DSXchange

I am trying to run the command below based on a technote on the IBM website. But I keep getting a prompt that I am not expecting and do not know how to respond. I need to be able to run the SELECT RTEMP command and need to set the pointer first. >LOGTO Salesforce >SET.FILE /software/IBM/InformationS...

I have DataStage 9.1.2 installed on a linux server and we are using the Netezza Stage to load data to Netezza. In the tmp folder on the server we are accumalating many nzlog files. Is there a way to have datastage automatically purge these files to keep only the most recent? Can the files be deleted...

I have tried deleting all entires I can find in the registry but I still cannot install the client on this machine.

The install displays many messages indicating that it cannot register entries. Many of them are DLLs and all under DataStage 7.5.2.

Any other thoughts?

Thanks for responding.

I left this laptop at home.

I will try cleaning 'all' registry entries after the un-install and see of this helps.

I am generally very conservative with the registry because I have had issues with it before (other machines).

The OS is Windows XP SP2. I tried un-installing, deleting reg entries under SOFTWARE, deleting the leftover directories c:/Ascential and c:/Program Files/Ascential, rebooting and re-installing but still get the same messages.

Are there other reg entries I may need to clean out?

I have attempted to install DataStage 7.5.2 client on a Windows laptop a few times but continue to get errors. First I get a popup with a message "String variable is not large enough...." and an Error code 401. Then several popups indicating the install is quiting. Has anyone seen this bef...

I am setting up a production environment for DataStage server. I want to our QA and Production environment to be protected. I was suprised to see that a dsadm cannot compile a job in a protected project. Is it a requirement that we export / import the executables to the QA and Prod env. I may use VC...

I am not that familiar with the language. When I review the code of the existing routines, it doesnt seem like I can just reverse the logic. For example, the current routines do not handle the implied decimal. The packed number is converted as a whole number then divided by a number to move the deci...

I have to admit that I have not had much success with reversing this code. The original code hardly makes sense. I am now using iconv with an MP coversion code to pack a number. I ftp the file to the mainframe as binary with no conversion to ebcdic. Negative whole numbers are sucessful, but I can't ...

I will need to convert several numeric data types. My target is the mainframe. I will be creating files using DataStage server for use on an IBM mainframe. That means I would need the ability to conver numbers to: PIC S9(9)V99. PIC S9(9)V99 COMP3. And others possibly. In the end the file needs to be...

I tried this but the CFF stage does not support Stream Input links. It appears that is only accepts output links (reads data ).

Does anyone have any other suggestions, other than creating mainframe programs to pack data.

I have a requirement to create COMP3 fields in a DataStage Server job. The files need to be sent to the mainframe for processing. Is there a way to create a COMP3 field in a server job, then transmit it to the mainframe? It sounds like its possible base on the dsxchange entry but I cant figure how i...

The problem is corrected in the odbc 5.1 drivers. I downloaded them from the IBM website.

The correct syntax for the Address entry is
Address=servername,port

Where the port number is the port of the target instance.

FYI: I also tested the 5.2 drivers but they did not work.

I am being told by IBM that this problem has been corrected in the new 5.1 odbc drivers for REDHAT. There is also a set of 5.2 odbc drivers for REDHAT.

I will download/test and put the results in a reply.

How do I disable named pipes for the named instance? Is this something a server admin needs to do, or is this something that can be done within DataStage?