DSXchange

narasimha

All I can say is ooooo .........

narasimha

How would you want the in coming null to look in the sequential file. Based on your need change the transformation to reflect for individual columns or at a global level

IF ISNULL(DSLink.xxxx) THEN 'How you want it look in your sequential file' ELSE DSLink.xxxx.

narasimha

A search on "Output file full" gave me 22 results.
The majority of them point to disk space filling up.

narasimha

Looks like IBM has not documented/published the migration path from Version 8.0.1 and later to Version 8.5 Below is the installation instruction provided in the "readme-migration.txt" To migrate from IBM InfoSphere Information Server, Version 8.0.1 and later versions, see http://www.ibm.co...

narasimha

Can you provide the exact, complete error message. This can give more insight to what is happening.

narasimha

The default behavior while executing sql scripts is - To continue and take no action when a SQL error occurs.
In the beginning of your sql file put the below line to exit the script with failure.

Code: Select all

WHENEVER SQLERROR EXIT FAILURE;

HTH

narasimha

Another way of doing this is to run sftp -b (Batch mode)
In the batch file you can write a get/mget followed by a rm

narasimha

Is this a one-time message or do you get this message consistently?

narasimha

The below should work for you. Note the semi-colons.

Code: Select all

BEGIN DBMS_STATS.gather_table_stats(XXXXXXX); END;

narasimha

Try using CALL instead of EXEC as below (without the semi-colon in the end)

CALL DBMS_STATS.gather_table_stats(XXXXXXX)

narasimha

I had the same issue - The solution is to introduce a new Environment Variable - DS_NO_FF/String/Variable/1

narasimha

I was thinking if there could be anything introduced in the "Filter Command" of the sequential file stage or something on that lines.
Looks like there is no option, but to invoke the job as a different user.

Thanks again

narasimha

I would want to implement this in existing server and parallel jobs.
Probably I should have been more clear/specific with my design, it is using sequential files (Majority of them are Server Jobs)

Source_Sequential_Stage -> Transformer -> Target_Sequential_File

narasimha

We have a secure location on our DataStage Server where I use the below approach to work with a file. sudo -u <user_thats_owns_the_location> touch <file_name> I can do it directly on the OS, but I would want to write to this location using a simple job. Source_File -> Transformer -> Target_File Any ...

narasimha

If only I were an "IBM Information Champion"

DSXchange

Search found 1215 matches

create a file using sudo