DSXchange

Comparing the two DSX files with diff does show you what's changed.

Remove Duplicates requires sorted input. In increasing order of inefficiency, try the following techniques.

Sort within DataStage, specifying "Unique"

Filter through sort -u

Remove Duplicates stage

Access through ODBC and specify DISTINCT in the SQL statement.

DSguru2B wrote:It keeps giving me the literal "MILLION" for input 0,1,99999.
I presume its not the complete routine, is it?

That zero is not supported was stated explicitly. But it would be a minor change.
Surprised by the other two results - will have to test and check the logic.

I missed the 16 nodes, 15 sessions per player bit. Can your Teradata installation accept (16 x 15) sessions? Maximum sessions is number of TPAs multiplied by number of AMPs per TPA. If you exceed this number, then many of your connections may simply be waiting for others to be freed. For example, wi...

Typically an after-job subroutine executing ExecSH to run a mv command that uses the result of a date command in whatever format you desire for your timestamp.

Search - it's been documented in more detail previously.

Try logging in again, so that .profile is re-executed.

If there had been a resolution it would probably have been posted here sometime in the past two years.

Have you asked your support provider?

These tables' metadata are deliberately incomplete.

DSguruji's answer will show you the publicly-available column names.

Any attempt to UPDATE these tables may void your warranty and may also damage your Repository.

Please post exact and complete warning messages.

"Something like this" can contain misleading information and waste our time.

Incremental design.
Thorough testing.

Buy LOTS of hardware from your friendly IBM rep.

One of the above is not as serious a comment as the other two.

All Data Sets preserve the partitioning with which they were written, and must be read with the same degree of parallelism on the same nodes.

Which XML stage?

In general you remove duplicates using sorting (with the "unique" option selected) or with a Remove Duplicates stage.

Repartitioning in a cluster/grid environment uses TCP sockets. The error message has included a port number. Try tracking down what that port number is used for.

We've helped you to solve the "orchadmin" command not found problem (PATH). We've helped you to solve the "liborch686.so" library not found problem (LD_LIBRARY_PATH). What "permissions" are you talking about? If PATH, LD_LIBRARY_PATH, APT_ORCHHOME and APT_CONFIG_FILE are correctly set, you should be...

Welcome aboard. :D The short answer is no. Can you CAST the BLOB as LongVarChar of a suitable size and work with that? In general BLOB is not supported for ETL, because there's no "T" (transformation) possible. You can extract and load BLOB, but if you want to access their contents, that's a whole n...