FI Parked Doc workflow in 4.6c

[Susan R. Keohan]

Hi group,
 
First, thanks to those of you who offered help with debugging this problem.
As it turns out, the check FMs in question do an import from memory to get
the FIPP ID (probably to avoid problems with timing...), and the export to
needed to be done in an update task.  In 4.5, we did not need to do the
export in an update task.
 
So, as a heads up, if your imports from memory are failing, check out how
the export is being done.  I can't say that it always needs to be done in
an update task, but that certainly was the case here.
 
Cheers,
Sue
 
>Greetings Workflow-ers,
>
>Before you stop reading, this is not my typical plea for speakers or
>presentation ideas for our next ASUG conference... but IF you are
>interested, please send me e-mail at skeohan at mit.edu !!  Time is running
>out for planning this conference, and there are some sessions that need
>speakers at this time.
>
>We are upgrading from 4.5 to 4.6c.  We have workflows which should kick
>off when an FI Parked document is created.  There are 3 entries in the
>event linkage tables for this event (FIPP CREATED).  For each entry, there
>is a Check Function Module, which will determine what the document type
>is, and based on that value, the correct workflow will be started. (Only
>one of three should start for any given doc type).
>
>So, the documents are created, sometimes manually, sometimes via a feed.
>The event FIPP Created is raised.  The check function modules are
>executed, but all three fail when they attempt to access the document.
>
>And, to make it more interesting, this problem does not occur in all of
>our 4.6c environments, but most noticeably, in the integration system.
>
>These events, function modules, workflows, etc., all work fine in our 4.5
>environment, and in our 4.6c development environment.
>
>Personally, I suspect that the FI documents (table VBKPF) are not actually
>in the database at the time the check FMs run.  I know that the workflow
>engine has had some performance improvements made.  But I am having a hard
>time convincing my basis friends to take a closer look.
>
>What fun!
>
>Sue