DSXchange

What did your Search reveal?

There is no 7.6 version. The final 7 version is 7.5.3.

Try a pipeline of commands then. What you're currently doing is trying to give the pathname of a directory as a command, so it's no wonder you're not allowed to execute. No-one can execute a directory. It's simply not permitted (as well as being meaningless). Try this pair of commands. The &&...

Me? No idea whatsoever!

YOU are in control. If you need everything to run in sequential mode, you can specify this in a number of ways. The default is otherwise, so you will need to take some action.

Welcome aboard.

The best person to tell you (since you don't appear to understand the COBOL terminology) is the person who wrote the specification.

You don't run a sequencer, which is a component for bringing logic together in a Job Sequence. The Job Sequence runs in its own process. Every job it invokes through a job activity starts one or more additional processes.

The target table already has rows in it. @OUTROWNUM always starts from 1. You need to determine the current maximum key value before your job processes any rows, perhaps loading it into a hashed file with a constant key, and add this value to @OUTROWNUM.

Have you tried this? It can't do any harm to try (if it fails, it will "fail safe").

If both sources are sequential file you can set the DRS to use ODBC and have a DSN set up using the text file driver. But a far more sensible (and fast) approach would be to read the file into a hashed file and do the lookups and comparisons in a Transformer stage.

Well I can not see where you mentioned it anywhere in this thread.

I suspect that the script picks up the remainder of the command line and executes it. To mind this is one unnecessarly level of abstraction, and the touch command itself may be executed directly. However, there may be in-house rules at that site, or the script may do something else first, such as sw...

It's not a bug. And there is no 7.6 version. It's an alert. Only Entire partitioning can make use of shared memory for exchange of keys, or even needs to. All other partitioning algorithms don't need to share keys. That you have chosen not to use Entire means that the use of shared memory for exchan...

If source is a sequential file, even though it has the date field in it, you have three options: to read through the file if you're using a Sequential File stage: you can filter the records downstream of that to use a filter command in the Sequential File stage to select based on that date field (yo...

Both inputs identically partitioned using a key-based partitioning algorithm (hash or modulus) on the key column(s) used for the join.