DSXchange

Looks like C++ compiler is not installed or not configured correctly. Search on "the PWD environment variable is not defined".

Search on "is used as output multiple times". Are you using the same DataSet twice in the same job?

Search on "Vertical Pivot". It had returned me 160 results.

Do you have write permissions to the scratch disk? Buffer operator first writes the data to RAM. Then writes to the scratch disk if the RAM gets full(i think the word is Overflow). So check if you have the write permissions on your scratch disk.

Do a search on "When checking operator: Defaulting".

##I TLIC 000002 23:50:52(000) <APT_LicenseCheckOp in APT_LicenseOperator,0> Node 'etldev1' has 12 physical processors. ##W TLIC 000003 23:50:52(001) <APT_LicenseCheckOp in APT_LicenseOperator,0> DataStage has found more CPU's on your system than your current license allows. ##W TLIC 000004 23:50:52...

In the log you will find an entry reading "This step has xx datasets". Open that entry. Should be able to find entries like "eEntire<>eCollectAny". In this case its using partition as Entire.

Search on "will partition despite the
preserve-partitioning flag"

Sort order might remain but i would have Removed the sort from the SQL and done it in DataStage on stream 1 before the remove duplicates stage.

Posting the same question thrice in the wrong forum with different subjects will not get you the answers fast. Delete the other two posts. Post this question in "IBM DataStage Enterprise Edition (Formerly .........)" forum. And reading about the Aggregator stage will help.

You will find two transformers in the palette for a parallel job. Transformer, Basic transformer. You will have to use a Basic Transformer in parallel jobs to use a basic routine. If you cannot find the Basic Transformer then customize your palette and add it.

how are you triggering these jobs? Job activity? One of the ways to check the status of the first job is through a basic routine to retrieve the job log and search through it for the error (to the best of my knowledge). And then trigger the down stream jobs.

SKIP_INDEX_MAINTENANCE=YES,SKIP_UNUSABLE_INDEXES=YES did you try tweaking any of these properties? I believe this is causing a problem during restart. We have a similar problem in load option with rejects. Though it is a functional success the job aborts. so we check the job log for "BAD"...

Whats your job design? Are you trying to load a table or read from one? Looks like there is a lock on what every you are trying to access, and the job failed because of that. Post your job design and also is "xfmjobname" a database stage?

Look at the stage, ecom_order_proj_dtl_SRC_SEL. there is possibly an implicit conversion. Check the data types for the source and the target columns and their mappings.