Column and stage variable order not preserved when saving DataStage job
The DataStage Designer client in Information Server Version 8.5 and Version 8.5 Fix Pack 1 does not properly save the column order and stage variable order as part of the job design information in the repository.
The DataStage Designer client in Information Server Version 8.5 and Version 8.5 Fix Pack 1 does not properly save the column order and stage variable order as part of the job design information in the repository. The next time the job is reloaded from the repository, the columns or stage variables may appear in a different order. This can result in the following problems:
- Jobs that write to sources or targets where column order matters (such as sequential file stages) might produce a file that would be unreadable by another job or application that was expecting a certain column order.
- Jobs that contain nested stage variable derivations (one stage variable depends on the value of another) would return incorrect results.
These problems would not necessarily manifest in an error or warning message, thus allowing the problem to go undetected to the user. IBM has classified these as s High Impact (HIPER) APARS. IBM recommends all users to install the patch for JR39983 which will prevent these problems listed above. The patches must be installed on each client workstation computer where the DataStage Client software has been installed.
|Information Management||InfoSphere DataStage||Windows||126.96.36.199, 8.5|