Seeing those two XML files since the Essbase installation… Everytime I just open it, browse it & close it. These files looks to define the outline metadata of BSO & ASO. Always wanted to edit the files & see if the change really gets into effect. In case that happens, we can define the default settings for our cubes’ outline.
For example, by default the duplicate members in BSO/ASO outlines is disabled. I see this section in the file, what if I change defaultValue to true & restart Essbase service? Will it start accepting duplicate names by default?
But I do not have enough courage to test that on Prod.
Q: Where are CDF/macro definitions stored on server?
A: They are stored in a XML file at the location below…
Q1: Where are smart slices stored? Is it on client or server? What is the location? How much space do they take up?
Q2: How do we secure smart slices?
Note: Version is 22.214.171.124
A1: Smart slices are stored on the server. They are stored as XML files which hardly take up any space (usually in kb). Its like the data slice definition is stored in XML files. These files are stored at <hyperion_home>\products\Essbase\aps\data\cubeviews\
A2: We cannot secure smart slices, they can be seen by all the users who have access to the database. But when retrieving data, it may show up #No Access if the user doesn’t have access to data.
Here is the link to a discussion on oracle forum
The user logging information can be extracted from the essbase.log file. However its is difficult to parse the log file & get the information out of it. Here is a method to get that information.
Note that I was using Essbase v126.96.36.199
As we already know there is a relational database involved with Essbase… we can query the table serverlogdetail & get the information required. In the query below, “Essbase” is the schema name that is usually provided while configuring the EPM.
This query lists all the users logged into Essbase in December 2010. When we generate the LogCharts in Essbase, these tables are populated. So, before we run this query, we have to generate the log charts.
What is that we are achieving after all? Once we get the output from the query to excel, we can pivot it & analyze the users’ usage, like which user is most active, which user has logged in how many times etc.
Whether useful or not, this is some good info to know… Information is wealth!! isn’t it?
select * from essbase.serverlogdetail where
extract (year from entrydate) = ’2010′ and
extract (month from entrydate) = ’12′ and
msgtext like ‘Logging in user%’
Gurus out there: If any of the information above looks to be wrong, pls write a comment to correct it.