DSXchange

What stage are you talking about? If the stage is still supported in version 9.1 (and most are) then you won't need to re-compile. If you want to add new stage types then, obviously, you are changing the job design and would need to compile after the change has been wrought. None of this discussion ...

Hence my second question.

Although the GRANT command is executed in a project, it has universal scope. It can also be executed from the UV account.

Note also that, by default, local user and domain\user are treated differently by the GRANT command. The system can be configured so that they are treated identically.

You could probably use Transformer looping to process this. Initialize a stage variable to the number of comma-delimited elements in the parameter value. The loop exit condition compares @ITERATION system variable with this value. Use the Field() function to extract each value and then perform the t...

I think ssnegi has missed the vital point that the list will be dynamic.

Hashed files (and please note it's not "hash" file) destructively overwrite records if the key (as identified in metadata) is the same.

Do you have duplicate keys coming from source?

Are there any other constraints in your job design that might limit the number of rows to 20?

You can't, because it's not.

Done properly, DataStage merely stores what the metadata retrieval mechanism for the database reports.

As Arnd suggested, you can write a script (using dsjob calls) or a job to traverse all the logs of all the jobs in the project. You can also do this with a routine (creating a select list of job names). It would make sense only to traverse the most recent execution of each job from its log - use the...

You will be exporting your projects anyway. When you import them into the new system DataStage metadata will be taken care of automatically.

1. The SQL might have an ENCRYPT function (or you could write one yourself). 2. You could export from source and pipe through an extraction utility before sending to DataStage. 3. You could perform the encryption in DataStage as the extracted data stream in (for example using External Filter stage t...

I concur entirely with Stuart. IA has no inherent concept of meaning; it gives one frequency distributions of values, formats, etc., but it is not the tool even for identifying similar spelling (though sorting by value in the analysis results may help somewhat in making these visible). You (and/or a...

Repartitioning theoretically requires to move rows to other nodes (partitions). On MPP/cluster/grid environments this may mean moving rows over the network (TCP/IP), which degrades performance somewhat. However, on an SMP ("share everything") environment, one copy of the data are kept in s...

Can you compile other jobs with Transformer stages? This error often suggests that you don't have a supported C++ compiler or that the license for the same has expired.

Basically a Column Generator stage adds columns to an input stream. It cannot generate rows out of thin air. For that you need a Row Generator stage.

Use a fork join design, with one side of the fork counting rows (grouped by keys). Filter the result so that only keys with a count of 1 are transferred to output.