DSXchange

Hi,

I am getting the following in my job output:
cs}}},2: Fatal Error: Tsort merger aborting: Scratch space full [sort/merger.C:870]

I know that increasing the scratch disk will resolved the problem but what I would like to know after reading on similar topics in this forum is; is it only the SORT stages that generates the Tsort merger aborting errors or does other stages like the join could also do it?

Thanks.

I think even stages like Change Capture and others which use the tsort operator internally, i.e if the input data is not sorted generate the same error. Because they have the tsort operator in them.

Yes other stages like join, remove duplicates,... also can generate 'Tsort merger aborting errors'. Take a look at the score by adding APT_DUMP_SCORE parameter in your job. You can see which all operators insert Tsort (Datastage automatically insert OSH Tsorts in certain situations)