Phase Preparation of Tables for Data Transfer Aborted

[Former Member]

We are trying to perform the solution "Full Transfer of Client-Specific Data for SAP ERP" using the following systems:

Control/Central System: NetWeaver 7.3 ABAP instance

Sender System: ECC 6.0 EhP5 IDES

Receiver System: Shell System created from Sender System

O/S: Windows 2008 and MSSQL 2008 on all three systems

Kernel: 720 patch 401 on all three systems

The activity "Preparation of Tables for Data Transfer" is aborted after a runtime of almost 5 hours. On the Central System, we looked at SM37 and found that the job CNV_MBTCNV_MBT_PEM_START had failed with only "Activity Preparation of Tables for Data Transfer execution 0002 in package 90002, phase TDBAS_PHASE_PACKAGE_SETTINGS has been aborted" as the error message.

Does anyone know how else I can troubleshoot this issue?

[Former Member · Accepted solution]

Hi ,

It seems one or more subactivities under phase 'package settings' has failed.

Please check the status of all subactivites under this phase.You can do this by going to 'extended process monitor' and selecting the aborted phase.You can find the logs and error message against the aborted sub activity.Also,check ST22 for any dumps.

hope this helps.

[former_member65049 · Accepted solution]

Hi,

You can try checking the job log in SM37 in sender and central system.

Also, check the extended process monitor and see what the activity log says about the aborted activity.

Regards,
Rupam

[Former Member · Accepted solution]

Hi Kartik,

You are correct. As I explained, the subactivity "Preparation of Tables for Data Transfer" has failed.

In the logs for this subactivity, the error message was "Activity Preparation of Tables for Data Transfer execution 0003 in package 90002, phase 'Package Settings' has been aborted."

No short dumps were found on the Sender System sadly.

[Former Member · Accepted solution]

Hi Rupam,

On the Central System, the job CNV_MBTCNV_MBT_PEM_START failed with the error message  "Activity Preparation of Tables for Data Transfer execution 0002 in package 90002, phase TDBAS_PHASE_PACKAGE_SETTINGS has been aborted".

On the Sender System, the job CNV_MBT_CNV_TDMS_06_GET_USER_TAB ran 5 times and was cancelled each time, running for 2562, 2760, 5007, 4161 and 4062 seconds each. No short dumps were found on the system during that time.

Here is the log of the job log of the first job:

Date       Time     Message text                                                                                  Message class Message no. Message type

15.02.2013 17:29:59 Job started                                                                                        00           516          S
15.02.2013 17:30:04 Step 001 started (program CNV_TDMS_06_GET_USER_TAB_X, variant &0000000000010, user ID DHEUNG)      00           550          S
15.02.2013 18:12:41 Job also cancelled due to process termination. See the system log                                  00           554          A

In the system log of the Sender System between 17:29:59 and 18:12:42, we see the following:

Time     Type Nr  Clt User   TCode Priority Grp N Text

18:12:16 BTC  022 800 DHEUNG                BZ  Y Unexpected return value 1 when calling up DbSlR
18:12:17 BTC  022 800 DHEUNG                BY  J Function ROLLBACK on connection R/3 failed
18:12:18 BTC  022 800 DHEUNG                R3  9 Error in DB rollback/SyFlush, return code 016384
18:12:18 BTC  022 800 DHEUNG                BY  4 Database error 0 at UPD access to table TST01
18:12:18 BTC  022 800 DHEUNG                BY  0 >
18:12:18 BTC  022 800 DHEUNG                F6  H Database error: TemSe->XRTAB(3)->1 for table TST01 key
18:12:19 DIA  011 800 DHEUNG                EI  Q BP_JOB_CHECKSTAT: Failed to correct status of job. Reason:
18:12:19 DIA  011 800 DHEUNG                EF  H > Function module was called with an invalid dialog mode ( )
18:12:21 BTC  022 800 DHEUNG                Q0  2 Stop Workp. 22, PID 3332
18:12:25 RD                                 Q0  I Operating system call recv failed (error no. 10054)
18:12:32 DIA  011 800 DHEUNG                EI  Q BP_JOB_CHECKSTAT: Failed to correct status of job. Reason:
18:12:32 DIA  011 800 DHEUNG                EF  H > Function module was called with an invalid dialog mode ( )
18:12:33 WRK  000                           Q0  Q Start Workp. 22, Pid 2776
18:12:41 BTC  022 800 DHEUNG                F6  F TemSe object JOBLGX17295300X02098 was closed remotely

Filesystem has 50GB free. No space issue here.

Errors in the Windows Event Viewer list the same ones as shown in the log file.

I did a quick check on the error messages and it looks like this thread http://scn.sap.com/thread/1846681 may be related. Edit: After running the consistency check in t-code SP12, we found a large number of inconsistent objects and deleted them all. The Spool Consistency Check found no errors. Hopefully this should resolve the errors with the Table Preparation step.

[Former Member · Accepted solution]

Had one of our Basis colleagues look at the logs and he recommended digging into the WP trace since they were stopped. In the dev_wpXX files we saw errors regarding no heap available and that the limit set by the parameter abap/heaplimit was reached.

In the end, we increased the value of abap/heaplimit from 40000000 (default) to 80000000 and for good measure, increased RAM from 4GB to 12GB and doubled paging space from 20GB to 40GB. After running the job again, it completed successfully in 1.5 hours. One thing we noticed during the runtime of the job CNV_MBT_CNV_TDMS_06_GET_USER_TAB was that one disp+work process was running in Task Manager and it consumed 8GB of memory! This is something we will have to make a note of.

[Former Member · Accepted solution]

Good to know that problem is resolved.Thanks for sharing the solution in detail.