DSXchange

Is environment idle for both the case? You server might be loaded on the later case.
There may be a chance where the additional node might be be easily accessible by dataset to write the data into. For testing, you can try the give single node where the sequential file is created.

I guess Ray should be more happy, as this is his favorite database, as I could recall from some of his previous post.

Fine. Option suggested by Ray would be more simple, when compare to others.

You can use it for any audit purpose about jobs. We used it once.

DSExport.exe and dscmdexport.exe both are client component. I dont think we have an option to export from Unix server side.

Just curious, why do you need to use ODBC for file? Is it not flat file?

There is a bug in Version 7.0. Which will just give out an message in director that the job is been started at the scheduled time. It looks like "DS.SCHED: Started at.....". But actually the won't start.
If this is the case, you need to approach support. But still, even they can't help much.

SIYF(Search Is Your Friend.) Simply use "SQL Server stored procedure" as keyword, you ll get many hits.

Since you are using PX edition, the order is not guaranteed. Unless you are processing in sequential mode. Since the records are processed in different CPU's, based on the resource of each CPU the records will be processed, and the order will be changed. If you strictly need the order to be maintain...

Version 7.5.1 minimum requirement is version 2.8.
But couldnt find anywhere mentioned about the latest support version. Perhaps IBM website might have been updated. Searching overthere could be worth.

You can optionally read all the fields in a single field using Varchar datatype and eliminate the empty lines using transformer.
Optionally you can use FIND witn /V option to eliminate the empty lines. Iam just looking for the command to find the newline. If I get it, I ll update you.

What version of datastage?
Version 7.0 supports only till 2.7, 2.8, 2.9. I ll get you the latest as well. But you could tell which version you are into.

Iam working in Version 7.0.
Same path. Worked with dsadm.

If you make a search on 'Range lookup' you can get an idea of how to go about this. One way is you can populate the hash file in such a way that the values are continues, ie., 100 to 200 will point to 200 and 200 to 500 will point 500. As you say, you can use Universe stage for join query for the ra...

Don't you find any other error/warning message along with this?