DSXchange

You use the @NULL system variable.

That's just a matter of setting the Lookup Failure property to 'Continue', is it not?

Tune your query. As Ken notes, the time to the first record is NO indication of the overall time it will take. Check your explain plan for nested loops and hint to 'use_hash' joins if needed. Also consider 'full' and 'parallel' hints and 'alter session' statements before sql to get the maximum dump...

Did it immediately failed after the upgrade or did it work for awhile and then start crapping out? Regardless, that 'user api layer' error is odd enough that I'd suggest you open a case with your official support provider and then let us know what the resolution was.

Assumed so, thanks for the confirmation.

Output is a dynamic array here, not Ans or Status, which is why I found this syntax odd. Technically it works cuz DataStage is smarter than us but tells me the coder didn't really understand what they were doing.

Or as noted in the original post, make sure the pathname parameter value has a trailing separator: p_FoldName /dw/dsadm/ p_File MyFile.Txt #p_FoldName##p_File# Makes it more portable if that is a consideration. However, I put the slash between the parameters as there's no chance of going from UNIX t...

Ray told you how to do that.

As Ray notes, substring the date out then convert. IMHO, neither field should be declared a Timestamp in your job as neither fits that definition, use Varchar for the source and decimal/integer/numeric for the target.

Well, now that I'm up from sleepy time... This is oddly coded as I don't know why one would treat 'Ans' as a dynamic array for no good reason, which is what the angle brackets are doing: they denote the element of the array to access. So <1> means the first element and <-1> is special magic to mean ...

There's no 'Remove Duplicates' stage in Server. Well, there is but it's called the Transformer.

Welcome aboard! :D

Search the forums for "USERSTATUS", a mechanism to pass values from one job to another. You could also land the data, say in a flat file, then use a routine to pick it up and then make your decision.

Define what "has not worked" or "will not run" mean.

Interesting first post.

girish312 - you need to include your questions in the body of the post, no-one can see your subject until they reply. And most won't even notice it, I'd wager.

girish312 wrote:Were you able to reslve the answer to schema def issue here?

Welcome aboard. :D

Odd as it sounds, in automatic mode no such files are created. The only information you have is what ends up in the job's log.

Hand typing leads to errors so I use the 'helper' when I can. For example, environment variables have only one dollar sign in them, not two.