DSXchange

It is not a general query regardless of engine, because the solution is different depending on whether you're using a server job or a parallel job. As Sanjay points out (in highly annoying text messaging abbreviations - what does your DataStage documentation look like, Sanjay?!!), there is a Remove ...

Don't think so, Denzil. The post to which you referred argued for clearing the status file: if you can't open the status file there's no way you're going to be able to clear it. Sounds like the RT_STATUS1711 hashed file (the status file for job number 1711) has been deleted or there is a permissions...

No, not clear at all. A hashed file will take all records, but only the last will remain, since every update to a hashed file is a destructive overwrite. A UV stage will only take matching records if you use one of the slow, double-operation, rules (insert or update, or update or insert, or replace ...

Reset the job, which will also reset the wave numbers, and this problem should go away. DataStage works with the concept of "resources" (in the RT_STATUS hashed files in the Repository, which is viewed by the Status view and Monitor view in Director, and by "show performance statistics" in Designer ...

If you're trying to use dsdadm as the administrative user, you will need to enable impersonation mode. Read up on this in the installation guide.
Otherwise, try using dsadm as the user ID. But, even then, you should properly have installed as dsadm.

You're right. The only way to execute a query against a hashed file is if that hashed file is known (in the VOC file). The SETFILE command can be used to establish the pointer, or you can write it directly to the VOC file (I know that you know the format of an "F" entry in VOC). Once that's there, y...

Teej has been a bit quiet on ADN too. Could it be, now that he's a salary earner, that he's on (gasp! shock! horror!) vacation?

Guys, I think the target here is Teradata, and the stage type is the bulk loader. So column names are needed, because they are used to create the control script. That's why I suggested that the single-table (and therefore updatable) view might work. (Can you bulk load a view? Never tried.) Another t...

Here's a stored procedure that creates an entry in the Batches table (one of the tables used for overall control of DataStage processing). At the end another stored procedure (uspBatchesSelectBatch) is executed to return the row that was just created. CREATE PROCEDURE uspBatchesCreate @BusinessUnit ...

Not even all employees of Ascential have access to this information. Ascential's take is that you don't need to know - it works OK and that's all you need to know. Plus, they reserve the right to change it at any version. And it has been changed over time. If you are a first-line support person, the...

Nothing in DataStage. Can you create views with the "correct" column names for the three errant tables in Teradata, and access the views rather than the tables from DataStage?

Welcome aboard! :D
Thanks for posting the solution. Hopefully it will help someone else in the future.

Is the original question something like how to drop a column from a database table? In this case, the knowledge is outside of DataStage or even ETL, and will depend on the specific database under consideration (in particular whether that database supports a DROP COLUMN or equivalent option with its ...

What do you mean by "external" hashed file? You could certainly use a SETFILE, thereby gaining a VOC pointer (without which none of Kim's suggestions will work). Note, too, that Kim has used the Pick variant of the SELECT statement, and has not encapsulated it in a DSExecute call (or EXECUTE or PERF...

Why do you want to know? VOC, in case you were unaware, is the vocabulary for the DataStage project. Not only the file pointers, but also every word or token (including references to routines) needed by DataStage are stored in the vocabulary. What you have asked - to be able to see the contents - is...