cancel
Showing results for 
Search instead for 
Did you mean: 

What is your Best Practice for Refreshing Data in Repositories?

Former Member
0 Kudos

For those of you who have IT personnel asking for a refresh of production data from one repository to a non-production repository:

What have you found as the best practice for performing these data refreshes? Is it possible?

Thank you,

Jessica

Edited by: Jessica Salinas on Jul 9, 2008 10:16 PM

Edited by: Jessica Salinas on Jul 9, 2008 10:17 PM

Accepted Solutions (1)

Accepted Solutions (1)

Former Member
0 Kudos

Hi Jessica,

The best way is you can take archive copy of the repository and you need to unarchive the archive copy of repository in to non-production environment.

you can get all the data from production to non production environment.

Hope it helps you.

Regards,

RDNPrasad.

Answers (3)

Answers (3)

Former Member
0 Kudos

Hi Jessica,

Using the Archive and Unarchive method:

While archiving your production rep and unarchiving it as a new non production rep ,you have an option to do this With Data and Without Data,if you say with data the along with your rep structure while archiving ,the data in the rep will also be archived,if you say without data then jsut the rep structure will get archived and you can then populate the new rep with new refreshed data differnt than the production sytem data.

You can also use the XML schema Export /Import function if you just want to copy the rep design and not the rep data.

Master /Slave rep Method:

This method works fine when you want two similar Repositories to be used with one acting as a master and teh other the slave.The two rep are such that,If you wnt to make certain changes in the master rep which needs to be reflected later on in the slave rep after confirmation then Master slave concept can be used.

Or when you want a back up such that what ever is updated in the master rep the slave should follow on command.

Hope it helps

Kindly reward points if found useful

Thanks

Simona

Former Member
0 Kudos

Hi Jessica,

According to me, the best approach would be to use the Master and the Slave concept.

This is what SAP has to say about master and the Slave repository concept.

MDMu2019s master/slave feature enables u201Cpull-basedu201D replication of MDM repositories. Through this feature, you can create one or more read-only copies of an existing MDM repository. Each copy can be on the same MDM Server as the original or on a completely different platform in a completely different location.

The original, or master repository, keeps track of all of its subsequent data and schema changes. Each copy, or slave repository, can update itself with these changes by sending a synchronization request to the master, which replies with the updates.

Hence, for your case, you can keep the Production Repository as the master Repository.

And the Non- Production reposiotry as the Slave repository. Then when ever you want to update the non-production repository as per the produciton rep, you can simply synchonize the slave. Automatically, the non-produciton rep will be updated as per the Produciton rep.

For more information, please refer to the link below:

[http://help.sap.com/saphelp_mdm550/helpdata/en/8e/9f9c427055c66ae10000000a155106/frameset.htm]

Kindly go to-> Repository Administartion->maintaining Master and Slave repositories

You will find ample of information

Hope it helps.

*Kindly reward points if helpful

Thanks and Regards

Nitin Jain

Former Member
0 Kudos

Hi Nitin,

If roles differ in master and slave repositories, then how is that handled? I do not wish to overwrite roles which differ between production and non-production repositories.

Thank you,

Jessica

Edited by: Jessica Salinas on Jul 10, 2008 4:22 PM

Former Member
0 Kudos

Hi Jesica,

If you want to enhance the data in one or other way from production then, You can take all the data into excel sheet (Data manager -> Record Mode->file->Export to excel->Select all fields and qualifiers). You can import this data manually to the other repo.

If you want exact data and schema of production as your non-production repo, you can archive the production repo and unarchive it on the other repo (To replace the existing or to create a new one).

Archive file will get stored in C/program files/SAP MDM 5.5/Server/Archives folder of your server machine. If you are using the same server for the second repo then its fine otherwise you have to move this archive (.a2a file) from this location to the same location for other server. Right click on the server name ->Unarchive repo. Provide DB user id/Pwd. Give the repo name (You want to create) and choose the archive name from drop down and say ok.

After some time it will show you a success message.

Hope this is helpful.

Regards,

Dheeraj.

Former Member
0 Kudos

Hi Dheeraj,

If we do an archive/unarchive option, is there a way to retrieve roles and users automatically from the previous repository to the newly unarchived one? Is it possible to write scripts for MDM?

Basically we found that using this archive/unarchive option will bring over production users, production roles, and those role to user assignments. However, we do not wish to bring users and roles over from production.

Please share your thoughts.

Thank you,

Jessica

Edited by: Jessica Salinas on Jul 10, 2008 4:28 PM