If I understand correctly, based on logged in user you want to select a date from a table in another database that is associated with that user id? If so, then you need a connection in the RPD to this other database and you need to set up an initialization block that fetches the date based on the user id into a session…

Your ask is a little unclear - what is the end game / what are you expecting OBIEE to do?

If you had a snapshot of the rankings at regular intervals it would be a simple case of row-level filtering and not having to worry about a dynamic ranking. To boot, you'd get the ability to show snapshot-over-snapshot increase/decrease in the rankings ...

The basis of a KPI is AS AT ... needs 'inventory style' measurement ... The KPI then has a context and is highly defined. Being able to change the date range (or any other filtered attribute) at run-time for a KPI makes it a different KPI. That converts to a management and monitoring nightmare. Sales for the past 2 weeks…

In addition to your other issues ... you are burdening the OPV (which is not a performance approach in itself) ... 1)OPV Field - LAST_DAY(TO_DATE(TO_CHAR(TO_DATE(A.REC_CALNDR_PRD,'YYYYMMDD'),'MM/DD/YYYY'),'MM/DD/YYYY')) AS MONTH_ENDING You don't need half the functions... LAST_DAY(TO_DATE(A.REC_CALNDR_PRD,'YYYYMMDD')) AS…

mini dimension with the distinct values ...

That might be it as well ... your restart may be running from a cached query and/or result set on the database. Good thinking in the short term!

[nQSError: 12017] Unexpected socket read timeout: connection terminated by network, e.g. by the firewall. ^ error is telling us what the issue is ... why or root cause I'm not sure of. It's interesting that at 9 minutes you get a fail ... is anyone monitoring WHILE you are running the first pass (that fails at 9 mins)?

yes - to get the perspective you will need at least: ITEM NUMBER - business key mapped to BOTH LTS ITEM COUNT - mapped to one LTS ITEM UPDATED COUNT - mapped to second LTS Both LTS set at same aggregation content level

ok ... you had me concerned there for a second ... still you need to take the attributes out of the fact - simplify it so you can get it working. So now ... * make a count(item id) measure from the ITEM MASTER FACT and call it Item Count - set the aggr level appropriately * make a count(item id) measure from the ITEM…

"Indicating that your model isn't set up correctly ..." as I suspected... What is your fact table? Which is your dimension? Fact table holds relationships between dimensions (keys) and additive elements ONLY ... From the picture you have dimensional attributes in your fact table ... so your Logical diagram has a date…

Indicating that your model isn't set up correctly ... perhaps posting a few images of your model and LTS setup would allow us to help you out a bit more.

+1 to @Gianni Ceresa OBIEE is not a reporting tool ... it's an online analysis tool. There is a lot of web content on conditionally hiding columns using BI Publisher.

[nQSError: 12017] Unexpected socket read timeout: connection terminated by network, e.g. by the firewall. Random and intermittent errors? Perhaps your network guys can provide some insight into what is happening between the BI Publisher Server and the BI Presentation Server at the time of your scheduled run... seems to me…

Why are you using opaque views? These are performance killers (no predicate filtering). This should be easily done in the BMM with physically mapped logical columns - the OPV is standing in your way.

Looks like MAX is missing 'BY Po Ordh Order No' ... and the 'Po Ordh Release No' will affect your query grain. Edit: Better yet, create a a set of logical columns (PO Min Date & PO Max Date) that uses something similar to following in the physical mapping MIN(Start date) over (partition by PO Number) MAX(End date) over…

Don't join facts to facts ... feels wrong because it is. Rules of Star Modeling: * Every Dimension is related through a Fact * Every Fact Measure has a Dimensional context Why don't we join facts? Violates Rule 2 ... you end up with facts that don't have dimensional context - they only have factual context ... then you…

"when I try to use the second date" ... the whole idea of canonical time is to use a generic date ... Example: User picks 3/17/2017 and thus Ship Count is count of rows from Fact where the join was fact.shipdate = calendar.date Order Count is count of rows from Fact where the join was fact.orderdate = calendar.date Make…

"don't think someone else will need it." - good luck! This is not pollution of the RPD - it's logical manipulation of data which is the purpose and intent of the RPD. As for 3 people sharing a single access to the RPD - this is development mismanagement -- not to mention risky. I suggest setting up a MUD environment and…

Why not build a logical column in the RPD that achieves this? Build once -- use many.