Advertisements
RSS

Tag Archives: Oracle

Error in getting XML input stream with Oracle Business Rules 12.2.1

When trying to compile a Oracle ACM/BPM 12.2.1 project (with Oracle Business Rules) the following message throws up: “Error in getting XML input stream”

Jdev_Error

When Oracle Business Rules 12.2.1 generates it’s default XSD it uses the full system path instead of a relative path for it’s imports. So make sure to manually change the import configuration.

XSD_file

Advertisements
 
2 Comments

Posted by on 12-02-2016 in Oracle

 

Tags: , , ,

Oracle Service Bus 11g – Another session operation is in progress. Please retry later.

Since our Bamboo pipeline uses the same account for deployments we sometimes get this error “Another session operation is in progress. Please retry later.” during deployment.
So 1st of all, don’t use the same user (like weblogic) to deploy when your with a lot of developers / build pipelines / etc.
Time for us to use multiple users for each agent (or think of something else really smart), but in the meanwhile we sometimes have to discard the changes from an open session.

sbconsole -> View all Sessions -> Select session -> Discard all changes

2

1

However sometimes the sbconsole does not allow this action and we need to restart the Admin + Managed server.
Usually this fixes the issue, however this time it didn’t.

Luckily for us the great Pierluigi has a blog which tells us to clear the OSB domain session folder, so:

  • shutdown the Admin + Managed servers
  • clear all the SessionXXX folders in the ${DOMAIN_HOME}/osb/config/sessions
  • start the environment

The content of

[oracle@server sessions]$ ls -l
drwxr—–. 5 oracle oinstall 4096 Jan 13 13:22 SessionScript1452682586592
drwxr—–. 5 oracle oinstall 4096 Jan 13 13:22 SessionScript1452687725653
[oracle@server sessions]$ rm * -rf

References

 
Leave a comment

Posted by on 13-01-2016 in OSB

 

Tags: , ,

Using the Weblogic External Listen Address to support Network Address Translation (NAT) firewalls

When trying to connect or deploy from JDeveloper 12.2.2 to our Oracle Fusion Middleware 12.2.1 domain in the Amazon EC cloud I keep having connection problems. Contacting the consoles is not a problem, however extending the IDE Connection results in this error:

Pic0

t3://127.0.0.1:7011: [RJVM:000575]Destination 127.0.0.1, 7011 unreachable.; nested exception is:
java.net.ConnectException: Connection refused: connect; [RJVM:000576]No available router to destination.; nested exception is: java.rmi.ConnectException: [RJVM:000576]No available router to destination.Dec 08, 2015 9:50:35 AM oracle.tip.tools.ide.soabrowser.LogUtil logStackTrace

And deploying an artifact to the server results the same

PicError

Weblogic configuration

I couldn’t find anything regarding the error on Oracle Support, but luckily my collegue Daljit Singh had the answer. Since the Amazon EC2 uses a public IP (which we use to connect to the admin server) the internal passthrough to the Managed Servers fails. To solve this we should use the Weblogic “external listen address” configuration. The external listen address and port are used to support Network Address Translation (NAT) firewalls. These should match the IP address or DNS name that clients use to access application on the server.

Go to the Weblogic console -> Environment -> Servers -> Managed Server -> Configuration -> General -> Advanced

Make sure the public ip-adres is stored in the External Listen Address here

Pic2

Your managed server requires a restart afterwards. But then the connection issue is solved.

restart

 
2 Comments

Posted by on 08-12-2015 in Weblogic

 

Tags: , , ,

JPS-01050: Opening of wallet based credential store failed

After installing a new Oracle Fusion Middleware 12.2.1 domain on a Ubuntu server (for development purposes) and starting the AdminServer I get the following error:

<Critical> <WebLogicServer> <BEA-000362> <Server failed. Reason: There are 1 nested errors: oracle.security.jps.JpsException: JPS-01050: Opening of wallet based credential store failed. Reason java.io.IOException at oracle.security.jps.internal.config.OpssCommonStartup.preStart(OpssCommonStartup.java:334)
at oracle.security.jps.JpsStartup.preStart(JpsStartup.java:286) at oracle.security.jps.wls.JpsBootStrapService.start(JpsBootStrapService.java:80)

However, when checking the cwallet file it is there with proper access rights

ubuntu@ip-10-0-1-170:/opt/oracle/config/domains/rbx_dev/config/fmwconfig$ ls -l cwallet.sso
-rw——- 1 ubuntu ubuntu 194 Dec 1 13:11 cwallet.sso

So when searching we found this Oracle Support Doc ID 1923395.1

Unable Start AdminServer: JPS-01050: Opening of wallet based credential store failed. The FMW WebLogic Server (WLS) installation has been configured to use a non-default Java temporary files directory, i.e. the following has been set in the WebLogic startup or setDomainEnv.sh script:

EXTRA_JAVA_PROPERTIES=”-Djava.io.tmpdir=/appl/oracle/temp_java_files ${EXTRA_JAVA_PROPERTIES}”

Reference: How to Change the WebLogic Server Location for Temporary Files (Doc ID 1336002.1)
When the Middleware home was restored the directory specified by java.io.tmpdir parameter was missing,
Therefore an IOException occurred when opening the wallet and WLS was unable to initialize the OPSS successfully.

The description however is not completely accurate for our specific problem, but pointed us in the right direction. Since in our case the default /tmp folder is owned by root on Ubuntu and the “normal” ubuntu:ubuntu user/group running the Weblogic scripts has no access.

So we could fix the issue in 2 ways:

    1. Using a custom tmp folder in our setDomainEnv.sh script which the ubuntu user had access
      ## CUSTOM FOR RBX_DEV ##
      EXTRA_JAVA_PROPERTIES=”-Djava.io.tmpdir=/opt/oracle/tmp -Djava.security.egd=file:/dev/./urandom ${EXTRA_JAVA_PROPERTIES}”
      export EXTRA_JAVA_PROPERTIES
      ## CUSTOM FOR RBX_DEV ##
    2. Giving access to the default /tmp folder for our ubuntu user
      sudo chmod o+rwx /tmp
 
1 Comment

Posted by on 08-12-2015 in Weblogic

 

Tags: , ,

How to use dynamic validate in Oracle Service Bus 12c

One of the new features introduced in Oracle Service Bus 12c is the ability for dynamic validation. A feature which can be used to validate a message against a WSDL or XSD schema file which is both explained here by Oracle. The example on the Oracle website shows this XML code to validate against a XSD:

<validate xmlns="http://www.bea.com/wli/sb/context">
<schema>default/MySchema</schema>
<schemaElement>
<localname>MyElementType</localname>
</schemaElement>
</validate>

However I tried “playing” with the new feature but couldn’t get it to work. The error I got was:

<soap:Fault>
<soap:Code>
<soap:Value>soap:Receiver</soap:Value>
</soap:Code>
<soap:Reason>
<soap:Text xml:lang="nl">OSB-382524: Failed to perform validation</soap:Text>
</soap:Reason>
<soap:Detail>
<con:fault xmlns:con="http://www.bea.com/wli/sb/context">
<con:errorCode>OSB-382524</con:errorCode>
<con:reason>Failed to perform validation</con:reason>
<con:location>
..
</con:location>
</con:fault>
</soap:Detail>
</soap:Fault>

After contacting Oracle Support I received the following info:

  • BUG 20367846: Internal Documentation  – Validate action to dynamically select a schema – documentation is not clear
  • BUG 20380158: Validate action – Dynamic validation throws NPE

A patch for BUG 20380158 was then released for Oracle Service Bus 12.1.3, which then could be downloaded from Oracle Support:

ppatch

I downloaded patch 20380158 and used opatch to install it on my DEV environment.


..\p20380158_121300_Generic\20380158&amp;gt;opatch apply
Oracle Interim Patch Installer version 13.2.0.0.0
Copyright (c) 2014, Oracle Corporation.  All rights reserved.

Oracle Home       : C:\ORACLE\middleware_12.1.3
Central Inventory : C:\Program Files\Oracle\Inventory
from           : n/a
OPatch version    : 13.2.0.0.0
OUI version       : 13.2.0.0.0

OPatch detects the Middleware Home as &quot;C:\ORACLE\middleware_12.1.3&quot;

jul 09, 2015 12:48:49 PM oracle.sysman.oii.oiii.OiiiInstallAreaControl initAreaControl
INFO: Install area Control created with access level  0
Applying interim patch '20380158' to OH 'C:\ORACLE\middleware_12.1.3'
Verifying environment and performing prerequisite checks...
All checks passed.

Please shutdown Oracle instances running out of this ORACLE_HOME on the local system.
(Oracle Home = 'C:\ORACLE\middleware_12.1.3')

Is the local system ready for patching? [y|n] y
User Responded with: Y
Backing up files...

Patching component oracle.osb.server, 12.1.3.0.0...
Patching component oracle.osb.server, 12.1.3.0.0...

Verifying the update...
Patch 20380158 successfully applied
OPatch succeeded.

After the patch the error was still their but I received a more detailed error message in my logging:

<soap:Fault>
<soap:Code>
<soap:Value>soap:Receiver</soap:Value>
</soap:Code>
<soap:Reason>
<soap:Text xml:lang="en">OSB-382571: Schema element: GetCaseRequestMessage not found</soap:Text>
</soap:Reason>
<soap:Detail>
<con:fault xmlns:con="http://www.bea.com/wli/sb/context">
<con:errorCode>OSB-382571</con:errorCode>
<con:reason>Schema element: GetCaseRequestMessage not found </con:reason>
....
</con:fault>
</soap:Detail>
</soap:Fault>

After more contact with Oracle Support it indeed appeared that the Oracle documentation was not correct (as mentioned earlier with their reference to BUG 20367846). The XML structure example shown by Oracle also requires an nameSpaceURI element.  Which means the correct XML input for dyanamic validation for an XSD is:

<validate xmlns="http://www.bea.com/wli/sb/context">
<schema>mySBproject/schema/myXsdSchemaFile</schema>
<schemaElement>
<namespaceURI>nl.rubix.messages/caseservice/1.0</namespaceURI>
<localname>getCaseRequestMessage</localname>
</schemaElement>
</validate>

Example Oracle Service Bus Project with Dynamic Validation

Here is an example of my SB project:

pipeline

I first use an Assign action in my pipeline to create the $dynValidate variable with XQuery which holds the XML structure for the validation. (You could also use the xquery directly on the validate action).

assign_xquery

Here is an example XQuery to generate the required Dynamic Validate XML structure:


xquery version "1.0" encoding "utf-8";
(:: OracleAnnotationVersion "1.0" ::)
(:: pragma parameter=”$i_operation” type=”xs:string” ::)

declare function local:dynValidate($i_operation as xs:string) as element()
{
<validate xmlns="http://www.bea.com/wli/sb/context">
<schema>mySBproject/schema/myXsdSchemaFile</schema>
<schemaElement>
<namespaceURI>nl.rubix.messages/caseservice/1.0</namespaceURI>
{
if ($i_operation = "getCase")
then <localname>getCaseRequestMessage</localname>
else if ($i_operation = "insertCase")
then <localname>insertCaseRequestMessage</localname>
else ()
}
</schemaElement>
</validate>
};

declare variable $i_operation as xs:string external;
local:dynValidate($i_operation)

After we have the $dynValidate variable we can use it as input for the Validate action. (As mentioned, you can also directly use the Xquery here instead of an expression with the earlier generated variable.

validate

Hope it helps!

References

 
Leave a comment

Posted by on 29-07-2015 in Oracle, OSB

 

Tags: , ,

How to withdraw tasks and handle them in Oracle BPM

If you want to withdraw a Human Task in Oracle BPM there are multiple options. This blog post will first show 3 options and then explain how to model your Oracle BPM process to anticipate on the result of a withdrawn task.

1. The BPM Workspace

Using the default Oracle BPM WorkSpace to withdraw a task:

withdraw5

2. Using the TaskService

If you use a custom front-end (instead of the default Oracle BPM Workspace) you can use the Oracle SOA Suite TaskQueryService & TaskService to handle your tasks. In the TaskService there are operations named withdraw and withdrawTasks to either withdraw 1 or multiple tasks. The example below shows a withdrawTask operation request message.


<soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/"
 xmlns:tas="http://xmlns.oracle.com/bpel/workflow/taskService"
 xmlns:task="http://xmlns.oracle.com/bpel/workflow/task"
 xmlns:com="http://xmlns.oracle.com/bpel/workflow/common">
 <soapenv:Header/>
 <soapenv:Body>
 <tas:withdrawTask>
 <com:workflowContext>
 <com:credential>
 <com:login>weblogic</com:login>
 <com:password>welcome1</com:password>
 <com:onBehalfOfUser>weblogic</com:onBehalfOfUser>
 </com:credential>
 </com:workflowContext>
 <tas:taskId>a71a4c2b-b121-4cde-91af-79872dbda522</tas:taskId>
 </tas:withdrawTask>
 </soapenv:Body>
</soapenv:Envelope>

This solution is implemented in our Adaptive Case Management project in the case a specific (power)user is logged in and wants to withdraw the current task (and activity) where he/she is working on.

3. Using events/signal

When you want to withdraw a task without actually opening a task (or having access to it’s taskId) you can use a signal subprocess to receive an event.

withdraw4

The event can trigger an Update Task action which calls the withdraw operation from within the BPM process.

withdraw3

The Oracle BPM Process modeling

When using the withdrawTask the task state is set to “WITHDRAWN” when the task is closed. So make sure to use the data association on the output tab of the task to store the state in a process data object.

withdraw1

In the BPM process flow we can use an exclusive gateway to model the logic by checking the value of the taskState process data object.

withdraw2

Often we use gateways in Oracle BPM processes to check the outcome of a human task and model the outcome to different process flows. However keep in mind that the state can also inform about e.g. EXPIRED, ERROR or WITHDRAWN tasks.

withdraw7

References

 
2 Comments

Posted by on 18-06-2015 in BPM, Oracle

 

Tags: ,

How to customize your Service Bus 12c pipeline templates

One of the new features in Service Bus 12c is the ability to use pipeline templates. Usually the Oracle Service Bus pipelines in an environment have many common steps. Think of the re-use of logging, error handling, alerts and pattern + naming convention for your stages. In practice with OSB 11g we often used a “template” or existing OSB project which we then copied and modified. With the “clone” option of 12c this task is already easier, but the use of pipeline templates is even better. Since templates and concrete pipelines (generated pipelines from a template) remain linked we can update our services easier with new insights. For example, when you want to change your default logging or fault handling behavior.

Index:

To use pipeline templates to their fullest potential we can customize them to our own needs. For this we have multiple options.

Generic configuration

The essential configuration of most actions can be left empty in the template without any problem. As soon as they are implemented in concrete pipelines the actions there will come into an ERROR state. The example below shows the Routing which is empty in the pipeline template. When developing the concrete pipelines we can then easily set the correct business service.

Empty

Many actions, like Validate, allow us to use generic configuration of the Location instead of specific elements from the service datamodel. By using generic configuration as ./*[1] we can tell the Validate action to validate the 1st element in the SOAP body. Which normally would be your getMessageRequest or getMessageResponse element. The correct element and schema  to validate against should still be selected in the concrete pipeline.

validate

Dummy values

Unlike most actions, the problem with the Replace action seems that when you leave the XSLT or XQuery of choice empty, the template will turn into an error state. To prevent this from happening I use a dummy XQuery file in my CommonObjects project and assign it from all Replace actions I use for transformation of messages..


transform

Locking actions

We can set each individual action in our template to a “locked” state. With this preventing the developer of the concrete pipeline to change certain locked logic of the service. This is quite usefull for default logging and fault handling you want to enforce in your services. Right-click on each specific action you want to lock and choose “Lock Action”. Sadly there is no graphical representation to quickly show you which actions are locked.

logging

Using template placeholders

We can use template placeholders to allow developers using the template to customize the concrete pipeline to the extend that we determine in the template. We can allow customization of unlimited stages, nodes and route, but in most cases you will probably use the most strict placeholder “Actions”. In the example below we show a stageEnrich which allows the developer to add unlimited amount of actions in this stage. Actions like Service Callouts, Transforms, Logging, etc to make sure we can extend our service. With this, creating enough flexibility for the developer to customize the logic for the concrete pipeline, without changing the default we want to enforce for logging, error handling, etc.

d01

 

Enrich

References

 
4 Comments

Posted by on 09-06-2015 in OSB

 

Tags: , ,