JMS
  • 14606921 - When importing an FTP or SFPT proxy service, the pollerTargetManagedServer property is not updated
  • 14589368 - The JMS expiration header value is incorrect when passed from a proxy service to a business service
  • 14562380 - Business service stops responding during JNDI connection
  • 14498581 - Schema validation fails for a schema that references a group from another schema
  • 14469822 - SOAP faults are always encoded as UTF-8 regardless of HTTP header encoding type
  • 14404259 - Derived caches are initialized on first request, causing latency
  • 14397074 - No error is logged if the File transport fails to delete and archive a polled file
  • 14382007 - The domainupgrade.py script processes config.xml without multi-byte character consideration
  • 14380470 - A null pointer exception occurs when creating a Tuxedo transport proxy service in a clustered environment
  • 14344870 - A replace action in a publish node in OEPE generates an error
  • 14334299 - When processing strings containing "&", the test console response is not well-formed XML
  • 14309178 - Sample socket transport is not available in OEPE
  • 14183066 - Mismatched MFL data type causes the Service Bus console and managed server to stop responding
  • 14180879 - An MQ business service fails when the remote queue properties are changed
  • 14151128 - An incomplete error message is generated when transforming MFL XML to non-XML
  • 14148620 - Custom SOAP faults are not returned when using the Direct Binding Adapter to call a proxy service
  • 14139066 - Korean character encoding is not parsed from the Name field when using the Email transport
  • 14125945 - The Process Security Headers option is missing when using a complex WSDL and OWSM policy
  • 14090958 - The Service Bus console does not display groups when a search returns more than 500 results
  • 14080712 - Simultaneous invocations of JEJB transport fail
  • 14067063 - Replacing or assigning an empty XML node fails with an unexpected end-of-file
  • 14019559 - The Work Manager configured for the JMS outbound response is not being used
  • 14003894 - OEPE loses references to proxy message flow transformations in the "Select an XQuery Resource" window.
  • 13996499 - The Process WS-Security Header option is missing when using a complex WSDL and OWSM policy
  • 13990209 - When Transaction Required is enabled, proxy services cannot retrieve any Tuxedo error codes
  • 13968461 - An InitialContext object is created for each JMS message send with static username/password
  • 13902323 - The bea-inlinedxml() function trims whitespace when parsing text to XML
  • 13891198 - No response is generated for an HTTP 204 error code, resulting in a timeout
  • 13890572 - The documentation is missing the -dmiddleware.home parameter for performing a command line export
  • 13845112 - Login IDs using special characters result in an invalid session name on the console
  • 13834019 - SAP Adapter proxy service throws a null pointer exception for multi-threaded synchronous request/response
  • 13832579 - Update the documentation to show that the MQ connection max wait time is defined in milliseconds
  • 13825284 - The description field cannot be updated for a project
  • 13817055 - When WebLogic Tuxedo Connector encounters a null pointer exception, it continues to send the null buffer request to Tuxedo
  • 13811425 - OWSM custom assertions fail to catch faults in Service Bus
  • 13808765 - A null pointer exception is generated when you create a second Tuxedo transport business service
  • 13801026 - Excessive copies of deployment plans are created for services that use the DeploymentUtils API
  • 13799324 - Using an MFL transformation with a zoned decimal field generates an error
  • 13791584 - The FTP transport is unable to overwrite files with a ".a" extension
  • 13788002 - After WebLogic Server is redeployed, the EJB transport throws NoSuchObjectException and EJBException
  • 13738356 - Provide the setSaveCDataEntityCountThresh/setSaveCDataLengthThreshold options as global options
  • 13698881 - Using the WebLogic Server Administration Console to configure the Service Bus console throws a class cast exception
  • 13652069 - When you undo the changes from a previous session, that session's description disappears
  • 13627312 - An illegal state exception is thrown when updating global variables from a split-join catch block
  • 13619680 - A stack overflow exception occurs when harvesting Service Bus artifacts
  • 13546226 - When importing a WSDL that references more than one schema, an import error occurs
  • 13543942 - A connection error sometimes occurs when a report action is executed for the first time
  • 13528323 - A JMS message is processed by the service continuously because it is not removed from the queue
  • 13508576 - Cannot activate a Tuxedo business process with two remote access points
  • 13495349 - Exporting via Ant does not export system resources
  • 13489458 - The MFL format tester fails when a field defined as zoned decimal length
  • 13483567 - Running customization for a JMS business service throws an error
  • 13434404 - An MFL transformation throws a literal mismatch exception when processing large messages
  • 13428060 - OWSM custom assertion fails to catch faults
  • 13411631 - 30-second delay when request uses an If-None-Match HTTP header
  • 13408550 - A general web service security error exception is re-thrown and unhandled
  • 13395466 - Threads become stuck when message tracing is enabled for proxy services
  • 13392863 - Service Bus console performance is impacted by diagnostics file size and activity
  • 13348259 - The JMX monitoring API causes the number of certain Java objects to increase
  • 13331811 - The format builder creates invalid XSD with COBOL Copybook import
  • 13080759 - A socket stays in CLOSE_WAIT status after an endpoint returns a 500 error
  • 13079080 - MQ proxy services start polling and processing messages before the server starts up
  • 13077476 - The error count value for JMS proxy services in the service health table does not increase
  • 13071345 - FTP proxy services incorrectly process files with a ".stage" extension
  • 13066878 - A "no transaction found" error is thrown for a "transaction required" proxy service
  • 13042292 - HTTPS calls to an HTTP proxy service through a Web server fails
  • 13027129 - Web service was detected to be vulnerable to reflected cross-site
  • 12989085 - JMS proxy and business services cannot use TIBCO XA generic connection factory
  • 12965730 - If a Tuxedo business services returns a null buffer, an exception is thrown
  • 12940253 - The mail.smtp.timeout property is missing when configuring an SMTP server
  • 12910439 - MFL transformations generate an incorrect result for zero-length strings
  • 12852520 - Proxy services with WS-I compliance enabled do not refresh after a WSDL update
  • 12839462 - File proxy services generate a polling error when a member of the cluster is down
  • 12814364 - Cannot add a reply action in a split-join error handler
  • 12759297 - With access authorized to a proxy service based on the HTTP host name, permission is denied
  • 12663297 - Attaching OWSM policies on an MTOM-enabled service throws an error
  • 12585136 - The Email transport generates multipart/related emails and not mulitpart/mixed
  • 12407693 - Need a configuration mechanism to force an MTOM response from a proxy service
  • 12347300 - The SOA Direct transport forces a transaction rollback for any fault from a proxy service
  • 11848647 - Error when creating an EJB business service with JAR file dependencies
  • 11829825 - ADF error on Enterprise Manager when accessing the OWSM policy page
  • 11714474 - An exception is thrown when importing a resource that was changed from SOAP to Any XML
  • 11700111 - Response encoding does not affect encoding of XML if XOP/MTOM support is enabled
  • 10350668 - A conditional branch value will not accept a string value
  • 10234522 - Service Bus does not create JMS queue for a proxy service using SSL ports
  • 10203651 - An error occurs when testing an XQuery with an input of type xs:date in OEPE
  • 10155280 - An MFL transformation gets stuck in a loop when input data is invalid
  • 9534799   - Not all unused connections are destroyed by the MQ connection timeout
  • 8178762   - The business service timeout does not take into account the connection timeout

 

 

JMSExpiration Header Represents an Absolute Time, Not a Time-To-Live


Bug: 18084317
Added: 18-February-2014
Platform: All

The documentation for the JMSExpiration header in Service Bus incorrectly describes the JMSExpiration header as a relative value that is added to the current time to determine the expiration for a message; in other words, a time-to-live value. The JMSExpiration header is actually an absolute value that indicates the specific time that the message should expire. You can use an XQuery expression to calculate the precise expiration date and time for each message. Note that this only applies to release 11.1.1.7; previous releases perform as documented.


Stuck Threads Seen When Using the Same Work Manager for Both Proxy and Business Services


Bug: 16943591, 17937271
Updated: 21-March-2014
Platform: All

A stuck threads situation can occur when a proxy service and the business service it invokes are both configured to run on the same Work Manager and the proxy service invokes the business service from within the pipeline using a service callout, publish action, or routing with exactly-once QoS. When threads are stuck, responses are no longer processed and client sockets enter the CLOSE_WAIT state. This occurs primarily with business services configured with an HTTP or JMS transport.

This issue is more common when there is a publish action in the pipeline and the configuration results in a blocking call, as in any of the following scenarios:

  • QoS set to exactly-once

  • QoS is set to best-effort, the messaging pattern is one-way, and the retry count > 0
    Note: With an HTTP business service, the call is non-blocking in the above case because a response code is still returned. It is not treated as a one-way messaging pattern.

  • Transaction is required (or any transaction flag)

Workaround: The Work Manager (dispatch policy) configuration for a business service should depend on how the business service is invoked. If a proxy service invokes the business service using a service callout, a publish action, or routing with exactly-once QoS, use different Work Managers for the proxy service and the business service. Do not use the same Work Manager for both. For the business service Work Manager, configure the min-thread-constraint property to a small number (1-3) to guarantee an available thread.


Slower Performance and Exceptions Seen with Certain JDK Versions


Bug: 16188448, 17860263
Updated: 18-February-2014
Platform: All

Oracle Service Bus performs more slowly when running on certain versions of the Java Platform, Standard Edition Development Kit (JDK). Also, in certain versions of JDK, there is a limit to the number of XML readers created (64,000), causing the StAX XMLInputFactory to throw an exception on reader creation. This affects the SAX parser used by Service Bus.

Workaround: To ensure best performance, install Oracle Service Bus with JDK 6 Update 71 (or later) or with JDK 7 Update 51 (or later).


OEPE Installation Required for Upgrades to Oracle Service Bus 11.1.1.5 and 11.1.1.6


Bug: 17436892
Added: 18-February-2014
Platform: All

When upgrading Oracle Service Bus to release 11.1.1.5 or 11.1.1.6, an OEPE installation is required even if the Oracle Service Bus IDE for Eclipse is not used. Otherwise, the upgrade process fails validation.

Workaround: To upgrade without installing OEPE in the upgraded environment, run the installer with the -novalidation flag, as shown in the command line example below. This skips the OEPE validation and all other validations during the installation.

./runInstaller -jreloc /home/oracle/java -novalidation


Cache Tuning Properties are Documented Incorrectly


Bug: 17967700
Added: 17-December-2013
Platform: All

In the Oracle Fusion Middleware Performance and Tuning Guide, the cache settings listed under "Tuning OSB Operational Settings" include an extraneous property. The property described in "Initialize Router Cache for Proxy service Runtime Data," com.bea.wli.sb.pipeline.initializeRouterCache, should not be used. Instead, set the property listed under "Cache Tuning for Proxy Service Runtime Data" to preload the cache, as shown below.

EXTRA_JAVA_PROPERTIES="-Dcom.bea.wli.sb.pipeline.RouterRuntimeCache.preload=true


Specifying Filters for Reporting


Bug: 17482269
Added: 21-October-2013
Platform: All

When you filter run time reports by index and use multiple indexes for filtering, the indexes are joined by an OR clause in the query. For example,if you filter by AccountNo=1001,PolicyType=1002,Level=1, the resulting report will include all entries with an account number of 1001, a policy type of 1002, OR a level of 1.


Location of Known Hosts File


Bug: 16983796
Added: 21-October-2013
Platform: All

In Section 49.4.2, "Use of the known_hosts File" in the Developer's Guide for Oracle Service Bus, the location of the known hosts file for the SFTP transport is incorrectly listed as DOMAIN_HOME/osb/transports/sftp. The location should be DOMAIN_HOME/config/osb/transports/sftp.


Harvester Documentation Clarification


Bug: 16682224
Added: 21-October-2013
Platform: All

When harvesting an Oracle Service Bus configuration JAR file using Oracle Enterprise Repository, the harvester handles all proxy services, business services, and flows. Where a dependency found, the harvester also handles additional artifacts such as WSDL and schema resources. The harvester does not automatically harvest these additional artifacts found in the configuration JAR file. It only harvests the additional resources when a proxy service, business service, or flow has a dependency on them.


Required Patches for Installing or Upgrading to Oracle Service Bus 11.1.1.7


Added: 29-January-2013
Platform: All

Before installing or upgrading Oracle Service Bus to release 11.1.1.7, make sure you have installed all required patches to WebLogic Server and WebLogic Portal. You can obtain the patches through the Patches and Updates page on Oracle Support. Review the Read Me files for installation instructions and additional information. The following patch numbers are required; check for additional patches on Oracle Support.

WebLogic Server

  • 13961904
  • 13573621
  • 14030073
  • 14322089

ExaLogic

  • 13642828
  • 13829888

WebLogic Portal

  • 8173628

Note: This list includes the patch described under Performance Patch below.


XQuery Expression Drag And Drop Is Not Saved On Safari 5.1.7


Bug: 16211923
Added: 29-January-2013
Platform: All

When the Oracle Service Bus Console is running on the Safari 5.1.7 Web browser, the drag and drop feature of the message flow XQuery/XSLT Expression Editor does not work as expected. When you drag and drop an XQuery expression from the palette in the left panel to the expression field, the expression is not saved when you click Save All. The XQuery expression also disappears when you click Validate. This results in a conflict when you save the message flow.

Workaround: Instead of using the drag and drop feature, do the following:

  1. Select the XQuery function or variable you want to use in the palette.
  2. Select the text that appears in the Property Inspector, right-click the selected text, and click Copy.
  3. Right-click in the expression field and click Paste.

The expression can then be validated and saved.


Assign Activity Produces an Empty Message Body if a New Line Character Exists


Bug: 16205232
Added: 07-March-2013
Platform: All

The resulting message body is empty if a proxy service processes an XML string with a newline character after the XML declaration (<?xml version="1.0" encoding="UTF-8"?>), the message flow includes an Assign activity that uses fn-bea:inlinedXML() to assign that XML string to a variable, and the message flow uses the variable to replace the contents of $body. This is related to issue 13902323, caused by the fn-bea:inlinedxml() function trimming whitespace when parsing text to XML.


Certain HTTP Transport Properties Cannot be Globally Updated


Bug: 16091796
Added: 29-January-2013
Platform: All

Two properties for the HTTP transport, Connection Timeout and Read Timeout, are not operational or environment values. This means they are not included in the customization file and their values cannot be updated using scripts.

Workaround: These values can be updated manually in the Oracle Service Bus Console. If you need to do a bulk update of these values, check with Oracle Support to obtain a patch.


After Upgrade, Oracle Service Bus Console Throws Exceptions


Bug: 16039792
Added: 29-January-2013
Platform: All

After upgrading from version 11.1.1.3 to 11.1.1.7, the Oracle Service Bus Console throws exceptions when performing tasks such as viewing projects, exporting configuration JAR files, and creating a session.

Workaround: Restart the WebLogic Servers from a new command or terminal window.


Refreshing the Runtime Cache Under Load Might Cause Latency


Bug: 16000058
Updated: 04-March-2013
Platform: All

This release introduces a new system property, RouterRuntimeCache, that you can use to change the way derived caches are initialized. You can read about this change in "Cache Tuning for Proxy Service Run-Time Data" in Oracle Fusion Middleware Performance and Tuning Guide. If you enable this property, the cache is also purged and reloaded any time an operational change is activated after system startup. Depending on the extent of the changes, this might prevent messages from being processed for the duration of the reload process and cause additional latency. We are enhancing this feature to minimize the reload latency, and a patch will be available from Oracle Support.


Updating Global Settings Under Load Might Cause Latency


Bug: 15972130
Added: 29-January-2013
Platform: All

Under a heavy load, updates to the global settings might lock message processing threads for an extended period, eventually degrading Oracle Service Bus performance. A patch will be made available from Oracle Support to increase the update speed.


Oracle Service Bus WLST Script Attempts to Create Sample Domain


Bug: 15971368
Added: 29-January-2013
Platform: All

If you run the wlst.sh script from the OSB_HOME/common/bin directory, the script attempts to create a sample Service Bus domain each time. If the sample domain already exists, the script attempts to overwrite the existing domain.

Workaround: Run the wlst.sh script from WEBLOGIC_HOME/common/bin instead of from the Service Bus home directory.


Pass-Through Pipeline Might Marshall and Unmarshall Message Body


Bug: 15908859
Added: 29-January-2013
Platform: All

If the message flow in a pipeline is configured to simply pass messages through with no modification, the messages are still parsed by the XML parser if the message content is accessed and read during processing. This means the original text is materialized into XMLBeans and then back into XML. For example, the message might be processed in a pass-through scenario, but be accessed and read by a logger during processing. In this case, the message is materialized into XMLBeans. When the message body includes decimal or hexadecimal representation of characters, the XML parser resolves these representations to the actual characters. Note that only the character representation of the information is changed and not the information itself.

Workaround: If there are character representations in the message content and you want to preserve the original character representations instead of converting them to the actual characters, either make sure the message body is unread during processing to ensure it is truly pass-through, or specify the representations inside a CDATA section before the message is received by Service Bus. For example:

<typ:NameElement>
   <typ:name><![CDATA[&#x6C38;&#x697D;&#x4F4F;]]></typ:name>
</typ:NameElement>


Importing a Database Adapter Into OEPE Causes an Error


Bug: 14842617
Added: 29-January-2013
Platform: All

When you create a Database Adapter that uses native sequencing in Oracle JDeveloper and you import the adapter into Oracle Enterprise Pack for Eclipse (OEPE), a TopLink mapping error occurs. The error message is:

Invalid Toplink Mapping XML. Error Message: Attribute not allowed (no wildcards allowed): table in element sequence-field@http://www.eclipse.org/eclipselink/xsds/persistence Location of invalid XML:
<xml-fragment table="<table_name>"
xmlns:xsd="http://www.w3.org/2001/XMLSchema"
xmlns:eclipselink="http://www.eclipse.org/eclipselink/xsds/persistence"
xmlns:xsi="http:// www.w3.org/2001/XMLSchema-instance"/>

Workaround: In your Eclipse working directory, navigate to the generated TopLink mapping XML file in the imported project. Open the file in an XML editor, and remove the TABLE attribute from the sequence-field element. Save and close the file. Do not validate the XML file against the XSD.


JCA Polling Proxy Services Generate Struck Thread Errors


Bug: 14831608
Added: 29-January-2013
Platform: All

Proxy services configured for JCA polling adapters (such as the JCA File Adapter, DB Adapter, AQ Adapter, and so on) might generate errors in the server log indicating that the thread is stuck, which causes the server's health status to change to WARNING. The errors are caused by the single listening thread created in the default Work Manager. This thread is non-terminating so the services can continue to poll for new events, but this means the thread necessarily remains busy longer than the configured maximum time. The error message indicates that the thread has been busy for "xx" seconds working on the request, which is more than the configured time (StuckThreadMaxTime) of "xx" seconds. The errors can be ignored, but might also fill up the server log.

Workaround: Because these error messages do not indicate a true problem, they can be suppressed. To do this, create a new Work Manager with the following setting:

<ignore-stuck-threads>true</ignore-stuck-threads>

Then modify the service's Dispatch Policy in the transport settings to use the new Work Manager. Any new stuck thread messages are written to the new Work Manager instead of cluttering up the default Work Manager.


Unable to Create Groups in the Oracle Service Bus Console


Bug: 14806859
Added: 29-January-2013
Platform: All

When you try to add a new security group on the Summary of Groups page of the Oracle Service Bus Console, an exception is thrown and the group cannot be created.

Workaround: Add new security groups using the WebLogic Server Administration Console, or obtain the associated patch from Oracle Support.


Error Message Generated for Public Key Algorithm Does Not Give Context


Bug: 14502051
Added: 29-January-2013
Platform: All

When an invalid public key algorithm is specified for the SFTP Transport in a business or proxy service, the resulting error message does not give a clear indication of what the error is.


Harmless Warning for JEJB 3.0 Proxy Service


Bug: 14321779
Added: 29-January-2013
Platform: All

After a JEJB 3.0 proxy service is created and activated, the following warning message appears in the managed server logs:

<Warning> <WebLogicDescriptorWL> <BEA-2156000><"config-root"/<path_to_ear_file>/<filename.ear>/plan was not found>

This warning is harmless and can be ignored.


Changes to Carriage Return Handling in XML in 11.1.1.6


Updated: 01-September-2012
Platform: All

An update was made to Apache XMLBeans, which corrects handling of carriage return characters in an XML document. The update was made because XMLBeans did not properly escape carriage return characters (\r). If the document contained &#13;, it was correctly unescaped to \r when parsed into XMLBeans; however, \r was not escaped when it was streamed out. This resulted in an invalid XML document because \r is invalid in XML. After the fix to XMLBeans, \r is correctly escaped to &#13;.

If you upgrade to version 11.1.1.6 and currently have a workaround for the above issue in place, you need to modify how return characters are handled. For example, if the workaround was to escape &#13; a second time to &amp;#13; in order for it to be streamed back out as &#13;, you need to remove the second escape to avoid validation errors.


Service Bus Installation Guide Incorrectly States that 64-bit IDE Not Supported


Bug: 13573801
Removed: 29-January-2013
Platform: All

This issue has been resolved in Oracle Service Bus 11.1.1.7. The details are available in Oracle® Fusion Middleware Installation Guide for Oracle Service Bus.


Performance Patch


Bug: 13573621
Last Updated: 29-January-2013
Platform: All

Changes in Oracle WebLogic Server may impact performance for Service Bus services. A patch has been created to address this. Oracle recommends applying the patch for all platforms, including Exalogic deployments. You can access the patch from the Oracle Service Bus downloads page. Accept the license agreement and expand the Prerequisites & Recommended Install Process section. The link to the patch is under the Oracle WebLogic Server download. Make sure to review the readme.txt file before installing the patch.

Note: This patch disables the Concurrent Request Manager capability in WebLogic Server. It is included in the list of required patches under Required Patches for Installing or Upgrading to Oracle Service Bus 11.1.1.7 above.


IBM WebSphere MQ Support


Bug: NA
Updated: 15-March-2012
Platform: All

Support for IBM WebSphere MQ client and server software versions 5.3 and 6.0 will be deprecated according to IBM statements regarding the support lifecycle. To view information about IBM support for WebSphere MQ, see http://www-01.ibm.com/software/support/lifecycle/index_w.html.


OWSM Performance Impacts


Bug: 13364492
Updated: 13-January-2012
Platform: All

New features have been introduced to Oracle Web Services Manager (OWSM) that may have minor impact to performance for Service Bus services that use OWSM policy. A patch will be made available and recommended.


Upgrade to Service Bus 11.1.1.6.0 Requires Upgrade to WLS 10.3.6 to Ensure IDE Support


Bug: 13147687
Updated: 17-January-2012
Platform: All

Oracle Fusion Middleware 11g Release 1 (11.1.1.6.0) supports both Oracle WebLogic Server release 10.3.5 and 10.3.6. However, if you have already configured Oracle Service Bus 11g Release 1 (11.1.1.5.0) with Oracle WebLogic Server 10.3.5 and you want to update Oracle Service Bus with Oracle Service Bus IDE to 11g Release 1 (11.1.1.6.0), then you must also update Oracle WebLogic Server to release 10.3.6. Oracle WebLogic Server 10.3.5 is not supported in an Oracle Service Bus IDE 11g Release 1 (11.1.1.6.0) environment.

Workaround: If you must update Oracle Service Bus to 11g Release 1 (11.1.1.6.0), and you want to keep your Oracle WebLogic Server release at 10.3.5, then you must disable Oracle Service Bus IDE by manually deleting the following file:

MW_HOME/oepe_11.1.1.8.0/dropins/oracle.osb.ide.lnk


HTTP Business Service Connection Timeout Not Honored on Windows 2008 64-bit


Bug: 13097895
Updated: 13-January-2012
Platform: Windows 2008 64-bit

Connection timeouts set on HTTP business services ("Connection Timeout" property in the HTTP transport configuration) are not honored in Windows 2008 64-bit environments.


Error Indicators in the OEPE Project Explorer Appear Only at the Configuration Level Directory


Bug: 13097882
Updated: 13-January-2012
Platform: All

In the Oracle Enterprise Pack for Eclipse (OEPE) Project Explorer view, when an error exists in a resource, an error indicator appears only on the configuration-level directory. No error indicators appear on directories or resources within the configuration to help pinpoint the error.

Workaround

  1. If you do not plan to work on WLST development, disable/uninstall the Oracle WebLogic Scripting Tools, which uses the PyDev feature that causes the issue. The scripting tools are located at OEPE_HOME/plugins/oracle.eclipse.tools.weblogic.scripting*.jar. Delete the scripting JAR files.
  2. Uninstall the PyDev feature: In the OEPE menu bar, choose Windows > Preference > Install/Update, and click the Uninstall or update link. In the Installation Details window, uninstall the PyDev feature.

Updates to Message Protection Handling


Bug: 12972692
Updated: 13-January-2012
Platform: All

This issue affects Service Bus services that use message protection when you upgrade from previous versions of Service Bus 11g to Service Bus 11.1.1.6.0.

In previous releases of Service Bus 11g, business services always enforced message protection using the keystore.recipient.alias property that identified a public certificate stored in the domain-level keystore.

In Service Bus 11.1.1.6.0, if a proxy service implements a message protection policy, the service's base64-encoded public certificate is published in the effective WSDL. The WSDL-bound certificate is the service's public key by default, as determined by the Encryption Key defined for the proxy service. Any business service created from that WSDL now uses the WSDL-bound identity certificate for message protection instead of the keystore.recipient.alias property.

If no certificate is bound to a WSDL, the keystore.recipient.alias property is used, as in previous releases.


Existing Server Extended with Oracle Service Bus for Developers Template Fails to Start


Bug: 12343853
Updated: 29-April-2011
Platform: All

After extending an Oracle WebLogic Server domain with the "Oracle Service Bus for developers" domain template, and using Derby as the data source for the JMS reporting provider, the extended server fails to start because of errors related to Derby database initialization.

Workaround: Set the DERBY_FLAG to true in your DOMAIN_HOME/bin/setDomainEnv.cmd(.sh) script. For example:

set DERBY_FLAG = true

For Derby best practices and troubleshooting information, see "Running Multiple Development Domains with Derby" in the Oracle Fusion Middleware Development Guide for Oracle Service Bus at http://www.oracle.com/pls/as1111/lookup?id=OSBDV1892.


Console Logout Does Not Work when Logging In Through OAM


Bug: 11729325
Updated: 10-February-2011
Platform: Linux

When using Oracle Access Manager 10g or 11g to log in to the Oracle Service Bus Administration Console, clicking Logout does not end the session. It returns you to the Oracle Service Bus Administration Console home page, still logged in, rather than logging you out and returning you to the Oracle Access Manager login page.

Workaround: To end the Oracle Service Bus Administration Console session, close your browser.


Starting Managed Server with IPv6 Host and Port Throws BEA-000000 Error


Bug: 10029204
Updated: 14-January-2011
Platform: All

In the development environment, when starting a managed server by passing the managed server's IPv6 address and port, Oracle Service Bus throws a BEA-000000 "Cannot contact Admin server" error.

To work around this issue, create a new key-pair, then reconfigure SSL on the Admin server so that the key-pair in the server-side SSL certificate uses the host name of your IPv6 managed server. Then restart your managed server.

For more information on SSL, see "Configuring SSL" in Oracle Fusion Middleware Securing Oracle WebLogic Server at http://www.oracle.com/pls/as1111/lookup?id=SECMG384.


Session Activation Error Message in Firefox


Bug: 9958876
Updated: 14-January-2011
Platform: All

When activating an Oracle Service Bus session in Firefox, session activation takes longer than three minutes and you see an error message stating, "Another session operation is in progress. Please retry later."

The error message is harmless, and you can ignore it. The server is still processing the session activation, and you can wait for activation to finish.

The error message occurs when the Firefox network connection is set to "Manual proxy configuration" and the "HTTP Proxy" value is used for all listed protocols. The message is triggered by Firefox's auto-refresh occurring in conjunction with the proxy settings.

If you want to ensure the error message does not appear rather than ignoring it, change the Firefox proxy settings to another option, such as "Auto-detect proxy settings for this network" or "Automatic proxy configuration URL." To access these options in Firefox, choose Tools > Options > Advanced > Network > Settings.


Password Field Not Editable in Domain Template on Linux


Bug: 10394900
Updated: 14-January-2011
Platform: Linux

On Linux, when creating a new domain using a developer template in the Oracle Fusion Middleware Configuration Wizard, the password and confirm password fields are not editable, and you cannot enter a password to create a domain.

To work around this issue, do one of the following:

  • To work around the issue each time it occurs: Click the close window "x" button in the upper right corner of the Configuration Wizard. In the confirmation dialog that appears, click "No" to return to the Configuration Wizard. Now you can enter the password and confirmation password.

    or
  • To fix the issue permanently:

    1. Kill a scim processes. For example, kill "pgrep scim"
    2. Modify (or create) the file ~/.scim/config to include the following line:
      /FrontEnd/X11/Dynamic = true (follow case sensitivity)
    3. If you are running VNC, restart the VNC server.
    4. Run the Configuration Wizard again.

JAVA_HOME Incorrect in setDomainEnv After Domain Upgrade on 64-Bit Install


Bug: 10404384
Updated: 14-January-2011
Platform: All

On 64-bit installations of Oracle Service Bus, after upgrading a domain from 11g Release 1 Patch Set 2 (11.1.1.3.0) to 11g Release 1 Patch Set 3 (11.1.1.4.0), the JAVA_HOME value in the upgraded domain's setDomainEnv.cmd/sh file is incorrect.

The reason JAVA_HOME is not set correctly on 64-bit installations is because the 64-bit installer does not include a JDK.

To work around this issue, edit the setDomainEnv.cmd/sh in the upgraded domain with the correct JAVA_HOME for the JDK you are using.


Oracle Application Adapter Installation Failure on Windows


Bug: 10424957
Updated: 14-January-2011
Platform: Windows

When installing the Oracle Application Adapters for use with Oracle Service Bus, such as the Oracle Application Adapter for SAP R/3, you receive the following error:
"The ORACLE_HOME doesn't have the right version of SOA or OSB. The adapters cannot be installed."

Because the Oracle Application Adapters installer gets the JAVA_HOME value from the Windows registry, any space in the JAVA_HOME value in the registry causes the installation failure. For example, if the JavaHome value is C:\Program Files\Java\jre6, the space between "Program" and "Files" causes the failure.

To work around the issue:

  1. Back up your Windows registry.
  2. In Windows, choose Start > Run, enter regedit in the Run dialog box, and click OK.
  3. In the registry editor, go to the following registry folder:
    HKEY_LOCAL_MACHINE\SOFTWARE\JavaSoft\Java Runtime Environment\version_number
  4. Modify the JavaHome and RuntimeLib values so that they point to a path that does not contain spaces in the directory names.
    Changing the value may require you to either point to a different Java home without spaces in the directory names or rename the path of a Java home to remove spaces.

Inbound Adapters Throw Error on Payload Size Limit


Bug: 9969535
Updated: 14-January-2011
Platform: All

When using the Oracle JCA adapters with Oracle Service Bus, Oracle Service Bus throws an error when the inbound payload size (DOM size) exceeds what is allowed by the payloadSizeThreshold adapter endpoint property.

To work around this issue, increase the size of the payloadSizeThreshold property on the JCA adapter endpoint configuration.


MQ Transport in Binding Mode Throws Error on Linux


Bug: 10020344
Updated: 14-January-2011
Platform: Linux

When Oracle Service Bus and IBM WebSphere MQ are installed on the same Linux system, configuring the Oracle Service Bus MQ transport to run in binding mode causes an error.

To work around this issue, perform the following steps:

  1. Make sure Oracle Service Bus and IBM WebSphere MQ are installed using the same user credentials.
  2. Add the MQ Java library path to the Oracle Service Bus DOMAIN_HOME/bin/setDomainEnv.sh file. For example:
    -Djava.library.path=/opt/mqm/java/lib
  3. Oracle JRockit: When using Oracle JRockit on Linux, add the LD_PRELOAD property to setDomainEnv.sh as well. For example:
    LD_PRELOAD=${JAVA_HOME}/jre/lib/i386/libjsig.so

Publishing Services to IPv6 Server Fails


Bug: 10032879
Updated: 14-January-2011
Platform: All

Publishing services from the development environment to a domain that uses IPv6 addressing results in a MalformedURLException error.

To work around this issue, when specifying the host name in the "Define New Server" window in the Servers view of the IDE, use either the server host name or specify the IPv6 server address in square brackets. For example: [2001:0db8:ffff:ffff:ffff:ffff:ffff:ffff].


Resources Disappear When Server Encoding Changes


Bug: 9488927
Updated: 14-January-2011
Platform: All

After creating resources in the Oracle Service Bus Console, then changing the server encoding and restarting the server with the different encoding, only English language resources (if any) appear in the Oracle Service Bus Console. Non-English resources do not appear.

For example, if you start the server using ja_JP.utf8 encoding and create resources in the Oracle Service Bus Console with Japanese, Korean, and English names, then restart the server using a different encoding such as ko_KR.euckr, only the English-name resources appear in the Oracle Service Bus Console.

In this situation, the missing resources still exist but are not displayed because of the encoding change.

To work around this issue, switch back to the original server encoding and restart the server. Oracle Service Bus does not support server encoding changes.

 

 

 


Back to top


Healthcare Integration


MLLP 2.0 Negative ACK Not Sent


Bug: 16034415
Added: 21-December-2012
Platform: All

Oracle Healthcare does not send MLLP 2.0 Negative ACK if the database commit for an inbound message fails.


Auto Stack Handler and Auto Stack Interval Are Deprecated


Bug: 16074089
Added: 25-January-2013
Platform: All

The features, Auto Stack Handler and Auto Stack Interval, are deprecated in Oracle SOA Suite 11.1.1.7.


Auto Stack Handler and Auto Stack Interval Should be Disabled Before PS6 Upgrade


Bug: 16070511
Added: 25-January-2013
Platform: All

If you are using Oracle SOA Suite for healthcare integration release 11.1.1.5.0 Feature Pack or 11.1.1.6.0 and need to upgrade to 11.1.1.7, then you should explicitly disable the Auto Stack Handler and Auto Stack Interval properties. These properties are deprecated in the 11.1.1.7 release.


Filters Cannot Be Set for Oracle Healthcare Audit


Bug: 16013010
Added: 21-December-2012
Platform: All

The audit feature of Oracle Healthcare provides filters so that you can perform audit based on certain conditions. In the current release, though the Filter option is available, it does not work even if you configure it.


Oracle Healthcare Does Not Correlate FA with Outbound Message


Bug: 15855317
Added: 21-December-2012
Platform: All

In case of an MLLP 1.0 server with multiple client connections, when an MLLP 1.0 client sends an ACK, then the server needs to correlate the FA with the outbound message and te outbound message typically goes to the complete state.

However, if the server and client reside on the same computer or have the same IP address, then on the MLLP server side, the outbound message remains in the WAIT_FA state. This type of configuration is not supported.


Oracle Healthcare Goes into a Loop if It Receives Wrong MLLP 2.0 Commit Ack


Bug: 14683762
Added: 25-January-2013
Platform: All

You need to decide on the same Commit Ack characters before using MLLP 2.0 in Oracle Healthcare. If a Commit Ack character is changed, then the the sending system treats the message as an HL7 message rather than an Ack. In this case Oracle Healthcare goes into a loop and keeps in creating 1 byte messages.


Dispatcher Messages Are Not Getting Logged


Bug: 14477178
Added: 21-December-2012
Platform: All

The dispatcher log messages do not get logged when the logging is set to <oracle.soa.hc.enginecode>trace:32 in the Oracle Fusion Middleware Enterprise Manager console. They only get logged when oracle.soa.b2b.engine logging is set to trace:32.


Retry or Resubmit Ineffective in Case of Multiple Client Connections


Bug: 14398136
Added: 21-December-2012
Platform: All

In case of multiple client connection, if the client reestablishes the connection using a new dynamic IP address or port, then the response cannot be transmitted, and you may encounter a transport error.

You may encounter the same issue with retry or resubmit of messages. If the client reestablishes the connection using a new dynamic IP address or port, then messages cannot be delivered using the retry or resubmit mechanism.


AQ Is Not Supported in Oracle Healthcare


Bug: 14396417
Added: 21-December-2012
Platform: All

Oracle Healthcare does not support the backend interface using AQ queues. Only the JMS Queues option (along with default composite in memory interface) is supported.


Send to Internal Channel/Receive from Internal and Healthcare Adapter Have Different JMS Message Type


Bug: 14394653
Added: 21-December-2012
Platform: All

In Oracle Healthcare, the default JMS message type in the case of Internal channels (Send to Internal and Receive from Internal) is Bytes. However, in Oracle JDeveloper, the Healthcare Adapter JMS option, the default JMS message type is Text Message.

If you use default values and configure an Oracle Healthcare inbound or outbound message with the JMS Send to Internal or Receive from Internal channel, and the Oracle JDeveloper Healthcare adapter with the JMS integration type to use the default JMS message type, then it throws the following error:

 

<Error> <oracle.soa.adapter> <BEA-000000>
<JMSAdapter ADTInwithJMSProj0620 JMSMessageConsumer_ConsumeMessage: Could not
construct inbound adapter message - removing message in error - [destination
= jms/b2b/B2BTest1 (payload = 1)] [BINDING.JCA-12158
ERRJMS_UNEXPECTED_TYPE.
ERRJMS_UNEXPECTED_TYPE.
Received a JMS Bytes message, while expecting to receive a JMS Text message.
Please examine the log file to determine the problem.
]> 

 

So, it is recommended that you should change the default JMS message type values appropriately.


Functionality of b2b.rowlockingforcorrelation in Case of MLLP 2.0


Bug: 14321786
Added: 21-December-2012
Platform: All

In case of MLLP 2.0, when an outbound message is sent to an endpoint, the Business Message state changes from MSG_WAIT_TRANSMIT to MSG_WAIT_ACK until the endpoint receives a commit ACK. On receiving a positive commit ACK, the state changes to MSG_WAIT_FA or MSG_COMPLETE depending on whether or not an FA is required.

However to improve performance, the state change from MSG_WAIT_TRANSMIT to MSG_WAIT_ACK is not done by default. By default, even after sending outbound message to endpoint, the Business Message reflects the MSG_WAIT_TRANSMIT state. However, the Sequence Manager state of the message is change to ACQUIRED from PROCESSED.

So, if you are ready to compromise on the performance lag, but in return, want to display the correct Business Message states, then only set b2b.rowLockingForCorrelation=true. This ensures that the Business message state changes to MSG_WAIT_ACK when the outbound message is successfully sent to the endpoint.


Intermittently, Purging Instance Data Throws SQL Exception


Bug: 14281762
Added: 21-December-2012
Platform: All

In the Repository Management page, when you select Purge Control Number and click the Purge Instance Data button, you get the following error:

 

Purge instance data problem encountered: Internal Exception:
java.sql.SQLException: ORA-00060: deadlock detected while waiting for
resource Error Code: 60 Call: ALTER TABLE B2B_BUSINESS_MESSAGE DISABLE
CONSTRAINT B2B_BM_NDS_FK Query: DataModifyQuery()  Please consult server logs.

 

This error is thrown when you try to purge data within an active message flow.


Flow Trace Page Is Blank if Composite Is Not Deployed


Bug: 14241959
Added: 21-December-2012
Platform: All

If SOA composite is not deployed, then clicking the JMS Send To Internal Channel link in Oracle Healthcare reports opens a blank flow trace page. Flow tracing is not reserved for Composites only, but can also be displayed for standard Oracle AQ Internal Delivery Channel.


Provisioning Users for Viewing Payload


Bug: 14186354
Added: 21-December-2012
Platform: All

Oracle Healthcare console does not provide the document type provisioning capability. So, this has to be done using the Oracle B2B console. Once the user has been provisioned for specific document types in the Oracle B2B console, the details is preserved when just editing the user's role in the Oracle Healthcare console. However, if the user is removed from the Oracle Healthcare console User Provisioning page, and then and added back at a later time, all previous document type provisioning is lost, and you need to redo the provisioning using the Oracle B2B console.


Start JDeveloper in Preview Mode


Bug: 13801385
Updated: 1-March-2012
Platform: All
Release: 11.1.1.6.0

To ensure access to the healthcare adapter in JDeveloper, start JDeveloper in preview mode. For example:
./jdev -J"-DPREVIEW_MODE=true"


Unable to Discard Beyond the 25th Message in Sequence Viewer


Bug: 13777933
Added: 29-February-2012
Platform: All

In Oracle SOA Suite for healthcare integration, you cannot use the dashboard sequence viewer to discard messages beyond the 25th message.


Resubmitting Failed Messages


Bug: 13745321
Added: 29-February-2012
Platform: All

When you try to again resubmit a failed message sfter resubmitting it once, the message goes into the MSG_WAIT_TRANSMIT state and is treated as a new message.

Workaround: If you want to resubmit a failed message for the second time, you should always resubmit the resubmitted message (the message that was resubmitted subsequent to a failure) and should not try to resubmit the failed message directly. The resubmitted message cannot be easily tracked from the reports page, so you can use the dashboard sequence viewer to find the newly created resubmitted message.


Resubmit Count Is Not Incremented


Bug: 13745311
Added: 29-February-2012
Platform: All

When you resubmit an application message, the resubmit count is not incremented and shown as zero. However, commandline utilities enable you to use exclresubmit and maxcount to resubmit messages. So, if you use maxcount and resubmit only a subset of failed messages, you will be unable to resubmit the messages using commandline with exclresubmit.

Workaround: You should not use exclresubmit when performing commandline resubmit of messages.


Establishing TCP Reconnection for Client Endpoints upon Server Restart


Bug: 13729924
Added: 29-February-2012
Platform: All

For Client endpoints (outbound), whenever the servers are restarted, to re-establish the TCP connection with the external server (Listener), you need to:

Disable and re-enable the Use JMS Queue as Default option under Settings > Runtime in the Administration tab of Oracle SOA Suite for healthcare integration console.

This preceding step is mandatory. If you do not perform the step, once the servers restart, messages will not be sent successfully, although the Client endpoint is displayed as enabled in the Oracle SOA Suite for healthcare integration console.


Audit Policy Is Supported Only for the Default User


Bug: 13723449
Added: 16-February-2012
Platform: All

Currently, the audit policies in Oracle SOA Suite for healthcare integration is supported only for the default user.


Message Processing Overview Not in Sync with Message Volume Chart


Bug: 13719712
Added: 14-February-2012
Platform: All

In a dashboard, after clicking the Refresh icon, the data displayed under Message Processing Overview is not in sync with the information displayed in the Message Volume chart.


JMS Settings for Inbound Messages in Case of Server Crash or Fallover


Bug: 13699603
Added: 13-February-2012
Platform: All

In the case of a server crash or fallover, messages do not get forwarded from the inactive JMS Queue(because of server shutdown) to the active JMSQueue. An additional JMS Queue setting is required to enable the inbound messages to complete processing in the case of server crash or fallover.

You need to do the following:

  1. In the Weblogic Server console, navigate to Services->Messaging->JMSModules->SOAJMSModule->dist_B2BEventQueue_auto->
    Configuration->General
    .
  2. In the Forward Delay field, enter any value (in seconds) greater than 0 .

This is the time (in seconds) that the Queue waits for a consumer to attach a message before forwarding the message to another member queue with an active consumer.


Test to Production (T2P) Does Not Update Large Payload and Callout Directories


Bug: 13696574
Added: 13-February-2012
Platform: All

T2P does not automatically update large payload and callout directories. You have to manually update these post T2P in the production environment by using the Oracle SOA Suite for healthcare integration console.


Outbound Messages Remain in MSG_WAIT_TRANSMIT State


Bug: 13695933
Added: 13-February-2012
Platform: All

At times, you may see that run-time messages remain in the MSG_WAIT_TRANSMIT state. This can happen if the Dispatcher is not started or it has stopped working.

Workaround: In these cases, restart the Dispatcher.


Test to Production (T2P) Does Not Update FTP User Password


Bug: 13696558
Added: 13-February-2012
Platform: All

Typically, FTP endpoint parameters, such as FTP host and user name, automatically gets updated from test to production environment

However, currently, T2P does not automatically update the FTP user password. You have to manually update the FTP user password post T2P in the production environment by using the Oracle SOA Suite for healthcare integration console.


Internal Receive from Channel Gets Disabled Post Upgrade


Bug: 13686347
Added: 13-February-2012
Platform: All

When you upgrade Oracle SOA Suite for healthcare integration from version 11.1.1.5.0 to 11.1.1.6.0, the internal receive from channel (internal listening channel) gets disabled.

You need to enable internal receive from channels from the Oracle SOA Suite for healthcare integration console, else you may encounter runtime errors.


Default Dashboard and Report Settings Are Changed in Production Environment


Bug: 13696484
Added: 13-February-2012
Platform: All

In the Oracle SOA Suite for healthcare integration console, in the production environment, the settings for the default dashboard is specified as None. In the Dashboard tab, you need to specify the name of the dashboard that you want to launch by default.

Similarly, in the production environment, the settings for the default report is specified as Last 24 hours. In the Reports tab, you need to set the type of report that you want to launch by default.


Any Change in Document Details Causes Redeploy of Endpoint Documents


Bug: 13696117
Added: 13-February-2012
Platform: All

When you modify the document parameters under the document version, type, or definition in the Oracle SOA Suite for healthcare integration console, the change causes an immediate redeployment of the supported documents for associated endpoints without any warning or confirmation prompts.


Messages Getting Stuck in the MSG_WAIT_FA State


Bug: 13656089
Added: 13-February-2012
Platform: All

With sequencing enabled, when inbound messages are sent to an Oracle SOA Suite for healthcare integration Server endpoint, if the connection is dropped, some messages get stuck in the MSG_WAIT_FA state due to the failure in sending Functional Acknowledgement because of the lost connection.

Workaround: You can process the stuck messages by using either Purge or Resubmit in the Report tab to enable the flow of messages to move forward.


Server Restart Required for Oracle B2B Server Property to Take Effect


Bug: 13644558
Added: 13-February-2012
Platform: All

You need to restart the mamaged server (SOA Server) for Oracle B2B Server property oracle.tip.b2b.edi.binaryDocType to take effect.


Coherence Initialization in a Cluster Setup


Bug: 13620702
Added: 13-February-2012
Platform: All

In the Windows operating system, in a cluster environment, the Endpoint Coherence Cache of Oracle B2B should start up in a distributed mode. To achieve this, you need to:

  • Set the b2b.HAInstance property to true. This allows the Oracle B2B Coherence Cache to start up in a distributed mode.
  • While setting up the nodes of the cluster, you should configure the WKA settings for coherence distributed cache set up. For more information, see the Setting Up a Cluster chapter in Oracle Coherence Developer's Guide.


Oracle Healthcare Commandline Utilities Does Not Have Validation for Batch Commandline Parameters


Bug: 13601735
Added: 13-February-2012
Platform: All

Oracle SOA Suite for healthcare integration does not provide validation for batch commandline parameters. You need to provide the valid parameter values, otherwise you may encounter unexpected behavior.


Usage of Oracle B2B Console and Oracle SOA Suite for healthcare integration Console Together Is Not Supported


Bug: 13598854
Added: 13-February-2012
Platform: All

As of the current release, there is no support for using both Oracle B2B and Oracle SOA Suite for healthcare integration simultaneously for single installation of Oracle SOA Suite. The metadata created by an Oracle B2B user is not supported for running Healthcare-related scenarios in the Oracle SOA Suite for healthcare integration console and vice-versa.


Date/Timestamp in Oracle SOA Suite for healthcare integration Audit Log Does Not Contain Timezone Information


Bug: 13589357
Added: 13-February-2012
Platform: All

The date/timestamp in Oracle SOA Suite for healthcare integration audit log, which is available in fmw_home/user_projects/domains/domain_name/servers/managed_server_name/logs/auditlogs/SOA-HCFP/audit.log, does not contain the server timezone information.


Acknowledgement Mode Values Are Not Localized


Bug: 13589326
Added: 13-February-2012
Platform: All

When you create endpoints in a localized instance (localized in any of the nine Admin languages) of Oracle SOA Suite for healthcare integration, the values for the Acknowledgement Mode list are in the English language (not localized).


Name and Description Labels in the Document Creation Page Are Displayed Vertically


Bug: 13589289
Added: 13-February-2012
Platform: All

When you access the Oracle SOA Suite for healthcare integration console in Korean (Ko), simplified Chinese (zh_CN), traditional Chinese (zh_TW), or Japanese (Ja) browser languages, while creating a custom document, the Name and Description field labels are displayed vertically. However, the issue is not reproducible in any of the European languages.


Dispatcher Count Value Resets to Zero on Purging the Repository


Bug: 13579035
Added: 13-February-2012
Platform: All

By default, Oracle SOA Suite for healthcare integration users need to use dispatcher count values as 1. The postinstall script, which is required to be run to set up Oracle SOA Suite for healthcare integration, changes the dispatcher count from 0 to 1. However, this count resets to 0 after you purge the entire repository.


Failure Condition Check Boxes in Oracle SOA Suite for healthcare integration Audit Policy Page are Enabled


Bug: 13573090
Added: 13-February-2012
Platform: All

The audit policy page for Oracle SOA Suite for healthcare integration in Oracle Enterprise Manager Fusion Middleware Control console enables you to audit both Success and Failure conditions for each event (such as User Login and User Logout). You can configure the audit policy by selecting or deselecting the Success or Failure condition Enable Audit check boxes for these events.

For the current release, only Success conditions are tracked. However, the Oracle Enterprise Manager Fusion Middleware Control console still allows you to select or deselect Failure conditions check boxes even though Failure conditions are not tracked.


Oracle SOA Suite for healthcare integration Throws oracle.mds.naming.InvalidReferenceTypeException


Bug: 13563887
Added: 13-February-2012
Platform: All

When you apply OPatch to Oracle SOA Suite and start the managed server before completing the postinstallation tasks, Oracle SOA Suite for healthcare integration throws the following exception:

<Warning> <oracle.mds> <BEA-000000> <Error
creating the checksum provider
oracle.mds.naming.InvalidReferenceTypeException: MDS-01159: Expecting
absolute document reference. Reference 
"/soa/b2b/seed/protocol.xml#GRABTRANS.results" has local element reference
portion.
	at oracle.mds.naming.ResourceName.splitAbsRef(ResourceName.java:468)
	at oracle.mds.naming.DocumentName.create(DocumentName.java:628)
	at oracle.mds.naming.DocumentName.create(DocumentName.java:262) 
	.....

Messages That Are Blocked in Sequence Are Reported As Sent


Bug: 13532714
Added: 13-February-2012
Platform: All

In Oracle SOA Suite for healthcare integration dashboard, the status of even those messages that get blocked (which are actually not sent successfully) in the sequence are displayed as sent.


Enabling FTP Sequencing


Bug: 13526698
Added: 13-February-2012
Platform: All

To enable FTP sequencing in Oracle SOA Suite for healthcare integration, you have to configure the FTP listening channel with the following parameters:

  • Sequencing
  • TimeStamp Format
  • TimeStamp Offset
  • TimeStamp Source

The following are sample values for the parameters:

  • Sequencing: true
  • TimeStamp Format: 43,55,'MMM d yyyy'
  • TimeStamp Offset: +0000
  • TimeStamp Source: MMM dd yyyy

Please note that selecting only sequencing flag does not work.


Business Message Detail Displays Interchange Control Number and Group Control Number


Bug: 13508343
Added: 13-February-2012
Platform: All

The Oracle SOA Suite for healthcare integration console displays Interchange Control Number and Group Control Number in the Business message details. However, these numbers should not be displayed because these are EDI-specific data and are not present in HL7 messages.


Changes Made to Dashboard Are Lost


Bug: 13463337
Added: 13-February-2012
Platform: All

When you customize a dashboard in the Oracle SOA Suite for healthcare integration console by modifying the available settings, the changes made to the dashboard are lost when you switch to other tabs, such as the Designer or the Reports tab, in the console.


Number of Messages Shown in Dashboard Is Wrong When Endpoint Is Disabled and Then Enabled


Bug: 13426326
Added: 13-February-2012
Platform: All

The number of messages displayed in a dashboard does not reflect correctly when an endpoint is disabled and subsequently enabled.

For example, the dashboard for a given endpoint, which is confogured to show the messages published in the last hour, displays five messages. You then disable the endpoint and subsequently enable it. The dashboard does not display any message.

Here, "Last Hour" means messages processed in the current clock hour, and not the message transactions within 60 mins of current time. For example, if a message was submitted at 4:55 PM and the endpoint was re-enabled at 5:01 PM, the last hour includes only transactions processed since 5:00 PM and will not included those submitted at 4:55 PM.


Resubmit of Inbound Message, Which Failed Due to Duplicate Transaction, Passes


Bug: 13383138
Added: 13-February-2012
Platform: All

A Wire message of an inbound 850 that had previously failed due to detection of the message as a duplicate passes on resubmit. Typically, a duplicated message should throw an error on Wire message resubmit, unless the original message is purged from the system.


Change in Default Behavior of Dispatcher


Bug: 13261298
Added: 13-February-2012
Platform: All

In the current release of Oracle SOA Suite for healthcare integration, the outbound/inbound dispatcher count is 0 by default. This causes an issue in Oracle SOA Suite for healthcare integration, because by default, sequencing is enabled in MLLP and TCP Healthcare transport protocols. So default values of the dispatcher and sequencing flags in transport are not in sync.

Workaround: After you you install Oracle SOA Suite, run the postinstall script to set up Oracle SOA Suite for healthcare integration that among other actions, sets the Dispatcher count to 1.


Endpoint Dashboards Are Not Imported


Bug: 13255434
Updated: 25-January-2012
Platform: All

If you import a healthcare integration export file from a previous version, the endpoint dashboards created in the earlier version are not included in the import file. The dashboards need to be recreated in the new environment.


11.1.x Database Requires a Materialized View Patch


Bug: 13255212
Added: 13-February-2012
Platform: All

Oracle Database 11.1.x associated with Oracle SOA Suite for healthcare integration requires a Materialized View patch. Without this patch, the materialized view refresh is slower.


Intermittently, Inbound Messages Get Delivered to the Incorrect Back-end application


Bug: 12992887
Added: 21-December-2012
Platform: All

If a SOA composite is deployed for a particular doc type (such as ADT_A04), then by default, all the ADT_A04 inbound messages received on any endpoint will be delivered to the deployed SOA composite.


Avoiding Stackoverflow Exception During Import


Bug: 12951614 and 13863165
Added: 11-September-2011
Platform: All

When trying to import a big block of healthcare metadata, such as endpoints to an already-existing large metadata, you may encounter the Stackoverflow exception.
For example, if you try to to import 200 new endpoints to an existing set of 600 endpoints, you may encounter the exception.

Workaround: You need to add the following JVM argument in the setSOADomainEnv.sh file located at $DOMAIN_HOME/bin:

PORT_MEM_ARGS="-Xss512k -Xms768m -Xmx1536m"


Location of HL7 Document Configuration Export ZIP File


Bug: 12943051
Added: 14-February-2012
Platform: All

The HL7 document configuration export ZIP file is required to populate the HL7 document structure. Once you run post-install script to set up Oracle SOA Suite for healthcare integration, you can find this file in the following folder:
SOA_HOME/soa/thirdparty/soahcfp/hl7_doctypes


Runtime Settings Might Not Appear on the Healthcare Integration User Interface


Bug: 12904895
Updated: 12-September-2011
Platform: All

The runtime settings might not appear on the AdministrationSettingsRuntime page of the healthcare integration user interface.

Workaround: If the runtime settings do not appear, follow the steps below to export the repository data, purge design-time data, and reimport the repository data.

  1. Launch the healthcare integration user interface in a web browser. The URL is http://hostname:port_number/healthcare.
  2. Under the Designer tab, click the Administration sub-tab and then double-click Repository Management.
  3. Click Export and then click Continue on the dialog that appears.
  4. Save the file to a temporary directory.
  5. Click Purge Design Metadata.Note: If you have any enabled endpoints, you need to disable them before you can purge the metadata.
  6. In the Import section, browse to and select the file you exported in steps 3 and 4 above, and then click Import.

Oracle B2B Mapsets Are Visible on the Healthcare Integration User Interface


Bug: 12859777
Updated: 12-September-2011
Platform: All

When you create a mapset using the Oracle B2B Console, those mapsets also appear, and can be edited, on the healthcare integration user interface. This include mapsets for messaging types that are not valid in the healthcare integration user interface (such as HIPAA). These mapsets can be viewed, but should not be modified, from the healthcare integration user interface.


The Map EMS File Created from Specbuilder Does Not Contain Trigger Event


Bug: 12854042
Added: 11-September-2011
Platform: All

When you create a map EMS file for mapping between a source and a target HL7 documents by using the Oracle SOA Suite for healthcare integration user interface, the GuidelineInfo section of the map file contains only the message type (such as ADT) information. It does not capture information regarding the trigger event, such as A01. This creates a mismatch in the display of the document type in the healthcare integration user interface, because the DocType is set as ADT and not ADT_A01. So, a correct map file that is uploaded using the user interface will not automatically detect the source and target document definitions used in the map file. If there is no available ADT document, then the definitions will not get preselected in the user interface. Even if an ADT document is available, then the document type ADT will be selected and not the ADT_A01 as expected.

Workaround: You need to apply a hotfix after installing the Oracle Document Editor that is bundled with the Oracle SOA Suite for healthcare integration Feature Pack release. You can download the hotfix from support.oracle.com under the patch number 12797702.

Please note that because Oracle Document Editor is available only for the Windows platform, the hotfix is only applicable on the Windows platform.


Exception Queue Cannot be Created and Specified in the Same Session


Bug: 12850898
Updated: 12-September-2011
Platform: All

When you create an internal delivery channel using the Send to Internal function, you cannot specify that queue as the exception queue in the runtime settings. The queue does not appear in the drop-down list of options for the exception queue.

Workaround: Log out of the healthcare integration user interface and then log back in again. The newly created queue appears in the drop-down list for the exception queue.


Reports Appear to be Missing Messages


Bug: 12846032
Updated: 12-September-2011
Platform: All

The reports on the healthcare integration user interface only display messages that have generated an associated business message. If an error occurs before a business message can be generated, then the message does not appear in the reports.


Limited Number of Dynamic Tabs for Dashboard and Reports Can Be Opened


Bug: 12845568
Added: 11-September-2011
Platform: All

When the maximum number of dynamic tabs are open in Dashboard or Reports tab, then you cannot open any more new dynamic tabs. The maximum number of dynamic tabs that can be opened at a time is 15.

Workaround: You need to close few tabs to be able to open new ones.


Oracle SOA Suite for healthcare integration Document Protocol Links Throws Error After Upgrade


Bug: 12777111
Added: 13-February-2012
Platform: All

When you upgrade Oracle SOA Suite 11gR1 from version 11.1.1.2.0 to 11.1.1.6.0, extend the Oracle Enterprise Manager domain with all the SOA components, and start all the servers, all the applications work fine. However, when you try to access the document protocol folders in the Oracle SOA Suite for healthcare integration console, you get errors such as the following:

<Error>
<oracle.adf.share.el.VariableResolverELContext> <BEA-000000> <The variable
resolver, oracle.adf.model.binding.DCVariableResolverImpl, was used to 
evaluate expression, payLoad.  All variable resolvers have been deprecated.  
Please consult the documentation for the variable resolver and modify the
expression to not depend upon the variable resolver.> 

 


Back to top