Issue: attributes change run triggers the rebuilt of 0IC_C03 aggregates

Hello Guru,
I am facing to a problem that I don't understand.
Here is below the detail of the messages of a '3-infoobjects attributes change run' step of a Master Data process chain.
22:06:14     Job started
22:06:14     Step 001 started (program RSPROCESS, variant &0000000231235, user ID RFC_BW)
22:06:16     Starting process type ATTRIBCHAN with 3 objects...
22:06:16     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BICHNG_D8FC4Q04UD7NGLDMZ0QUBLX4K with ID 22061600
22:06:16     Editing of aggregate 100139; background job BICHNG_D8FC4Q04UD7NGLDMZ0QUBLX4K; start time 22:06:16/07.07.2010
22:06:18     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BICHNG_D8FC4Q04UD7NGLDMZ0QUBLX4K with ID 22061601
22:06:18     Editing of aggregate 100143; background job BICHNG_D8FC4Q04UD7NGLDMZ0QUBLX4K; start time 22:06:18/07.07.2010
23:02:29     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BICHNG_D8FC4Q04UD7NGLDMZ0QUBLX4K with ID 23022900
23:02:29     Editing of aggregate 100140; background job BICHNG_D8FC4Q04UD7NGLDMZ0QUBLX4K; start time 23:02:29/07.07.2010
23:52:38     Starting process type MDHIERACT with 2 objects...
23:52:39     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BIACTI_D8FCR9Y1NWOP2T4F76FYE0HDG with ID 23523900
23:52:39     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BIACTI_D8FCR9Y1NWOP2T4F76FYE0HDG with ID 23523901
23:53:10     Aggregate 100139: Repair indexes
23:57:14     Aggregate 100140: Repair indexes
00:17:23     Aggregate 100143: Repair indexes
00:50:09     3 aggregates are condensed
00:50:09     Starting process type CONDAGGR with 3 objects...
00:50:10     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BICOND_D8FD3GTNF0KF92R34346SW6PG with ID 00501000
00:50:10     Editing of aggregate 100143; background job BICOND_D8FD3GTNF0KF92R34346SW6PG; start time 00:50:10/08.07.2010
00:50:10     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BICOND_D8FD3GTNF0KF92R34346SW6PG with ID 00501001
00:50:10     Editing of aggregate 100140; background job BICOND_D8FD3GTNF0KF92R34346SW6PG; start time 00:50:10/08.07.2010
00:50:16     Program RSBATCH_EXECUTE_PROZESS successfully scheduled as job BICOND_D8FD3GTNF0KF92R34346SW6PG with ID 00501600
00:50:16     Editing of aggregate 100139; background job BICOND_D8FD3GTNF0KF92R34346SW6PG; start time 00:50:16/08.07.2010
00:50:22     Hierarchy/attribute change successfully carried out
00:50:23     Event RSPROCESS with parameter D59BE8XWZMHY0IFP2VLEFP671 successfully triggered
00:50:23     Job finished
As you can observe, the step takes almost 3 hours !!!!!!! After anlyses BICHNG_* concerns
22:06:16 Job started
22:06:16 Step 001 started (program RSBATCH_EXECUTE_PROZESS, variant &0000000891873, user ID RFC_BW)
22:06:18 Aggregate 100139( 0IC_C03) is being adapted / rebuilt
22:06:18 Statistic UID of the aggregate process: 'D8FC4Q9KSAMRDWQ1RDPPKMF3O' (RSDDSTATAGGR)
22:06:27 Aggregate 100139: Read (1) from table /BI0/E0ICC03_ (max 500000 records)22:06:27 SQL: 07.07.2010 22:06:27 RFC_BW
I don't understand what is the link between the '3-infoobjects attributes change run' (BBATCH, 0MAT_SALES, 0MAT_PLANT) and the rebuilt of 3 0IC_C03 (Material Stocks/Movements) aggregates (why does the 'attribute chan run' trigger the rebuit of aggregates???).
The rebuilt of the 3 aggregates takes 1 416, 3 366, 3 009 secondes...
Does anyone understand my problematic? If not I can give you more details.
Thanks in advance for your help.
Samuel

1/ Exactly Fred
Here is the definitions of the aggregate:
100139:
0PLANT
0STOR_LOC
0MAT_PLANT__0VAL_CLASS
0STOCKCAT
0STOCKTYPE
BMODIFIC
BMVTSTOCK
0MATL_TYPE
0CALxxx
2/ For my information in the aggregate 100140 , does the presence of BBATCH (one of the 3 IO) trigger the rebuilt or is it only 0MAT_PLANT__0VAL_CLASS ?
100140:
0PLANT
0STOR_LOC
0MATERIAL
0MAT_PLANT__0VAL_CLASS
BBATCH
0STOCKCAT
0STOCKTYPE
BMODIFIC
BMVTSTOCK
0MATL_TYPE
0CALxxx
3/ In the same process chain, I have another Attributes Change Run: 0MATERIAL. Several active aggregates using 0MATERIAL__0MATL_TYPE are built on a CO-PA cube.
Why did the '0MATERIAL Attributes Change Run' rebuild no aggregate? (see below)
22:06:04     Job started
22:06:04     Step 001 started (program RSPROCESS, variant &0000000231238, user ID RFC_BW)
22:06:06     Activation of master data for 0MATERIAL
22:06:07     154 Data records deleted
22:06:11     154 Data records deleted
22:06:12     154 Data records updated
22:06:13     154 Data records updated
22:06:13     0 aggregates are condensed
22:06:14     Hierarchy/attribute change successfully carried out
22:06:14     Event RSPROCESS with parameter D59BEEE069HKRCRU0ZPYFUPB1 successfully triggered
22:06:14     Job finished
Samuel

Similar Messages

Attribute change run process doesn't run in the chain

Hello everyone,
I have implemented a process chain with multiple parallel processes (12, like the batch processes in the application server) and I have put for each stream a loading of master data followed by the attribute change run process. When I run my chain then I find that one o two Attribute change run process doesn't run (the InfoObject chosen seem to be random!). The chain remain in yellow color. Can you help me anyone about this question?
Regards,
Francesco

Couple of thing you should keep in mind while loading master data.
if two loads are targeted to the same infoobject they will collide and as a result the later load will fail.
it is a good practice to use only one Attribute change run in the end before you load master data. but if there are dependecies you can use several in series
Also when you set up Attribute change run make sure that Attribute change run will pickup only the particualr infoobject not all the objects in the Attribute change run list.

Attribute change Run Failed/Scheduled

Hello All,
For a process Chain Attribute Change Run has failed in BW 3.5,and I Repeated the Change Run in Chain only
where Repeat option is ,now the problem is in RSA1 it is now that change run is scheduled
and sm37 it is showing that Canceled.
Now there is Other Process Chain in BI7.0 which picks data from 3.5
Should I repeat the Master data Chain in BW 3.5 followed by in BI 7.0
Or any other way is possible as in BI7 the chain has run successfully..
Thanks,
Kiran

Hi Kiran.......
BI 7.0 and BW 3.5............totally two seperate systems..............here BW 3.5 is just act as a source system of BI 7.0.............if change run fails in BW3.5...............it is not going to effect BI 7.0.............
As ajeet already said............it will effect the SID s in 3.5.............because attribute change run activates the master data...........and SID get generated during the activation of master data.......
Also it will effect the aggregates............because aggregates also get adjusted during change run.......if any of the aggregates contains the master data.........which are getting activated in the change run....
Now u r saying........that in SM37 it is showing that the job got cancelled............So it will not locked the master data..............terminated change runs generally lock the master data............then we hav to repeat the chain run to release the lock.........and u hav already repeated the change run............so no issue........
Hope this helps..........
Regards,
Debjani........

Attribute change run problem

Hello Gurus,
I have deleted all the master data tables(P,S,X,T). Deleted the PSA table entries too. I am loading data frm R/3.
No data exists in 0VENDOR ,when i click manage master data.
When the moment I load data in 0VENDOR_ATTR, without doing Attribute change run(ACR), the data is getting displayed in Maintain master data.
Please let me know why this happens.
Thanks,
Sonu

Hi
According to my understanding of your issue, there is no master data in 0VENDOR, i.e., all the Dimension tables along with SID's are deleted.
But SIDs of the 0VENDOR_ATTR(whcih is the attr of 0VENDOR and also acts as characteristic) exists.
Sow hen you load data the master data is written into 0VENDOR and hence teh SID's of the attr are not deleted do teh master data will exists for 0VENDOR_ATTR.
Please let me know incase i got your question wrong.
Regards, Hyma

Attribute Change Run and Editing BI data

Hello Experts,
I have two questions.(They are not related to each other).
Can BI data(Infocube data) be edited?
For eg GL account in R/3 is changed from 100 to 2100. I want to know how can this be handled in BI.
Attribute Change Run in the process chain ends with error.
"Master data characteristic ZACCTNUM activated in 6 seconds. Error while activating characteristic   ZACCTNUM ". There are no more information.
   The BWREMOTE user has the authorisation to Execute Change Run.
   What could be the possible reasons for this failure.
Thanks in advance.
Regards
Meenakshi

Hi Meenakshi,
Regarding your questons:
1. You can't edit the data directly in the cube. If you have incorrect data in the cube and the data is still in the PSA you can edit the incorrect data in the PSA and you can then delete the incorrect requests from the cube. You can then reload the correct data from the PSA. If its not an option to delete individual requests you could delete the incorrect data using the selective deletion option (if you can select on the incorrect data in the cube). You could then reload the correct data from PSA or source system.
2. Please check the infoobject ZACCTNUM is active in RSD1. If it is there may be an inconsistency with the data in the infoobject ZACCTNUM, please check if the solution part of the SAP note 1033721 can help you in this case.
Best Regards,
Des.

Attribute Change Run in Process Chain

I'm making a process chain purely for attributes. Every time I add an infopackage for attributes I get prompted to make an attribute change run. Do I need an attribute change after every attribute infopackage or can I just have one at the end?

Hi,
You can have one attribute change run at the end of the process chain for all the attribute loads.
Cheers,
Malli.

Attribute Change Run- Serial or Parallel

If we are building a chain for Attribute Change Run for all Master data objects, should we:
1. load Masterdata A, followed by attribute change run, load masterdata B followed by attribute change run, load masterdata C followed by attribute change run....
2. Load Masterdata in Parallel followed by AND with ONE attribute change run.
The first one seems grossly redundant but then if we have only ONE attribute change run, wouldnt that lock up all the objects for a LONG time?
Also, does an attribute change run by default include ALL objects which have been loaded anew and for which the change run has not run???

Thanks!
So, having just one attribute change run for all these objects is a better option?
Wouldn't it lock up the objects for a long time?
Also, say I have loaded object A , B and C. And in my process chain I have Attribute change run only for Object C.
Will the Attribute Change Run include newly loaded A and B by default?
I ask this because we saw something like this happening
Edited by: CC on Apr 9, 2010 8:33 PM

Attribute Change Run in Process Chain Texts?

Dear Colleagues!
A simlple question: Do you have an Attributes change run in the end of a Process Chain only containing text? (Master Data text) Or i s it sufficient to have it after attributes?
Regards
Silje

Hi,
Attribute chage run is meant only for Attr/Hierarchies.
No need to include it for texts.
For the texts changes were automatically reflected with out change run.
Regards
Jonn

Attribute Change Run terminating - repeated several times - not locked

I had a failure in my master data chain tonight. The attribute change run that ran yesterday ran for 11k seconds and was cancelled (it says manually). When the chain ran tonight, the process failed again. I checked the log and it says "UNCAUGHT_EXCEPTION" and it created a short dump. I tried repeating to see if it did anything and it gave me the same message.
I went into the change run monitor and reset the status and manually tried to run the change run again. Same result "UNCAUGHT_EXCEPTION".
I then checked table RSDDAGGRMODSTATE and there was no blank CNSID entries, but there was a blank "completed flag" entry. I don't have change access to the table in prod so I cannot manually change it to X (meaning closed). I reset the status again in the chagne run monitor and the record with the blank "completed flag" disappeared. I repeated the change run and it failed again and a new record with the "completed flag" being blank appeared in the RSDDAGGRMODSTATE table.
I checked SM12 and there are no locks there. I'm on 7.0 so I can't manually right click on 0CUSTOMER and activate the master data either.
As a last resort I did run the function module RSDDS_AGGR_MOD_CLOSE but that short dumped too.
Any ideas? Please? This will fail my other chains that have 0CUSTOMER in the aggregates. Thanks.

HI,
Firstly try this FM RSDDS_AGGREGATES_MAINTAIN
if not possible
Try doing it manually
Uncaught exception may come due to many reasons
IN RSA1 go to tools and select attribute change run
Check the blog it shows you how to manually do attribute change run step by Step (for BI7.0 also)
/people/debjani.das/blog/2009/03/30/attribute-change-run-acr
Regards,
MADhu.
Edited by: MadhuMarupudi on Aug 21, 2009 5:49 AM

Attribute change run problem - infoobject is not listed in the list

Hi folks,
When I load master data for 0customer, the infoobject 0customer doesnt come in the infoobject list for attribute change run. Also, when I try to activate master data manually it says 'Master Data already active'. and the reality is M version of Data is not active.
Can you guide me.

Try running the program RSDMD_CHECKPRG_ALL for this infoobject. You can select all the options when running this program.
Try running this program multiple times, coz my experience showed that initial run may not always fix this issue.
Good luck!

What is the back ground job when we run an Attribute Change run ?

Dear One´s,
What is the back ground job when we run an Attribute Change run ?
Thanks in advance,
Raj

Hi Raj,
1. If the attribute change run is triggered through process chain the job name is 'BI_PROCESS_ATTRIBCHAN'
2. If the change run is triggered from RSA1> tools?Appl Heirarchy/Attribute changerun then the job name starts with BI_STRU*
3. If you trigger the program RSDDS_AGGREGATES_MAINTAIN from SE38 then the job name will be RSDDS_AGGREGATES_MAINTAIN
Hope this helps...

Attribute Change run issue

Hi All,
I have a Info Object 0MEASERE, we have maintained master data for that and ran attribute change run... but unfortunately it was not activated, so we ran change run again now we get the message saying that it is locked in the terminated job.
But when i check st12 there are no locks against the user.
Can someone please tell me what should i do to activate this?.
Regards
BN

Hi,
Chk this thread. It should help ur issue.
Hierarchy is locked by a terminated change run
Hope it helps. Assign points if useful.
Cheers.

Attribute change run fails - "Error when deleting the data record /B05/X.."

Hello,
I get the error message "Error when deleting the data record /B05/X.." on doing the change run.
I have already done RSRV on this objects and found no errors, but still the activation fails.
please give your suggestions ASAP to fix the issue as it is on production system.
Thanks.

Attribute change run could fails due to 2 reasons-
(1) attribute xchange run is already runnnig for any toher object(monitor lock) (check by thi prog-RSDDS_CHANGERUN_MONITOR)
(2) data is loading deleting from the object on which attribut changerun is running.(go in a manage tab and check)
check this two conditions and rerun the attribute changrun again from program - RSDDS_AGGREGATES_MAINTAIN and give the master data name in the selection.
Hope it helps you.

Hi i have 50 infoobjects as part of my aggregates and in that 10 infoobjects have received changes in masterdata.so in my process chain the Attribute change run in running for a long time.can i kill the job and repeat the same.

Hi i have 50 infoobjects as part of my aggregates and in that 10 infoobjects have received changes in masterdata.so in my process chain the Attribute change run in running for a long time.can i kill the job and repeat the same.

Hi,
I believe this would be your Prod system, so don't just cancel it but look at the job log. If it is still processing then don't kill it and wait for the change run to complete but if you can see that nothing is happening and it is stuck for a long time then you can go ahead and cancel it.
But please be sure, as these kind of jobs can create problems if you cancel them in the middle of a job.
Regards,
Arminder Singh

ABAP program to add in the process chain before the attribute change run

Hi
I need a ABAP program to add in the process chain before the attribute change run(process) to avoid the locks, can any one send me the program
Regards,
Satish

I am not sure of any standard program .... but you can write your own... should be a small one..
you can check if the table RSDMDENQ has entry for your InfoObject... if yes, then its locked.. if not.. its not locked

Issue: attributes change run triggers the rebuilt of 0IC_C03 aggregates

Similar Messages

Maybe you are looking for