Development vs Execution Repository

Hi
Is it always a good idea to make the Production Environment an Execution Repository. Is there a good case for making it a Development Repository?
Regards
Geoff

user964802 wrote:
Hi
Is it always a good idea to make the Production Environment an Execution Repository. In my opinion , it is better to have Production Environment always an Execution Repository
In this case , only scenario (compiled code) can be imported from a dev environment , thus reducing the scope of unwanted changes .
Is there a good case for making it a Development Repository?
I am not aware of any .. please do share if you find a good case.
Regards
GeoffThanks,
Sutirtha

Similar Messages

Can we change the Work Repository from Development to Execution?

hi,
Can we change the Work Repository from Development to Execution? if yes then how can we change it?
Regards,

AMSI wrote:
hi,
Can we change the Work Repository from Development to Execution?No .
if yes then how can we change it?
Regards,

Do we need Scenario for all to deploy to EXECUTION repository

I am just trying to find this piece of information all over and could not find it(probably my search word may not be correct one)
I have package (main one and generated scenario for that) which has scenario of other packages, some procedures as well as some interfaces. It's big junk of all the combination of objects
My question is do we need to generate Scenario for all interfaces and procedures in order to promote to Execution repository
OR just the main package's scenario and the child scenario is enough.
I do not have scenario for Interfaces as well as odi procedures, do I need to generate scenario for all objects,
Please any help would be greatly appreciated.
I just found this link
ODI Export Scenario, only main?
As per the link I do not need to worry interfaces nor ODI procedures, which do not have a scenarios in DEV repository when go to EXECUTION repository.
Please make me correct, if I am wrong
Thanks,
Edited by: user12236189 on May 11, 2012 4:12 PM

You only scenarios to be taken to the execution environment.
If a scenario calls other scenarios, then you need those child scenarios too.
All the interfaces/procedures are not needed since they are already merged into your master package scenario

ODI Console - Execution Repository

Hi All,
Using ODI11g - One master repository , one DWR and one EWR.
When i login into odiconsole using SUPERVISOR using work repository option.. i can execute scenarios in the Development Work Repository(DWR). But How can i execute scenario in Execution work repository(EWR) using ODIConsole ???
-Henry.

I think this how it works -
When ODIConsole is installed it creates 2 jdbc connections on weblogic. It points to master and work repository using this jdbc connections.
Hence you can only connect one work repository using ODIConsole.
I change the JDBC to point to EWR and it started showing up objects from that work repository.
It would have been nice if there were option to connect to any work repository while logining into odiconsole. Now the only way around is changing JDBC connection to point to repository of your choice.
-Henry.

JDev 10.1.3.0.4 - Different source path from development to execution?

Hi,
I've just migrated to JDev 10.1.3.0.4.
Getting a java.lang.NoSuchMethodError in execution while class is correctly available in development IDE.
I'm almost sure this is a bug, I want to investigate this issue otherwise I can't migrate.
I think embedded ClassLoader is loading from a different path then used in development.
This is my first step in debugging:
Given an object instance how can I get the URL used by ClassLoder to load the class?
Object o = ...
ClassLoader cl = o.getClass().getClassLoader();
...need the class system path...
thanks for any help
aldo

Aldo,
the error is not that the class cannot be found but the method you call. This could be subject to a change in the signature of the method
Frank

SQL Developer Data Modeler Repository

Hi,
I would like to know how to save all my applications into the Data Modeler Repository instead of doing it piece by piece and having to create a dmd file for every single application I imported into Data Modeler.
In Oracle Designer, everything is in the repository and I can run reports against the tables and views belonging to that repository. So, I can produce a pdf documentation out of it. Can we do the same with Data Modeler?
Thank you!

Hello Anonymous,
I'll start with your first comment - "I would like to know how to save all my applications into the Data Modeler Repository ". The Repository is just for reporting purposes, as Dimitar points outs, so saving all your applications in the Data Modeler is merely File - Save. This action saves the current design (with all its models) to the file system. The main premise here is that the Data Modeler is a file based product and all models and designs are saved locally to the file system. This is a positive move for many customers who are now building applications in Java or using any of the application development environments that work with a variety of files. They are taking all the artifacts and placing them under version control using open source versing tools like Subversion. Using the Data Modeler, they can do the same.
For reporting you have a choice, one of which has also been explained. The action of creating a schema before you export the design is a one off step. Following on from this you can export new versions of the model to the same repository. The reason we did this was that many of the Designer audience wanted to write and run their own reports. So here you have a set of tables that you can write whatever SQL query you want to use.
The other reporting option is to use the integrated reports in the product and you do not have to export the design or open another product or write SQL.
Have you used the Design Rules in the Data Modeler? These are a set of "reports" for quality assurance purposes. There are many of these reports at all levels of the design which verify that your model is defined according to a set of database rules and general business rules and if you like, you can add your own rules.
I attended ODTUG last week in LA and was really pleased to listen to a number of talks by well seasoned Designer users who have successfully transitioned to the Data Modeler and who had many positive experiences.
Yes, the tools are different - but the one tool was not designed to replace the other - Data Modeler does none of the application design and generation that Designer does, those customers who have wanted to replace one tool (people are moving to the data Modeler from different tools) with the Data Modeler, have reported to me that they are pleased with the change.
Regards
Sue Harper
Product Manager

SQL Developer Data Modeler, Repository or file based?

As announced, in a next release of SQL Developer Data Modeler it wil be repository based. Ok that is nice. But what does it mean for the current possibility to store your work in files. Will the possibility to store in files be supported in new releases when also a Repository is available?

Will the possibility to store in files be supported in new releases we are not going to remove that run and use aspect of the tool.
Philip

Repository based development vs non Repository

hi *,
as in caps 6 it is possible to create projects outside the repository. one BIG question for me is when to choose repo based development and when not to choose repo based development. has anyone some thoughts to share about this topic?
secondary question for me is how long will repo based projects be supported? will 7.0 or 6.5 also be able to import 5.0.5 JCA projects?
please share
regards chris

Hi chris,
JavaCAPS6 isn't released so there isn't any best practices yet available in terms of repos vs non-repos projects.
Sun has a 5-7 yr end of life policy for its software (I can't remember the exact number). That is, after something is announced "end of life" Sun continues to support it for that period of time. JavaCAPS6 hasn't yet been released, let along announced "end of life", so repos projects supported by JavaCAPS6 will be supported for at least that EOL time frame and most likely much longer.
Jason.

Agent Execution Repository

How does a agent determine which work repository to execute scenerios from and write execution logs to.

This is not what I have observed. I started the agent in "agent" mode. Then I ran the startscen.bat to execute a scenerio. The agent will then refer to the odiParams.bat file to determine which work repository the scenerio exists in and write the execution information to that repository. Perhapse we are saying the same thing but I do not specify the work repository connection information "In packet initiating the execution", it is taken from odiParams at the time of execution. I have not tested this as a dynamic parameter, each time I have changed it I have restarted the agent.
T

Win development qnx execution

Hi,
Most of the drivers for a given OS are developed and copiled under given OS. How to solve the problem of software (and driver) development on a computer
with a different OS thatn the target machine? For example software development in LV on Windows for QNX OS. I know that it is possible to use EDM mod
(Microprocessor SDK) to compile for other OS and architectures (I've done that and it works). But where do hardware drivers compiled for QNX fit in?
Even if you use MHDDK (if I understand correctly) one compiles the driver (which is written in C) on the machine with QNX OS.
How am I able to use it in a software that I'm developing under Windows?
Is it possible to include it in some form as a "external" code by using System Exec, CIN or CLFN?
If I'm going in a wrong direction please suggest the correct one
Just to clarify. I develop software under Windows and want to communicate with Hilscher Profibus card on a machine with QNX Neutrino.
regards
souske

Hi!,
I've been working with LabView 8.2 and CompactRIO, now i'm evaluating Lv Embedded runing on a QNX target. I don't have any problems building a vi if this doesn't contain vi's like those in Signal processing and Mathematics (actually only runs tools from Elementary & Special Functions, from the Mathematics section), however if i try to build a vi containing a block like those a 'Linker error occurred' and appears the next message:
C:\Documents and Settings\Administrator\My Documents\LabVIEW Data\Labview_Regulator\Code Generation\Code Generation\QNX_Console\QNX_Application\NI_AALBase_lvlib_Real_Dot_Product.o: In function `NI_AALBase_lvlib_Real_Dot_Product_BlockDiagram':
C:\Documents and Settings\Administrator\My Documents\LabVIEW Data\Labview_Regulator\Code Generation\Code Generation\QNX_Console\QNX_Application\NI_AALBase_lvlib_Real_Dot_Product.o(.text+0x9c5): undefined reference to `LVThreadClose'
cc: C:/QNX632/host/win32/x86/usr/bin/ntox86-ld caught signal 1
Anyway, i tried including NI_AALBase_lvlib_Real_Dot_Product.o into Extra Linker Flags but a similar error occurs and shows me next:
C:\Documents and Settings\Administrator\My Documents\LabVIEW Data\Labview_Regulator\Code Generation\Code Generation\QNX_Console\QNX_Application\NI_AALBase_lvlib_Real_Dot_Product.o: In function `NI_AALBase_lvlib_Real_Dot_Product_InitFPTerms':
C:\Documents and Settings\Administrator\My Documents\LabVIEW Data\Labview_Regulator\Code Generation\Code Generation\QNX_Console\QNX_Application\NI_AALBase_lvlib_Real_Dot_Product.o: In function `NI_AALBase_lvlib_Real_Dot_Product_BlockDiagram':
C:\Documents and Settings\Administrator\My Documents\LabVIEW Data\Labview_Regulator\Code Generation\Code Generation\QNX_Console\QNX_Application\NI_AALBase_lvlib_Real_Dot_Product.o(.text+0x9c5): undefined reference to `LVThreadClose'
cc: C:/QNX632/host/win32/x86/usr/bin/ntox86-ld caught signal 1
I've attached the vi which i referred above, i tried with different vi's and it was the same result. I have installed QNX Momentics 6.3 and LabView in the same computer.
Can anyone help me?
Regards
OsS
Attachments:
Test III.vi ‏11 KB

Activating subscription service on a custom developed read only repository

Hi Experts,
We have a custom implementation of a read only KM repository manager which just expose documents, but does not have any editing functionality. Now the client wants to activate subscriptions on that repository manager. The current implementation of the manager is not sending any emails. Do you know what should be implemented by the manager in order to get the subscription service to work? We are looking for a daily notification as it is clear that because it is a read only manager there is no way to get notification for all events.
Regards,
Tania

Hi,
See the Javadoc package description com.sapportals.wcm.repository.service.subscription for:
Detailed explanation of the API and underlying concepts
UML diagram of interfaces
Code sample: Creating a subscription
https://media.sdn.sap.com/javadocs/NW04/SPS15/km/com/sapportals/wcm/repository/service/subscription/package-summary.html
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/com.sap.km.cm.docs/library/kmc/Knowledge%20Management%20and%20Collaboration%20Developers%20Guide.html
Patricio

How to use a Work Repository when type is set to 'Execution' ?

Hi All,
I was just trying my hand at importing and exporting work repositories , I need to know how we can use a certain work repository when its type is set to 'Execution'
I want to import some packages (which are in some other Work repository ) into that Work repository but I am missing something somewhere.
Can somebody tell me what are the exact steps to import packages or Entire Work Repository into this newly created 'Execution' type Repository ??
Regards.

Hi,
An work repository which is of EXECUTION type cant be used to import packages/interfaces etc.
You can able to import only SCENARIOS in Operator component (from scenarios tab) and "just" execute it. As the term execution repository stands, u can able to execute objects created in Development repository.
Moreover, u cant access designer itself if its a EXECUTION repository.
Thanks,
Guru

Error in Migration Repository using Sql*Developer Migration Workbench

To migrate a MS access mdb database to Oracle using Sql*Developer Migration Workbench, I try to create first the the Migration Repository in SQL*Developer.
The repository is created, but under PACKAGES I could see that package MD_META has been compiled with errors.
PACKAGE BODY AO_PERSONAL.MD_META@DMCAPX (AO_PERSONAL):
Error(27,2): PL/SQL: SQL Statement ignored
Error(30,23): PL/SQL: ORA-00904: : ungültiger Bezeichner
Well, when I ignore this and continue with the next step of the migration process "Capture Microsoft Access Exported XML file" to create the Captured Model, I get the following error message:
ORA-04063: package body "AO_PERSONAL.MD_META" enthält Fehler ORA-06508: PL/SQL: aufgerufene Programmeinheit : "AO_PERSONAL.MD_META" konnte nicht gefunden werden ORA-06512: in "AO_PERSONAL.MD_PROJECTS_TRG", Zeile 3 ORA-04088: Fehler bei der Ausführung von Trigger 'AO_PERSONAL.MD_PROJECTS_TRG'
Target database RDBMS is Oracle Database 10g Enterprise Edition Release 10.2.0.3.0. So the proposal of note 427916.1 to create the Repository in a 10.2 RDBMS does not solve the problem.
I even tried to use the Quick Migrate feature in SQL*Developer, but got the same error message.

"inserted value too large for column"
Is there a LOGON trigger on the database or user ?
I'd be surprised at SQL Developer trying to insert data into anything on connect.
I'd suspect code being executed by the database, maybe using a MODULE or PROGRAM from v$session or USERENV / SYS_CONTEXT.

How to deal with more than one work repository vividly?

I create one master repository with two work repository , now how can i vividly build up ODI interface in differently work repository?
anyone reply will be appreciately,thank you very much!

There are two kinds of work repositories - Development and Execution.
Usually you have one Development repository which you use to develop packages, interfaces, procedures, variables etc.
As soon as you object is ready to deploy you can create scenario from it. Afterwards you can move scenario to any execution repository you have using export/import.

Conect to Work Repository from Open Tool

Hello.
I'm developing a new Open Tool plugin, and I would like to know if there's any way to connect to the work repository through ODI Java API.
That's it, I could add JDBC connection parameters to the plugin so the user can put there the driver, the JDBC URL, etc. for connecting to the Repo. But, if it's possible, I'd prefer to connect through some Java functions. Another option would be to add a schema in the Topology for the work repository DB, and read the connection data from it, so the user only had to enter the name given to the schema.
In a few words, my aim is to make user's life easier and make the plugin looks well integrated into the ODI platform.
Also would be great to know about a good advanced Open Tool development tutorial.
Thanks and best regards,
Carlos.

AMSI wrote:
hi,
Can we change the Work Repository from Development to Execution?No .
if yes then how can we change it?
Regards,

Development vs Execution Repository

Similar Messages

Maybe you are looking for