DSXchange

What exact 8.x version do you have and are you current on fixpacks / patches / whatever? And for grins, please confirm that when nothing is missing from the file you do not get any errors.

So... you are saying that sometimes the elements in red may not exist in the file? I don't recall that being a problem. What exactly do you mean by it is "not able to parse incomplete file"? What error(s) are you getting?

What is your runtime warning limit - any chance you could set it to 1?

ray.wurlod wrote:The solution is most easily wrought using a pair of stage variables

Seems to me that a single one would do the trick.

Sure, I've seen them too. But more interested in what our friend Chetan.C has to say on the subject. It's their table after all.

It doesn't need it to be sorted but it will appreciate it.

However, if you tell it that it is sorted you best not be wrong which is what your "out of sequence" error is saying.

"Affect" as in change the result? No. It would mostly just be a waste of time as the Aggregator would need to undo all that work and "sort" the data itself back as a string.

Probably, yes maybe. No clue if it would be faster / better / slower / worser or even if it would be all that different as there are too many variables at play. Honestly, the only way to properly answer the question would be to try both ways on your system with your data and see. And hopefully the v...

I'd be curious what the purpose of the table would be. You've captured them all in one table... and then what? How you would leverage it would drive the design, it would seem to me.

Sorry, but is there a question here? Was this supposed to be part of your other post on this option?

Typically that would be because the values do not in fact match. It could be as simple as extra spaces at the end of one value: trying to match "A" to "A " for example. I'd start with that.

The only way to "share" like that is to pass the value as a job parameter to all of the jobs. If this job is being run concurrently then it must be a multi-instance job and if that's the case you should be able to include the Invocation ID as part of the filename so there's no collision.

DB2

Sounds like you need a "pre-process" to delete the file when the day changes, then the job can simply overwrite each time.

As I said, pass it in as a job parameter. Use something in a Sequence job to read the file and craft the string that needs to be passed - a routine can be used or it may be as simple as an Execute Command stage that does a 'cat' and passes the result to the parameter. All depends on the structure of...