Issue with Upsert option

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

antonyraj.deva
Premium Member
Premium Member
Posts: 138
Joined: Wed Jul 16, 2008 9:51 pm
Location: Kolkata

Post by antonyraj.deva »

devidotcom wrote:This issue is yet open.. can anyone help. Now values i used for the environment variables did not work for jobs which had record count >100000.

Please help.
Which environment variables are creating the error? Provide the complete error log.

--Tony
devidotcom
Participant
Posts: 247
Joined: Thu Apr 27, 2006 6:38 am
Location: Hyderabad

Post by devidotcom »

How do i give u the complete log? Below are the fatal errors.. I agree with you it must be the environment variaibles...

ora_Sample_load,1: Operator terminated abnormally: received signal SIGBUS
main_program: APT_PMsectionLeader(2, node2), player 7 - Unexpected exit status 1.
APT_PMsectionLeader(1, node1), player 8 - Unexpected exit status 1.

Below are the parameters from the log:
_INIT_RUN_NPREV=0
_INIT_UTS_ISA=sparc
_INIT_UTS_MACHINE=sun4v
_INIT_UTS_NODENAME=usbdz1gipdst801
_INIT_UTS_PLATFORM=SUNW,T5240
_INIT_UTS_RELEASE=5.10
_INIT_UTS_SYSNAME=SunOS
_INIT_UTS_VERSION=Generic_142900-06
_INIT_ZONENAME=usbdz1gipdst801
APT_COMPILEOPT=-dalign -O -PIC -library=iostream -c
APT_COMPILER=/opt/SUNWspro/bin/CC
APT_CONFIG_FILE=/dsproj001/is8/Projects/ERP_US/grp01_config_2x_ERP_US.apt
APT_DISABLE_COMBINATION=1
APT_ERROR_CONFIGURATION=severity, !vseverity, !jobid, moduleid, errorIndex, timestamp, !ipaddr, !nodeplayer, !nodename, opid, message
APT_LINKER=/opt/SUNWspro/bin/CC
APT_LINKOPT=-G -library=iostream
APT_MAX_DELIMITED_READ_SIZE=1000000
APT_MONITOR_MINTIME=10
APT_OPERATOR_REGISTRY_PATH=/dspodup1/is8/Projects/ERP_US/buildop
APT_ORACLE_LOAD_OPTIONS=OPTIONS(DIRECT=FALSE,PARALLEL=FALSE)
APT_ORACLE_PRESERVE_BLANKS=1
APT_ORAUPSERT_COMMIT_ROW_INTERVAL=5000
APT_ORAUPSERT_COMMIT_TIME_INTERVAL=2
APT_ORCHHOME=/dspodup1/IBM/InformationServer/Server/PXEngine
ASBHOME=/dspodup1/IBM/InformationServer/ASBNode
BELL=^G
DS_ENABLE_RESERVED_CHAR_CONVERT=0
DS_NO_PROCESS_METADATA=1
DS_OPERATOR_BUILDOP_DIR=buildop
DS_OPERATOR_WRAPPED_DIR=wrapped
DS_TDM_PIPE_OPEN_TIMEOUT=720
DS_TDM_TRACE_SUBROUTINE_CALLS=0
DS_USERNO=-7507
DSHOME=/dspodup1/IBM/InformationServer/Server/DSEngine
DSIPC_OPEN_TIMEOUT=30
FLAVOR=-1
HOME=/
LC_COLLATE=en_US.ISO8859-1
LC_CTYPE=en_US.ISO8859-1
LC_MESSAGES=C
LC_MONETARY=en_US.ISO8859-1
LC_NUMERIC=en_US.ISO8859-1
LC_TIME=en_US.ISO8859-1


Please advice
antonyraj.deva
Premium Member
Premium Member
Posts: 138
Joined: Wed Jul 16, 2008 9:51 pm
Location: Kolkata

Post by antonyraj.deva »

devidotcom wrote:ora_Sample_load,1: Operator terminated abnormally: received signal SIGBUS
main_program: APT_PMsectionLeader(2, node2), player 7 - Unexpected exit status 1.
APT_PMsectionLeader(1, node1), player 8 - Unexpected exit status 1.
The "SIGBUS" error is a unix memory access error that occurs when the job tries to point to an invalid memory address or when time-based monitoring is enabled. If you enable row based monitoring, it should fix this problem.

SIGBUS may also occur when the job is attempting to access a value larger than a byte at an unaligned address in the server.

Try to set the APT_MONITOR_INTERVAL=5 (default is 5 - this controls time-based monitoring) and run the job after recompiling it.

Note that if you have the _INTERVAL variable set to anything but 5 (default) it automatically overrides the _SIZE variable setting and reenables time-based monitoring.

See if this resolves your issue.

--Tony
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

I've also seen that SIGBUS error get thrown by buggy Oracle client versions. Mentioned earlier that I thought it would be prudent to see about upgrading your Oracle client.
-craig

"You can never have too many knives" -- Logan Nine Fingers
devidotcom
Participant
Posts: 247
Joined: Thu Apr 27, 2006 6:38 am
Location: Hyderabad

Post by devidotcom »

This option mentioned below did not work.

We have some bitmap indexes on the table, we tried dropping and then ran the job. The job executed perfectly.

We want to use this option now of drop and recreate indexes but the Open/close and Rebuild/Maintenance options are only present when the write method is Load. Any suggestions on how to implement this when the Write Method is Upsert?

Thanks
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Only the obvious one - in the Before and After commands.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
devidotcom
Participant
Posts: 247
Joined: Thu Apr 27, 2006 6:38 am
Location: Hyderabad

Post by devidotcom »

Instead of dropping indexes will making the indexes as Unusable is the same as dropping indexdes before the load and then rebuild them once the load is completed. Will this work the same.
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

I don't believe so, but that's really a question for your DBA.
-craig

"You can never have too many knives" -- Logan Nine Fingers
ljulienne
Participant
Posts: 10
Joined: Tue Jun 09, 2009 3:14 am
Location: Lyon, FRANCE
Contact:

Post by ljulienne »

I don't know if case is still open but I had this kind of problem (It was SIGTERM error). Job stopped for huge reason in upsert with update only.

In fact, that was due to parallel update. I changed the OCi stage to sequential and the problem was solved.
Laurent JULIENNE
SYNCER
Post Reply