DSXchange

We have the following set up. Our online application is bound with Datastage RTI jobs through Text over JMS bindings. Each RTI job has its own input/output destinations and corresponding MDBs deployed in WebSphere.

Recently we started getting the following error causing it to abort:
**********
main_program: Internal Error: (d->nPartitions()== nPartitions_): datamgr/segment.C: 276
Traceback: pureAssertion__13APT_FatalPathFPCcPCci() at 0xd898ddec
setPartitionDescriptor__13APT_DMSegmentFPC16APT_DMDescriptor() at 0xd833e8f4
setPartitionDescriptor__16APT_DMDescriptorFv() at 0xd833a3d8
readDescriptor__17APT_DMFileDataSetFv() at 0xd83e40d8
describe__17APT_DMFileDataSetFR16APT_DMDescriptor() at 0xd83e32d0
setFileMode__14APT_DataSetRepFRC11APT_UStringQ2_11APT_DataSet9DirectionQ2_11APT_DataSet10OutputFlag() at 0xd80dee2c
__ct__11APT_DataSetFRC11APT_UStringQ2_11APT_DataSet9DirectionQ2_11APT_DataSet10OutputFlagRC11APT_UStringQ2_11APT_DataSet6DSType() at 0xd80d845c
instantiateAndInitializeDataSet__22APT_OSL_SIL_DataEntityFP20APT_OSL_SIL_StepSpecR12APT_ErrorLog() at 0xd8b6e04c
instantiateAndInitializeStep__20APT_OSL_SIL_StepSpecFR12APT_ErrorLog() at 0xd8b69e6c
createAndCheckStep__7APT_OSLFP20APT_OSL_SIL_StepSpecR12APT_ErrorLog() at 0xd8b2ea30
************

Anyone ever faced the above error? Any inputs would be greatly appreciated.

Thanks

Are you accessing a Data Set that was created with a different configuration file (and therefore with a different number of partitions)?

ray.wurlod wrote:Are you accessing a Data Set that was created with a different configuration file (and therefore with a different number of partitions)?

Thank you for replying Ray -
We are having this issue with multiple jobs - one for a RTI job that aborts with this error as soon as it is enabled and the other for a parallel job invoked through a sequencer. The RTI job has been using the same configuration file for a long time but the other parallel job has been using a different config file from what it is using currently.

Also, per the post in http://dsxchange.com/viewtopic.php?t=11 ... 5aa52e5b94, we did try removing the datasets last created by the parallel job when it aborted, but that didnt seem to help.

Can you try this - change the name of the dataset in your job and rerun?

John Smith wrote:Can you try this - change the name of the dataset in your job and rerun?

Yes, that didnt work either. Getting the same exception.

IBM's response
There is issue with Jobmon if the number of records crosses some limit.

To avoid that, please set the following:

APT_MONITOR_SIZE=100000 or more
APT_DISABLE_COMBINATION=true

One of the parameters that was set when the problem was resolved was, APT_NO_JOBMON to True, this would also workaround the issue with Jobmon.