DSXchange

It really is a mismatch, if you cut-and-paste your SQL to your favorite tool and try to run it, you might see the problem. typically, if your SQL contains "CAST(MyColumn as VarChar(32))" then you will get this kind of error, you would need to write "CAST(MyColumn as VarChar(32)) as My...

Do you need the OBJECTID column downstream of your Change-Data-Capture stage? If not, then remove it from both input links. If you need that column you can remove it from one input (assuming you always want the value from the link which still contains that column) or if the value depends upon the CD...

You could write a job with 4 sequential file stages going into a funnel stage, or you could make one sequential file stage read multiple files, or you could CAT the 4 files together and read one file in datastage. I'm sure that there are numerous other possibilities, but these are the first that cam...

That is indeed a strange problem.

What method are you using to promote your jobs between the environments?

Are these 3 distinct DataStage installations or one installation and 3 projects?

Is there anything special about this job - seldom used stages, for example? I've had hidden bad characters and the like before, but in those cases the import would always fail. Can you think of any differences betwen prod and the other machines? Perhaps NLS is enabled in the other machines but not o...

What are the warnings that you are getting?

Unfortunately this looks like a problem that needs to be addressed to your service provider and fixed by IBM and I can't think of an easy workaround that doesn't involve changing your SQL. I wonder if the parsing problem persists if you move the wildcard strings to an included SQL file?

The status of your main multi-instance job, the one without a ".instancename" suffix, should be "compiled". The job should not be started without an instance name; if this job is running then until it is completed any attempts at trying to start a parallel instance will result in...

Most often I see this error when taking a VarChar() column into an integer column and the string value in some records cannot be implicitly converted. This usually occurs in transform stages, but if you follow the instructions posted earlier regarding disabling operator combination in order to locat...

It isn't a documented parameter, but obviously there.

What about \%\P?

Can you replace the "%" with "\%" to see if that makes a difference?

If you have a tab delimited file, are using parallel jobs, and don't need the double quotes, I'd simply add a filter to your sequential file read stage with the command "tr -d '\"'"

There are actually 2 step involved, first you need to parse the string value into the internal date format, then output that internal date as a string OCONV(INCONV(UserVariable[1,4]:'-':UserVariable[5,2]:'-':UserVariable[7,2],'D4-YMD'),'D4-DMY[2,A3,4]') I'm not at a PC where I can check the syntax, ...

Since the FTP protocol is a pretty simple one, unless there are huge amounts of transmission errors, you probably won't be able to do much. Normally the correct approach is to compress the file prior to transmission, but then the DataStage FTP stage isn't the correct one to use due to the binary nat...

I use the term "dummy sort" as well for sort stages that specify "don't sort, already sorted".

I just realized that this thread has split, as I answered a question on another thread. We should terminate one of the threads or merge them.