DSXchange

The issue is not the number of columns in the table, but a mismatch in your Oracle stage between the number of columns you have defined and then number of columns the generated SQL statement is trying use. This can happen when you define your own SQL (and perhaps later on add a column, forgetting to...

I couldn't find the IBM DataStage exam in the list at Pearson training and asked them about it, and just got the answer 415 is not the Exam Number for any of the IBM exams that we offer. I've also checked through our expired exams, and it never has been. I would suggest you check with IBM as to the ...

Did you import the metadata using the same database userid (perhaps the one you are using from DS doesn't have sufficient privileges)?

The reason I asked is that if you are using your own query it might be causing the issue, whereas it is unlikely that a generated query will cause the error you are having. Which one of the 3 types are you using?

What happens when you try the same thing on that system DSN with MSQuery, is it successful?

I just wanted to narrow down the cause to the DRS stage. Can you run the job? Do you have custom SQL in the DRS stage?

...Is there a problem in DSJ.ParamNameList? I don't know, since it is not a DataStage mnemonic. By default variable and constant names are case sensitive, so you should use "DSJ.PARAMLIST", as the documentation specifies. The compiler can't know that the variable name you used is undefined in the f...

Define an ODBC stage to the same datasource as you are trying to access from DRS and see if you get the same error message. Have you used DRS successfully in this project before?

I think it is a misprint as well; getting 75 out of 79 that would equate to a score of about 95% and from the sample questions posted they would seem to want a written answer (not multiple-choice) -- thus some responses won't even be 100% correct because the examiners deem the answer to not be compl...

Looks like you have an extra 30th column in your metadata within DataStage while your table only has 29 defined.

Zeevs1970, first off, welcome to DSXchange. Could you explain your problem a bit more, since writing to a hashed file doesn't directly relate to using upsert into a table - unless you are doing it to an ODBC data source that links to a hashed file. Generally it is much faster to split your updates a...

Combining the two statements together makes it a condition, which returns only a TRUE or FALSE (1 or 0), thus the value of something like 'Hello World' AND 1=2 would be 1. What you want to do in this case is a simple Trim(Ereplace(InDerivation, "nt", "") ,";"...

legendkiller - <sigh> did you actually read the information on $APT_ORACLE_PRESERVE_BLANKS? This is exactly what this setting does.

kris9999, in the transform prior to the stage make another output link to a sequential reject file, then use the drop-down list in the constraints to specify that rejected rows should go down there and add the columns that you are interested in to that output stream. this is standard error handling ...

Read up on the $APT_ORACLE_PRESERVE_BLANKS setting in PX.