DSXchange

The sample data you have listed is somewhat unclear - you have 3 rows with differing number of columns. A "group" in this context is normally a set of records where one column contains the same value. Is it "Col1:-" in this case? Or have you displayed only values for "Col1"? Could you explain just a...

As DSGuru2B has stated, the ODBC system DSN needs to be created on the Windows DataStage server machine.

In my experience the problematic portion of this cross-platform migration is in finding all the places where paths have been hard-coded in Windows format and need to be changed to UNIX. In addition, if any user-coded functions use OS calls to get information or trigger work these need to be identifi...

Can you connect to the ODBC on B using MSQuery? If that works then the DataStage connectivity should also work. Have you tried to manually type in the ODBC DSN in the metadata import?

your colleague already posted the same question at this link

Executing from DS opens up a new shell, so when that exits back to DataStage the variables and their values disappear as well.

The datastage job is a separate process is not started with a copy of your user's environment, which is why this isn't working as you expect. You will need to explicitly set the environment variable "sbpath" in your job for this one job. If all DataStage jobs need to have this value, you can have th...

If you assign the function value to the stage variable "X" and then use that variable in 20 derivations within the transform it will be much more efficient since you've only computed the function one time.

If you only use the "X" value one time then you are better off calling the function directly.

Check this current thread out for more information on your probable problem.

No problem. This particular error caused me several days of work to trace down and submit to Ascential as a bug so it is a very familiar issue to me. This call should fail immediately instead of waiting for 30 minutes on some semaphore before aborting.

I cannot supply you with detailed information. You can search this forum for more information or, alternatively, log in to the Ascential Developer Network site (you or someone in your organization should have a login ID) and search for the patch and download it as well as the description. I think th...

In almost all cases where you can specify a parameter you can use the concatenation operator to put together constants and variables. You haven't specified exactly where and how you are doing this, so I can only answer that "yes, probably it will work".

I just noticed that you are under UNIX and not Windows. Please read up on the DSExecute() command, particularly the first parameter

Add a CALL DSLogInfo('NT Command output was "':Output:'"','') to your job to see what is happening.

The DataStage server concatenation operator is the colon ":" character, not the plus "+" sign.