Best methodology to have dev and production?

HTML DB users-
A customer asks what are some of the better methodologies to have a development and production environment with HTML DB? They are used to having controlled releases of new features and fixes into the production environment and want to do that as well with HTML DB.
We would like your thoughts...
The customers exact question:
Our thinking is that we develop in a development workspace and app and then migrate via export and import to a production workspace and app.
What we have encountered is several issues:
1) After the workspaces have been set up the user can only point to one at a time.
2) We get integrity violations in the export an import.
Is there a better methodology/solution?
You help is greatly appreciated.
Cheers,
-david

david,
Different options for separating production from development environment are discussed in this best practices whitepaper:
http://www.oracle.com/technology/products/database/htmldb/pdf/Oracle_HTML_DB_Best_Practices.pdf
Do you have more details on the two issues that were raised about the current approach:
1) After the workspaces have been set up the user can only point to one at a time.
2) We get integrity violations in the export an import.
Why can the user only point to one at a time? What does that mean? Can you describe the steps used to export and import and show as error messages?
Sergio

Similar Messages

Dev and Production BW systems not in sync.

Hello All,
We have been using two BW systems for our implementation project. One is the development and second one is production BW system. I observed that few BW objects have different configuration (like update rules, user exits, queries , web templatesetc) in the production compared to the development system. Now my requirement is to find out the all the BW objects in the production that having different status than development. So that I can make corresponding changes in the development system as well in order to sync dev and production systems.
1. Is there any tool/transaction is available by which I can retrieve the list of the BW objects and may be delta changes between dev and prd systems.
2. Or any standard process exists to sync the both versions.
Any help to this will be highly appreciated.
Thanks in advance.
Regards
Harsh.

Hi Harsh,
For the Transfer rules/update rules you can do a remote compare of the Generated Program which would give you an idea whether they are in sync or not. Likewise you can do remote compare of the User exit fn module.
Bye
Dinesh

ONE BI publisher for DEV and Production

Hi,
is there anyway I can setup ONE BI publisher for both the dev and production environment. the dev and production are in
separate boxes? the reason I need to do this is that oracle charges licenses fees for the development as well.
Thanks,
Odeh

Hi,
Iis technically possible to have two different databases pointing at BI Publisher, but you want to make sure the users can access the correct reports. But what I would say is that it isn't a particular good way to have your environments setup. The point of having a dev and a prod environment is to ensure that things are tested and proof of concept can be carried out without affecting the live systems. I would suggest that the separate licenses is worth the cost from my own personal experience, otherwise you risk having a live environment that is quite unstable.
Suzanne

Org.structure differs on dev and production

Hi,
I'm developing a workflow and in one of the steps (activity step), I assign agent using organizational structure (position from HR). The problem is, that the org.structure differs on dev system and on production system. Mainly the IDs of position posts differ. How to solve this? Thanks in advance.
Best regards,
Tomas

>
Tomas Linhart wrote:
> thanks for answer, I guess that's the best way of handling such situation. I'm going to creat custom table holding position used in workflow mapped to real position ID from org.structure. Or more generally, holding WF post mapped to org.structure type (position, job, ...) with corresponding ID in org.
I don't see why you need a custom table?
The org issues aside, all that is needed is to assign your org unit to the task. Then leave the agent assignment blank in your WF. Job done.
The other option is to create a dummy rule, which you thought was to be avoided. I don't agree, it takes 5 minutes, less effort than a table, less non-standard stuff ==> less explaining to people how to maintain & easier to troubleshoot.
Edit: Forgot to answer your other question
> I'm not sure if you consider org. replication the same as transport of org. I was thinking of transport myself, just didn't know the way how to. I've found transaction RE_RHMOVE30 (also acessible from customizing), that should do that, but am not sure, if that's the right one and/or what options to select when transporting. Also, I'm not sure if this would ensure the position/job IDs are the same on both systems.
Replication would be carried out via transports. The RHMOVE* reports can also be used, or you could even set up an ALE connection between DEV and PRD. However I still would not recommend it. The development system is where you would need to create test users and/or org units. The other alternative is to have a master data client. All of these are a great deal of work which only make sense in certain types of environments - certainly not for the sake of a workflow.
Edited by: Mike Pokraka on Sep 27, 2009 9:52 PM

Protal in DEV and Production

I am a little confused. I want to be able to develop prolets in DEV (development) database
and then deploy it to production. Is there an easy way to accomplish this.
Thanks
null

Thanks for the tip. Does this mean that I have to install the heavy duty repository on Production as well as Development machine. Seems to me a big overhead. If only I could connect to the production repository to retrieve the data at runtime. Has anyone run into this issue or am I the only one ??.
I was hoping that it would run somewhat like Oracle Designer Product. Can somebody from Oracle confirm this
Thanks
null

Connection string in dev and production

I have a several servers that i want to pull data from.
I put the connection string in a OBJECT variable that come from configuration table and use it in the for loop mapping the value to local variable called connectionstring, this variable is the expression for the connection string property.
So far everything is working as expected.
my problem is that on my development i put my local server in the connection string variable so i will be able to connect to the server so i will be able to map source to destination.
When i deploy the solution to production the data in the configuration table is different obviously but i get a connection error that saying that it try to connect to my dev server.
Any idea why?
Thanks.

Seems like you haven't configured your package properly
Here are some important consideration for SQL Server Configuration
Now, let’s go through some important considerations that will help you to understand this configuration type better, and to avoid common implementation issues:
SQL Server configurations are very flexible, as you can store multiple configuration properties/values in the same table.
Multiple packages can use the same configuration table and filter.
You can have a configuration filter for each entry in the configuration table, or you can use the same filter to group a set of entries. If you use the second approach, make sure that all properties under the selected filter exist in the package
being configured. Otherwise warning messages are generated for those properties in the configuration filter that do not exist in the target package.
You can use existing database security and backup policies to protect the configuration table contents.
A drawback of this configuration type is that it relies on an SSIS connection manager to get access to the configuration table, but does not have built-in support for updating its connection string. In other words, the connection string that points
to the configuration table is hard-coded inside of the package, and you have to take extra steps in order to update it (for example, through an extra package configuration placed at the top of the Package Configurations Organizer, or in SQL Server 2008, by
using the /CONN option of DTExec).
The /CONN option of DTExec utility can be used to override the connection string used by the configuration only in SSIS 2008. This is because SQL Server 2008 Integration Services reloads configurations after applying DTExec command line options.
The configuration wizard can create the table and insert the required rows when they do not exist, but only if the direct method is chosen. The wizard never includes any sensitive data in the ConfiguredValue column of the table, and you would need
to manually update the table to add the required sensitive data.
If you use the direct method, and provide an existing configuration filter, the wizard will prompt you to either reuse the configuration information in the table or to override it with the property values from the current package. Make sure that
you understand this behavior, since choosing the wrong option could change the contents of the configuration table and produce unexpected execution results.
When you use the indirect method, the environment variable must contain the name of the configuration connection manager, the name of the configuration table and the configuration filter. The syntax should be:
“ConfigurationManagerName”;”Schema.ConfigurationTableName”;’ConfigurationFilter”
As with the direct method, you have to account for extra logic if you need to modify the connection string inside of “ConfigurationManagerName”

Problem in converting ASCII value in Dev. and Production

Hi...
The ASCII values for # differ in the development and the production system.
The code below (value 0009 ) populates # in the variable lv_sep.
DATA: lv_sep TYPE x.
FIELD-SYMBOLS : <field> TYPE x.
ASSIGN lv_sep TO <field> CASTING TYPE x.
<field> = 0009.
The the development # = 0009 and in production # = 1000.
Need to know why is this happening.

How do you get the XML document? Do you use XSU? You can use:
String str = qry.getXMLString();
to get the result XML document in String.
XSU will escape all of the < and >. Or the the XML document in
one of the column will make the result XML doc not well-formed.
Not quite understand your problem. You can send me your test
case.
i have an application that needs to send an XML document
over the wire. For this reason, I need to convert the
doc into a String at the sending side and back to Doc
at the receiving side. This document is stored in a "CLOB"
column in a table in the database. I use XDK to retrieve
this entire row (including the CLOB - hence this is an XML
document which has a column that itself is an xml document in
string format - this is just the clob read in by XDK).
Thus the row looks like
<ROWSET>
<ROW>
<col1> A <col1>
<CLOB_COL> ..clob value of an xml doc..</CLOB_COL>
</ROW>
</ROWSET>
When I convert this document into String and back, one of the "<"
tags in the clob column document gets changed to "<" and hence
the parsing fails! I have used the latest label of the XDK build
to get the latest parser jar but still i have the same problem.
I am using the following routines for the conversion.
/* for converting document to string */
public static String convertToString(XMLDocument xml) throws
IOException
StringWriter sw = new StringWriter();
PrintWriter pw = new PrintWriter(sw);
xml.print(pw);
String result = sw.toString();
return result;
/* for converting string to document */
public static XMLDocument convertToXml(String xmlStr)
throws
IOException,SAXException
ByteArrayInputStream inStream = new
ByteArrayInputStream(xmlStr.getBytes());
DOMParser parser = new DOMParser();
parser.setPreserveWhitespace(false);
parser.parse(inStream);
return parser.getDocument();

Execute SQL Tasks Failing for Duplicate Syntax Between DEV and Production DB

Newbie here...be patient with me!
I added tasks to refresh two tables (delete from, insert into, update) to an SSIS project . I have them running from the WinXP scheduler. The issue:
In dev the tasks integrate and execute successfully from scheduler
In prod I can right click and execute each of the three tasks without any problem, but the same tasks cause my project to fail when executed from scheduler
Questions:
Any ideas about what I am failing to see?
How do I get a meaningful log messages from the tasks that are failing?
Thanks for your ideas...
Installed Edition: IDE Standard
SQL Server Analysis Services
Microsoft SQL Server Analysis Services Designer
Version 9.00.1399.00
SQL Server Integration Services
Microsoft SQL Server Integration Services Designer
Version 9.00.1399.00
SQL Server Reporting Services
Microsoft SQL Server Reporting Services Designers
Version 9.00.1399.00
Microsoft Visual Studio 2005
Version 8.0.50727.42 (RTM.050727-4200)
Microsoft .NET Framework
Version 2.0.50727
Installed Edition: IDE Standard
SQL Server Analysis Services
Microsoft SQL Server Analysis Services Designer
Version 9.00.1399.00
SQL Server Integration Services
Microsoft SQL Server Integration Services Designer
Version 9.00.1399.00
SQL Server Reporting Services
Microsoft SQL Server Reporting Services Designers
Version 9.00.1399.00

Newbie here...be patient with me!
I added tasks to refresh two tables (delete from, insert into, update) to an SSIS project . I have them running from the WinXP scheduler. The issue:
In dev the tasks integrate and execute successfully from scheduler
In prod I can right click and execute each of the three tasks without any problem, but the same tasks cause my project to fail when executed from scheduler
Questions:
Any ideas about what I am failing to see?
How do I get a meaningful log messages from the tasks that are failing?
If i understand correctly you are trying to run SSIS package from Window XP task scheduler. The reason for the failure may be due to permission while accessing resources required in the package. However it is difficult to suggest without
looking at the error message.
Scheduled tasks maintains a log file (Schedlgu.txt), in the c:\Windows folder. You can view the log from the Scheduled Tasks window by clicking
View Log on the Advanced menu.
The log file size is 32 kilobytes (KB), and when the file reaches its maximum size, it automatically starts to record new information at the beginning of the log file and writes over the old log file information.
Refer http://support.microsoft.com/kb/308558
http://msdn.microsoft.com/en-us/library/windows/desktop/aa383604(v=vs.85).aspx
Regards, RSingh

Best practice to have cache and calclockblock setting?

Hi,
I want to implement hyperion planning.
What should be the best practice to set essbase settings for optimize performance?

Personally, I would work out the application design before you consider performance settings. There are so many variables involved that to try to do it upfront is going to be difficult.
That being said each developer has their own preferred approach and some will automatically add certain expressions into the Essbase.cfg file, set certain application level settings via EAS (Index Cache, Data Cache, Data File Cache).
There are many posts discussing these topics in this forum so suggest you do a search and gather some opinions.
Regards
Stuart

Report working in different ways in BI Dev and BI production

Hi Expert,
I have created a Report in which i am giving input to the variable(Date input) and using this i am displaying all the values that is till the date that i entered, and it is working fine in BI dev, but in BI production it is showing only data that is equal to supplied date only.
Although i had checked all the restrictions and variables are as it is transported.
if you have an idea please share your knowledge please...
Regards
Prakash

Hi,
Are you using this date variable in the global filter or in any of the local KFs?
Are you using the operand 'less than or equal to' for getting all the values less than or equal to this date?
Is this operand same in both the systems?
Is there data in production prior to the date range that you have entered?
Can you check the information on the report output to find out what value is captured in this date variable?
What is the last changed date of the queries in dev and Production?
I hope any of the above should give a clue to solve your issue.
Thanks.

Key figure diffrence in Dev and Pro

Hi Experts,
I have one cube in Dev and Production but key figure Current Price out put is difference in both
like dev it comes 30.12 and in production getting like 30,12.
but make sure i am using same keyfigure.
Thanks
David

Check the user default settings in SU01D tcode in dev and pro.
SU01D -> enter ur user id -> goto Default tab and check the decimal notation.
I guess it is 1,234,567.89 in DEV and 1.234.567,89 in PRO.
Please make sure that this setting is same in DEV and PRO.
you can ask your basis consultant to changes the settings....
--- Thanks...

Development and Production Requirements for a Forms/Report application

Dear all,
We have an application witch run in client/Server mode using Forms/Reports 6i For 300 users.We want to migrate it on Forms/Reports 9I using Oracle 9IAS and Oracle 9IDB.
I would like to have advises about material requirements of both development and production environment((developer deskop, development appServer and databaseServer, end user desktop,production databaseServer and production AppServer).
Thank u for giving more details about my request.

I have my dev and production on 2 different databases/instances, and the workspaces have a different name aswell (think 'dev' and 'prod').
I do reuse my application ids, which makes the whole process a lot more manageable: easier for me to identify and replacing a whole application.
However, when i want to replace only 1 page, apex stumbles over the workspace. Bummer.
So i took an export of a page on both workspaces, and opened the files.
prompt Set Credentials...
begin
-- Assumes you are running the script connected to SQL*Plus as the Oracle user APEX_040100 or as the owner (parsing schema) of the application.
wwv_flow_api.set_security_group_id(p_security_group_id=>nvl(wwv_flow_application_install.get_workspace_id,1377027056414870));
end;
wwv_flow_application_install.get_workspace_id,1377027056414870));
I took both workspace IDs and keep them in a simple txt in my installation-files folder. Each time i need to import a page from dev->prod i go in the exported file from dev and change the workspace id to the id of my production workspace id. Everything works magically after that.
If your application ids also differ, you can try changing the id in the exported file aswell, but i can't tell if it works or not since i've not done that yet.
prompt Set Application ID...
begin
   -- SET APPLICATION ID
   wwv_flow.g_flow_id := nvl(wwv_flow_application_install.get_application_id,130);
   wwv_flow_api.g_id_offset := nvl(wwv_flow_application_install.get_offset,0);
null;
end;where of course you'd change around nvl(wwv_flow_application_install.get_application_id,130);

Dev and Prod Environment Sync.

Hello Experts,
I would like to ask for your kind help in the following matter: How to resynch the dev and production environments.
For some (inextricable) reason, the production and development servers are out of sync. Mainly the SLD content is in different versions. The production one is older. Oh and there is no QA environment.
Besides that the business systems were created manually in each server and have inconsistencies. No SLD information was transported between them, only the repository and some directory information. Other directory information was created in the production environment directly. And I am affraid some development took place in production server directly too.
So I have been asked to produce a plan to get all this mess into shape. So far at a macro level I have these tasks planned:
10.- Backup prod and dev servers.
20.- Load latest CIM model and CR content on both servers (and see what happens).
30.- Connect the SAP systems as SLD data sources to prod SLD.
40.- Redo configuration in prod to point to automatically created business systems (and retest).
50.- Cleanup of manually created business systems the prod environment.
60.- Config the prod SLD as data source for dev SLD.
70.- Redo configuration in dev server to make it match prod SLD.
80.- Cleanup of manually created business systems the dev environment.
90.- Create in the dev server the repository objects that were directly created in prod.
100.- Test everything for consistency.
Please help me improve the task list/order.
-Sam.

Hi Sam
You are planning a lot of activity on a production environment. Two coments:
1. When you have your dev SLD configured as you require, with CIM, PI Content & Business Systems updated etc .. perhaps you should do a level of regression testing on this environment before embarking on changing the Live env.
2. Once you are happy that the Dev env It is possible to Export and import manually the SLD content from the Dev to the Live to align both systems, refer to link:
https://websmp104.sap-ag.de/~sapidb/011000358700000315022005E.PDF
This is available through the Export Link in Administration page.
Thanks
Damien

Crystal Develop and Production Guidelines in SQL Server (Crystal 9)

We have development and production environment on 2 different machines.
It seems when we want to migrate a report from dev to prod, we need to reset the DSN or connection and crystal then forces us to respecify all field names in a report.
It is a big pain. Has anyone seen this?
Thanks
Ken Light
DBA Interior Health

Hi
When you migrate a report, you need to
- Create a DSN with same name on Production as it was on Test.
- Do a set Datasource location.
- Verify database to ensure all changes made in the database are reflected.
- If you have used stored procedure with parameter in your report, then uncheck the above option and then refresh the report.
Hope this helps!!!
Regards
Sourashree

Dev and QA pointing to Same SLD

Hi Experts,
We have a situation in our landscape. Just want to confirm if it is right approach or will be face any issue in future.
We have DEV and QA box of SAP PI in our landscape. (PRD under construction)
1. The DEV PI and QA PI are installed on the box and has the same host name. xyz.com
2. When I execute the SXMB_IFR from Dev PI it opens the url like xyz.com:54000.
3. And when I click the SLD it opens the SLD with url xyz.com:54000/sld
4. When I check the SMICM-->Services the http port is 54000
Until this it is fine, but when I login to QA box the following happens
0. QA PI is the system copy of DEV PI.
1. When I execute the SXMB_IFR from QA PI it opens the url like xyz.com:55500 (as already mentioned xyz.com host is same for dev and qa)
2. When I click the SLD link of QA box it opens the url xyz.com:54000
-- it opens the SLD of Dev PI were as i guess it should open the url like xyz.com:55500
3. http port for QA PI is 55500
I can see something is wrong here, but not able to understand it.
Though we are not facing any immediate issue what problems can we expect in future.
How can we rectify it?? any sap note available for this??
Thanks
Prasanna

Hi Prasanna,
As Rajasekhar said ,
Your Project has one SLD for Dev/QA and mounted that SLD in Dev and one SLD for Production, some of the client prefer this approach.
Even in our project also we followed the same.
Since SLD is mounted in Dev ;when you clicked on the SLD it opens the SLD with url xyz.com:54000/sld.
There is nothing wrong in it .
Cheers
Regards
Venkat

Best methodology to have dev and production?

Similar Messages

Maybe you are looking for