I have a simple job, its just extract an load,
but when i run the job it hangs.
no log is recorded, after some time the job batch gets aborted,
i have no idea why this happens.
Please guide me through to solve the problem.
Remove the 'load' part, make the job write to /dev/null. If it still hangs, you know it is an issue on your extract, perhaps a very lenghty select. If it runs, then the issue is probably on your load stage.
Some more details on the stage types involved in the source and target would make solving the problem a lot easier as well.
Remove the 'load' part, make the job write to /dev/null. If it still hangs, you know it is an issue on your extract, perhaps a very lenghty select. If it runs, then the issue is probably on your load stage.
Some more details on the stage types involved in the source and target would make solving the problem a lot easier as well.
I've been getting network timeouts all day here - but I don't remember trying to re-post this one. Too late to delete the duplicate now as well. I suppose that this is one way to rack up some serious points.
i tried "view data" in the source stage, i got result in 5 min, there are about 15 lakhs records, but the stage hangs in the select query, when job is run.
and is there any other way other than removing the load part and writnig to dev/null
Both stages are Oracle8 , Source stage has one select query, Transformer has 3 lookups and writes to oracle table and rejects are written to sequential file.
please let me know in case you need anything.
Put a constraint of "1=2" onto the transform stage output writing to your Oracle stage and run the job. Better yet, make a copy of your job and keep only your source stage and write the result to a sequential file at /dev/null to see how long your select takes to execute in the real job, not just the view data.
</a>
