We have security tables setup to restrict external client access within a custom web app to only the data related to that client.
We are preparing to test reports built against a 6.5.1 Universe to see whether we can use the existing security tables in BO.
My theory is to take the exisitng security tables (which exist on the transactional system), copy them to DB2, set up the external clients as users and implement row-level security on each user in Supervisor.
We will ultimately have hundreds of external users and, being new to both Supervisor and to BOB, I am wondering whether I am being too simplistic or am setting myself up for a maintenance nightmare.
I have searched the forum for specific posts on how to implement this type of scenario, but don’t think I’ve read one yet. Please feel free to request more information, if necessary.
Any response/advice is welcome as I have to get a strategy in place ASAP. Thanks in advance.
Anita, my apologies as I know this is not the place for basic BOB instruction, but I did find a post that was helpful and would like to reference it here.
You provided me with two links to other posts in your reply. How did you do that? (I’ve been searching the BOB site for instruction, but can’t find any).
Funny you should reply. Yours is the previous post I was referencing:
Your suggestion is really what we want to implement, but in doing so, do we have to create unique BO IDs for each external client? Ideally we would have a Single Sign environment wher ethe users would enter the web app with their existing ID with a link to WebI in the site. The BO ID would be the same as the web app, which, in turn, is the same as what’s in the security table.
If we have generic reports built that require users to answer prompts about, say, date ranges and products, how exactly do we implement the security table so that BO appends the appropriate client codes in the where clause.
I hope I’m making sense here, but I’m relatively new to this.
(BOB member since 2002-06-17)
!