Error running match stage in parallel mode

[ashok]

Hi,
getting errror while running match stage with more than one pass in parallel mode,

the results are good when I have only one pass.

error message is:

ualityStage job TEST02(414238) exited with status 1.
STEP logs are in /QSHOME/Logs/TEST02.*.log.



/QSHOME/Logs/TEST02.0.log:##E TFSU 000154 16:12:27(000) <UNI(1),1> fork failed, Resource temporarily unavailableUNI(1) /QSHOME/Logs/TEST02.0.log:##E TFSU 000154 16:12:27(000) <MAT(1),1> fork failed, Resource temporarily unavailableMAT(1) /QSHOME/Logs/TEST02.0.log:##E TFSO 000089 16:12:27(001) <MAT(1),1> The runLocally() of the operator failed.
/QSHOME/Logs/TEST02.0.log:##E TFPM 000040 16:12:27(005) <MAT(1),1> Operator terminated abnormally: runLocally did not return APT_StatusOk /QSHOME/Logs/TEST02.0.log:##E TFSU 000154 16:12:27(000) <UNI(0),1> fork failed, Resource temporarily unavailableUNI(0) /QSHOME/Logs/TEST02.0.log:##E TFSU 000154 16:12:27(000) <MAT(0),0> fork failed, Resource temporarily unavailableMAT(0) /QSHOME/Logs/TEST02.0.log:##E TFSO 000089 16:12:27(001) <MAT(0),0> The runLocally() of the operator failed................

Thanks

[ray.wurlod]

First guess is that your system is struggling to start enough processes quickly enough. There is an environment variable with TIMEOUT in its name that may help (I'm not doing DataStage this week, so haven't any means to check the exact name - I think it's APT_PM_NODE_TIMEOUT.

[ashok]

when I added APT_PM_NODE_TIMEOUT=150 it is working for small number of records, and try to run with big number of records in file, it is still failing