DSXchange

Your samples indicate bad design for the extract source. There's no two ways about that. I often harp on such things. I "grew up" in a Cobol environment, where delimiters are avoided and fixed-length fields make reading and writing an exact science. For this, I would point to how some colu...

Sample: Loop X times to run FTP session, one at a time, and provide for restart from point in list where error happened. Generic job sequence which reads parameterized loop list name: 1) Loop list is indexed numerically, one (1) to last (X). 2) Execute stage finds maximum index, setting last loop va...

This is the right view. Macro-based variables always evaluate at runtime, regardless of previous values in a restart scenario. Any such value that you must retain on restarts needs to be saved and referenced where needed upon restart. How you do that will be complicated. Please post about your effor...

Craig, thanks for the praise. Sree, I suggest starting from the beginning. My FAQ assumes some understanding of mainframe systems, so I'll get very basic. The Cobol FD is not always 100% imported to DataStage. There are a couple of issues with it from v8.x that may be true in v9.x, and are still tru...

Your solution is one I hoped you would be able to use. Please remember that you were handed a file that is poorly designed at its source. The proper "correction" is for them to fix the design of the file. As you pointed out, that wasn't an option for you. There is standard practice in Cobo...

Sorry, I should have asked first which approach you are using. If you are going to parse the input record in the same job, you'll have to use Char for the one column, and on the Format tab set the under Decimal Allow all zeroes to yes. Tell me which approach you prefer, and I can be more specific. E...

Use binary type, and preserve the EBCDIC encoding in the format. That is best if you will then read the file with CFF.

From your example, I'm lead to assume that the first byte in each record will be different along this line: numeric 1 for header, alphanumeric "0" for details, character "T" for trailer. If that's true, then the suggestion in my EDIT will work. Sequential file -- one column of re...

That last part of splitting the files is the idea that I had but I also think it is where I am getting hung up. Could be an unsolvable problem. Are the three record types all of the same length? If that's true, you can (preferably on the mainframe) split them by record length. Do you have access to...

The first answer is negated by your "this is a shared file" comment: the layout is a flawed (actually, I'd call it defective) design, and you will simply not be able to use CFF for a full-file read. I suggest splitting the original file into three parts with internal consistency: one file ...

I've seen something that looks similar. I offer it for comparison. We invoke a generic script from our scheduler, which builds and sends the dsjob command line. We don't modify any command line parameters to rerun an aborted job, such as you show for -mode. We depend on the checkpoint to issue a res...

I'm wondering if Sequential File is the best stage for this. Have you tried File Connector?

I take your scenario a bit differently. What you describe is a human controlled system incident, sort of like the old days when the janitor tripped over the power cord and pulled it out of the wall socket. If you are initiating the actions as you describe, you are (and I write this not knowing why y...

That's like what I tell people when I'm teaching them a folk dance: "No! The other left foot!"

Not sure without more details about your job sequence design, but you might need to issue a RESET command before trying to rerun.

However, you could also enable checkpoints instead, and let DataStage handle reset and restart internally without needing to do it yourself with separate command lines.