Hi,
You have to increase the scratch disk space..
Else You can achive the same join condition in database Query...
Before that try to remove unwanted datasets in unix path..
Thanks
Manoj kumar
And if there is no possibility to increase scratch space is there any other cases to obtain current issue? Can we somehow to configure job and tell to JOIN stage not to use tsort operator, because actually scratch directory contains tsort files.
Achiving join condition in data base query is possible when the tables from one db schema. In my situation tables from one schema, so i did it in one query like a work around, but when schemas are different then it does not work if we have not dblinks, synonyms, ...
The Join stage requires sorted data, so you can either let the job do the sorting for you or you can ensure the data arrives sorted. Since you don't seem to have the space to do the former, let the original database sources do the sorting by ordering the output properly. Then you can put a Sort stage before the Join and tell it Don't Sort, Already Sorted so the job knows you've pre-sorted the data.
-craig
"You can never have too many knives" -- Logan Nine Fingers