SIGSEGV Error when trying to load LOB data

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
dsuser7
Participant
Posts: 41
Joined: Sat Nov 29, 2008 2:21 am

SIGSEGV Error when trying to load LOB data

Post by dsuser7 »

Hi,

We have few jobs which reads SQL Server table and loads another SQL Server table. The source and target tables are in different databases.

Job Design:

ODBCStage_Sourcetable ->Transformer ->ODBCStage_Targettable

ODBC Target stage - WriteMode: Insert; Table Action: Append; Generate SQL : Yes

The source and target tables contain 1 CLOB field, the data in which is html tags for few records. The job just hangs when triggered.

So we removed the CLOB field and ran the job, but the job still takes 2hrs or more and aborts with the following error.

Operator terminated abnormally: received signal SIGSEGV
ORCHESTRATE step execution terminating due to SIGINT


We set the APT_DEFAULT_TRANSPORT_BLOCK_SIZE to 1048576 but the job just runs for a while to abort with the above mentioned error.

Any suggestions how to make the job work?

Thanks!
priyadarshikunal
Premium Member
Premium Member
Posts: 1735
Joined: Thu Mar 01, 2007 5:44 am
Location: Troy, MI

Post by priyadarshikunal »

Any other warning or error? Do you get something like "From previous run" when you reset the job? Try searching as there are zillion post on sigsegv error.
Priyadarshi Kunal

Genius may have its limitations, but stupidity is not thus handicapped. :wink:
dsuser7
Participant
Posts: 41
Joined: Sat Nov 29, 2008 2:21 am

Post by dsuser7 »

Here are the warnings:

odbc__STAGE_QC_TEST__Ld,0: from /usr/lib/hpux64/libc.so.1
odbc__STAGE_QC_TEST__Ld,0: #1 0xc000000000560d70:0 in __nanosleep2+0xd0 () from /usr/lib/hpux64/libc.so.1
odbc__STAGE_QC_TEST__Ld,0: #2 0xc00000000040aa10:0 in sleep+0x3f0 () from /usr/lib/hpux64/libc.so.1
odbc__STAGE_QC_TEST__Ld,0: #3 0xc0000000218ec110:0 in jobmonRoutine () at api/operator_rep.C:7126
odbc__STAGE_QC_TEST__Ld,0: #4 0xc00000000013fe80:0 in __pthread_bound_body+0x190 ()
odbc__STAGE_QC_TEST__Ld,0: from /usr/lib/hpux64/libpthread.so.1
odbc__STAGE_QC_TEST__Ld,0: (gdb) Detaching from program: /opt/IBM/InformationServer/Server/PXEngine/bin/osh.exe, process 15757


Thanks.
vibwipro
Participant
Posts: 40
Joined: Thu Jan 24, 2013 7:34 am

Re: SIGSEGV Error when trying to load LOB data

Post by vibwipro »

Hi

Try to Test Connection as well as Try to view data in both Source and Target stage.
dsuser7
Participant
Posts: 41
Joined: Sat Nov 29, 2008 2:21 am

Re: SIGSEGV Error when trying to load LOB data

Post by dsuser7 »

Yup, did that, it shows data and Connection is successful.
vibwipro
Participant
Posts: 40
Joined: Thu Jan 24, 2013 7:34 am

Re: SIGSEGV Error when trying to load LOB data

Post by vibwipro »

A SIGSEGV signal is caused by an invalid memory reference or a segmentation fault. You are probably trying to access an array element out of bounds or trying to use too much memory.

So what transformation logic is implemented in Transformer ? Is any Surrogate Key being generated or some thing like that where DS Buffer come in picture ? This error generally deals with limited buffer size.
dsuser7
Participant
Posts: 41
Joined: Sat Nov 29, 2008 2:21 am

Re: SIGSEGV Error when trying to load LOB data

Post by dsuser7 »

The Transformer is used to assign some constant values to few fields and for CurrentTimestamp() and DSJobName. There are no transformations/surrogate key generate.

Only the source has CLOB fields - which might cause the memory issue? But not sure how to handle it.

Btw, the job runs fine if only 10 records are selected from source (Top 10).
Post Reply