cancel
Showing results for 
Search instead for 
Did you mean: 

Extraction data 0CLM_INVOICE works partially, then discontinuous.

Former Member
0 Kudos

Dear BI Experts/Gurus,

I have searched endlessly on the forum and found lots of helpful information, however I can't seem to find a solution to the problem I am experiencing:

I am having trouble extracting the data (delta initialization) from the R/3 System to BW for the FSCM 0CLM_INVOICE datasource. Dependent on the company code selected for extraction, the data is extracted either without any problems or with errors. Half of the company codes I have tested can be loaded without any problems. I am loading data from the Q-system, which is a copy of the productive system (few months old); a system test/connection check has been done and does not display any errors. All other datasources/extraction jobs work, only this datasource (OCLM_INVOICE) is causing problems.

Here a short summary of the steps:

1) Running infopackage (limited to a small company code with records <1000) u2013 Data extraction u2018freezesu2019 after loading 810 records) - This problem occurs for half of the company codes with records ranging from <1000 to >1,000,000

2) RSA3 u2013 Extractor checker in ECC6 shows total records are 892.

3) SM50/SM66 (ECC6.0) (Global process overview) u2013 System still reading data (slowly but surely)

4) Process Monitor (BW) u2013 Yellow triangle u2013 not yet finished (After several hours, the last 82 records are not extracted)

5) The next day u2013 Process overdue error

Here is a more detailed overview of actions taken and error messages:

The problem that occurs is that the data is only loaded partially, for example 5xx,xxx records and then the job just gets stuck on processing (yellow triangle u2013 not yet finished).

When I check the process sm50 / sm66 it is still loading data, but if I let the job continue it will automatically stop after sometime due to u2013 process overdue error.

In ST22 (runtime error analysis) the error msg was u201CNo more storage space available for extending an internal tableu201C.

I tried to fix the problem through decreasing the load size in the infopackage from 20,000 to 10,000. This however still does not solve the problem, and instead displays the following message u2013 u201Cselected number does not agree with transferred numberu201D (Yellow triangle u2013 not yet finished).

The process monitor shows that data records retrieved are 810/0 but in the ECC (RS3 u2013 extraction checker shows no errors) the same datasource displays a total of 892 records. It seems as if the system cannot load the last 82 recordsu2026

I have additionally tried to (re)activate the datasource in the ECC through RSDS_DATASOURCE_ACTIVATE_ALL and (re)activate the transfer structure through RS_TRANSTRU_ACTIVATE_ALL. Iu2019ve replicated the metadata in BW for this datasource and reactivated it u2013 I did not get a single error message doing this.

I do get an error message when trying to compare the active datasource version with the content version = u201CVersion comparison with content version not possible (source system XXXXX)u201D --- Message no. RSDS169. I cannot find any information regarding this error message, so Iu2019m not sure if this has anything to do with the problem.

Does anyone know what the problem could be? I am having a hard time finding a solution!

Any help is much appreciated

Thanks in advance!

Hugo

Edited by: Hugo00 on Sep 13, 2011 10:50 AM

Accepted Solutions (0)

Answers (1)

Answers (1)

Former Member
0 Kudos

No ideas? I searched the forum, and tested everything I could find that seemed relevant, hence the detailed problem description...If anything is not clear, and you think you can help, please post any questions.

Any clues, hints, tips would help.

Thanks,

Hugo

Former Member
0 Kudos

Hi Hugo,

After the Job fails in SM37, what log you are getting.

can you share your job log?

while load is running, check you have any TRFC stuck in SM58, if yes execute them manually.

Regards,

Venkatesh

Former Member
0 Kudos

Hi Venkatesh,

thanks for your reply, here is the log. I don't think that there are any TRFCs stuck, I am not getting any data in SM58.


Time     Message text
16:04:21 Job started
16:04:21 Step 001 started (program SBIE0001, variant &0000000086499, user ID ALE_WTS)
16:04:21 Asynchronous transmission of info IDoc 2 in task 0001 (0 parallel tasks)
16:04:21 DATASOURCE = 0CLM_INVOICE
16:04:21 *************************************************************************
16:04:21 *          Current Values for Selected Profile Parameters               *
16:04:21 *************************************************************************
16:04:21 * abap/heap_area_nondia......... 6000000000                              *
16:04:21 * abap/heap_area_total.......... 8000000000                              *
16:04:21 * abap/heaplimit................ 20000000                                *
16:04:21 * zcsa/installed_languages...... DE                                      *
16:04:21 * zcsa/system_language.......... E                                       *
16:04:21 * ztta/max_memreq_MB............ 2047                                    *
16:04:21 * ztta/roll_area................ 6500000                                 *
16:04:21 * ztta/roll_extension........... 4000000000                              *
16:04:21 *************************************************************************
16:04:32 Call customer enhancement BW_BTE_CALL_BW204010_E (BTE) with 812 records
16:04:32 Result of customer enhancement: 812 records
16:04:32 Call customer enhancement EXIT_SAPLRSAP_001 (CMOD) with 812 records
16:04:32 Result of customer enhancement: 812 records
16:04:32 PSA=1 USING SMQS SCHEDULER / IF [tRFC=ON] STARTING qRFC ELSE STARTING SAPI
16:04:32 Asynchronous send of data package 1 in task 0002 (1 parallel tasks)
16:04:32 IDOC: Info IDoc 2, IDoc No. 14994880, Duration 00:00:00
16:04:32 IDoc: Start = 07.09.2011 16:04:21, End = 07.09.2011 16:04:21
16:04:32 tRFC: Data Package = 1, TID = 0A00140F67704E6779F0019C, Duration = 00:00:00, ARFCSTATE = RECORDED
16:04:32 tRFC: Start = 07.09.2011 16:04:32, End = 07.09.2011 16:04:32
16:04:32 Asynchronous transmission of info IDoc 3 in task 0003 (0 parallel tasks)
----------------------2 days later it automatically canceled the job---------------------
01:17:05 Internal session terminated with a runtime error (refer to ST22)
01:17:05 Job cancelled

ST22 - runtime error belonging to the above job log.

 
Runtime Errors TSV_TNEW_PAGE_ALLOC_FAILED

 Short text
     No more storage space available for extending an internal table.

 What happened?
     You attempted to extend an internal table, but the required space was
     not available.

I don't know why there is a memory error, there weren't that many (global) processes running, if i'm not mistaken 3-5 including my own according to (SM66), the storage shouldn't have been an issue considering the few amount of records that weren't loaded (<100)...or am I missing something?

Thanks,

Hugo

Former Member
0 Kudos

Hi Hugo,

Try to reduce the data packet size to 500 and run your load, if it fails again ask your basis person to increase the PSA size.

Regards,

Venkatesh

Former Member
0 Kudos

Hey Venkatesh,

I tried it and it seems to take the extraction past the previous 812 records, currently to a bit over 2000, which should not be possible, since according to the extractor checker rsa3 there are only 898 records for that particular company code.

I am letting it run, to see what error message I receive. I checked the trfc which are green so there doesn't seem to be a problem on the ECC side.

Thanks for the input so far.

Best regards,

Hugo

Former Member
0 Kudos

Hi Hugo,

If you are getting 892 records in RSA3, then how your IP loaded 200 records till now.

can you please recheck once

1)type of load in IP(full or delta)

2)are you extracting the data in RSA3 with the same selections

check the company code which you gave in selections

Regards,

Venkatesh

Former Member
0 Kudos

Hey Venkatesh,

Thanks for all your input!

I'm running an IP with delta initialization and the exact same company code as selected in RSA3 (update mode C - init delta - 100 records per call / 10 calls) records loaded 892.

The IP has loaded up to 2661records in the meantime, but I think it will not load anymore and eventually cancel.

I'm not sure why I am getting different amount of records.. (I did change package size to 500 for delta init, since then it loaded more than 812 but also more than the 898 )

I've tried:

- replicated the metadata in BW

- reactivated the datasource

- created new infopackage

all worked properly w/no errors.

I also tried RSDS_DATASOURCE_ACTIVATE_ALL in SE38 -> this is for 7.0 datasources right? Because i'm getting a return message with "There is no DataSource with these attributes" in the ECC system. I've double checked the source system that I entered it seems to be correct (example R3QAND100 = system nameANDclient#). So i'm not sure why I am getting this return message.

When using RS_TRANSTRU_ACTIVATE_ALL in SE30 - I thought this is only for 3.x, but tried it anyway. I get the result that no transfer structure selected and no errors or warnings. This is probably since the datasource is 7.0.

Thanks,

Hugo

Edited by: Hugo00 on Sep 14, 2011 7:14 PM

Former Member
0 Kudos

Hi Hugo,

while checking your extractor in RSA3 do not use update mode other than F, because some times it will update the delta time stamp in table ROOSPRMSC.

In RSA3 give update mode as F and in selections give your company code then execute and check how many records you will get.(Increase the packet size from 100 to 10000)

RSDS_DATASOURCE_ACTIVATE_ALL -- this is for BI 7.0 and you need to execute this in BI system not in ECC system.

ask your basis person to increase the PSA table size of your data source.

check below thread

/thread/1848389 [original link is broken]

Regards,

Venkatesh

Edited by: Venky1903 on Sep 14, 2011 11:46 PM

Former Member
0 Kudos

Hi Venkatesh,

Thanks for the quick reply! 

I will check it out tomorrow and post back with the outcome.

Thanks again!

Best regards,

Hugo

Former Member
0 Kudos

Hey Venkatesh,

I've ran rsa3 with various data package sizes, and received a different amount of records each time. I also checked and saw that there is indeed a timestamp for the detla initializations I ran w/rsa3. Furthermor I did some more research online and came across a SAP note (note 1448185) that seems very helpful, i think you were right with the memory allocation.

I am having it implemented, and I will post back with the results.

Thanks for all your help so far.

Best regards,

Hugo