DSXchange

Interesting. A couple of points... Does it work if you add a semi-colon (or whatever SQL Server uses) statement terminator after each piece of dml? I've never had a select '*' work out well and you're just asking for trouble selecting something that technically can change into a fixed number of colu...

Automatic? No. You'll need to either take the job parameter route for use downstream in common code or shared containers or simply code the tablename in any other sql you use.

You could create two jobs - one to fetch the values and pass them as parameters to the second job to do the creation. Use UtilityRunJob() in the first job to run the second job once per record read.

Sure, parameters can be used pretty much anywhere that text can be entered, including for the schema and/or tablename. And you could use that same job parameter in any of the before/after sql wherever you needed it. We kept the schema ('owner') and tablename as separate parameters and dictated that ...

Anyone could be monitoring the job and causing this issue, not just you. 'Cleanup Resources' from the Director will at the very least show you who and maybe also be used to solve the problem, depending on the permissions involved.

When you say "on" a sequential file, do you actually mean *in* a sequential file? You won't be able to do pure "string manipulation" as you can't just rearrange the original value to get the desired result. So, hack off the time and then use the normal "date conversion funct...

If you want to apply the query in the source itself, use the condition below - to_char(your_date_column, 'YYYYMMDD') BETWEEN '20090306' AND '20090312' Wait... you can use the between like that after the to_char() to cause it to generate multiple values? In all my born days I've never seen the like ...

Wow... never knew I was privy to the complete list.

They are purposefully not published.

Since for whatever reason you don't seem to have "Insert then Update" as an option, then yes - go manual and include only the insert dml.

As noted, either will work but "Insert then Update" would be the smarter choice.

It's 0x20 (not 20x0) for a space and you can override the environment variable in just this job, no need to change your project-level default.

Sure, but the question is in what order are the operations performed? Either way will work as desired - provided the updates are filtered out - but best to perform the insert attempt first so that you don't burn two database operations for every record. I don't have any kind of access right now, do ...

I'm not sure you need to set both as I believe one overrides the other but yes - zero should mean only on successful completion. I'm unsure what the defaults are when those are not set, however... perhaps leaving them unset means the same thing?

1) Parameters, yes - wildcards, no.