DSXchange

The correct spelling of the second person personal pronoun in English is "you" in the nominative case and "your" in the possessive case. Please strive for a professional standard of written English in DSXchange. Among other things, it helps those whose first language is not Engl...

My guess is that it's a limitation of TOAD.

Which, to be fair, was never originally written with the intention of displaying anything but ASCII characters. Hence the existence of derivatives.

Please mark this thread as Resolved using the green button at the top of the page.

It can't not show it.

Does your Lookup stage have two (or more) input links and two output links? What do the output links look like?

I am trying to get the audit of a particular job using DSGetLinkInfo(DSJ.ME,DSJ.ME,DSJ.ME,DSJ.LINKROWCOUNT),but I am getting series of count(like 1,2,3,4,5,6..till max of link count) :cry: instead of only one max value. can u please guide me in getting those info like no of records read from source...

Go to Kim Duke's website and download his (free) ETLstats package. It will do everything you ask and more.

Create a reject link on the Lookup stage and feed it into a Sequential File stage. Change the Lookup Failed rule in the Lookup stage so as to direct stream input rows for which the lookup failed to the reject-handling link.

Moderator: please move to parallel forum

Try disabling operator combination to work out what's really happening.

Inability to write to RT_CONFIGnnn (hashed) files only occurs when compiling a job. When a job runs, it only reads data.

Please run the following command from Administrator client (command window) and let us know the outcome.

I am not premium reader so couldnt read your full reply. Do invest in a premium membership. It's not expensive, at less than 30c (Rs12) per day. Premium memberships are one way that the hosting and bandwidth costs incurred by DSXchange are defrayed. So, by taking the premium membership, you're help...

aninashh wrote:I heard from one of my techincal head as "Performance of big Hash files is bad and better way is to redesign the Job without using so big hash files".

Please let us know on what evidence your technical head makes that assertion.

The Hashed File stage can not create a 64-bit hashed file. You need to do this in a before-job subroutine (ExecTCL) or before the job starts (for example from a job sequence). Delete the hashed file in an after-job subroutine (ExecTCL again) or after the job finishes. Why not include ORDER BY clause...

The easiest way is to import the table definition of the sequential file (specifying it as fixed-width, of course, and specifying the field widths) then, in the Sequential File stage, click Load while on the Format tab. That way the metadata remain untouched by human hands and should be reasonably e...

You heard? From where? With what evidence? What do you mean by "full"? Your approach is valid. If you have to process more than 2GB of data, create your hashed file with 64-bit pointers. If you have to process more than approximately 1.9 million TB of data you probably don't want to use a ...

Then your SQL is a UNION from 4000 different status tables. (That's what Director has to do, though programmatically rather than using SQL.)

Or your script will loop through 4000 separate dsjob commands.

Enjoy.