DSXchange

Wait till the job finishes.

You don't need the link row count while the job is running - if you do, you can insert a Transformer stage and invoke a system variable (@INROWNUM or @OUTROWNUM).

You can create one or more values files for the Parameter Set. Write values into the values file before invoking the job or sequence that consumes the Parameter Set values. When you run the job, specify one of these file names as the "value" of the Parameter Set, rather than (as pre-define...

No.

Exception handler doesn't do anything. Its output is triggered if an activity in the sequence fails. And that's all it does.

Beware that, over time, you may need to download an updated version of the Update Installer itself. The documentation will tell you what version is needed. You can check the version by running it with the -version command line option.

Use the looping construct in the version 8.5 Transformer stage. It's an ideal usage for this functionality.

It can not "run fine in the other environment". A passive stage in a parallel job, as noted, can have at most one non-reject output or at most one input, and can not have (non-reject) input and output.

Yes, you can loop through all the jobs using the -jobinfo (and maybe some of the log interrogation options). That information will not, however, be available by inspection of the sequence's properties or log.

What's the source database? Display functions differ between databases.

In a Transformer stage in DataStage it's easy - just use Field() functions specifying ":" as the delimiter character.

You're on the right track thinking about RCP. There are examples here; Search is your friend.

I believe the limit on Solaris generally to be 32K subdirectories per directory.

Craig recalls correctly. DataStage does not recognize user DSNs - only System or File DSNs.

Are you using a Sequential File stage or a File Set stage to read this File Set? Are you specifying the Schema File property correctly?

It depends to some extent upon what flavour of UNIX you are on, and on the limit (if any) on the number of subdirectories that can be created in a directory. 1000 jobs is probably OK - you can go up to somewhere near 4000. However, performance problems start kicking in in areas like job startup and ...

You don't have any hash files. As I have noted myriad times in the past, the correct term is hashed file. There are at least four ways to delete hashed files, but only one of them is correct, and will depend upon how the hashed file was created. I'm sure that there's a post somewhere on DSXchange th...

Workaround is to set Array Size property to 1.

Even better would be to force generated SQL to be re-generated, for example by switching from generated = Yes to generated = No and back again.