DSXchange

How could you not find the "answer"? You have 70000+ posts to read thru, you could trace the history from version 1.0 to 7.5 plus discussion on the Hawk release.

I think what you mean is that you can't find your answer in one single post.

Since the DS Engine was installed on the coordinator node, and the other 17 nodes for the database also were PX nodes, you'd see tremendous degradation in Designer when queries were swamping the coordinator node. Jobs took a long time to open, close, and edit when the coordinator node was busy doing...

Consider: 1. A stored procedure that takes a fixed number of inputs for the WHERE clause, with a code value that tells the SP how to construct the proper dynamic SQL and then issues the appropriate SQL. 2. Break your existing job into 2 parts. Job one is ODBC-->XFM-->SEQ where the SEQ file contains ...

Won't work. You're saying that the WHERE conditions vary for each row being processed? Meaning each row constructs a WHERE clause, which may include 1, 2, 3, or more columns? The query is declared as a cursor, so there's fixed input variables. You can't suddenly use 2 columns for a WHERE condition o...

More like 5

Why do the last 4 posts from you in the space of 5 minutes look like job interview questions?

Why do the last 3 posts from you in the space of 5 minutes look like job interview questions?

Why do the last 3 posts from you in the space of 5 minutes look like job interview questions?

I can't figure out your context but here goes: Pushing is your local server sending data away. Pulling is your local server bring data to itself.

If something pushes data to you, it means their server is sending data to your server. If you go get that data, you're pulling it over.

Something about that coalesce ... OVER strikes me as funny. Your ORA message is about a function window, I suspect that's what it is. Maybe DS messes the query up somehow.

You seem to be trying to update the primary key value of a parent that would orphan children. Shame on you. You should create a reject link and spool the rejected rows to a file. Then, research why you think those rows should be allowed to update when there are children.

Hmmm, that shouldn't have made any difference. With generated SQL, that matters because it is writing the query for you and would add a GROUP BY clause with your columns. You would also have to put something into the DERIVATION column to MIN/MAX/SUM/COUNT/etc the non-grouping columns. It sounds like...

You should try recreating the job using just ODBC-->XFM-->SEQ. If you're on DS 5, there could be something flaky where the job isn't correctly put together underneath the covers. You don't need to use the "as sum_salary" because DS doesn't care. It runs the query as a cursor and just reads from the ...

Custom job control, use a Batch job, use DSExecute API to browse a directory for a list of files, parse the screen output and loop running your jobs, probably doing something like tagging/moving the file after processed so that it doesn't get handled again. Check out Batch jobs from Director, they'r...

I know you're new here, so welcome aboard. Please post your exact SQL because soooo many times the "just like this SQL" isn't just like the SQL.