Purge Global and Local Cache - OBIEE 11g

If you have clustered node environment using a Global folder to store cache or is a one node environment with Cache stored locally following process will be able to help purge cache both locally and in a global folder. This post is specific to a Linux environment and for a multi node environment this process needs to be repeated in each node.

1) Create a text file with below text. This is the same code you would use to clear cache from presentation (Administration > Issue SQL >)

Call sapurgeallcache()

Save this file at path - /instance/instance1/bifoundation/OracleBIApplication/coreapplication/setup

-- for second node it will be /instance/instance2......


2) Create a .sh file (How to create .sh file - http://www.linfo.org/create_shell_1.html) with below code and you can probably name it as cachepurge.sh

/<OBIEE_HOME>//instance/instance1/bifoundation/OracleBIApplication/coreapplication/setup/bi-init.sh

/<OBIEE_HOME>/Oracle_BI1/bifoundation/server/bin/nqcmd -d AnalyticsWeb1 -u <Admin Username> -p <password> -s /<OBIEE_HOME>//instance/instance1/bifoundation/OracleBIApplication/coreapplication/setup/<file from step 1>

**Replace values in <> with your environment specific values**

3) You can setup a crontab on .sh file so that it is scheduled and executed at intended time. If you want to run it manually then use below in each node

./cachepurge.sh

Setting up job in crontab - http://www.cyberciti.biz/faq/how-do-i-add-jobs-to-cron-under-linux-or-unix-oses/

This process should clear local cache and global cache, Global Cache will be cleared based on the frequency interval you have setup


Hope this Helps!

Exporting Data from OBIEE

As I mentioned in my previous post, exporting a detailed report with columns having more than 15 digits in to excel was an issue in OBIEE 10g. Thanks to guys at Oracle, this is now fixed with OBIEE 11g.

However in case you have a big file with more than 10000 records (I know its not ideal but sometimes Users would like detail data), I follow below process which is also efficient than downloading a file with bigger size in excel format.


> Export > Download data as csv
> Save the file as a text file with .txt extension
> Open New Excel file and go to Data tab
> Click 'from text'
> Select Delimited, click Next
>Select Delimiter (Comma or Tab), click Next
> Select the columns (with more than 8 digits) and select Text on top in Data Format
> Click Finish


Excel I believe can accept data up to 1 Million records.

Hope this is helpful