Multiple data sources with UME

Hi,
I want to run several (web) applications on my SAP WebAS. And these applications should have different user stores. That is, one should be a database, one a SAP ABAP System and another should be another ABAP System.
How can I configure the UME correctly?
Do I have to configure different data sources? Or do I have to write my own loginmodules?
And what's the situation if I have WebDynpro Apps? There I cannot use custom loginmodules, can I?
In the SAP Library it says that "UME can be configured to read and write user-related data from and to multiple data sources".
http://help.sap.com/saphelp_nw04/helpdata/en/e5/618a3eacd49076e10000000a114084/frameset.htm
Any hints?
Best regards,
Kilian.

Hi Michael, hi John,
All these 3 apps are independent, i.e. they have different users no "overlapping" users. We need for every independent, homogeneous group of users an own user store.
As John stated, we can edit a UME configuration XML file and define multiple data sources in it. However, this is still a server-wide user store.
Is there no way to assign an application to a user store or a UME data source and not have a server-wide user store?
And if not where can I get a description of these "initial namespace-attribute-value triples" mentioned in the SAP Library on User-Based Data Partitioning (see http://help.sap.com/saphelp_nw04/helpdata/en/cd/eafc3f8fc2c542e10000000a1550b0/content.htm).
Many thanks,
Kilian

Similar Messages

Bursting with multiple data sources

Hello all,
I am working in a retail environment and we are building reports for the stores. There are several reports and they all receive the same reports. We are using BIP Entreprise with the bursting feature and it works very well.
The problem is that there are too many reports and we would like to combine all reports of the same store within the same PDF. We use the multiple data source feature and was able to concatenate all reports together.
Using this new report and template, we can not make the bursting functionality to work properly: the bursted report do not contain all the required report components.
For example, I have 2 queries A and B each producing reports for 3 stores: 2, 4 and 5. Without bursting, the output is as is:
Query A - Store 2
Query A - Store 4
Query A - Store 5
Query B - Store 2
Query B - Store 4
Query B - Store 5
With bursting into a file using the store number, I get the following:
File 1 - Store 2 - Query A
File 2 - Store 4 - Query A
File 3 - Store 5 - Query A
+ Store 2 - Query B
+ Store 4 - Query B
+ Store 5 - Query B
I would like to have the following output:
File 1 - Store 2 - Query A
+ Store 2 - Query B
File 2 - Store 4 - Query A
+ Store 4 - Query B
File 3 - Store 5 - Query A
+ Store 5 - Query B
The main question is: can this be done using BI Pub Enterprise? We using version 10.1.3.3.2
If it is, can you provide me with help on how to configure either/or the requests and template to accomplish this task?
I create an SR and Oracle Support does not have an answser and suggested that somebody in the Forum might help.
Thanks in advance,
Minh

I would like to have the following output:
File 1 - Store 2 - Query A + Store 2 - Query B
File 2 - Store 4 - Query A + Store 4 - Query B
File 3 - Store 5 - Query A + Store 5 - Query B
the bursting level should be give at
File 1 - Store 2 - Query A + Store 2 - Query B
so the tag in the xml has to be split by common to these three rows.
since the data is coming from the diff query, and the data is not going to be under single tag.
you cannot burst it using concatenated data source.
But you can do it, using the datatemplate, and link the query and get the data for each file under a single query,
select distinct store_name from all-stores
select * from query1 where store name = :store_name === 1st query
select * from query2 where store name = :store_name === 2nd query
define the datastructure the way you wanted,
the xml will contain something like this
<stores>
<store> </store> - for store 2
<store> </store> - for store 3
<store> </store> - for store 4
<store> </store> - for store 5
<stores>
now you can burst it at store level.

Adhoc Query Requirement with Multiple Data Source

Hi All,
I have a Adhoc Query Requirement with Multiple Data Source. Is there any way to achive it. Other than Resultant set and bring into Model.
Thanks
SS

You can compare stuff in the EL, but I don't think this is what you need.
You can just use Java code in the backing bean class for all the business logic. You can use DAO classes for database access logic. Finally for displaying you can use the JSF tags such as h:outputText.

Web Analysis : populate the same table with multiple data sources

Hi folks,
I would like to know if it is possible to populate a table with multiple data sources.
For instance, I'd like to create a table with 3 columns : Entity, Customer and AvgCostPerCust.
Entity and Customer come from one Essbase, AvgCostPerCust comes from HFM.
The objective is to get a calculated member which is Customer * AvgCostPerCust.
Any ideas ?
Once again, thanks for your help.

I would like to have the following output:
File 1 - Store 2 - Query A + Store 2 - Query B
File 2 - Store 4 - Query A + Store 4 - Query B
File 3 - Store 5 - Query A + Store 5 - Query B
the bursting level should be give at
File 1 - Store 2 - Query A + Store 2 - Query B
so the tag in the xml has to be split by common to these three rows.
since the data is coming from the diff query, and the data is not going to be under single tag.
you cannot burst it using concatenated data source.
But you can do it, using the datatemplate, and link the query and get the data for each file under a single query,
select distinct store_name from all-stores
select * from query1 where store name = :store_name === 1st query
select * from query2 where store name = :store_name === 2nd query
define the datastructure the way you wanted,
the xml will contain something like this
<stores>
<store> </store> - for store 2
<store> </store> - for store 3
<store> </store> - for store 4
<store> </store> - for store 5
<stores>
now you can burst it at store level.

BI+ Reporting With Multiple Data Sources

Hi -
We've been using BI+ Reports with multiple data sources on separate grids. We are wondering if there is a better way to ensure the user has their point of view bar in sync between the the data sources. In our case, both Data Sources are separate HFM applications with all the dimensions identical except the account which is defined on the report. We are not using Metadata Management.
I'm aware of the "Merge Equivalent Prompts" in the Preferences > Financial Reporting window, but this does not always ensure the users have the same point of view.
What practice are other companies using to make sure the user runs the reports accurately with both point of view bars pulling the same entity/year/period.
Thanks in advance.

Thanks for your answer, however my question - I know it looks quite messy - is not directly related to data templates, my problem is that at the moment I am using a stored procedure following - almost exactly - what another procedure does in oracle EBS to call a BI publisher report.
1. This procedure populates some temporary table with the xml data where the concurrent request picks it up and uses it to populate the BI publisher template.
2. However I can't go and specify the XML location in my data template as this is not an option for us (i.e. reading the xml from the temporary table), so I am looking for another way to populate the BI publisher template (Note: I can't as well include the SQL queries in my Data template).
Thanks ;)

Increase performance in a multiple data source report

Post Author: ArturoFromPeru
CA Forum: General
Hi everybody.
I have a big problem, when i make a report that have multiple data sources the performance decrease considerably, i still don't know the reason. Is it true that using multiple data sources is too slow?.
I tell you a fact: I made the report connecting via "Field Definition File", and it take at least 2 minutes to show itself but it only have 170 records, and i'm very sure that the sql statements to each data source were executed very faster.
I even heard about index, but i think they only are useful when i connect directly to the database.
All my partner told me, that Cristal Report
Thanks you in advance
Regards.
Arturo

Post Author: ArturoFromPeru
CA Forum: General
Thanks you, but I explain better what i was doing.
I've made a report called "Kardex de Producto" which show the behavior of the products according to its sales, purchases.
The end result which is showed by the report is correct. I used three data source in the report, so links between each table is correct. The only problem i have is the performance because i still don't know why is too slow. In fact the sql statement is very heavy, but i think it doesn't matter to the report because when it takes the data from the database (connecting via Field Definition File) even if the very heavy sql statement returns twenty record Cristal Report must show itself very quickly. I don't want to believe that Cristal Report is very slow when it works with multiple data sources.
Do you mind if i give you my report? if not please leave me your email address.
Email: [email protected] / [email protected]
Thank you in advance
Regards
Arturo

Report Viewer and Multiple Data Sources

I know that it is possible to create a report in Crystal Reports using multiple data sources. But is it possible to use just the Free Report Viewer to view a report with Multiple Data Sources?
Our company uses Crystal Reports XI. I do not use the program myself, I am in the IT Department. I have limited knowledge of Crystal and do not do any of the report writing.
If this is possible, and if someone could help me out with what would need to be done, or point me in the right direction of a Knowledge Base or Help Topic that explains this, that would be great.
Thanks.

I believe it can but both data sources need to be set up. Ask a report designer to help you use the Designer to test this and see what is required.

Any examples of a data template using multiple data sources?

I'm looking for an example report using multiple data sources. I've seen one where they do a master/detail but I'm just looking to combine results in sorted order (sorted across all data sources). The master/detail used a bind variable to link the two defined queries, I'm thinking what I want won't have that, so I'm lost on how to make that happen. I have reports using multiple sql queries and there is a way in the data source pulldown to tell it to combine the data sources. It appears to be a more manual process with data templates, if it's even possible.
Any pointers/links would be appreciated.
Gaff

Hi Vetsrini :
That's just it. Mine is simpler than that. There is no master/detail relationship between the two queries. I have the same exact query that I run in two databases and I want to merge the results (ordered, by, say eventTime) in one report. So I think my results are going to be two separate groups (one for each data source) which I'll have to let BI merge vis XSLT or whatever it uses. That's fine for small result sets but for larger ones, it would be nice if the database did the sorting/merging.
Gaff

Multiple Data Sources In One Logical Table

I am new to OBIEE and I have came accross an issue. I appologize if this information is in the forum somewhere but I have searched and cannot find it.
My situation is that I would like to have one logical table that contains multiple data sources which have all the same columns. I already have session variables setup to differentiate the user's security through a row-wise variable for a specific column and a session variable for another column which determines the user's association to the data source in which they belong to. This security works well when the data sources are seperated in the Business Model and Mapping portion but the issue that arises is that the user's cannot share reports when the data sources are seperated in the BM&M.
I have dragged and dropped a table from the Physical model to the BM&M, I then dragged the second data source (with same meta data structure) over to the "Sources" folder in the first data source table in the BM&M. On the Content tab or each data source table I have defined the WHERE clause as such, where VALUEOF(NQ_SESSION."SCHOOL") session variable is my row-wise column filter and the VALUEOF(NQ_SESSION."GROUP") filter is my data source determinative:
sandbox."".SANDBOX.OBIEE_CROSS_ENROLLMENTS.HOME_SCHOOL = VALUEOF(NQ_SESSION."SCHOOL") AND sandbox."".SANDBOX.OBIEE_CROSS_ENROLLMENTS.DATA_SOURCE = VALUEOF(NQ_SESSION."GROUP")
Unfortunatley this only returns values in the BI Answers for the first drag and drop Pysical table in the BM&M layer and not the second Physical table drug into the "Sources" folder. I have also tried to create a new logical table and drag both tables into the "Sources" folder to no avail. I have experimented with the Fragmentation content on the "Content" tab of the seperate logical tables, checking the "This source should be combined with other sources at this level", which gives me an error in BI Answers that a column does not exist which really does.
What could I be missing? Advanced thanks to those who reply.
Thank you,
Kyle

Stijn,
Thank you for the article link. That was very helpful! It seems that I had a few things off as you do need the "This source should be combined with other sources at this level." checked. In my two table source columns for DATA_SOURCE I defined a literal ('086496' and '085597' for the other) in the Column Mapping tab. I pasted the following in the Fragmentation content, checking the "This source..." box on the Content tab:
eSIS.SANDBOX4_SCHOOLS.DATA_SOURCE = '086496'
And pasted the following into the WHERE clause, checking "Select distinct values" on the Content tab:
sandbox4."".OBIEE.NWOCA_SCHOOLS.SCHOOL_CODE = VALUEOF(NQ_SESSION."SCHOOL") AND sandbox4."".OBIEE.NWOCA_SCHOOLS.DATA_SOURCE = VALUEOF(NQ_SESSION."GROUP")
This took care of my user's security, utilizing the session variables in the WHERE clause. I am now able to generate reports that only one user can access from one data source and share that same report with another user who can only see data from the other data source.
Many thanks!!!

Single .rar file for multiple data sources

Can we configure multiple data sources into single .rar file? . Currently,
we are creating a seperate .rar file to configure each datasource by
'only' modifing ra.xml file. This is proving to be a tedius operation in
some cases when the application is integrated with multiple data sources.

Siva-
Unfortunately, no. One of the limitations of the resource adapter spec
is that it doesn't provide a means of specifying multiple separate named
configurations, so you can only have a single configuration.
If you need more flexibility, you may want to consider programatically
building up the PersistenceManagerFactory and then binding them into
JNDI.
In article <cqscdo$n1o$[email protected]>, Siva wrote:
Can we configure multiple data sources into single .rar file? . Currently,
we are creating a seperate .rar file to configure each datasource by
'only' modifing ra.xml file. This is proving to be a tedius operation in
some cases when the application is integrated with multiple data sources.--
Marc Prud'hommeaux
SolarMetric Inc.

LDAP as data source for UME

Trying to use a SSL enabled LDAP (Sun) for data source for UME. It seems that I can't use SSL directly from GRC CUP 5.3. Followed the instructions in saphelp, but when I test the connection, it gives me "Connection test with user path failed". The following is the connection data in UME Config:
Server Name: 10.56.17.20
Server Port:     62636
User:                cn=GMACApp_001,ou=Applications,dc=gm,dc=com
Password:       <correct one entered>
User path:        ou=People,dc=gm.dc=com
Group path:      ou-Groups,dc=gm,dc=com
Use SSL for LDAP Access is checked
Use Unique Attribute is not checked
I can connect to the LDAP using the same credentials with Softerra browser....Any ideas?

Opened a message with SAP....the response was less than helpful..."we don't support SSL". When I pushed them with the responses I recieved from the forum, the replay was "we have never done this". There must be a way. I can't be the only person on the planet that has to connect to a corp LDAP with a secure port!! I have tried the trick of conencting a LDAP as a data source for UME, but with limited success. Seems when the LDAP + db is enabled, the UME URL is not available (error 503). So that's not working so well either.
Any help will be appreciated.

Configure a single ODI install for multiple Data Sources in ERPI

I am setting up multiple data sources in ERPI to point to different EBS instances. How would I go about setting up ODI to support that without multiple installations of ODI? or is multiple installs the best way to go, from an 'ease of management' perspective?
Thanks!

First off - Have no knowledge of ERPi - Just googled it.
If its any help, we have multiple instances of EBS on the same box, we have seperate DWH (and ODI Work / Master Repos) environments aligned to these instances. An ODI Agent can only communicate with one ODI Work repository, hence we have Multiple Agents so we can use the EBIZ -> DWH pairs simultaneasouly.
Maybe someone will come along and tell us how it could work differently, until then - its works for us!

Multiple data sources in OC4J 10g

I've read through the documentation, searched metalink and searched the posts on the forums. There's so many examples out there that I've gotten myself confused. I haven't been able to locate an example for me, though.
After all that reading, I still don't quite understand JNDI and data sources. There's 2 problems that I'm running into that are somewhat related.
First off, most of the examples I've found seem to indicate that I can make up whatever name I want to put in for location. I seem to be missing something, because then my application cannot find that data source.
The second problem, is the basis for the first problem. I was able to create one data source which used the location jdbc/OracleCoreDS. However, since I work in an environment with multiple oracle databases, I wanted to create data sources for those other databases, the only difference being the URL. Enterprise Manager will not let me create another data source with the same location.
I was hoping someone could explain to me what else needs to happen to link up my own value for location so that my application can find the data source.
Thanks

Hello,
A data source entry is for one single database, so if you have multiple database to connect into your application you must create 1 entry per database (eg: jdbc/database1 jdbc/database2)
If you want to manage transactions over multiple databases (2 Phases Commit) you should the distributed transaction coordinator (DTC) provided within OracleAS.
For these 2 points (multi DB and transaction) I am inviting you to read the chapter 4-Datasources) on the "Oracle Application Server Containers for J2EE: Services Guide"
Regards
Tugdual Grall

CFTransaction & Multiple data sources

We have the need to update to multiple data sources under
CFTransaction. We have two MSSQL databases so I was able to create
a view in one to the other and that works (unless someone knows of
a problem that could cause). But we also have a MSAccess db that
gets hit too under the same CFTransaction. CF does not allow you to
have multiple data sources under one CFTransaction. If all the
tranasction control is on the db side, why does CF care if there
are different data sources?
Thanks.

As I said in my original post, in order for a transaction to
span multiple databases, you need to do what's called "2-phase
commit" or "distributed transactions". This is not a trivial thing
to do, nor is it something that is often needed (especially in
Coldfusion, which is 99% of the time talking to a single database).
Why do you need to do 2-phase commit? and why isn't it
trivial?
"2PC is an acronym for 2 Phase Commit. This is a protocol by
which data being committed to a database is committed in two
phases. In the first phase, the transaction processor checks that
all parts of the transaction can be committed. In the second phase,
all parts of the transaction are committed. If any part of the
transaction indicates in the first phase that it cannot be
committed, the second phase does not occur. ODBC does not support
two-phase commits."
http://docs.openlinksw.com/virtuoso/twopcimplementation.html
I don't think Access supports 2-phase commit. Access isn't
really intended for the kinds of things that need 2-phase commit,
which Microsoft readily admits (hence MS SQL Server).
> My thought is that it is not allowed because CF only
remembers one data source per cftransaction therefore can only send
one commit or rollback instruction. If that's the case, it sounds
like a simple enhancement.
No, this isn't allowed because distributed transactions is a
non-trivial thing. This isn't all handled on the database side, as
you reference. If you're dealing with one datasource, yes, it does.
Once you pull in a 2nd datasource, you need something that can talk
to all databases involved (a transaction coordinator) and 1) check
to make sure everything can be committed safely then 2) commit
everything (hence 2-phase commit).
Basically, you're trying to do something that's non-trivial,
you don't understand why it's non-trivial, so you're getting
frustrated. I realize you're not a Java shop. I was simply trying
to explain to you why it was non-trivial, and a method to work
around that if you so desire.

Data Extraction from Multiple data sources into a single Infoprovider

Hi Experts,
Can anyone send me links or examples on how to extract data from multiple data sources into 1 Cube/DSO.
Can anyone send me example scenarios for extracting data from 2 data sources into a single Cube/DSO.
Thanks
Kumar

Hi Ashok,
Check the following link from SAP help. Ths is probably what you are looking for.
[ Multiple data sources into single infoprovider|http://help.sap.com/saphelp_nw70/helpdata/en/e3/e60138fede083de10000009b38f8cf/frameset.htm]
Data from multiple data sources which are logically related and technicall have the same key can be combined into single record.
For example, if you have Sales order item and sales order item status .
These both have the same key - sales order and item and are logically related.
The Sales order item - provides information on Sales order item - material , price, quantity, plant etc.
The item status - povides information on delivery status, billing status, rejection status....
These are two different data sources 2LIS_!1_VAITM ad 2LIS_11_VASTI.
In case you have few master data attributes coming from different systems ie tow data sources in different systems say completely defines your master data. Then you could use a DSO or infoobject to combine the records.
If you want to see aggregated data you use a cube.
ie say you want to analyzae the customer revenue on a particular material and a particular project.
The project details would come from Project syatem and Sales details fom sales flow.
The data is then combined in the DSO with key cutomer and sales area say.
Then it is aggregated at cube level...by not using sales order in the cube model...so all sales order of the same customer would add while loading the cube.. to give direct customer spend values for your sales area at aggregated level.
Hope this helps,
Best regards,
Sunmit.

Multiple data sources with UME

Similar Messages

Maybe you are looking for