DSXchange

DataStage itself offers nothing. Some posters on DSXchange have created generic reporting off the local repository which may provide some guidance. In your case you'd be specifically interested in a query against the DS_JOBOBJECTS table.

Presumably you can report this information from wherever it's held in the metadata repository. But you're on your own as far as determining where that might be is concerned; the structure of the metadata repository is deliberately undocumented by the vendor.

You could use a routine to trigger ExecSH (via a Call statement) but why bother? You get the same effect, with one less layer of software, but calling the DSExecute routine from your routine. (ExecSH is simply a wrapper to DSExecute with the Shell argument set to "SH".)

I did not answer that question, so "did you mean?" is unfair. The description in the Parallel Job Developer's Guide is quite clear on this question.

Why ask me? I know next to nothing about Teradata. Maybe there's some setting you can make in Teradata to prevent the generation of informational messages.

You can't (in DataStage), because everyone is logging in as the same user. Login to DataStage uses the credentialled user, not the original user name.

Now, if everyone was credentialled to a different user ID, that would be a different matter...

I think you'll find that the Timestamp parameter type does not support fractional seconds.

That indicates that &PH& is Type 19 and you should not be having any file names in it truncated to 14 characters.

StringToDate() function with extended properties on the input set to Unicode.

The values are whatever you set them to be. Default values can be inspected in the stage properties. Copy = input2 is the same as input1 Insert = row exists on input1 but does not exist on input2 Delete = row does not exist on input1 but does exist in input2 Edit = row exists on both inputs (based o...

This is some redefinition of the term "working fine".

Windows XP is not a supported operating system for any but the client tier.

What you "know" isn't true. DataStage does not have a built-in calendar. This is a singularly easy task to accomplish. You need parameters of start date and end date. Use a before-job subroutine to write all of those dates into a file, and the job itself processes the dates. (An after-job ...

Conversion to 64-bit would certainly prevent these corruptions.

I'd also be asking why the files are growing so large in the first place - is there any way to reduce the number of logged entries?

Try it (the Connector). Let us know.

I do not believe that is possible in a parallel job. Take a look at what link variables are available; in server Transformer stage the link variable LASTERR would give the error text. I don't believe there's an equivalent in parallel jobs, but it couldn't hurt to investigate.