DSXchange

Please prove your assertion that they don't all start at the same time. Have player processes dump their PIDs into the log at startup time and compare the start times in the log.

No surprise really. Just look at the "Microsoft time" estimates when Windows is copying large files.

UniVerse databases can be accessed via ODBC.

The server engine was originally built on UniVerse, so it was easy to provide a "native" interface. The parallel engine was built from scratch by Torrent Systems and their predecessors, and they knew nothing about UniVerse or UniData.

BASIC Transformer has in-built Ereplace() function.

1. Not possible in version 7.
2. It's totally different. A different mindset is required. Read Parallel Job Developer's Guide as well as the connectivity guide for any database to which you want to connect.

Be aware that unlimited warnings can cause your project to become corrupted, or the log to become corrupted, if too many entries are logged.

Prefer to write jobs that don't throw warnings.

So, to clarify, you want to know the next (or, perhaps, current) value without actually incrementing the state file?
This will, of course, be different on different nodes.

To clarify Craig's answer, there are more columns in the Columns grid. Scroll the Columns grid to the right to find the "Contains terminators" property.

Note that, for this to work properly, character strings containing terminators must be quoted.

Write jobs that never abort.

Do you have the DataStage deadlock daemon (dsdlockd) running?

It's a mystery. You're getting Type1 behaviour but without the .Type1 indicator file. Have you actually resized &PH& to Type19?

Jobs are never executable files. Can you please clarify what kind of object you are discussing?

Server jobs have their "compiled" information in RT_CONFIGxxx table, while parallel jobs have their osh scripts in RT_SCxxx subdirectory.

I wonder what the resolution was?

Link should be OK now.

Search DSXchange for the pxEreplace function, which will handle your requirement even if there are spaces and commas in your data.

Otherwise do what SURA was attempting to suggest - Trim() to get rid of the spaces and Convert() to change the commas to semi-colons.