DSXchange

my doubt is in database are these columns ,LEVEL1,LEVEL2,LEVEL3. null or not null, if it is not null then you can use "Treat Warning as error message checkbox", but then your job will not run.

if they are null we can find some solution for your implementation.

i hope your date is in format yyyy-mm-dd, then in required target column give input as, hard coded values are just an example.I am using Oracle OCI stage and it works.

OCONV("2007-09-18","D-YMD[4,2,2]") : OCONV(00:00:00,"MTS")

if you are using basic code then following will help

ErrCode = DSSetJobLimit(JobHandle, LimitType, LimitValue)

go to Director ---> Job Run Options---> Limits --> in warning frame just increase the Abort after jobs(Warning values) from 50 to limit you want.

hello you can just do the following


ExecDos sqlplus -s oracle/oracle@BIODS.WORLD @#directory#\in\file.sql > #directory#\in\file_log.log

all your error will be caught in file_log.log

it works fine with decimal and varchar column lookup.

to implement your functionality you can use ExecDos (Before job routine) and give input values for moving a file.

ExecDos move #directory#\file1.txt #directory#\file2.txt

Please have a look on the input tag of sequential file stage, and check for input name should be different, i think in your case input name is same.

please write custom sql ( i don't have db2 stage but that is what we do in OCI (ORACLE) stage.

in oracle OCI stage you have input tab in which you will find "Treat warning as fata error" and with this in tranformer you use rejected link. you will get all error codes for oracle.

one hint is you have to use Stored_Procedure stage.

please search for forum you will definitely get a solution

if you are inserting records in OCI stage then you can "Treat warning as Error" put this check box (checked) you can get some message that will help you.

i have a similar case but it is for insert and update i have a values like Array Size : 1500 and Transaction Size: 50 which works fine for lakhs of records, you can try it out.

if your input file is a hash file with C1 and C2 as key then i think you get unique records like C1 C2 A 1 A 5 A 10 B 2 B 1 C 51 Now follow below steps. your job design will be like H/C file(hash input file) ----> transformer---> hash(H/C) file(output) 1. Create 2 stage variables like APPEND_C2_VALU...