DSXchange

No-one meant to imply any kind of VOC action was required. The assumption has been you are using a Values File to hold the default values and you want to overwrite that text file. Is that not what your goal here has been?

You must have missed the exactly part. What stage - transformer? And post your exact derivation so we can see what you are doing.

Just to be pedantic, how exactly are you populating the ACCOL_IND field? Stage, derivation, etc.

Not in this thread, this one is about ODBC for DB2 / AS400 tables. You already have another thread on that very subject, so ask any questions you have on that technique over there.

Not directly you don't - that could turn out to be about the slowest way to accomplish that task. Look into building a hashed file and using a UV stage over it instead if performance becomes an issue.

So, that part should be fine, so now we just need the gory details of what it is you are doing and seeing. How is this mysterious column defined, both in the job and in the target - data type, size/precision, nullability, etc? What error messages are you seeing in the log? Post them here unedited if...

Why ODBC? Stick with the native stages for DB2 access, fall back on ODBC for things you don't have native drivers for. For ODBC, there's an entire chapter on them in the Server Job Developer's Guide pdf, start there.

Show us the full name of the environment variable you set and what exactly it was you set it to.

A Date in DataStage isn't the same as a DATE in Oracle. Both are internal formats specific to their source. Hence the need for passing the value in as a string from an external source and using Oracle's TO_DATE() function to do the conversion during the load process.

Issues, plural? Not without a great deal more detail about them. 'Getting NULL values' isn't necessarily an 'issue' unless you explain to us why it is, what transformations you are doing, stages between the DRS stages and what errors you are seeing. Also please don't make us guess what 'first value...

What exactly would a Decimal(120) be?

I don't know. I would think there should be some way to do that, but I don't have any kind of DataStage access to test things out. Yes, DataStage is notorious for stripping quotes from job parameters but I don't recall anyone who couldn't find some kind of escape mechanism that works... eventually. ...

So, what happens when you substring them? Is it causing a problem or are you just worried it might cause a problem? Extensive testing should answer your question.

And? What have you done to setup the ODBC drivers? What driver are you using? People would need details to provide any help.

Only the UV or ODBC stages support 'multi-row result sets'. For the former, load an account-based hashed file with enough keys to ensure uniqueness or for the latter use the text drivers.