DSXchange

Does your database support auto-generated key values?

I do not believe that this is possible.

"All zeroes" is technically not a valid value for a decimal. There is a setting that allows DataStage to treat it as such, but I can't recall what it is at the moment. Hopefully someone else will contribute that before I get to check.

Maybe nothing is wrong with your job. One of its processes received an interrupt signal from somewhere. One possibility is that someone (or some code) issues a "stop job" instruction.

In a server job I'd imagine that the variable would be in an unassigned state. But you can check the generated Transformer stage code to verify this.

You can create deployment packages using Information Server Manager which has the istool command line interface. When you get version 8.5 you will be able to check in and check out these packages through an Eclipse plug-in for Subversion or via your own code using the provided API.

Welcome aboard. You could do this, but don't expect fantastic performance. Also, you have not indicated what you want to do with the output from these queries. Basically you need to construct a simple routine that invokes the shell to execute your commands, using the DSExecute() routine or the Perfo...

Because you've constructed a "here script" you need some way to end the nested command stack. It doesn't have to be EOF, you can use any string. The nested command stack begins with <<token and ends with token.

First develop an attitude to quality, such as ensuring you have spelled it correctly. You are nowhere near being ready to learn about pattern-action language. Learn the fundamentals of QualityStage first: investigation, standardization, matching and survivorship. Then learn about tuning those proces...

So you never bothered to inform us that you were constructing this command in a routine, a piece of information that was somewhat important.

It is not possible by any means other than overwriting the Data Set with only the rows that you want to preserve. A Data Set is not the same thing as a database table. Data Sets are intended for staging data between jobs; they are not intended to be manipulated by other means.

NextWeekdayFromDate() function.

Sort them and run them through a Head or Tail stage. Run in sequential mode (or perhaps on three nodes, but this is not guaranteed to be accurate). You don't need an Aggregator stage - there is no solution involving an Aggregator stage.

Welcome aboard. Setting PARAM1 to 7 should give +/- 7 days. If this is not what you are getting, please advise what you are getting.