First job Server Task

palmeal · Post by **palmeal** » Mon Feb 28, 2005 9:48 am

I have a few questions regarding set up and getting started that I would appreciate any help with.

My client software is UK based with the datastage server in the States. Connectivity to the datastage server is via telnet or through the datastage software itself. I wish to be able to connect to both Sybase and MSSQL servers that are resident in the UK.

For the sybase server, the interfaces file on the datastage server has been updated with server details. I am now able to telnet onto the datastage server and isql onto the sybase server.

However, when I try and create a job server task in datastage I don't see any mention of the sybase server.

I tried using the Dynamic RDBMS within job server and selected a DBMS type of Sybase. I then entered in the details for BKP_ANY2 but no connection was made to the server.

(1) Do I have to set up a data source name in datastage somewhere in order for me to see the sybase server - is this what the uvodbc.config file is for ? Does this file get attached to a specific datastage project ?

I'm not sure about how to set up the MSSQL server but assuming that I have my datastage source name for the sybase server and MSSQL server set up then I would like to do the following :

Connect to the Sybase server and extract data into a transformation process before forwarding it onto the MSSQL server. This sounds easy enough but looking at the types of tasks within job server it doesn't seem that the available options are going to help me much - when I select one of the sybase options when creating a server job the Sybase options are "Sybase BCP Load", "Sybase IQ 12 Load", "Sybase IQ Load" and "Sybase OC" - none of these jump out to me to say that I could connect to a sybase server and extract some data. Should I be using the Dynamic RDBMS.

(2) Could someone point me to what I should be using for this please ?

It is difficult to get these probable easy problems ironed out if I was working in the same time zone as my counterparts.

Thanks for any help.

roy · Post by **roy** » Mon Feb 28, 2005 9:58 am

Hi,
what is the SYBASE version you try to conncet 2? (probably not IQ or that would sound good for you)
SYBASEOC (Open Client is probably what you need)
read the docs regarding what you need for each stage (also click on help when inside the stage properties will give you info about the options and usage of the stage properties).
you have the Docs directory in your client installation that contains PLUGIN specific info and the servjdev.pdf to help using them and guid you in your job design.
more over you have the choice of using ODBC cnnections (usually slower then plugin) then you need to configure .odbc.ini and uvodbc.config in your server install + uvodbc.config in each project directory you want to use that DSN from.

IHTH,

ArndW · Post by **ArndW** » Mon Feb 28, 2005 10:01 am

Hello Palmeal - thanks for posting your issues and questions in such a clear manner! It is a welcome change from some of the postings that really don't clarify their issues apart from a "it doesn't work, how do I fix it" query. Even though I don't have recent Sybase installation experience there are people here who will certainly be able to help you.

Connectivity to both the Sybase and MSSQL databases must be done through your UNIX server, as you correctly surmised. You can connect to Sybase via the direct OC and to SQL Server (plus also Sybase) via ODBC, which do need to be set up in the uvodbc.config file in your project directory.

You can start with looking into your client's directory which contains the documentation (default should be "c:\Ascential\DataStage7.5\docs") and opening the SybaseOC.pdf document, which does go into detail on what needs to be entered where in order to get the OC connection working.

There are numerous threads in this forum, if you do a search on SQL server and ODBC you will get some hints on what to do, plus the documentation does go into great detail on setting up ODBC connectivity on UNIX (I really wish it were as easy as doing it on a Windows machine).

ray.wurlod · Post by **ray.wurlod** » Mon Feb 28, 2005 3:19 pm

The easiest way to remember the architecture is this simple statement.

DataStage Engine is the client to the database servers.

When DataStage runs a job, it is just another client application as far as your database servers (Sybase and MS SQL Server) are concerned.

palmeal · Post by **palmeal** » Fri Mar 04, 2005 3:32 am

Thanks for the answers - I now have access to both my sybase and MSSQL servers and am now starting to get sruck in.