DSXchange

venkatesh

see if this link helps...viewtopic.php?t=125768

cheers

what is the datatype for column NAME?
when you do a view data on the input file... does it display the ' correctly?

what is the exact error message that you get?

why cant you write into two streams.... anything that passes goes into your actual output while the failure link is dealt with appropriately rather than aborting the job...rather than writing unix scripts etc

Pradeep You never mentioned about it being a file or perhaps i missed it, if you were expecting to read them as column b, c, d, e ...so on.... with 'a' being your key column i mean would it be like a, b, c, null, null, null, null a, b, c, d, e, null, null a, b, c, null, null, null, null a, b, c, d, ...

does it work any better using an active stage in between the two read and write processes ...

Anthony Note that this would even strip off the potential white spaces between two words eg: address fields, if you have any. Using a trim(string) or trim(string, ' ', T) should strip any trailing white spaces... However can u see the output or the input on a database ? try to see the dump of that f...

Hi All, We have been using an ODBC stage to extract data from a remote DB2 database. Due to the magnitude of data we are receiving we feel that using a DB2 plugin would fasten things for us. However we would need some clarification in understanding if a DB2 client tool should be a licensed product o...

your not losing data because of that.... could you please post the complete log...

Deepak

all you have to do is select Enable Job administration in director option in admin.

that would let any user clean up resources or clear status files on jobs.

cheers
VM

Ankita


while you wait for the job to finish did u try querying your final table to see if it has the data that you were expecting?

does your job hang even when you run it without passing any data through it??

what kind of a lookup is it that you do into the lookup table??

cheers
VM

you could change the netezza stage execution properties to sequential. did you try that?

does your log look clean?? do any of your warning suggest something is wrong?

i would imagine that you might have already double checked your datatypes and nullable properties..

i tried replicating that and looks alright to me... all i see is nulls.

what is your lookup? table, dataset, file??

Did you double check if those failures are valid. ( the key doesnt find a match)

So you are saying whenever a lookup fails instead of a null it brings a value like 0001-01-01?

hello

can you create a local parameter defaulted to $PROJDEF that you pass into the transformer rather than getting the env value itself?

This might help if you wanna run that job alone with a different set of values unlike the ones in your environment without editing your code.

cheers

Puli

Are you saying datastage is defaulting the null to that 0001-0-01?

Is that how it exists in your lookup source?? if its consistently giving you a value like 0001-0-01 can you not use a transformer to convert that into a default value which is more acceptable?

cheers
VM