DSXchange

That is odd behaviour. I think that taking out the write to the datasets will let you see if it might be related to your database read part of the job; that is the only thing I can think of at the moment.

snassimr, the method of defining and using routines in DataStage is project-centric. The actual methodology used is called local cataloging in UniVerse terms. You can define routines as being "globally cataloged" but this would get overwritten when you recompile the routine. Overall it makes more se...

Ever since the first versions of DataStage, the number of jobs in a project, and later on within a Category in a project, adversely affected the performance in listing jobs in the Director, Designer and Manager. Although this has gotten a bit better over time, it still remains an issue at every sing...

Hi Rleishman, TRANS1 is not the name of the Transformer. Anyways, The key that is generated has a stage variable which is not initialised. But the problem is if thats the case, the key field should contain the null value in the output of the sequential file which is not... Shyju, who incorrectly to...

Craig, I remember back in Prime Information days we had a customer who complained about their performance in reading/writing to Information. After looking at just about every attribute on the OS I finally ended up having to go to the customer site to try to find out the cause. Lo and behold, after p...

Roy, is this performance degradation consistent across runs? There shouldn't really be a change in rows-per-second in typical PX jobs that I have seen. It might be worth reading from Oracle and not writing anything out at all and seeing if the performance stays constant throughout the run. What is y...

zam62,

it would be beneficial if you would tell us what the error message is.

akhelif,

can you confirm that the e-egu accented character really is the character you are expecting by outputting the SEQ(Arg1) and looking in the ASCII table for it?

Daniel,

I recall having had a similar problem sometime in the past; it had to do with the way that the custom SQL was interpreted - so my question would be whether or not you are using your own SQL in the stage or using generated code? If it is custom, could you perhaps post it?

I don't know where you are reading that only Windoze platforms are supported for EE - they are a more recent add-on onto the various UNIX platforms. If you go to the Ascential web site you can find a full and complete list of supported platforms.

In DataStage you would have to read through the whole file to get that number. In UNIX it would be "wc -l" and if you have DS installed on your workstation that command will also work in Windows from a DOS shell.

bapaju, if you always have 2 digit codes and the numbers you have represent the position in the alphabet then you can use the CHAR(64+{number}) to get the appropriate letter. Since you said your string could be 2, 4 or 6 then the one-line derivation could be: IF LEN(In.String=2) THEN CHAR...

Gokul,

Do you have columns going into the hashed file stage? Is at least one of them identified as a key? Can you try a View-Data on that hashed file stage?

I haven't used the GUI to remove datasets, but I would be surprised if that doesn't do it correctly. If you remove a .ds dataset descriptor file and then re-create the dataset in a program then you will have extraneous files in the scratch area from the previous set that will never get removed. If y...

Have you tried it at UNIX - "orchadmin rm {dataset.ds}"? It sounds like somehow just a unix "rm" was issued. Do you get any error messages from the dataset management program?