Page 1 of 1

"Check DS is set up correctly error"

Posted: Wed Aug 16, 2006 1:44 pm
by jdmiceli
Hello all!

One of my cohorts is having a very difficult time getting things going on her new project. We are running DataStage Server Edition, v. 7.5.1.A running on Windoze (soon to be going to AIX Unix). Her source is Informix and she is using the Informix CLI driver utilizing a DSN that is already working correctly in other projects. Her target is SQL Server 2000 using the OLEDB stage on a server that is also aready successfully being used in other projects.

Once we get the Informix CLI source stage set up with it's connection information, a transform stage put in place and the SQL OLEDB target connection created, and then take the step of going into the source stage and pressing 'View Data' (I know that's a specific order, but that order is what got us the best results), then we get an error box with this text:

DataStage Repository Interface
Error calling subroutine: DSD.Browse (Action=3); check DataStage is set up correctly in project ALIR
I tried rebuilding her job in my project, which has never had this problem, and the same thing happened. I'm leaning toward this being an Informix connection issue of some sort, but I have no idea how to go about fixing it.

I would appreciate any advice anyone has for this situation. Thanks much!

Posted: Thu Aug 17, 2006 8:02 am
by ray.wurlod
This message can indicate that indexing in the DataStage Repository has become out of synch. When you can get exclusive access to the project, execute the DS.REINDEX ALL command (through the Administrator client Command window perhaps) and service is likely to resume.

Where is the administrator command window?

Posted: Thu Aug 17, 2006 8:56 am
by jdmiceli
Hi Ray,

Thanks for the response. I must confess to some ignorance yet as to where certain things are and how to perform command functions. Where would I find the Administrator Client Command window? I wish I had an administrator to go to, but I know more than he does at this point as he has only recently been given the responsibility and has no prior experience with the tool.

Thanks in advance for your advice and patience! :D

Bestest!

Belay that last question!!

Posted: Thu Aug 17, 2006 9:01 am
by jdmiceli
I found it!! Amazing what a little digging will do - preferably before opening mouth and moving feet freely! I ran the command and it showed lots of information that appeared positive. I am testing the stuff now to see if we have the same problem. I will post again if I have any more troubles.

Thanks again Ray! You are awesome as usual.

Bestest,

Still a problem with this

Posted: Thu Aug 17, 2006 11:59 am
by jdmiceli
Hi again,

I did what Ray suggested and then tried to recreate the problem she is having and ended up with the same problem. So then, I created an entirely different project and tried to write the same job from the ground up (Src --> xfm --> Tgt with no transforms - just a straight pull). We even built new DSN's for the new project and I still got the exact same results. I also noticed that it won't even save the jobs we create or allow for the creation of new categories to put them in.

So, I am well and truly stumped. Any further advice?

Thanks in advance!

Posted: Thu Aug 17, 2006 12:45 pm
by kris007
I am not sure if you have already tried doing the following but just sharing my thoughts:

1. Split your job.Create your Job with Informix-->>xfm--->>Seqfile and see if the error still exists.

2. The last time I saw that problem, I was using Oracle OCI stage and had to re-install the plug-in. You might want to give that shot and see if it helps.

3.Restart the Server. It works sometimes.But, I guess you might have already done that.

Our solution to this problem

Posted: Mon Nov 20, 2006 8:38 pm
by jdmiceli
Hi all!

In preparation to closing this out, we found that we had a bug in our installation. Once we got a patch from IBM, the problem went away.

Thanks!

Posted: Mon Nov 20, 2006 9:24 pm
by ray.wurlod
May we know what the bug was, what the patch was/does, and what the ECASE number is? That will help anyone in future who encounters the same bug. Hopefully.