DSXchange

I heard, DS repository is buit over 3rd party database like oracle in HAWK?
If that is the cant cant we 'hack' there too?

What if you view the file from unix using vi <filename> or
cat -e <filename>?
Are the dots comming up?
Was your requirement to convert space into to null charcters?

The query to create the view would be

Not only for past 2 years, but may be more or exponential, and that too after the IBMs Aqusition. I guess Just holding the number of user information in the static database wont affect the day to day performance. Atlest site can have more Gold customers using these numbers.

Does your version have Oracle9i stage?
Oracle 9i is teh compatable version for 10g. Advisable as well.
Do you extract any date field from the tables, was it in proper format?

It also depeds on where you database resides. If it is costly have carry huge number of rows from database server to DS server, it might be effecient to do the necessay aggregation steps from DS by either stroed procedure or using your SQL logic, and carry the necessay records for transformation to ...

In the transformer prior to the DB2 API stage, on the rejectred link, in the constraints, use LinkVariable --> Outputs --> (DB2Link) REJECTED.

Why cant you acheive this in user defined query?

I guess protedted project wont alow you to compile by normal user.
The other option is to update DS_JOBOBJECTS and update the READONLY flag as NRO. Which is what refered by Arnd as dangerous.

I am wondering, how it allowed you to write to a dataset in paralle. And not gave out all those errors that I got when I did the same.
Why cant you serialize the flow, unless requiered to write in parallel otherwise?

Look at the line numbers (above and below as well) in the generated osh script, if you have used any transformaion or datatype mismatch, it can be identified during this process.

Mark the output column of the aggregator as Nullable = yes. And handle_null in modify stage that been used.
Use decimal_from_decimal option for range limitation warning.

What if you run all the job serially? Will this error prompts again?

sb_akarmarkar wrote:Try with folder stage and aggregator stage...

Thanks,
Anupam

Do you mean writing the record id as file name for each set of group.

IPC would be a key through with the link level row buffering parameters (buffer size and timeout )can be altered or changed. (The value, if set in project level in row buffering)