DSXchange

Let us know if your topic has been resolved or if you still need some help with it...

Why a loop? Why not just leverage the File Pattern capability of the Sequential File stage and use a wildcard? Add the option to include the filename as a column in the output.

First, tell us exactly what you need to do with these files.

I don't recall Server jobs supporting milliseconds in timestamps. Just treat the field as a string and substring off the date or use Field() to get it.

Well, there you go. Stop trying, it's not a supported operating system.

Post a count of the number of files in the &PH& (Phantom) directory in each of your projects. And sure, logs can be saved but it's up to you to programmatically transfer the contents to less transient storage.

Not much help without a Premium account but Ray posted a complete OpenSequentialFile routine back in the day to illustrate all of the ins and outs of dealing with files in a routine.

Command line via dsjob -jobinfo or from a routine using DSGetJobInfo() can get you elapsed time and then you can compare the results.

Yes, that could work - a BASIC routine in all jobs to append the information to a flat file and then a one-time load of that to STAT post cycle.

So... another attempt at this topic.

If you think it would help, post the copy book.

Best to create a generic job that captures stats from any job and updates your STAT target. Also look into downloading ETLStats from Kim Duke's website.

Then you'll need to contact your official support provider.

Start by not creating two.

I lock one and we continue the discussion in the other... your choice on which one gets locked.

qt_ky wrote:I think you have another post on this same topic.

viewtopic.php?p=417699

Probably from people with a Server background as it only exists in the Parallel product. In any case, either way works just fine.