DSXchange

That's procedural code, not sql. Write a proc and then execute it in the ODBC or STP stage.

Short answer: no. You can look in the log of the job at the first log entry and it will note the Job Controlling job: DSJobController = Sequence2 The last message it will post will be one that again notes that the job was run under the control of another job: (Sequence2) <- job2: Job under c...

Sorry, but you are going to need to rephrase this. I for one have no idea what your issue is. Are you saying this 'flat file' needs to be used as a reference lookup and hence your 'I tried to do it with hash file' comment? If so, please post an example of what's in this file and how you need to use ...

My recollection is that there was some software we had on our PCs to allow for multiple versions to exist, and I am guessing there were some similar tweaks necessary on dev to allow for 2 server versions to coexist. That would be the MCM or Multiple-Client Manager. And there is a pdf in your Docs d...

No, read the reply again and check the help for the 'Null String' column option in the Sequential File stage.

Define 'modified'.

The Manager will show you "Date/Time Modified" but that doesn't necessarily mean anything changed.

No, it will just be informational. Here, one from my logs:

Huh... guess there's an echo in here here here.

But Count can, it seems... yes?

Basically answered in your other thread. Yes, if your data is problematic then you could need to validate / cleanse each and every field.

They are two different things. If you want spaces to be NULL in your target, explicitly convert them in your job.

Ray - I'm not sure that's all true. We seem to have had postings lately where SQL that "works fine" in their favorite tool cannot be handled by the ODBC drivers. Barbara's and Mr IfThenElse for example. And I know for a fact that the ODBC stage doesn't enforce the column binding like other stages do...

Maybe - but when you have a question, you should start your own thread. If you think it may be related to another issue, link to the other thread, don't hijack the other person's. Please.

Again, what database are we discussing here? Maybe I'm wrong, but that seems to me like a critical piece of information...

If your MV is set to refresh on every commit, then don't issue intermediate commits. Leave the Transaction Size at its default of zero, which means only commit once at the end. Doesn't sound like making it a 'manual' step is an option for you. You must commit the changes in your job which will then ...