DSXchange

A job's status is always its current status. That's all you mentioned in your specification.

To get previous status requires going to the job log, even assuming that you don't have auto-purge up to previous 1 job run enabled. It would be substantially different code.

As observed, two passes through the data. There is no out-of-the-box sort for a dynamic array, but it's readily enough done, most easily while the dynamic array is being populated. FUNCTION SortDynArray(MyDynArray,Seq) * Insertion sort on dynamic array, must not contain nulls. * Assumes sing...

Median requires two passes through the data. Can you please provide your algorithm for doing it in DataStage? I don't have one. (The same is true for Rank.)

madhukar wrote:removed the files in scratch space but still it gives 99% full (previously also)
Am i missing anything?

Yes, the 99%.
Use du -s to identify which directories are taking the resources.

Too difficult.

Just add a Column Generator, operating sequentially, downstream of the Sort stage (which presumably was also operating sequentially, or use a Sort Merge collector if not).

Use a Cycle with an increment of 1.

And that, dare I say it, is your problem, not DataStage's.

A record schema describes a record.

PRIMARY KEY is a table constraint, not a record constraint.

A record schema does not describe a table.

Sort IS available in the Aggregator; it's on the Partitioning tab of the Input link properties. Then you should also use "Sort" as the aggregator's operational method.

vinodhraj wrote:Hi,

I want convert datatypes inside the transformer.
Any soluton.

Thanks

Vinod

There ARE no data types inside the Transformer in server jobs.

Can you be more specific as to your requirement?

From the client software (Designer or Manager) you can't change the first component (50) other than by upgrading the version of DataStage. You can't change the middle component other than via some versioning tool, such as Version Control. But you can change the third component.

DataStage is not a reporting tool.

You do this in SQL. Hopefully in some variant of SQL that gives you a display function such as MEDIAN.

DataStage is not a reporting tool.

You do this in SQL. Hopefully in some variant of SQL that gives you some display functions such as RANK in its query language, and some means to restrict a query based on rank.

... or on the way home from the office, having been phoned (usually Friday afternoon, ahead of the weekend) to pick up some nappies. That's the one. Like a lot of things that one discovers using a DW, it makes sense in hindsight.

Business Glossary is another client that uses the common repository, so any business terms, etc., recorded in Business Glossary will be available to the other clients. In DataStage you could effect data elements to map the business terms to the underlying SQL data types. Metabrokers can still be use...

Reminds one of Ford's response to Bill Gates's declaration that cars, if built like software, would cost $50 and be better than Rolls Royces. The Ford executive retorted that you'd have to fix any problems by everyone getting out, closing all the windows, restarting then opening the windows again. T...

Welcome aboard. :D Do you comprehend what a record schema is, in the context of parallel jobs? From any table definition in the Repository you can click on the Layout tab then choose the Parallel option to view the equivalent record schema. Right click on the background area and you can save the rec...