DSXchange

ray.wurlod

SELECT SCH,CLAS,STDU,COUNT(*) T0FLAG
FROM #schema#.tablename
WHERE FLAG = 0
GROUP BY SCH,CLAS,STDU 
UNION
SELECT SCH,CLAS,STDU,COUNT(*) T1FLAG FROM #schema#.tablename
WHERE FLAG = 1
GROUP BY SCH,CLAS,STDU ;

ray.wurlod

We have one particularly useful pair of routines here that can log a message or a warning in the controller's job log. We use the former to capture gross elapsed timing of a job that is run in a loop to process data for different time periods. That way there's a nice adjacent summary in the job sequ...

ray.wurlod

Do it in extraction SQL with GROUP BY and COUNT and WHERE.

This way DataStage won't be required to process nearly as many rows and will therefore finish faster.

ray.wurlod

Once you have the controller name use DSAttach() to establish a handle to it. Don't forget to DSDetach() once you're done with it.

ray.wurlod

Probably not. The Information Server web console gives you overview of all DataStage projects (even if they're on multiple servers) but does not, as far as I am aware, give you the ability to invoke DataStage. I would be very surprised if it did, running inside a browser. But stranger things have ha...

ray.wurlod

Impossible to tell now that you've destroyed the evidence.

ray.wurlod

That is a deficiency in DataStage metadata management, in my opinion. The only way to reconnect the linkages is manually (or build a tool, which is the same thing) and that, in turn, requires a systematic methodology of preserving the linkage information before importing the new table definition.

ray.wurlod

With what character are the Char(18) fields padded?

ray.wurlod

Hint: your routine CalculateDate does not return 0.

ray.wurlod

Deliver the value of A on the link as a column called B and marked as Key.

Code: Select all

DELETE FROM table_name WHERE B = :1

ray.wurlod

I have not done this and don't propose to any time soon but, in theory, if all the Surrogate Key Generator stages use the same state file then "sharing" should be possible, and there ought to be no need to restrict execution to sequential mode. Try it in a small sample job and let us know the outcome.

ray.wurlod

How is Zeke detecting the job status?

Change -wait to -jobstatus then the exit status of dsjob should be the same as that of the job.

ray.wurlod

Check for terminators IN THE DATA causing this split. You may need to change your metadata to set the "contains terminators" for that field.

ray.wurlod

If you have this character AND a linefeed, then this character is the "final delimiter" and the linefeed is the record delimter ("UNIX style").

ray.wurlod

The same happens as described for temporary space. You just get more - or differently sized - temporary sort files on disk.

Search found 53125 matches