Tuesday, October 15, 2013

OBIEE 11g customise

The information related to style and skin can be found under u01/app/oracle/product/obiee1111/ORACLE_BI1/bifoundation/web/app/res  >> sk_blaf/b_mozilla_4

In this example, we are going to change the logo of BI

1. Added the following to instanceconfig.xml file:

/u01/app/oracle/product/obiee1111/instances/instance1/bifoundation/OracleBIPresentationServicesComponent/coreapplication_obips1/analyticsRes


before



2. http://verisklinux:7001/console. navigate to Deployments > Install. Deploy analyticsRes from

/u01/app/oracle/product/obiee1111/instances/instance1/bifoundation/OracleBIPresentationServicesComponent/coreapplication_obips1 (see screenhsots)

Activate Changes

 
 
 
 
 


 
 
 
 
 
 
 
 

 
 
 
 
 
 
 

 
 
 
 
 
 
 
 
 

 
 
 
3.Start AnalyticsRes. make sure that its status changes to Active.

4. To test my deployment, added test.txt to the analyticsRes folder and access it via
    http://verisklinux:9704/analytics/test.txt

5. Copy the oracle standard skins to my deployment folder
cp "/u01/app/oracle/product/obiee1111/Oracle_BI1/bifoundation/web/app/res/sk_blafp/" /u01/app/oracle/product/obiee1111/instances/instance1/bifoundation/OracleBIPresentationServicesComponent/coreapplication_obips1/analyticsRes/sk_verisk

6. Edit instanceconfig.xml and add the following:
    verisk

7. Restart BI Services

Tuesday, October 8, 2013

Efficient idea on how to implement a report that shows the impact of column (at all levels) on reports in OBIEE PS:

http://www.kpipartners.com/blog/bid/151016/Build-OBIEE-Reports-For-Impact-Analysis-Data-Lineage

Monday, March 18, 2013

Suspension

I haven't posted here in a very long time; not because I am not into OBIEE anymore but because I have taken the role of BI Lead Consultant in CapGemini FS and this translates into hardly having any time for blogging anymore. I have tones of articles that I could write about and I come across so many interesting situations that would be worth sharing but unfortunately time is so limited that probably I will post only once in a while

'Till free(er) times.

OBIA Software Requirements

Certification

Software Downloads on eDelivery

The software requirements for OBIA can be mainly be downloaded from edelivery.oracle.com or otn.com:
 
edelivery.oracle.com
 
  • Product Pack:     Oracle Business Inteligence
  • Platform:             Linux x86-64 / Microsoft Windows (32-bit)

Business Analytics Warehouse Tier   (Red Hat Linux)

1) Oracle Database 11g Release 2 (11.2.0.1.0) for Linux x86-64

otn.com
 

Download:

2) JDK 1.7.0+  ( http://java.sun.com/javase/downloads/index.jsp )

3) Oracle DB 11g R2 Client (11.2.0.1.0) for Linux x86-64 (source: otn.oracle.com)

Download:
 

4) Informatica PowerCenter Services - Powercenter 9.0.1 Hotfix 2

 

5) DAC Server 10.1.3.4.1 

 
 

6) DAC Patch 12381656

      Download link: www.metalink.oracle.com
 

7) Oracle Business Intelligence, v. 11.1.1.6.0 - for Red Hat Linux x86-64 (64-bit)


 

8) Repository Creation Utility

 
 
 

ETL Client Server  (Windows 32 bit)

1) Oracle Database 11g Release 2 Client (11.2.0.1.0) for Microsoft Windows (32-bit)

 

2) JDK 1.6.0_29+ 

 

3) Informatica Client

   Download link:  Informatica Client 9.0.1 Disk 1
 
  
 

4) Informatica Services

 

5) DAC Server and Client

 
 

6) DAC Patch 12381656

Download link: www.metalink.oracle.com
 

7) Oracle Business Intelligence 11g (11.1.1.6.0) for Microsoft Windows (32-bit)

 
 

8) Oracle Business Intelligence Applications 7.9.6.3 for Microsoft Windows

 

9) Repository Creation Utility

 
 

10) Oracle Business Intelligence 11g Developer Client Tools (11.1.1.6.0) for Windows (32-bit)

 

Sunday, November 18, 2012

Dashboard - Google Map integration

OBIEE Google Map with multiple addresses or with the MarkerClusterer API.
 
It is based on blogs:
http://obiee101.blogspot.com/2009/03/obiee-google-maps-multiple-addresses.html
http://blog.guident.com/2009/12/integrating-obiee-and-google-maps-with.html
Working example can be found on Beyond University BI Demo
To use Google Maps you need to sign up for a key on Google Maps website. In those examples we use geocodes but you may use address or postcode but geocodes are much quicker.
All of that html/javascript code is quite simple and self-documented so it is very easy to change.
Tips:
Don't put too much code to Narrative View, create javascript functions in additional file and reference that in your Narrative view (i.e. ).
In javascript don't use single-line-style comments (// comment), use multi-line-style comments instead (/* comment */). When coping or modifying Answers with single-line-style comments you might end up with code placed in single line comment after your comment - some actions remove end of the line sign from Narrative View.
 
 
Simple OBIEE Google Map with multiple addresses.
 
Create an BI Answer query, go to Narrative view, check box next to "Contains HTML Markup" and type similar code in following sections.
In the Prefix part put (in the first line put your Google map key):
 






 
Now check if it works in the compound layout.
 
 
 
OBIEE Google Map with multiple addresses using with the MarkerClasterer API.
 
Create an BI Answer query, go to Narrative view, check box next to "Contains HTML Markup" and type similar code in following sections. Put beyondmap.js somewhere on the server.
In the Prefix part put (in the first line put your Google map key, in second line put path to beyondmap.js file on your server):
 






 
Now check if it works in the compound layout.
 
 
 
Get GeoCodes for postcodes using Google Maps API in OBIEE.
 
Create an BI Answer query, go to Narrative view, check box next to "Contains HTML Markup" and type similar code in following sections.
In the Prefix part put (in the first line put your Google map key):
 







 
 
Now check if it works in the compound layout.

Friday, May 4, 2012

OBIA Architecture

  NOTE: It is recommended to set up all Oracle BI Application tiers in the same local area network.
             Installation of any of these tiers over WAN may cause timeouts during ETL mappings execution on the ETL tier

Tiers

Machine Summary

Tier machine name User Name IP Address
1 OBIA DW Tier OBIADW oracle

2 OBIA ETL Tier OBIAETL oracle

3 OBIA Presentation Tier OBIEE oracle

4 OBIA Client Tier WIN xxx
 
 
Business Analytics Warehouse
 
Machine Specification  
M5000 SPARC VII - Core Factor - 0.75
16 GB RAM 2 cores
Sun Solaris 10 (64-bit)
 
Disk 400 GB
Software
1. Oracle Database 11g EE 11.2.0.1 ( with patch 9739315 applied and 9255542 is bug not a patch)
 
 
ETL Server
Machine Specification 
SPARC T3 - Core Factor 0.25
4 GB RAM  8 cores
 Sun Solaris 10 (64-bit)
  Disk 100 GB
Software
1. DB Client 11.2.0.1.0 for Solaris
2. DAC Server 10.1.3.4.1 Patch 12381656 applied 
3. Informatica Server - Powercenter 9.0.1 Hotfix 2
4. Java Platform Standard Edition (Java SE) Development Kit (JDK) 6 (1.6.0_05)
 
 
 
ETL Client 
Machine Specification 
3 GB RAM 2 quad cores
Windows XP Professional SP3 (32 bit)
  Disk 1 TB
Software
1. Oracle Business Intelligence Developer Client Tool (MS Win 32 bit) 11.1.1.5.0
2. Oracle Database Client 11g EE 11.2.0.1 3. Informatica Client - PowerCenter 9.0.1 Hotfix 2
4. Informatica Services - PowerCenter 9.0.1 Hotfix 2
5. DAC Client 10.1.3.4.1 Patch 12381656 applied 
 
 
Presentation Server
Machine Specification 
SPARC T3 - Core Factor 0.25
8 GB RAM  8 cores
 Sun Solaris 10 (64-bit) (Update 8+)
  Disk 50 GB
Software
1. Oracle Business Intelligence Enterprise Edition 11.1.1.5.0
NOTE: certified with Oracle DB:
            - Oracle 10.2.0.4+
            - Oracle 11.1.0.7+
            - Oracle 11.2.0.1+ 
2. BI Apps 7.9.6.3 repository and Catalog deployed
 
 

Wednesday, April 25, 2012

OBIA Best Practices

Best Practices and Important Notes

DAC Notes
  1. To check the pmcmd command sent by DAC to Informatica to execute a task, click on the Status Description in the Task Detail window. We can run this from the Linux box (where the Informatica server resides)
  2. Use the Analyze Repository Tables (Tools > DAC Repository Management > Analyze Repository Tables) when it has been made a considerable change to the core metadata -> this helps to speed up the steps that follow
  3. Micro-ETL:
    - scheduled to run every couple of hours and it can be running even when users are using the OLTP system or running BI reports
    Steps:
      [] create a new Subject Area and add the workflows that we are interested in (i.e.: 1 fact or 1 dimension only) + Assemble
      [] create a new Execution Plan and add the Subject Area created before. Tick Micro-ETL. This is VERY IMPORTANT as the process is this: when Micro-ETL is ticked, the new data added in the OLPT comes into DW even when the users are still logged on and run reports, so no truncate of the staging or target tables happens. Also, the indexes are not dropped. The next time the Incremental is run (involving these tables), it brings in everything from the last incremental run (in a proper way), with dropping and recreating indexes, just to make sure that nothing was
    missed during the Micro-ETLs run and that the statistics are up to date.
    Note: It does not make sense to Analyze nor to drop/create indexes in the properties of the execution plan, all we want is to bring in those new records (so ideally we want the Micro-ETL to run in a couple of mins only - fastest as it can)
      [] Generate parameters and Save
      [] Build and Run 
  4. Deploy to PROD
  • The DAC client could be used to connect to many DAC servers.
  • In terms of migrating all the changes from DEV to PROD: export the custom container only (Logical, no other tick boxes checked) and import it in PROD. Note that when DAC server was created, an empty DAC repository has been created. so we only need to bring in the changes (ie custom container, custom subject areas, custom execution plans etc) and we setup the parameters and the connections to use the PROD connection details.
Export DAC customisations from DEV:     DAC > Tools > DAC Repository Management > Export  > tick only Logical -> Select container > Export.
Import DAC customisations into PROD:    DAC > Tools > DAC Repository Management > Import
 
Informatica Notes
  1. It is good practice to use versioning of the changes so check-in must be made under a specific revision.
  2. Before making any changes to the mappings, it is advisable to keep a copy of the original one. Export SDE > Save as XML and put it under a revision control system. This will ensure that if the SDEs get messed, we could always revert to the original by Importing it to the custom folder (address the conflicts with Replace)
  3. If we want to make add some conditions to the query only for the Full load, instead of updating the mapping (SDE), just change the Workflow query (Full) - as the mapping is used by both Workflows (incremental and full) and we only want the change in one of them so it does not make sense to create another mapping just to reflect our condition.
  4. Running Workflows directly from Informatica: when creating a Mapping as a copy of an existing one, right-click Mapping/Mapplet > Declare Parameters and Variables > always give an Initial Value for the params so that when running it in the Workflow Manager it will use the default values and it will not error.
In Informatica Workflow Manager: don't forget to replace the Parameters with the OLTP/DW actual values
- drag WF and on the task > right click > go to Properties
                      -- $Source Connection Value -> change to be whatever we want to point out to (Use Object and select object from the list)
                      -- $Target Connection Value -> change as above
               - NOTE there is always the option to revert to the original (Use Connection Variable)
               - Edit Task, go to tab Mapping > Connections > make sure there are no variables used but the actual values are passed. Save and run the workflow
 
DEV NOTE: to see large mappings just right click on the grey area in the Mapping Designer and select Arrange All Iconic