Geoprocessing Uses Old SDE Data After Web Edits

Anonymous User

Greetings!

I've implemented web editting using ArcGIS Flex API on a multi-versioned SDE data layer and I can confirm in ArcCatalog that the edits are made immediately as expected.  The problem that I'm facing though is that after the edits are complete I have a geoprocess that the user kicks off on the webpage to run some analysis on the updated data.  I've noticed in testing that often the result reflect the data in its state prior to editting.  Reading up on this suggested that a database "compress" might help flatten the various states of data so as to insure that the geoprocess runs against the latest data available (I've confirmed that the sde connection file used by the geoprocess uses the correct version of the database so that's not the issue).

Anyway, the compress command added to my geoprocess model seems to do the trick but has certain inherent limitations.  Mainly, I can't seem to rerun the process immediately without getting the following error thrown:

   Unexpected fault:
   [FaultEvent fault=[RPC Fault faultString="Error Executing Task. Error executing tool.: ERROR 000099: The dataset cannot be compressed.
   Failed to execute (Compress).
   Failed to execute (Owl_Habitat)." faultCode="500" faultDetail=""] messageId=null type="fault" bubbles=false cancelable=true eventPhase=2]

but just in general this seems to be a somewhat overkill approach.

Has anyone else experienced this kind of data refresh lag that I'm experiencing and if so, what approach are you taking to get around it? I feel like I'm using a hand-grenade to swat a fly here and it seems that there must be a much simpler approach to insuring that my geoprocess is working with the latest data updates.  Thanks in advance for any feedback/suggestions to this issue ...

- Bob

ps Wil cross post in Geodatabases and ArcSDE forum

Anonymous User

Have you checked in your database if this edit is in the base table?

Yes, that's part of the strangeness of it Eric - the edits appear when viewed in ArcCatalog but are stale when it comes to running my geoprocessing model! Very odd ....