DSXchange

It should be enabled by default. What makes you think it isn't?

Who knows? Start with a Null Field Value property setting of "" on the Format tab of the Sequential File stage.

Yes, I recommend using a server Sequential File stage.

Depending on the volume of data, use a server job (small to medium volume, say up to 10 million rows) or a server Shared Container containing just a Sequential File stage in a parallel job if the volume of data is large.

That's gotta be about the shortest correct response ever!

True? We can only choose to believe what you told us.

Check your job log for warnings, and check null field value in the target stage. Check, too, that the output column is marked as nullable.

The difference between the statements is B.YY. Does column YY exist in table B?

Look in the job log to determine where the abort occurred, and try to identify the line number. Because everything is buffered in parallel jobs this may not actually be possible.

DO try to think up meaningful subjects.

Look elsewhere for the problem - the error message is complaining about a LONG data type, not about a VARCHAR data type.

Why do people think they can change the client's data with impunity?!!

What if these linefeeds are meaningful to the client?

If the operating system behaves conventionally (and since DataStage jobs are processes) that's what I'd expect.

When you get version 8 you can search filtering on Date/Time Modified.

Row buffering can not be the culprit directly, but the timing issue might be related to clearing those buffers. But I don't think so - the stages should not close until the buffers have been cleared.

Is C:\WINDOWS in your PATH?