Deleted USER_DIM accounts and WORKFLOW_STATE

Jump to solution
Community Participant

Hi. We've just started using Canvas Data. I have a question about the USER_DIM table. 

When I create a test user via the Canvas web app, and then later delete that user via the Canvas web app, they seem to appear in the USER_DIM table with WORKFLOW_STATE=REGISTERED. Some appear as WORKFLOW_STATE=PRE_REGISTERED. This doesn't seem correct... The schema doc for Canvas Data does not seem to explain the workflow states in any detail. Can someone please elaborate? 

The reason why this appears to be a problem is that deleted Canvas users who have been set up as Parent Observers seem to still get email notifications regarding observee course activity, even though the user has been deleted. I was wondering if this is something to do with their workflow state not being set to DELETED - ? 

1 Solution

Are they showing as active enrollment, user, and pseudonym if you do this?

select * from enrollment_dim ed left outer join user_dim ud on = ed.user_id

left outer join pseudonym_dim pd on pd.user_id =


You can put the pd.unique_id or pd.sis_user_id or you can also use the user_id.global_canvas_id but you'd have to add your shard id to that number to make the query work.  

Are you using redshift hosted by Canvas or local data warehouse?  I would begin to wonder if your data files have actually updated.  If you are using redshift, I have had multiple occasions where the cluster we were on for AWS Redshift ran out of disk space, or there was a problem with the import process and our data did not update for a while.  They don't do much monitoring of the import processing.  If you want to check for sure, look for the most recent updated_at date in submission_dim or the timestamp in requests to see when the latest row was added.  

View solution in original post