Your font has to be an ansi font from previous painful experience with essbase.

You don't detail any physical joins that bridge the gap, are there any? i.e. you need a way to conform at least one of the dimensions to join BOTH fact tables, joined via a valid key.

Show sql view is also useful, as well as catching the values in a presentation variable and displaying in title view. Prompt area on analysis will interfere with drill in 'is prompted'

Like this for viewing variables on the page; - https://blogs.oracle.com/exalyticsobiee/how-to-pass-presentation-variable-to-the-title-view-using-dashboard-prompt

I the business will not agree but what the table shows on the right is spurious. Anyone with no prior knowledge would look at it and read that item 10 was ranked 10th this year (true) and ranked tenth last year (false - from your table on the left) I am guessing your client does this in excel yes? If you dropped the…

For row level security see => https://www.clearpeaks.com/obiee-11g-advanced-row-level-security/

Hi Saro, I don't think you are understanding. There are limits, but they are universal limits applied by a single file. You can put the numbers up, but it is for everyone, there is no user / analysis / dashboard 'level' based limits.

I have not experienced this but as well as Joel's suggestion below I would also suggest compare and contrast the messages you see in logs when only Node1 is up and log in is wholly successful with when only Node2 is up. Your 'random' behaviour would indicate that some users are lucky and land on Node 1, some are not and…

Can you try it without LDAP (revert to default security model) and see if it always works? Also, is there any pattern with which users work and which do not - do some always work and some always not, or is it seemingly random?

Is this what you are looking for => OBIEE 11g/12c and ODI12c: Mavg (Moving Average) Other alternatives are using the BY clause to manually create aggregations; - https://gerardnico.com/dat/obiee/obis/logical_sql/group_by You could also achieve this using filter function in a similar way. Finally; and arguably best; you…

What security model are you using? Is there active directory or similar in the security stack that could be your cause?

Could this be your issue => https://community.oracle.com/thread/1038714 Note also that 11g is not supported with later versions of sqlserver, you are really past the point of needing to upgrade!!

Ok, I would not change it from that, but it does look like a problem with your underlying driver, can you try my suggestion?

"I tried with datetime too" - what is the data type of the column in sqlserver? When you use the wizard to import the table what data type does OBIEE assign to that column?

Can you try connecting with; - "Data Direct ODBC, set of drives included with OBIEE, which enable connection to SQL Server, MySQL, Hive etc –The 6.0 data drivers are provided by Oracle and are installed with OBIEE."

What odbc connector are you using to connect to your physical data source? When you say checked the physical query do you mean without any other logic being applied in the analysis?

As to its purpose see this paragraph in particular, and the last part of it for the concept in a nutshell, so essentially it is like a bin (case) statement but with the following differences, you might also want to look at the plsql docs on width_bucket. Note the following differences between the BIN and WIDTH_BUCKET…

Try this, the blog is also good for some other relatively new functionality; - OBIEE in IL: OBIEE 12c Advanced Analytic part 2: BIN and WIDTH_BUCKET

Did we solve your issue? If so can you kindly close the question (assumed answered or if you want to point the answer out for the benefit of others 'Mark Correct'.) thanks, Robert.

Did you solve your issue? If so can you kindly close the question (assumed answered or if you want to point the answer out to others 'Mark Correct'. thanks, Robert.