DSXchange

Hmmm... I'm not sure you can have multiple admin users. The admin doesn't need to be 'dsadm' anymore, however, but still believe there can only be one and it is established when the product is installed. I'd have to check the docs. Perhaps someone else knows for certain. Curious why the perceived ne...

As noted, that would be because no solutions have been given. Contact your official Support provider, see what they say your options are. Your 'work-around' may very well be to do this outside of DataStage using Sybase command line tools via a script, for example. Of course, DataStage could run / mo...

Glad I could help, "vijaydev".

Depending on where you need this 'parameter', you could try having your before job script set an environment variable and then pick that value up in the Initial Value area of a stage variable. Emphasis on 'try' as I've never had a need to do this, but assume it could be made to work. If having it in...

And 'before' = 'inside'. Move it truly before, call the script in a Sequence job using an Execute Command stage and then pass the result to the required parameter in a downstream Job Activity stage.

I have checked. gdb and sed are installed and the directories where they reside are in the PATH. Just to be anal - they are in who's PATH? They would need to be in the path of the running job's environment, not you logging in as that same user checking from the command line. You can check the secon...

Wow, I think that's all one sentence. As noted in the linked thread, there's no good reason to have a Server job running continuously for any long period of time, let alone a year. Unless, perhaps, it is an RTI-enabled web service job and probably not even then. I have looping jobs that run every da...

That makes even less sense, what would one have to do with the other?

I'll have to play with it when I get a chance.

The Sybase OC stage does not support the Text datatype. Not sure what, if any, work-around exists. And unfortunately, there's no solution in that linked post either. At least not yet.

So, inside the job once for every record processed? Regardless...

You could 'abort the job' from inside your routine if you liked, or take the code sample I posted and run with it. In either case, abort by writing a Fatal log message.

Seems odd that you 'must include' an optional command line switch like /exec but it's been a long time since I've played with that. What happens when you leave it off?

Again - where exactly are you calling this batch file? And if you want to abort based on a stage variable, you can do that in another one. For example, something like this


Or something similar.

Wouldn't even have to be in the same schema, the same database would do. Heck, doesn't even need to be that.

... or you could specifically check for the 'F' versus 'P' and branch a Sequence job off to a Terminator stage when it sees 'F'.

Where are you running this 'bat file'? Typically, you'd just need to ensure the batch returned a zero return status when everything went well and a non-zero return status when there was a problem.