1. Using Integrations in Oracle Integration Generation 2
  2. Troubleshoot Oracle Integration
  3. Troubleshoot Integration Activations

Troubleshoot Integration Activations

Learn about troubleshooting integration activations in Oracle Integration.

Topics:

Troubleshoot Integration Activations

If your integration fails to activate, review the following failure issues and corrective actions.

Activation Failure Causes Corrective Actions

WSDL issues:

  • Duplicate schema issues in the WSDL

  • WSDL/schema validation failures

  1. Check the WSDL for duplicate schemas.

  2. Check that the WSDL is valid.

  3. If necessary, correct the WSDL in the application.

  4. Reconfigure the integration by regenerating the endpoints.

  5. Retry the activation.

Deployment time out issues.

Retry the activation because the server may be slow to react.

Database issues:

  • Reached the size limit:

    java.sql.SQLException: ORA-01654: 
unable to extend index

  • Database is down.

  • Connection pool issues:

    Could not create pool connection for 
datasource 'SOALocalTxDataSource'. The 
DBMS driver exception was: Listener 
refused the connection with the 
following error: ORA-12516, TNS:listener
could not find available handler with 
matching protocol stack

  1. Check that the database is running and has enough space.

  2. If not, the database administrator must increase the database size limit or connection pool size.

  3. Retry the activation.

A connection is down or was modified.

  1. Check if the service/application is accessible. If it is down, retry the activation later.

  2. If the connection is invalid, update the connection configuration.

  3. Reconfigure the integration by regenerating the endpoints.

  4. Retry the activation.

Managed server unavailable.

  1. The administrator must restart the server.

  2. Retry the activation.
If an integration contains too many actions (for example, it has 25 or more stage write actions), activation failure can occur. The diagnostic logs show a 51 code too large error:
RegularFileObject[/u01/data/domains/iscs
/servers/integration_server1/dc/338c94
f1-d3ee-45b9-bbf8-43c9203df851/SCA-INF/
bpel/StageCurrentGTHROFileForEachBpel/
src/orabpel/stagecurrentgthrofileforeach
bpel/BIN0.java]:51 code too large

Optimize the integration. For example, split the integration into multiple integrations.

If you attempt to activate an integration that includes a custom adapter, and the log file shows an IllegalStateException error, the metadata must be regenerated.

The activation issue is caused by mismatches in the metadata caused by the WSDL not being regenerated for some operations. The resolution is to update all operations directly or indirectly dependent on a resource when the metadata for that resource changes.

For custom adapters, either click through the adapter in the Adapter Endpoint Configuration wizard to regenerate the metadata or perform the following action:

  1. Open the META-INF/cloud-adapter.xml file.

  2. Add the following information parallel to sdkVersion. If the features tag already exists, add this one line inside it. 

    <features>
   <feature enabled="true">sdk.wsdl.regeneration</feature>
</features>

Delay in Message Processing of First Instance after Activation

After activation of an integration, there is a several second delay between when the instance is created and when the message is processed. This is the expected behavior.

On the first request for all integrations, the artifacts and backend processes are lazy loaded as a first step. This can make for a longer time period spent between the first two steps shown in the activity stream when compared to subsequent invocations. This is the expected behavior for the initial request, and by design.

For example, you may see the following in the activity stream:
Sep 3rd 12:43:16.442 PM UTC 2020 Message received: Instance create
Sep 3rd 12:43:23 Message processed by Trigger

Note the seven second delay between when the message was received and processed.