Master Data Duplicate Error

[Former Member]

Hi BW Gurus,

Would someone help me understand more about this error, "Duplicate Record Found". We get few lines of these. We noticed that in the past few weeks in our prod environment we get this error pretty often. Usually, we start the update immediately from PSA, then do RSRV and wait for the RSRV to do it's thing. But meanwhile, we have to stop some loads so that it will not interfere with the rebuilding of the aggregates and ultimately it delays all the users.

I was wondering if anyone has encountered and resolved this. Again, I'm not clear on how RSRV washes out the internal issues that we cant. Any help will be well rewarded.

Thank you,

Tia

Message was edited by: Tia Ismail

[Former Member · Accepted solution]

Hi Tia,

In your InfoPackage try checking the option for Ignore Double Data Records (Processing tab).

Hope this helps...

[Former Member · Accepted solution]

Hi Bhanu,

We thought of that, but currently we have

"PSA and InfoObject in Parallel (Package by Package)" option selected. Seems like Ignore Duplicate records is possible when "Only PSA" option is selected. Could I be wrong ?

Thank you,

Tia

[Former Member · Accepted solution]

yes, you need to choose the third option, i.e. Only PSA and subsequently it will update to the target.

[Former Member · Accepted solution]

What's the difference, between these two options in terms of data load performance.

I thought that "Psa and InfoObject in Parallel" is good for data loads. Please advise if not.

What kind of effects would we expect, if we select "Only PSA" ? I really appreciate your responses.

Thank you,

Tia

[Former Member · Accepted solution]

Hi Tia,

Simple speaking, PSA and InfoObject in Parallel is fast and also needs more resources.

The third option: PSA and Update Subsequently into Data Targets takes a bit more time, but consumes less system resources. Also with this you can take advantage of other fetures like error handling, checking for referntial integrity and ignoring double records.

On the processing tab of the InfoPackage, click on the ? icon (it is towards the right). This will help you understand how the data moves for each of these options.

Take a look at this link for complete details:

http://help.sap.com/saphelp_nw04/helpdata/en/b8/5b723b9b4f8f7de10000000a114084/content.htm

Hope this helps...

[Former Member · Accepted solution]

So, would this option probabaly result in not having to go to RSRV ? Because, at this point, since we resorted to RSRV, droping and rebuilding the aggregates is causing a lot of delays. Could you tell me more about RSRV, please ?How does it work, internally ?

Thank you,

Tia

[Former Member · Accepted solution]

Hi Tia,

You are right...if the load goes well then you will not have to run the RSRV test. To learn more about RSRV you can try this: In RSRV on most of the tests you can right click and choose Description. This will give you a fair amount of information about the purpose and functioning of the test.

Hope this helps...

[Former Member · Accepted solution]

Hi Bhanu,

Does this option -- "PSA and Subsequently update data targets" with ignore duplicate records, -- work for the time dependent navigational attributes ? From the link you had send i found a good resource but it talks about non-time dependent records. Please let me know.

Thank you,

Tia