Planning that new WebSphere Process Server v6.0 GA production deployment? I suggest you check out the End Of Service Dates published by IBM for your versions to ensure that you aren’t painting yourself into a corner before you even start. Beware clicking that last link, it’s a poorly created web page with every single IBM product and revision; the worlds longest HTML ever.

End of Marketing: IBM stops actively selling it to customers

End of Support: IBM stops answering the phone when you call to complain about it. Of course, you can always enter into an extended contract to continue support of mission critical installations.

Quick links:

WebSphere

WebSphere Integration Developer

I tried to get other quick links but it was too frustrating with all the product versions inlined together.

Related Posts

• javax.xml.bind.JAXBException: Unable to locate jaxb.properties for package {0}
• IBM APAR PK49507: PROFILE RECREATION Required!
• Versioning business processes and human tasks in WebSphere Process Server
• Does Commerical Enterprise Software Suck?
• WebSphere Application Server: Non-Deterministic Error when configuring Default Messaging Activation Specs

From DeveloperWorks, What’s new in WebSphere Process Server V6.1.2

IBM WebSphere Process Server is a powerful runtime engine that can be used as the heart of a Service Oriented Architecture. It is built on WebSphere Application Server, and includes WebSphere ESB, enabling you to run integration modules created with WebSphere Application Developer, mediation modules, and J2EE applications. This article examines the features new to WebSphere Process Server V6.1.2. Basic knowledge of WebSphere Process Server is required for this article.

Goes a step beyond the lis of features that I posted earlier with some user interface screenshots.

Related Posts

• What’s new in WebSphere Integration Developer V6.1.2
• What’s new in IBM WebSphere Enterprise Service Bus V6.1
• Handling unmodeled faults within WebSphere Process Server V6.1
• WebSphere Enterprise Service Bus does not support Event Sequencing
• Building clustered topologies in WebSphere Process Server V6.1

From DeveloperWorks: Configuring IBM WebSphere Process Server V6.1 with an Oracle Database

Redpaper, published: Thu, 17 Jul 2008 This IBM® Redpaper publication explains how to configure IBM WebSphere® Process Server V6.1 to work with an Oracle® database.

There's always a thin amount of documentation on integrating IBM products with non-DB2 databases. Here is a redpaper on using Oracle for WebSphere Process Server.

Related Posts

• Oracle SOA Suite 11g - Hello Competition!
• Integrating Oracle BPEL with an HTTP (REST) Service
• IBM Redbooks: WebSphere Process Server and WebSphere Integration Developer
• WebSphere Process Server and RPC/Literal Web Services: Beware Messages defined by ‘type’ instead of ‘element’.
• WebSphere Business Integration V6.1 Performance Tuning

The ability to roll back a update is a nifty idea. The theory is that you installed something that broke your platform. You can now just ‘undo’ the install.

In practice, it’s a feature that next to nobody uses. It requires a deep-faith in the programmers of the rollback logic to return the system to it’s initial state. If they missed a single file or java class, it could land up affecting the entire system. The user is then left with a question “Is the environment broken or is there a bug in the platform?”. In order to resolve this question, they will uninstall and re-install the product.

Therefore, if you know that you will never bother to rollback, why are you dedicating disk space for the installation manager to rollback? Yup, that’s the default setting. Every old version of every plugin in every product you’ve ever installed is still there needlessly. For people like me who work in virtual machines, disk space is precious.

You can turn this open off by clicking File -> Preferences.

You can turn off the check box to preclude future saving. You can also click the Delete saved Files button to remove any current old versions.

Before I clicked the button, it told me I had 1.6G of files saved.

Related Posts

• IBM Software Support Lifecycle - General Availability, End Of Marketing, End of Support Dates
• SCA, BPEL, Websphere Adapter for JDBC and Transactionality
• WB552: Random DataPower Thoughts Part 3
• Techniques for Managing Large WebSphere Installations
• Problem Determination and WebSphere Integration Developer

I started my update from 6.1 to 6.1.2 and the process took two hours (not including download times).  That’s unacceptable. Most of the time was spent installing the WebSphere Process Server fixes.

The UI appeared frozen for about an hour (still responsive, just that the progress bar didn’t move). When I jumped down to the logs, I saw that it was reporting it’s status properly 10%.. 20%. I have no idea why this information isn’t reflected back into the UI.

There’s no way a complete User-centered design cycle was completed on this installer. At the end of the day, I don’t really mind that it took two hours (a warning would be nice). But I absolutely hate it when I’m left wondering if my installer is dead or hung and if I should kill it for an hour.

Related Posts

• Venting some WebSphere Integration Developer Steam
• Upgrading WebSphere Integration Developer v6.1 - Installation Manager
• Close WebSphere Integration Developer when updating Fixpacks for WebSphere Process Server
• Getting Started with WebSphere Process Server and WebSphere Enterprise Service Bus Part 3: Runtime
• Seamless access to SAP services using IBM WebSphere integration tools

Cheers to David Currie for pointing out that while the WebSphere Integration Developer user interface allows you to specify an event sequencing qualifier inside a mediation module, if you attempt to deploy the solution to an WebSphere Enterprise Service Bus Server, you’ll get exceptions.

Interestingly enough, if you deploy the same Mediation Module to WebSphere Process Server, then it will work because Process Server is a superset of WESB and has the required runtime.

I highly doubt that this was an intentional UI design decision in WID.

Related Posts

• BPEL or ESB: Which should you use?
• WebSphere Enterprise Service Bus - Mediation Flow Component: No Refactoring for WSDL Faults
• What’s new in IBM WebSphere Enterprise Service Bus V6.1
• What’s new in WebSphere Process Server V6.1.2
• SOA Jam: Unify The Numerous Enterprise Service Bus Products

V6.1.2 has been released for WebSphere Integration Developer and WebSphere Process Server.

You can read the list of features. I’m going to highlight the ones that interest me:

WebSphere Process Server

• New out-of-the-box, ready-to-run, Web 2.0 BPM client for business users with configurable work lists and detailed work item views, including support for collaboration using prebuilt human tasks
• Enhanced flexibility for modifying the flow of in-flight process instances with the ability to skip activities, as well as to jump forward and backward between activities
• Enhanced trace to capture what is flowing through the server at runtime for later display in the WebSphere Integration Developer Integrated Test Client with the full power of the test client for problem determination and analysis of application logic
• Necessary database scripts for creating database instances on remote servers now available from the installation media directly, eliminating the need to install WebSphere Process Server to gain access to the scripts
• Adds new format support for the following additional message formats:
  • Delimited and full support for Comma Separated Values (CSV)
  • Fixed-width format
  • JavaScript Object Notation (JSON)

WebSphere Integration Developer

• Enhances visual tracing by showing more fine-grained events and a visual execution path within the implementation when testing state machines, processes, and mediation flows.
• Delivers a new capability for testing XML maps in isolation, using the local built-in XSLT engine for a rapid iterative development experience. Can also set breakpoints and debug XML maps locally using the Test client.
• Improved support for arrays in the XML map editor.
• Shows which files are pending on a publish to a server operation.
• Supports refactoring of interface parameter names.
• Helps improve traceability of human tasks and business rules. Supports the ability to compare and merge human task and business rule changes in a business process imported from WebSphere Business Modeler. Provides support for business process annotations from WebSphere Business Modeler.

WebSphere Enterprise Service Bus

• Improved error messages, logging, and First Failure Data Capture (FFDC) usage. FFDC can reduce defect resolution times and the number of times IBM Support asks a client to recreate a problem with different trace settings turned on.
• Adds support for manipulating MQCIH message headers for WebSphere MQ, which are exploited by the CICS bridge to enable interaction with CICS applications.

Related Posts

• Whats new in WID/WPS v6.1
• What’s new in WebSphere Integration Developer V6.1.2
• Taking component testing to the next level in WebSphere Integration Developer
• WebSphere Integration Developer: Struggling with the Business Object Mapper
• Versioning business processes and human tasks in WebSphere Process Server

A technote was published for customers who use the WebSphere Adapter for JDBC for inbound processing of an SQL Server 2000 database. Apparently, the adapter will lock the entire event table (bad!).

The ‘fix’ aka workaround is to use a view to the event table and then set a custom property inside the adapter.

Related Posts

• WebSphere Adapter for JDBC: SetObjectKeys NullPointer Exception
• JDBC Adapter Export with events that comprise multiple rows
• WebSphere Adapter for JDBC Export - ‘InvalidVerbException’
• SCA, BPEL, Websphere Adapter for JDBC and Transactionality
• WebSphere Adapter for JDBC: Superfluous RecordNotFound Exceptions

In a cluster where the messaging engine has failed over between nodes, a situation could arise where the transaction log (’tranlog’) has become corrupt or is unrecoverable. In this situations, you may see messages on SIB Queues that are in the LOCKED state but are unprocessed. If you factor in a sequential delivering of messages, we may see that the queue refuses to process any messages at all.
In this case, we have to look at the SIB Data store and run the following query:

select count(*) from sib000 where XID is not null

if you discover existing values in XID you should update XID field to null:

update sib000 set xid = null where xid is not null

This situation would only occur if the ‘tranlog’ on a server was unrecoverable and should not be considered a ‘normal’ step of server recovery or server startup.

Related Posts

• Event Sequencing Message Locking - Resolution!
• WebSphere Application Server: -recovery flag
• Tip: WebSphere Process Server - Do not delete the ‘tranlog’ on a production server
• Asynchronous SCA One-Way Operations And How To Recover
• WebSphere Default Messaging (SIBus) and WebService Error Handling

In development mode, one of the steps to ‘clean up’ from a wonky process server instance is to delete the wstemp, temp directories along with the ‘tranlog’. The ‘tranlog’ is an internal file that WebSphere Application Server uses to manage in-flight transactions and attempt to recover them should the server crash. When you delete the tranlog, you remove this information from process server’s memory. At development time, your transactions are transient, you don’t really care if they complete or not, a new one can be created at any time.

On a production server, though, each transaction should be considered an extremely precious resource that can’t be deleted. Therefore, the tip that some of us use to delete the tranlog should NEVER be done on a production server. Removing it may lead to an inconsistency in your production database which may be unrecoverable.

Related Posts

• WebSphere Application Server: -recovery flag
• WAS: Resolution to Messages in the SIB locked forever
• WebSphere Process Server Endurance
• SCA, BPEL, Websphere Adapter for JDBC and Transactionality
• BPEL: Beware the use of nested loops in a short running process

A handy resource for seeing what exactly is contained in a new release of software are the APAR fix lists published by IBM. They contain a list of every APAR that is resolved and you can click to get more information.

You can find them for:

• WebSphere Application Server 6.0.2
• WebSphere Application Server 6.1
• WebSphere Process Server 6.0.2
• WebSphere Process Server 6.1

Related Posts

• Pre-review: WebSphere Business Integration Primer
• IBM Installation Manager Tip: Don’t “Save Files For Rollback”
• Versioning business processes and human tasks in WebSphere Process Server
• IBM Redbooks: WebSphere Process Server and WebSphere Integration Developer
• Tip: WebSphere Process Server - Do not delete the ‘tranlog’ on a production server

From DeveloperWorks, WebSphere Business Integration V6.1 Performance Tuning

Draft Redpaper, last updated: Mon, 9 Jun 2008 - Learn valuable tips for tuning - Get the latest best practices - Try the example settings This IBM® Redpaper was produced by the IBM WebSphere® Process Server, WebSphere Enterprise Service Bus, WebSphere Adapters, and WebSphere Business Monitor performance teams in Austin Texas, Böblingen Germany, and Hursley England.

This book attacks performance on three levels, Architecture, Developerment and Performance Tuning. All key issues that need to be considered to create systems that can meet Service Level Agreements (SLA). I'm happy to see this kind of information get published so close to the public release of a version.

Related Posts

• IBM Redbooks: WebSphere Process Server and WebSphere Integration Developer
• Pre-review: WebSphere Business Integration Primer
• Configuring IBM WebSphere Process Server V6.1 with an Oracle Database
• WebSphere Integration Developer: Struggling with the Business Object Mapper
• WebSphere Process Server Endurance

I had a custom snippet in my map between two string attributes and a list of business objects. The idea being that each string from the source should becomes an entry in the business object list with a wrapper.

No problem. I wrote up the snippet to create the destination business object (since it does not get initialized for you), added my wrapper data objects to my list. Everything looks good. I have “Print to Log” (aka system.out) to look at my resultant business object and it looks perfect.

Except when I run it, the List doesn’t get populated.

The key to my problem was how I was running it. You can’t test a business object map on its own, it’s always a part of some grander module ecosystem. In my case, it was a mediation flow component wedged between two XSLT transforms.

What I discovered was that it was a missing map in my XSLT that caused my list to remain null. The business object map was fine the entire time.

What’s the story? Remember that there are some components that you cannot test in complete isolation.

Related Posts

• WebSphere Integration Developer: Visual Snippet Java Code Generation Is Non-Deterministic
• Business Object Maps and ‘Required’ Attributes
• Whats new in WID/WPS v6.1
• Taking component testing to the next level in WebSphere Integration Developer
• My Integration Test Client DeveloperWorks Article

From DeveloperWorks, Use ARM to monitor SCA invocations in IBM WebSphere Process Server

This two-part series shows you how to monitor Service Component Architecture (SCA) invocations using the Application Response Measurement (ARM) standard in IBM WebSphere Process Server V6.1. You can use an ARM implementation, such as IBM Tivoli Composite Application Manager for Response Time Tracking, to generate a graphic view of SCA invocations. This article, Part 1 of the series, starts by describing ARM and showing you how to debug synchronous scenarios using Tivoli Composite Application Manager for Response Time Tracking. In Part 2, you'll get an introduction to SCA invocation patterns and learn how to debug asynchronous scenarios.

Finally, something that can tell you at the SCA level how your components are running on a production server. This is an important step in the maturation of SCA.

Related Posts

• Web services with SOAP/JMS in IBM WebSphere Process Server
• WebSphere DataPower vs WebSphere Enterprise Service Bus
• BPEL: Beware the use of nested loops in a short running process
• Getting Started with WebSphere Process Server and WebSphere Enterprise Service Bus Parts 1 & 2
• Asynchronous SCA One-Way Operations And How To Recover

We had a WebSphere Adapter for JDBC export in our module, and it was throwing the following exception:

[5/27/08 17:55:52:315 EDT] 0000006b NUXInResource I PollEventManagerWorker run()
CWYBS0011I: Polling has started.  UserAction=No action is required.
[5/27/08 17:55:53:356 EDT] 0000006d NUXInResource E com.ibm.j2ca.extension.eventmanagement.internal.EventSender
 getObjectForEvent(Event) EventStore impl
 (com.ibm.j2ca.jdbc.inbound.JDBCEventStoreWithXid@6eec3b01)
 threw an unexpected and unchecked exception
javax.resource.ResourceException: EventStore impl
(com.ibm.j2ca.jdbc.inbound.JDBCEventStoreWithXid@6eec3b01) threw an
unexpected and unchecked exception
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventManager.getObjectForEvent(EventManager.java:658)
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventSender.getRecordForEvent(EventSender.java:365)
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventSender.doSendEvent(EventSender.java:237)
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventSender.sendEvent(EventSender.java:158)
      at
 com.ibm.j2ca.extension.eventmanagement.internal.EventListSender.sendEvents(EventListSender.java:113)
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventListSender.run(EventListSender.java:90)
      at
com.ibm.ejs.j2c.work.WorkProxy.run(WorkProxy.java:394)
      at
com.ibm.ws.util.ThreadPool$Worker.run(ThreadPool.java:1471)
Caused by: java.lang.NullPointerException
      at
com.ibm.j2ca.jdbc.inbound.JDBCEventStoreWithXid.setObjectKeys(JDBCEventStoreWithXid.java:1703)
      at
 com.ibm.j2ca.jdbc.inbound.JDBCEventStoreWithXid.getRecordForEvent(JDBCEventStoreWithXid.java:1449)
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventManager.getRecordForEvent(EventManager.java:680)
      at
com.ibm.j2ca.extension.eventmanagement.internal.EventManager.getObjectForEvent(EventManager.java:649)
      ... 7 more

What this exception is saying is that in the process of trying to get an ‘Object’ for the given event (the object is a record in a table for the jdbc adapter) it could not find the primary key.

The solution was to re-generate the XSDs for the table using Enterprise Service Discovery and set the primary key attribute. You then will have to uninstall the current ‘bad’ adapter instance from your server and restart your server. Install the new fixed adapter module and everything will work.

For more information, you can reference this thread from DeveloperWorks.

Related Posts

• WebSphere Adapter for JDBC Export - ‘InvalidVerbException’
• WebSphere Adapter for JDBC Export and SQL Server 2000: Table Locking while Polling
• JDBC Adapter Export with events that comprise multiple rows
• WebSphere Adapter for JDBC: Superfluous RecordNotFound Exceptions
• UPDATE: WebSphere Adapter for JDBC: Superfluous RecordNotFound Exceptions