So you managed to get master-detail to work? Or you managed to fix the dashboard prompt issue not filtering your analysis?

In the analysis you need to add a filter on the column as "is prompted". Because it's a real filter (so the query sent to the source contains a WHERE clause filtering the value, you change the value a new query is executed), while the one you used as example is just a visualization filter (so the query is executed once…

Your screenshot looks like a single analysis with 2 views, right? If it's the case the easiest way is to set a master-detail, so the table view where you have the year prompt will drive the other view on the left (a chart I guess) and change the year directly. If it's 2 analysis or you want it in a dashboard page than you…

If you look at agents you have a condition screen where you can set a condition, and if condition not met then the agent abort (in a way, it doesn't execute and send out anything). So do what Christian said and set it as condition for your agent, and set your agent to run daily: done!

Hi, The result is right, the calculation is doing exactly what you asked: INTEGER / INTEGER = INTEGER Just few hours ago the same topic covered: Remove default floor function while dividing a column by another in OBIEE You need to cast one of the 2 as double, so you have a INTEGER / DOUBLE = DOUBLE

Hi, Aren't you just facing the typical "integer division" ? When dividing or multiplying with integers the result will also be an integer. If your "failure" and "success" are integers you need to cast at least one as double or you will always be doing an integer calculation. If both are already double then use "100.0"…

Did you look on MOS? I guess I remember a bug related to that, where dashboard do not react to prompts anymore or something like that ...

Google how to embed D3js in OBIEE, there are many blog posts covering the topic. Also keep in mind D3.js is already available in your OBIEE 12c as VA uses it, so you don't have to deploy it yourself, just load the one embedded in OBIEE.

I couldn't find a 11g for the test, will need to find one of the old SampleApp to see how was the 11g world

Hi, You need to open the RPD with the Admintool in online or offline mode, on the right side of the screen you will see the physical layer and there you need to edit the connection pool to point to your database (so database connection string or TNS, username and password). If you do that offline you also need to re-upload…

Hi, Something really like what you posted isn't possible out of the box. The easiest would be to use something like D3js or any other JS library to render the chart yourself (and so you have full control over the visualization). Out of the box, if not wrong (I actually never used it in the last .... 10 years ) I guess you…

Don't you have an issue in the keys you set in the map layers? You can maybe concatenate the state and county if you don't have a county unique ID. For example this layer which is part of SampleApp use a concatenation of County and the State abbreviation to make counties "unique".

Oh, so you mean drill strictly on the map? I thought it was a more generic drill (so also on values in tables etc.). From a modelling point of view hierarchies for all the dimensions are needed in the RPD (it doesn't make it mandatory and allow to not have any, but you are just losing a lot of functionalities and also have…

Hi, Did you check the logical level key to make sure it's unique and it clearly identify each counties? It's a bit like if you think at years and months and for the months level in the hierarchy you set as key the month name (like "January"). Every year has a month named "January", so the logical level key isn't unique and…

It's difficult to tell you something precise as the concept works, the theory of dimensional modelling too. So ... try to keep it simple: practice with just a mini-star being that and find out how to make it works, then you can integrate it with your existing model.

Hi, This is generally known as degenerated dimension as it's mainly a fact table containing attributes than the opposite. But it end up being the same thing in the end. It must be modelled using the object twice to have it as dimension and fact as well. Not sure why you set the content levels to Total and for what, so you…

Hi, As a fact table must not have attributes inside but only aggregated columns (that's the theory of a dimensional model) then your idea 2) isn't really a good idea Which means you are left with idea 1) Depending on the relationship between NC Type and shipping you maybe also can use 2 separate fact tables defining…

As OBIEE isn't an opensource tool there isn't a public available recipe on how the IDs are cooked. But as Christian said they are generated when you add the columns in the analysis, their main usage is to uniquely identify columns in the context of the analysis itself. It doesn't look like these IDs are used outside the…

As it's part of the same functionality (go to the beginning, go to the next 30, go to the prev 30 and load all) not something you can enable with a checkbox I would say. Is that only for a unique single analysis or more something generic that you don't want to see that "load all" arrow never again?

Well ... what Christian said You use the 2 system variables to reference the user credentials and use that in the connection pool for the DB. For the first part it's about integrating your AD with OBIEE, the doc cover that as well as ton of blogs.