System Exception ACCESS VIOLATION

dataservices administration
1

Hello All,

I have recently upgraded from BODI 11.7.3.1 to BODS 12.2.3.3 successfully.
I executed one of the flows containing many WFS and DFS and got the below error

System Exception <ACCESS_VIOLATION> occurred. Process dump is written to <I:\Business Objects\Data Services
12.2.3.3\log\BODI_MINI20120830083151_5928.DMP> and <I:\Business Objects\Data Services
12.2.3.3\log\BODI_FULL20120830083151_5928.DMP>
Process dump is written to <I:\Business Objects\Data Services 12.2.3.3\log\BODI_MINI20120830083151_5928.DMP> and <I:\Business
Objects\Data Services 12.2.3.3\log\BODI_FULL20120830083151_5928.DMP>
Call stack:
0023:00F2335D, XProc_children::wait4threads()+0477 byte(s), x:\src\core\xchildren.cpp, line 0470
0023:00F2FFE2, XStep_pblock::execute()+0114 byte(s), x:\src\core\step.cpp, line 0740
0023:00F2FF4D, XStep_sblock::execute()+0029 byte(s), x:\src\core\step.cpp, line 0715
0023:00F32D2C, XPlan_spec::execute()+0348 byte(s), x:\src\core\plan.cpp, line 0082

I found that it crashed at a workflow having 3 DFs in parallel, I executed the workflow separately and it executed succesfully.
At job level, I changed the execution of these DFs from parallel to sequence and job did not crash this time.
These DFs have different source and targets, but still it crashed when in parallel.

Can you please let me know the cause and resolution (as my fix is just a workaround)?

Thank you.

kavita3110 (BOB member since 2012-08-30)

2

We had a simular issue and the error message you post is completely the same.

Problem is the message has no usefull info… but if you open te correct log-file on the server you might find the complete error message.

I our case it was always an …

I just solved it yesterday… turns out that 2 queries in the same DF or parallel DF are inserting/deleting lines from a table… this causes the table to be locked at some point … which sometimes gives us this error…

I have adjusted the DF so the deleting and inserting queries dont run in the same DF using the same table… and now it runs fine…
Good luck!

Nielsa

Vlampiep (BOB member since 2012-09-05)

3

Hello Experts,

We have now upgraded from DI 11.7 to DS 12.2.3 on production as well and many jobs failed giving the access violation error and creating dump files on the server. We also tried to switch to in-memory cache… In some cases it was succees but taking more time to load the data.

(12.2) 10-27-12 02:15:52 (E) (3140:9240) FIL-080105: |Data flow INIT_EKKN_DF|Reader INIT_EKKN_R3DF
A row delimiter was seen for row number <1173419> while processing column number <19> in file <//wprosg/P01/ZDX3098_EKKN.dat>.
The row delimiter should be seen after <20> columns. Please check the file for bad data, or redefine the input schema for the
file by editing the file format in the UI.
(12.2) 10-27-12 02:15:52 (E) (3140:9240) FIL-080105: |Data flow INIT_EKKN_DF|Reader INIT_EKKN_R3DF
A row delimiter was seen for row number <1173504> while processing column number <19> in file <//wprosg/P01/ZDX3098_EKKN.dat>.
The row delimiter should be seen after <20> columns. Please check the file for bad data, or redefine the input schema for the
file by editing the file format in the UI.
(12.2) 10-27-12 03:53:50 (W) (000:000) ABT-200201: Job with pid <8516> is killed.
(12.2) 10-27-12 03:53:50 (E) (8516:2280) SYS-170101: |Session IASCF_JOB|Work flow FACT_load_WF|Work flow EKET_Correction_WF|Data flow EKET_Correction_DF
System Exception <ACCESS_VIOLATION> occurred. Process dump is written to <I:\Program Files\Business Objects\Data Services XI
12.2\log\BODI_MINI20121027020259_3396.DMP> and <I:\Program Files\Business Objects\Data Services XI
12.2\log\BODI_FULL20121027020259_3396.DMP>
Process dump is written to <I:\Program Files\Business Objects\Data Services XI 12.2\log\BODI_MINI20121027020259_3396.DMP> and
<I:\Program Files\Business Objects\Data Services XI 12.2\log\BODI_FULL20121027020259_3396.DMP>
Call stack:
001B:00E2894F, vcharUserInterfaceRegular::allocVchar()+0079 byte(s), x:\src\eval\vchar.cpp, line 0511
001B:00E1C95A, XRow_data::get_addrAndAlloc()+0234 byte(s), x:\src\eval\row.cpp, line 2196+0137 byte(s)
001B:00E1D53B, XRow_data::get_addrAndAlloc()+0043 byte(s), x:\src\eval\row.cpp, line 2188
001B:00A693A7, Flat_file::getColumn()+2231 byte(s), x:\src\filespec\flat_file.cpp, line 3000
001B:00F4BEFD, RdrFile::getnext()+0621 byte(s), x:\src\reader\rdrfile.cpp, line 0963
001B:00F3414B, RdrAdapter::execute()+0027 byte(s), x:\src\reader\rdradapter.cpp, line 0787
001B:00E527F1, XTran_desc::execute()+0273 byte(s), x:\src\core\xtran.cpp, line 0588
001B:00D2C223, Rww_thread::main()+0083 byte(s), x:\src\rww\rww.cpp, line 0425
001B:01B4C20B, RWThreadFunctionImp::run()+0139 byte(s)
001B:01B3C9D2, RWRunnableImp::exec()+0226 byte(s)
001B:01B4C4DC, RWThreadImp::exec()+0076 byte(s)
001B:01B4D819, RWThreadImp::setTimeSliceQuantum()+0249 byte(s)
001B:77E6482F, GetModuleHandleA()+0223 byte(s)
Registers:
EAX=05FB842C EBX=04728F80 ECX=0000002D EDX=05FB8420 ESI=00000001
EDI=05FB62E0 EBP=00000022 ESP=0930F99C EIP=00E2894F FLG=00010206
CS=001B DS=0023 SS=0023 ES=0023 FS=003B GS=0000
Exception code: C0000005 ACCESS_VIOLATION
Fault address: 00E2894F 01:006E294F I:\Program Files\Business Objects\Data Services XI 12.2\bin\acta.dll
==========================================================
Please collect the following and send to Customer Support:
1. Log files(error, monitor_, trace_*) associated with this failed job.
2. Exported Atl file of this failed job.
3. DDL statements of tables referenced in this failed job.
4. Data to populate the tables referenced in the failed job. If not possible, get the last few rows (or sample of them) when

Please suggest how to resolve access violation error and long running of flows. need it urgently as it is on production.

Thanks & Regards
Kavita3110

kavita3110 (BOB member since 2012-08-30)