DSXchange

jiegao wrote:I seperate the job into 2 jobs...

- check your SELECT on the input to see if there is a limit
- check your transform constraints to see if you might have set a limit

can you 'touch' or 'mkdir' using your dsadm id in the project directory?

I'm no expert, but the message is pretty clear; you are trying to fit more data into a field than it can contain. The full error message will also contain the name of the column in question.

As far as I know "cd <dir> | ls" won't cause an error, but won't work either. You should change your code to read

Hello Bandish - that sounds like it might work, but in this case we have a potential of millions of rows per hour, so doing a sort is not an option for performance and resource reasons.

Could you post a specific example of a string and the StringToDecimal call? If you got that working correctly, perhaps the rest would fall into place as well.

DSRajesh - so many people have tried to help in this thread, and opened up a number of questions which you haven't answered or addressed, so there is little incentive for anyone to do more on the problem until you've done a bit of work on it as well. You now have a case where a write to sequential g...

How about StringToDecimal() instead of DecimalToDecimal() for CHAR fields?

I agree with Ray - get the IBM rep to convince you otherwise. Even better, write a simple job that outputs to a dataset (keep it simple with just one node) and then make a copy of the generated file, whose location you can discover with the data management tool. Then re-run the same job with the sam...

In the interests of this forum, could you post your solution so that in the future, anyone else having your problem and using the search mechanism can find the solution.

You might need to go into DS.TOOLS and reindex your files; or perhaps just a refresh in the director might work to make the jobs visible. Also, try going in to the Manager to view the jobs.

Do you have a peek stage in your job?

Ummm, the &PH& contents do not affect job runs unless you have a serious underlying disk issue on the system, since a new file is generated each run. It sounds like the ancillary information for the job is corrupt and gateleys suggestion will work. If not, there might be more problems in the...

Stay with VarChar, these characters are not really `special'. NLS is only needed if you need to manipulate non-latin characters, not if you only wish to move them around. What are your Oracle character set settings, it is possible that the conversion is happening there.

From the DataStage command line (either through uvsh or from the Administrator) enter "LIST DS_JOBS WITH JOBNO = <number> @ID".