Database physical layout

Dear All,
We are considering of building a database to be primarily used for datawarehousing. We need to know more about the physical layout: what type of raid configuration for the data files, what should be the ideal size of LUNs, and how much space to allocate for the FRA etc. The database size we are estimating now would be between 1 to 1.5TB. Any suggestion, direction, advise would be highly appreciated.
Thanks.

user6116850 wrote:
We are considering of building a database to be primarily used for datawarehousing. We need to know more about the physical layout: what type of raid configuration for the data files, what should be the ideal size of LUNs, and how much space to allocate for the FRA etc. The database size we are estimating now would be between 1 to 1.5TB. Any suggestion, direction, advise would be highly appreciated.Ito RAID, Oracle recommends SAME - Stripe And Mirror Everything.
Ideal LUN size? There is none. That can be a limitation of the storage array. On a pentabyte warehouse, you may want to use as large as LUNs as possible in order to reduce the number of disks seen by the o/s, as the o/s itself may have a limit on the maximum number of devices it support.
If the storage system does not support striping, then you want to size the LUNs and number of LUNs in such a way to do striping at ASM level (or whatever volume manager s/w used).
For performance, you want the fastest storage system and disks that fits your budget. 1TB is pretty small - that's the size of a standard SATA disk these days. And pretty inexpensive too. But then, using 10 x 128 GB drives and striping these could provide better performance as more disk controllers are available. The downside of that is that this also increases the number of potential disk failures.
You may not want to use a storage system at all, but the local drive bays in the server. And use the onboard SCSI RAID controller to mirror and/or stripe it.
So there's not really a pat answer to the questions you have raised. A lot of it depends on architecture and budget. The only issues that hold true the vast majority of times are that we never have disks that's really as fast as we would like and invariably we want more space... ;-)

Similar Messages

Oracle 8i physical layout

Hi, i am installing oracle 8i on NT
for my production database. I would like to
know optimal solution of oracle logical and physical layouts with 6 physical drives.
Also, how that ties with fault tolerance
i.e. RAID1 RAID5 (which I am thinking of implementing)
If anyone know any good sites with this kind
of architectural solutions, please let me
know.
thank you.
null

No-one have never seen these aspects? :-(
Edited by: Huge10 on 9-mar-2011 6.56

Planning the physical layout of the database

Our application keeps very large historical data for 1 year. We do not use the Partitioning Option but have created 12 different BIG TABLES,
one for each month. We are wondering if for performance reason or any other reason, whether it is better to put all the tables
into 2 physical file as show in option 1 or 24 physical files in option 2. Reason for option 2 is to reduce the size of the physical
files so that we can speed up the backup process by selecting individual files to backup on a monthly basis.
Does Oracle have any problem with creating too many Table Spaces and physical data files?
Option 1 - 2 physical data file
HISTDATA Table Space
HISTINDEX Table Space
Option 2 - 24 physical data files
JANDATA Table Space
JANINDEX Table Space
FEBDATA Table Space
FEBINDEX Table Space
MARDATA Table Space
MARINDEX Table Space
DECDATA Table Space
DECINDEX Table Space

From a performance standpoint, it will only have an effect if the different layouts allow you to utilize more disks or to spread the work more evenly among disks. If you have 10 different mount points on 10 different sets of disks, you would probably want files to be spread across all the mountpoints. If you have a SAN where everything is striped and mirrored, it doesn't make a difference.
From a managability and a backup and recovery standpoint, multiple tablespaces seem most appropriate, particularly since 11 of the 12 months can presumably be marked read-only and excluded from the backup.
Justin
Distributed Database Consulting, Inc.
http://www.ddbcinc.com/askDDBC

Move the database storage layout, please Help.

Hi Buddies,
My current database A (9i) 's storage layout is
File System SIZE
/yqa1/tech/oracle 8628 MB <-- Oracle software plus server's dump dirs only
/yqa7/appl/oracle 34521 MB <-data
/yqa8/appl/oracle 34521 MB <-data
we want to split existing 32G filesystem into separate chunks of 8G and map them into seperate filesystems
(for expamle: split /yqa7/appl/oracle into
/def1/appl/oracle [8G]
/ghi7/appl/oracle [8G]
/jkl23/appl/oracle [8G]
/xyz2/appl/oracle [8G]
for database A, finallly, we want to change the storage layout like this:
File System SIZE
/yqa1/tech/oracle 8628 MB <-- Oracle software plus server's dump dirs only
/yqa7/appl/oracle 8628 MB <-data
Is there any best practice to achive the purpose? or any frinds can provide practice documentation regarding to this?
Thanks a lot in advance.
Jerry

You'd need to take the tablespace offline
ALTER TABLESPACE tablespace_name OFFLINE;
before you move or copy the datafile to relocate it.
After completing the move or copy, you could use the
ALTER DATABASE RENAME FILE 'old_file_name' TO 'new_file_name' ;
for each of the datafiles in that tablespace.
Then, you can bring the tablespace online again
ALTER TABLESPACE tablespace_name ONLINE;

Streams Replication:Source database Physical or Logical Standby DB

Can the source database in streams replication be a physical or logical standby database ? If so, is the process of configuring streams the same as a regular database ? Are there any best practices or different configuration if the source is Logical or Physical standby DB ?
Thanks in advance.

Never done it, but I don't see any reason why it should not work.
Streams, at capture site, is only a data dictionary game and in a logical standby your data dictionary is open read write.
Streams, at capture site, never touch the source tables, in fact they may even not exists from Streams point of view,
as it deals only with the redo that are generated.
So Streams horizon is limited to the data dictionary, the log buffer, the archives and, in SYSAUX tablespace, all the LOGMNR_% tables. All these structures are read write in the logical standby. However, for the capture/propagation you may have to set to true the 'include_tagged_lcr' parameters.

Managing&Auditing Database Physical files.

Hi all,
Sorry if I posted this issue in this forum.
I have just this problem in monitoring and auditing oracle files and disk spaces in our Windows server.
We have a server farm of around 20 Windows server, for global centralized test servers.
Each server has different versions of Oracle DB, and maybe contain more than 10 database/intance (small test DBs).
So we monitor around 200+ Databases.
We run backup daily for each db using cold backup and maintain at least 1week retention files.
We we have around 5x10x200 backup zip files.
We also monitor all the Physical list of DB_home folders and it corresponding .dbf files and timestamp.
What I did is map all the hardrive from each servers on my laptop to I can read all and audit all the physical files.
I want to have a centralized list of the files associated with the database in spreadsheets and update it periodically.
My question is how can I run a list cascade command that will list all the files and sub-directories and saved in in .csv or excel file?
Is it like "ls -lt">list.log in linux.
Thanks

Please see the threads referenced in these links.
https://forums.oracle.com/forums/search.jspa?threadID=&q=Images+AND+HRMS&objID=c3&dateRange=all&userID=&numResults=15&rankBy=10001
https://forums.oracle.com/forums/search.jspa?threadID=&q=Images+AND+HR&objID=c3&dateRange=all&userID=&numResults=15&rankBy=10001
Thanks,
Hussein

Detach Database: physical file permission

Hi All. I am using windows account to login which is Sysadmin of the SQL server.
when I detach a database, the physical file permission will only have one entry, which is my domain ID with full control.
it using another windows domain account which is also sysadmin would like to attach the database again: FAILED since access denied....
how to avoid this happenn ??

When you detach a DB, the ACLS are reset and the permission is given to the logged in windows account user - the full control on the DB files as expounded here:
https://msdn.microsoft.com/en-us/library/ms189128.aspx/html. You can circumvent your problem by using "EXECUTE AS Domain\[Attaching User]" before detaching the DB using sp_detach_db
(this will retain full CONTROL permission to the attaching user)
However, starting from SQL 2005, you can use a more graceful method of moving files (on the same server) rather than using the attach/detach as shown below:
Take the DB offline
Move the physical files to the new location
Run the “ALTER DATABASE DBNAME MODIFY FILE (NAME = logicalfilename, FILENAME = physicalfilepath” command to provide the new path
Bring the DB online
However, if you are moving the DB to another server, then backup/restore will be best (you can use compression (SQL 2008 R2 standard edition or higher supports it) and this will move much faster than uncompressed MDF and LDF files
Satish Kartan www.sqlfood.com

Forward physical standby archivelog files to a logical standby database

Hi,
We have a production database (1) and we have a physical standby database (2) for it.
Is it possible to forward the archivelogs files from (2) to a logical standby database (3). We want to use the (3) as a UAT - Ad Hoc read only Database.
Thanks.

Hi,
The following Data Guard configurations using cascaded destinations are supported.
A. Primary Database > Physical Standby Database with cascaded destination > Physical Standby Database
B. Primary Database > Physical Standby Database with cascaded destination > Logical Standby Database
A physical standby database can support a maximum of nine (30 as of Version 11.2) remote destinations.
Physical Standby Forwarding Redo to a Logical Standby :
Advantages :
1. Users can connect to Logical Standby database and they can access
2. Instead of querying primary database they can use logical standby database.
3. without any additional overhead on your primary system, and without consuming any additional transatlantic bandwidth.
Disadvantages :
The following data types will not support in Logical standby database just check your application before implementing logical standby
a. BFILE
b. Collections (including VARRAYS and nested tables)
c. Multimedia data types (including Spatial, Image, and Oracle Text)
d. ROWID, UROWID
e. User-defined types
f. LOBs stored as SecureFiles
g. XMLType stored as Object Relational
h. Binary XML
Thanks
LaserSoft

Database layout crashes appleworks

good morning. I am using AW 6.2.9 on an iBook G4 w/ OS 10.4.11.
In one of my databases, one layout crashes the program whenever I try to view it in layout mode. I can see in it Browse, and I can see every other layout in both browse and layout. Obviously, it worked at when I first created the layout, so I'm not sure what happened. I've tried creating a new duplicate layout, but it crashes too (but not a new blank one). So other than recreating the whole layout from scratch, can anyone advise why this might be happening and how it can be resolved. Thanks!

Have you added any objects to the layout? The layout editing mode is actually a draw environment & it's possible an added object is corrupt. I don't know how you would be able to delete objects if you can't view it, so I think you might have to recreate it. One thing to try, before going through that, is to save the database as an AppleWorks 5 document & then open this AppleWorks 5 document in AppleWorks 6. There is the slight possibility that this will strip the corruption. You do not need to have AppleWorks 5 to do this.

Upgrading E1 from 8.12 tools 8.97.2.1 to 9.0.2 tools -need db table layouts

Where can I find a net change E1 812 vs. E1 9.0.2 document regarding specifically database table layouts when considering table conversion. For example, if the table layout for the F0911 table in E1 9.0.2 has changed as compared to the same in E1 8.12 with respect to added fields, deleted fields, changed fields, etc.
Thanks

Hi,
https://support.oracle.com/CSP/ui/flash.html#tab=KBHome%28page=KBHome&id=%28%29%29,%28page=KBNavigator&id=%28viewingMode=1143&from=BOOKMARK&bmDocDsrc=KB&bmDocID=705511.1&bmDocType=REFERENCE&bmDocTitle=JD%20Edwards%20EnterpriseOne%20Applications%20Programmer%27s%20Guide%20Xe%20to%209.0%29%29
Best Regards.
Bruno Condemi

How to reference MS Access database in ColdFusion MX 7 on 64-bit Windows

I am migrating customers from:
Windows 2000 Server (fully patched)
Internet Information Services (IIS) 5
ColdFusion 5
to:
Windows Server 2003 Standard x64 Edition (fully patched)
Internet Information Services (IIS) 6 (running in 32-bit
mode)
ColdFusion MX 7 (7.0.2) Standard
Unfortunately, some of these existing customers are still
using Microsoft Access databases instead of SQL Server. Having all
of these existing customers migrate to SQL Server is not yet
practical. On the old Windows 2000 / ColdFusion 5 server, they are
using OLE DB data sources to reference the MS Access databases
using the "Microsoft.Jet.OLEDB.4.0" provider. None are configured
to use ODBC drivers when referencing MS Access or SQL Server.
I have yet to figure out how to get ColdFusion MX 7 to:
- Create a data source that references MS Access databases on
64-bit Windows.
- Reference an existing MS Access data source created with
the 32-bit version of the ODBC Data Source Manager on 64-bit
Windows.
A bit of background information regarding MS Access on 64-bit
Windows:
- At the time of this writing, Microsoft has not
created/released 64-bit drivers for Microsoft Access and it's
unlikely they ever will, for understandable reasons.
- Contrary to seemingly popular belief, the Microsoft Jet 4.0
Database Engine does exist on 64-bit Windows. See the following
regarding version and file location information:
How to obtain the latest service pack for the Microsoft Jet
4.0 Database Engine
http://support.microsoft.com/kb/239114
- You CAN create MS Access based data sources within 64-bit
Windows, but they have to be made using the 32-bit version of "ODBC
Data Source Manager". I'm assuming these DSNs may only be used by
32-bit applications, but I have not tested that theory.
32-bit data sources:
ODBC Data Source Manager:
%SystemRoot%\SysWOW64\odbcad32.exe
Registry location:
HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\ODBC
Default file DSN directory:
C:\Program Files (x86)\Common Files\ODBC\Data Sources
64-bit data sources:
ODBC Data Source Manager:
%SystemRoot%\system32\odbcad32.exe
Registry location:
HKEY_LOCAL_MACHINE\SOFTWARE\ODBC
Default file DSN directory:
C:\Program Files\Common Files\ODBC\Data Sources
The 64-bit version of "ODBC Data Source Manager" (under
Administrative Tools) only show "SQL Server" and/or "SQL Native
Client" as options when creating new data sources, unless
third-party software is installed that supplies additional drivers.
The 32-bit version of "ODBC Data Source Manager" (by default,
there is no shortcut for this in the Start menu) shows the typical
drivers seen on 32-bit versions of Windows, including SQL Server,
SQL Native Client, Microsoft Access, Excel, FoxPro, Paradox, etc.
When using ColdFusion Administrator in ColdFusion MX 7.0.2 in
64-bit Windows 2003 (IIS is running in 32-bit mode to allow
ColdFusion MX 7 to function at all), two problems are encountered
when dealing with MS Access (and presumably other 32-bit drivers).
In both cases, ColdFusion is trying to reference the registry
location for 64-bit data sources instead of 32-bit:
1. Creating a new data source within ColdFusion
Administrator:
Data Source Name: whatever
Driver: Microsoft Access
CF Data Source Name: whatever
Database file: (physical path to whatever.mdb)
(other settings are irrelvant for this example)
Error generated when submitting:
Unable to update the NT registry.
Cannot open HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBC.INI\ODBC
Data Sources: Windows error number 5 occurred.Access is denied.
ColdFusion stores the data source, however a "verify"
generates the following error:
Connection verification failed for data source: whatever
java.sql.SQLException: [Macromedia][SequeLink JDBC
Driver][ODBC Socket]internal error: Data source name not found and
no default driver specified
The root cause was that: java.sql.SQLException:
[Macromedia][SequeLink JDBC Driver][ODBC Socket]internal error:
Data source name not found and no default driver specified
2. Creating the data source using the 32-bit version of "ODBC
Data Source Manager" then trying to reference it with ColdFusion:
ODBC Data Source Manager:
System DSN or File DSN:
Name: whatever
Driver: Microsoft Access Driver (.mdb)
Database: (physical path to whatever.mdb)
(creation is successful within ODBC Data Source Manager)
Reference the above DSN within ColdFusion Administrator:
Data Source Name: whatever
Driver: ODBC Socket
Error generated when submitting:
Error accessing available odbc datasources. - Cannot open
HKEY_LOCAL_MACHINE\SOFTWARE\ODBC\ODBC.INI\ODBC Data Sources:
Windows error 2 occurred.The system cannot find the file specified.
ColdFusion MX 7 can't find the 32-bit data source on 64-bit
Windows because it's looking for it in the wrong registry location.
Can ColdFusion MX 7 be configured to look in
HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\ODBC
instead of
HKEY_LOCAL_MACHINE\SOFTWARE\ODBC
on a 64-bit Windows machine?

Paul,
Thanks for the insight on this. Regarding the initial problem
I ran into with creating Access based DSNs, very early-on, I
noticed there were two different driver options for Access
displayed within ColdFusion MX 7 Administrator:
Microsoft Access
Microsoft Access with Unicode
For reasons I'm am still baffled over, I somehow completely
forgot about and through the wonders of apparent tunnel vision,
completely overlooked the "Microsoft Access with Unicode" driver
option when I was running through my tests.
At your seemingly-obvious suggestion, I created a data source
using the Microsoft Access with Unicode driver and it worked
perfectly fine with the one customer's web site I'm using for
testing (of course, this is a copy--the production site is still
housed on the old ColdFusion 5 server). This obviously doesn't
actually fix the underlying issue with ColdFusion looking in the
wrong part of the registry on 64-bit Windows, but at least when it
comes to dealing with Access databases, your suggestion of using
the Microsoft Access with Unicode driver appears to be a functional
work-around. So long as there aren't any surprises waiting for me,
hopefully this will also work for the remaining ColdFusion sites I
have to migrate that are using Access databases.
Since you mentioned JDBC drivers in comparison with ODBC, I'm
pointing out a couple of references in case they're helpful to
anyone following this thread:
ColdFusion MX 7 - About JDBC
http://livedocs.adobe.com/coldfusion/7/htmldocs/00001736.htm
Types of JDBC technology drivers
http://java.sun.com/products/jdbc/driverdesc.html
http://java.sun.com/products/jdbc/
I don't actually develop anything in ColdFusion, so I'm
dealing with all of this from a server administration standpoint
since we do have some domain hosting customers that do develop or
at least have existing ColdFusion based sites. I am all
too-familiar with the caveats of using Access in any kind of
production or heavily used environment and have spent years keeping
users away from ODBC. For ColdFusion users, that meant creating OLE
DB DSNs and often having to fix poorly written SQL statements and
fixing ColdFusion date-related values and boolean values (the
common problems encountered when switching CF / MS Access users
from ODBC to OLE DB). I can only recall running into problems with
a single SQL Server based ColdFusion site when migrating from ODBC
to OLE DB. That customer never did get their many coding problems
fixed, so they're the only ones still using ODBC in CF. I have no
idea what'll happen when that one site gets migrated over--if it'll
even work. Obviously a lot of things have changed between
ColdFusion 5 and the MX versions that have come along afterward.
Thanks again for the Access with Unicode driver tip and JDBC
info. It looks like I can finally start moving forward again with
the site migration process.
Josh

Discovery data couldn't be inserted to the database. This could have happened because of one of the following reasons:

Discovery data couldn't be inserted to the database. This could have happened because of one of the following reasons:
- Discovery data is stale. The discovery data is generated by an MP recently deleted.
- Database connectivity problems or database running out of space.
- Discovery data received is not valid.
The following details should help to further diagnose:
DiscoveryId: ae39d630-f82d-77ee-d5d9-dd4c4a2e7bc2
HealthServiceId: ab069276-c6dc-13d7-3ef5-28e780c34953
Microsoft.EnterpriseManagement.Common.DiscoveryDataUnresolvedClassPropertyNameException,Unable to resolve the class property name specified in the discovery data item. Try using a fully qualified name.
Class property name: 12f3e541-ec4d-7310-be57-fa3bde0bed06
Rule ID: ae39d630-f82d-77ee-d5d9-dd4c4a2e7bc2
Instance:
<?xml version="1.0" encoding="utf-16"?><ClassInstance TypeId="{6ce7ab22-1880-0eb1-40ba-86d9a8c3d5f5}"><Settings><Setting><Name>0b3944af-12ff-1e1f-ac5c-55485fc10671</Name><Value>MANAGED
REAL TIME APPLY</Value></Setting><Setting><Name>12f3e541-ec4d-7310-be57-fa3bde0bed06</Name><Value>rac2</Value></Setting><Setting><Name>18900065-cbab-1788-0c0e-2c97ef41d7d2</Name><Value>LGWR</Value></Setting><Setting><Name>1a42a2bc-9e87-029a-6db6-f512f8d00ca2</Name><Value></Value></Setting><Setting><Name>5c92d602-346f-9202-e2b4-7cfdc7226661</Name><Value>Common</Value></Setting><Setting><Name>5e763b8d-1979-d0e9-08bf-ea857c67fbaa</Name><Value></Value></Setting><Setting><Name>7dd3b6cb-8180-d58a-fae2-985730c3ffb8</Name><Value></Value></Setting><Setting><Name>7ecb288d-50c8-dd5b-419f-5d4712ef4b75</Name><Value>Database
Physical</Value></Setting><Setting><Name>8c62b4e5-a737-b8d7-9017-4e8ea4c5a4ac</Name><Value></Value></Setting><Setting><Name>99cdac55-6b00-2485-cb26-bed62d37c926</Name><Value>Common</Value></Setting><Setting><Name>9ed7463c-fa1e-00ea-6743-5e54b928b7e8</Name><Value></Value></Setting><Setting><Name>a606d019-eff4-c6b9-eec9-32d43913fa58</Name><Value>DatabasePhysicalMP</Value></Setting><Setting><Name>aa9fbc8a-e274-91e8-7731-ba324682cab9</Name><Value>DatabasePhysical</Value></Setting><Setting><Name>afb4f9e6-bf48-1737-76ad-c9b3ec325b97</Name><Value>rac2</Value></Setting><Setting><Name>b60a9e0e-5baa-9b0a-8201-6ddd697d060d</Name><Value></Value></Setting><Setting><Name>bf194186-a061-ccf1-94de-971988a79003</Name><Value></Value></Setting><Setting><Name>c2ec02fc-fa4d-65c9-82da-fa76476433eb</Name><Value>rac2</Value></Setting><Setting><Name>d87b0cd0-2149-5736-ab85-42ebddab12a2</Name><Value></Value></Setting><Setting><Name>e1165cbd-8778-5bc1-de07-af341d420068</Name><Value>GRD</Value></Setting><Setting><Name>f56fc436-4a31-48dc-3127-745e5a6d2800</Name><Value></Value></Setting><Setting><Name>fda3d28f-3f34-e51c-e1ad-b1413a70c4ac</Name><Value>GUARD</Value></Setting></Settings></ClassInstance>.

Try to reimport the MP. You should use the following sql to query the problematic MP
select d.discoveryid, d.discoveryName,mp.mpfriendlyname from discovery d inner join managementpack mp on d.managementpackid=mp.managementpackid where d.discoveryid='ae39d630-f82d-77ee-d5d9-dd4c4a2e7bc2'
Discovery data couldn't be inserted to the database: Event IDs: 33333 and 10801
http://social.technet.microsoft.com/Forums/en-US/operationsmanagergeneral/thread/5f826064-d3f5-43f0-a328-c2f3eb56e932
Insert Discovery Data Fails
http://social.technet.microsoft.com/Forums/en-US/operationsmanagerauthoring/thread/8b1387f0-ca4a-480b-a0f0-557d14c3fa89
Event ID: 10801 in RMS server
http://social.technet.microsoft.com/forums/en-us/operationsmanagergeneral/thread/5467A6A7-0B26-45C1-BF69-368E5A7AB841
Roger

Missing characters in german keyboard layout

I'm using Gnome 3 on a PC with a german keyboard with a physical layout like this one. The german layout is configured in the settings and mostly works as intended, but one key is mapped wrongly. The key next to the left shift key in the second row from below is supposed to have the characters "<" and ">" on it, but instead it produces "-" and "_", duplicating another key. As a result, there is no way to type the "less" and "greater" characters. The built-in keyboard layout viewer shows an US physical one (like this one, notice e.g. the "enter" key which occupies only one row instead of two in the german one and the wider left "shift" key), regardless of whether the german or an english (logical) layout is selected.
So, if there is a way to change the physical layout, how do i do it?

Hi, I have had that problem a while ago...the other way around. I'm from Germany and was missing German characters on my US keyboard. I wrote about how I solved this here: https://myubuntux.wordpress.com/2012/01 … d-in-xorg/
I'm pretty sure this will help you solve your problem...at least introduce you to the tools I used.

SID NAME FOR PRODUCTION & STANDBY DATABASE IN DATA GUARD

Hi all,
we are trying to implement Standby database (Physical) between two
sites. We have a specific question regarding this:
==> Can we have SID of both the Database SAME..? If not, why.?
-- We have two HOST & DB Details like this and would like to implement StandBy
Database:
Site-1:
Host OS: Solaris-10 3/05 SPARC
Host DB: Oracle10g R-2 10.2.0.2.0
Host DB1: IGFL
Site-2:
Host OS: Solaris-10 3/05 SPARC
Host DB: Oracle10g R-2 10.2.0.2.0
Host DB2: IGFL
Thanks & Regards,
Rishikesh

My 2 Cents:-
SID can be same.
http://download-east.oracle.com/docs/cd/B19306_01/server.102/b14239/create_ps.htm

OBIEE Issue - Physical Query not picking cascaded Conditions

In one the OBIEE Reports, I have given a set of NOT LIKE conditions as below:
Session Name is not LIKE (pattern match) '%AUDIT%'
AND Session Name is not LIKE (pattern match) '%audit%'
AND Session Name is not LIKE (pattern match) '%Audit%'
But when we refresh the report, it filters only records which uses AUDIT keyword as part of Session Names. When I checked the session log (attached for reference), I found that though logical table query takes those three conditions, while firing against database Physical Query contains condition only for AUDIT wild card as filter. Whether there is any limitation applied on using NOT LIKE conditions or is it a bug or the way the conditions are sequenced is wrong (in OBIEE sense).

Thanks buddy. Though this workaround will solve the issue, I am curious to know why OBIEE behaves such a way. Its peculier to see OBIEE admits all 3 conditions in logical query where the Physical query skips few of the conditions in session log itself.

Database physical layout

Similar Messages

Maybe you are looking for