experience with orphaned indexes ?

[LWagner]

Loading to a new app group, we had three partial load failures.  Data was not loaded but indexes were. On access to those record, the server has no record to retrieve.  Delete of the loadid returned 0 rows deleted, and the index rows are still present.

Would an SQL delete of the index rows in question have any negative impact on the the system ?  There seem to be 294 rows in the three loads.

[Justin Derrick]

I'd be more inclined to use arsdoc delete with an SQL command that deletes the index rows based on the doc_name field.

Are you on 8.5.0.6?  I'm seeing this behavior in a customer site as well.

-JD.

[Ed_Arnold]

Hi Larry -

Are you using the -f parm? 

-f
    Use to unload the data if the load process fails. If the database manager step fails, then Content Manager OnDemand should remove any index data that was added to the database. If the storage manager step fails, then Content Manager OnDemand should remove any storage objects that were copied to storage volumes.

Ed

[Justin Derrick]

Hey Ed.

When my customer has tried that, they still get zero rows unloaded, and the rows persist (and are returned in queries).

-JD.

[jeffs42885]

Bumping this back to life- JD did you resolve this by any chance? I am seeing this issue when unloading VIA system load folder.

[Justin Derrick]

I don't have a good answer -- I wasn't involved in the ultimate resolution.

I don't know if anyone had any success with the 'arsdoc delete' option with an SQL statement using the Doc_Name field though.

-JD.