DSXchange

8 dot what?

wahi80 wrote:Currently we are planning to move 800 plus server jobs from Datastage Server to SSIS.

Out of curiosity, what is driving this?

Sounds like a bug to me, perhaps you should see if a patch is available for the issue.

I have no idea, was just trying to get some clarity around what you've tried so far. At this point I would suggest you involve your official support provider.

DataStage will perform an implicit conversion from a string to a number and vice-versa. I don't believe this to be true in a Server job, that's more a feature of Parallel when moving data from one datatype to another. From what I recall, Server will happily move characters into a number field witho...

It just needs a VOC record, so you'd have to create it for a pathed hashed file otherwise it already has one. And then it can be accessed using the UV stage via 'localuv' from what I recall.

Which of the five scenarios did you determine fit your case? Exactly what software did you install?

Sorry, don't have a good way to merge duplicate posts so here we go:

prasson_ibm wrote:If your table is defined with UPI then how come table can have duplicate records.

Rather than ask in a four year old post let's get you into one of your own. Clarify your source - is it a string or is it already a numeric field? Actually, let's clarify both.

Another option: Write it out with X keys to a normal hashed file. Lookup up against it with <X keys using the UV stage. Enable the 'multi-row return' property.

That may not be true if you have a 'Server only' license.

Out of curiosity, what exact version are you running? And since you asked, for the record there's no missing setting or configuration for this.

Wouldn't help, I have no DataStage access so couldn't do much with it. Working on my fourth year using another ETL tool... I just do this for the fun.

Count the records. Typically 0 would mean empty but if you know there is a header record then 1 would equal empty. Easier on UNIX than Windows, however.

Of course it is, but that's not what you asked. An output file will always be created by a job. What you do afterwards is up to you - check for empty and delete it if you like after job.