DSXchange

From what I recall those timestamps can be updated when the hashed file is read / accessed, not just when content changes happen.

Same as before I would imagine.

1) I don't recall but the docs say "Runs the specified job and returns statistics from the job run" and "The statistics are returned in an array". I would start by looking at what the 'statistics' show when the called job fails.

2) Have you looked into a looping Sequence job?

Search. There are quite a number of conversations here on that topic, one here for example. Check the posted IBM pdf link there, as noted it contains an example of usage on page 155.

:evil: I am seriously considering deleting posts that literally contain nothing but input and output examples. We must put an end to this madness. I cannot say this enough - Spell out your requirements in words Tell us what you've tried Tell us what issues you've had Do not simply drop off your dirt...

Has anything DB2 related ever worked in Production? I'm guessing this is your first time attempting this and the answer is 'no'. If that's the case, then it would seem that whatever steps were done in Development to configure DataStage to be able to access DB2 - those steps were not done in Producti...

Is it better to re-write the jobs from scratch? Forgot about this question - no. Goodness no. Unless... perhaps... you have a very small number of jobs and you focus more on 'best practices' because you currently live in the Wild West and you need a lot of that refactoring you mentioned. :wink: If ...

Just to round this out, files (of any kind) are always created before being written to. As noted, you can delete them after job or... just not worry about it.

So... what's the goal here, the end game? Trying to dynamically generate the SQL needed to only update the column names that have changed? If so then it seems to me you'd need to spool that out to a file and then have something "after-job" process that in the database. Otherwise, if you ar...

You really need to consider finding it again.

There's nothing to 'define/change' about it as @DATE returns the current date in internal format. It's all about how you convert it to an external format and I've already pointed out your issue there.

You did not specify a date separator in the OConv function, which is not the same as telling it you are wanting a space. Explicitly declare it so you can explicitly remove it: If RUNDATE='1000-01-01' then Ereplace (oconv(@DATE-1,"D/YMD[4,2,2]"),"/","") else Ereplace(RUN...

Explain your rules for determining when something is a "Salary" and something is an "Alias Salary" so people can stop guessing. After that, it just seems to me that it would just be if-then-else derivations using those rules.

Not exactly what I was expecting to see when opening this - what does this have to do with masking? :? And do you only want a single output row? Or do you want to repeat the column names for every row as some kind of masking attempt? Regardless, they are generally known ahead of time so can be hard-...

Not sure what kind of help you need. Did you understand that when you said "the source records are updating" you are basically telling us you have no problem? Perhaps you did not understand the question - he was asking if between the 8.5 and 9.1 job runs more updates had been done and you ...