DSXchange

These commands have never been documented as part of Datastage. The server engine was cloned years ago from a database called UniVerse which was sold to IBM and now belongs to Rocket software. Much of the documentation is available at http://www.rocketsoftware.com/u2/products/universe/resources/tech...

I'd involve a DB2 DBA in this. Have them do runstats and perhaps monitor the execution plan of your upsert to see what is happening "underneath the covers".

Are you moving this field without doing any transformation to it? If so, then the incoming field is probably not nullable but contains a null. You can activate "$APT_PRINT_SCHEMAS" with "true" for one run of your job and check the log file to see what the actual runtime schemas b...

Ok, we've established that your settings are correct in DataStage and not limited. I just looked at the number in question, 7,052,615,504 -> 6.7Gb That means that your lookup set is huge and you should either change this stage to a JOIN stage or you could see if you can reduce the number of columns ...

Are you still seeing the word "IDENT1" in your job with a transform stage?

I'm not in a good position at the moment to offer an example. Can you do a simple wrapped stage calling a script that just give an echo of the command line? Then declare just one column in your wrapped stage Wrapped -> Interfaces -> Input mask with type "standard in" and likewise for your ...

You cannot add values as parameters in this form. You need to add dynamic values to Wrapped -> Interfaces -> Input; for example as a stdin and stdout.

What sort of a join do you want to do on these 2 files with no common keys?

Row 1 of file 1 combined with Row 1 of file 2, and so on?

The parameters and parameter values don't come from the data, they are to specify parameter values that don't change per data row but are used to configure the call.
You need to specify the Wrapped -> Interfaces -> Input for your input and output data.

Good news - there is a slowly-changing-dimension stage which covers your requirements.

Do you have an editor capable of macros (emacs, notepad++ or the like)? If so, you could create a .dsx export and then apply the macro to make these replacements and then re-import the .dsx file with the changes applied. As Ray has noted, it can't be done directly within DataStage.

And from which program are you getting this rounded value? I ask because very often the full value is actually being stored, but whatever mechanism is used for display doesn't show the full value.

The precision of 18 places looks very much like a single-precision value.

the "DS_TRANSFORMS" file does not contain the information you want. I would suggest using the (documented) BASIC API Calls rather than trying to work with the DataStage internal (undocumented) hashed files. Select all Projects, for each project: Select all Jobs, for each Job: Select all st...

Date and Timestamp columns have no intrinsic format. You only need a format when loading from another datatype or when converting to another datatype. In your case you need to convert the VarChar into Timestamp using something like "StringToTimeStamp(In.StringColumn,'%MMM %dd %YYYY %HH:%nn%aa')...

What have you tried? Within the same job sequence you can use the return values from a Routine Activity as a parameter value in calling another job.