DSXchange

I think in your case you are getting Nulls for all the fields because of the wrong comparison used in the Derivation. If the Empno Field is of Datatype "Integer" then the Transformer derivation should be compared to "0" if dslnk.left.empno=0 then dslnk.right.empno else dslnk.left...

Please convert it either to Date/ Timestamp depending on Target Data type. A column can have a single data type.

We load this in a sequence with Checkpointing set. Job 1 is loaded only once per load, hence the data extraction happens only once from the source. Then Job 2 will run in Loop. If for example DS1, DS2, DS3 and DS4 were created and if DS1 and DS2 were loaded and job failes during DS3 Load then we hav...

Hi Ray, Will this be a possible solution to eliminate the issue of re insertion / updation even a job loading huge volume fails. Job 1: 1. Partition the source data using round robin partitioning method into different datasets. Perferably to the number of nodes available to make even division. Job 2...

Hi Suja.somu,

Thats a message been given by Datastage as the Table definitions were not available during the design time, they were created at the run time as the RCP is been enabled. Hence you are seeing tha message. Its not a warning and its not going to impact your job.

Regards,
Harish

Please try a Remove duplicates after the Funnel stage based on the key colums to remove the duplicates.

Hi, Please refer the below link for Datastage and Oracle Datatype considerations. http://publib.boulder.ibm.com/infocenter/iisinfsv/v8r5/index.jsp?topic=%2Fcom.ibm.swg.im.iis.ds.parjob.dev.doc%2Ftopics%2Flimitationsdatatypes.html Please try using Decimal(22,10) in your transformer and let us know if...

Hi, A lookup stage will get the reference data set to memory before it stats matching the data. But while considering the data for lookup it will only select the first matching record from the reference dataset and drops all the other based on the Keys. Hence is the warning message "Ignoring du...

Hi Prabhu,

Please try this function.
IsValid('Date', StringToDate(<date Field>,"%yyyy%mm%dd") )=1 then its a valid date else its an invalid date.

Ray had put some comments, but i dont have a premium account i am not able to see it

please try this logic also


if( decimaltostring( input_decimal_field,"suppress_zero")=0) then '' else input_decimal_field

Yeah Ray, my bad missed that point. You need to hash partition the data so that similar key values falls in same Partition and above logic works.

Please check if the comparision is char to varchar as char padds extra spaces or pad characters which can fail the condition checks....

Create a script to create schema file which will import the structure of the files to both the links and enable rcp..

This can be achieved using stage variable's too..sort the data using your 2 fields svF1 is initially set to blank. If svF1=field 1 value then set svF1 to field1. So if the value changes then pass the value of firld2 to output....

Thank u craig