IBM Support

InfoSphere Information Server Pack for SAP BW version 4.3.2.3: Known limitations and workarounds

Question & Answer


Question

What are the known limitations and workarounds for the Pack?

Answer

InfoSphere Information Server Pack for SAP BW version 4.3.2.3 has the following known issues and limitations:


Installation

  • On AIX systems, installing in console mode generates a warning message about an invalid size 0 file, but the Pack installs successfully; referenced in technote 1416273.
  • When you upgrade from Pack 4.1, the InfoPackage setting in Load stages is not retained.
  • If you are installing the Pack by using the graphical installer, there is no notification if the Pack is already installed. However, the console installer does detect and report this. Both installation methods finish successfully in this case, with a warning message in the installer log (/opt/IBM/InformationServer/logs/sapbw_server_install.log).
  • When installing the Pack on a multi-tier InfoSphere Information Server installation with the Services tier on Windows, an error might occur as the result of a missing Windows registry key. For more information, see technote 1442306.
  • In multi-tiered Information Server installations where the Service tier is installed on a different computer than the Engine tier, you have to manually create the following folder before performing the Pack installation on the Service tier:
    <Information Server folder>/Server (e.g. /opt/IBM/InformationServer/Server)
    Otherwise, the installation will abort with the following error:
    ERROR! The IBM InfoSphere Information Server directory is not valid.
    Note: Installation on the Service tier is needed for deploying the Information Center (Help) files.
  • When you install the Pack on the Client tier of Information Server 11.3.x, the installation program does not create a shortcut to the Datastage Administrator for SAP in the Windows start menu. A shortcut is only created on the desktop. It is recommended that you manually copy the desktop shortcut to the "IBM Information Server" program group after the installation.

Running jobs

  • When sending data from an empty file to SAP using the Load stage, the results are determined by your SAP configuration. One packet is always sent by the stage.
  • In both the Extract and Load stages, a job may fail with the error "Unknown project name".
    Workaround: Select the correct Project in DataStage Administrator for SAP > Source System Properties > DS Job Options. If the error persists, delete and recreate the Source System that the job uses.
  • When the wait time for an InfoSpoke expires, Information Server starts the job. However, after the requested InfoSpoke finishes processing, the InfoSpoke starts the same IBM InfoSphere Information Server job, even though the job is not scheduled.
  • When a parallel load job is running in true parallel mode, several NULL records can be sent to SAP . The architecture of the parallel engine requires each node running the job to send a "LAST PACKET" indicator to SAP . Because parallel jobs provide no advance notice that a node has completed its work, the nodes send a ZERO row packet to SAP as the LAST PACKET. SAP BW considers the ZERO row to contain a NULL row.
  • Load jobs with more than one input link finish successfully, but the job log contains the following message:
    When checking operator: Passive stages should have either one input link or one output link.
    You can ignore the message.
  • JR36889: The Load stage jobs complete successfully, but data of the type "float" is not loaded to SAP, resulting in the following error message:
    Transformer_1,1: APT_Decimal::ErrorBase: From: the source decimal has even precision, but non-zero in the leading nybble, or is too large for the destination decimal.Record dropped.
  • JR38631: A Load job with a parameter used as the SAP password fails.
  • Job reset tool: The "SAP system number" (or instance number) always defaults to 0. This only affects Extract jobs. The issue is present in versions 4.3.2.1 through 4.3.2.3. Internal tracking number: 223135.
    Workaround: To clean up artifacts from failed Extract jobs, refer to technote 1441967. Complete the steps listed in the technote under Manual Cleanup, option B.
  • Temporary files are not completely deleted after a failed execution (both Load and Extract). This behavior is by design to allow detecting if the job was initiated from SAP BW or Datastage. In very rare instances, this may impact functionality and the job will be unable to start again after a failed run, even after it is reset using the Director client. Please refer to this technote for cleanup instructions. This issue will be addressed in a future release (internal tracking number: 218431).

Known limitations

  • JR42709: SAP BW stages cannot recognize SAP BW 7.x DataSources.
  • The Pack does not support multibyte metadata.
  • You cannot run two InfoPackages that use the same source system and InfoSource.
  • When you change a source system ID in SAP, the change is not reflected in the SAP BW stages.
  • Extract jobs on parallel canvas with nodes running in Grid Format (nodes on different host systems) are not supported.

[{"Product":{"code":"SSZJPZ","label":"IBM InfoSphere Information Server"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":"Pack for SAP BW","Platform":[{"code":"PF002","label":"AIX"},{"code":"PF010","label":"HP-UX"},{"code":"PF016","label":"Linux"},{"code":"PF027","label":"Solaris"},{"code":"PF033","label":"Windows"}],"Version":"4.3.2.3","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]

Document Information

Modified date:
16 June 2018

UID

swg21695278