DSXchange

resolved?

Tell me your job design. How are you handling multiple files?

SURA wrote:To achieve that , if i pass #$DB_PWD# in the default value of LND_PWD will it help me to get the value from $DB_PWD (Admin)??

Don't use #. Just pass $DB_PWD in the default value. You are all set.

pivot_column = 1 for file 1
pivot_column = 2 for file 2
pivot_column = 3 for file 3..

will this resolve your problem?

Ok... Chandra told and i am going to stick to this forever

"Reset if required then run" is better than "run". Ray has given you the reason as well. If a job aborts, it will be reset & run automatically during the next batch run. Got it? Otherwise, it is a trouble!!

Deepa, Your took a complicated route now.. @INROWNUM is the incoming ROW number from each parition. That's tricky. Your earlier approach was simple for your requirements. Just assign 1 to pivot column and process in transformer stage. When done, sort by your new (concatenated) column in desc order a...

yeah.. OP will find that "another" stage to use. She will refer the IBM doc for that stage & that's when she will ensure data is partitioned Her case is a simple one!!

Remove unwanted rows by adding another stage

It's not working...

then tell us what's happening or what error you are facing? did you check the director log to see what's getting passed actually?

That makes sense. Now, try using stage variables to achieve the same. In transformer method, you don't need count of records. All you need is to find when a group changes. Your case is pretty simple as the group does not change at all. i.e. always "1"

Did you install DataStage clients in your machine? If yes, you will see "Documentation" by going to Start -> Programs -> IBM Information Server... ;) You will see "oceanic" documents & may be 2012 will not be enough to complete reading, but you can just look for what you want...

Real? dsadm is the DataStage administrator account but that should not let them show magic like this ;) I guess when the delete session hung, the session got locked. May be dsadm released the lock and save/saveAs buttons were enabled. Again it's my guess only... "job can't be found" error ...

Deepa, Your sample (input/output) data does not look like vertical pivot scenario. Do you have a group by column or not?

You can achieve vertical pivot in transformer (using stage variables to determine group change & concatenate columns).

Read IBM doc for more clarity.

search this forum and you will see n number of posts.

No Chandra. To ensure, try a simple sample job. That's the best way to learn

SURA wrote:Is it documented any where?DS User

I don't know.

If your password is encrypted type, just make sure that you enter exactly what you need. e.g. $VALUE, $ORA_PWD etc