Import process for software knowledge base catalog data
The cron task instance that imports software knowledge base catalog data performs several tasks to retrieve and process data. Understanding these steps is useful when planning or troubleshooting problems.
Step 1: Retrieve the software catalog data file
The software catalog data file is retrieved from the directory specified in the TEMPDIR parameter of the SWCATALOGIMPORT cron task.
If you use the software catalog data file that is provided monthly by IBM, you must manually download the file to the directory specified in the TEMPDIR parameter.
If you have Tivoli® Software Knowledge Base Toolkit installed locally, you can configure Control Desk to retrieve the file from the knowledge base server automatically. The software catalog data file is retrieved from the server using HTTP. The file is saved in the working directory that was specified in the TEMPDIR cron task parameter. The file is first saved with a .zip extension, then unzipped to the name specified in the TEMPFILE cron task parameter. Each time the cron task is run, all files in the directory are overwritten. If the file is not refreshed, there is a problem with connectivity to the knowledge base.
Step 2: Perform XSL transformations
The data file is run through several XSL transformations, which produce the following additional files in the temporary directory.
- SwkbtManufacturers.xml
- SwkbtSoftware.xml
- SwkbtSwRel.xml
- SwkbtSwVariant.xml
- SwkbtSwCatalog.xml
- SwkbtSwPartNumber.xml
Each time the cron task is run, all files in the directory are overwritten. If the file is not refreshed, there is a problem with the transformations. To diagnose these problems, run the cron task with DEBUG logging enabled and send the log to customer support.
Step 3: Load files into Enterprise Services
Each file is loaded using the Enterprise Services configured to the TAMITEXTSYS External System. There is an endpoint configured to the external system, but it is not used in the loading of the files. You can test these loads by using the Data Import feature to process the files individually.
Step 4: Integration framework processing
The integration framework retrieves messages that are loaded into the JMS queues. The messages are processed individually against the business objects in the software catalog application, to make sure that the data conforms correctly to the latest business rules. This step is not controlled directly by the cron task instance for software knowledge base imports.