DSXchange

Can you access Oracle from the command line outside of datastage?

We are just guessing as to the likely cause without getting the detailed error message from you. The specific SQL or database error message. It could be your transaction log filling up or any one of a number of possible errors.

If you have just one transform stage in both scenarios there will be no difference between the two at runtime.

Is your CS_ORDER_FACT Table in schema CDW your intended target?

Can you make a quick test job to read the CDW.CS_ORDER_FACT table from DataStage to see if that works?

The "VOC" is a hashed file, which can be treated as if it were a table. This contains not only the keywords used by the engine itself, but also pointers to all files used by all jobs; i.e. by creating a new job you will implicitly make several changes to the VOC; thus there is little to be...

austin_316 - since the user can choose any location when creating a project, one cannot specify a specfic path as an answer to the original poster's question.

I still can't find a column labeled "INTEREST"

No problem in doing so, but you need to answer the following question first:

If your data contains "Hello There", what should the resultant string value be?

You were on the right track and close - the project directory is where the files are stored.

DataStage has no mechanism for internal transfer of DataStage data from one server to another, so the only way to read/write a dataset is for it to be visible to an "open" command and that means it needs to be visible on the file system.

What is the "Column Metadata" for the COMP-3 and the COMP columns? Note that ANSI COBOL does not specify a fixed format for COMP - the compiler implementor can choose the internal representation. Thus, unlike COMP-3, whose format is fixed, DataStage cannot know what the COMP binary format ...

Please start a new thread as this topic is closed and resolved and it very doubtful that you error is identical, otherwise the solution here would have solved your problem as well.

Double precision can store more values than single, so DataStage is warning you that there may be cases where this conversion can fail.

I don't see an "INTEREST" column and the length of the schema doesn't add up to 8000. Did you cut off anything?