DSXchange

Hello,

We have just migrated our DataStage 8.5 architecture to an 11.5 system.
Unfortunately I have a big problem with a (server) shared container

We exported the server container from the 8.5 server and imported it to 11.5. A job uses this container, but aborts with the following error:

ContainerUmsatz_PN03,0: Unable to open project 'da1' - -1002.

da1 is the DataStage Project. I get this error on every node.

Any Ideas where I have to search for the problem?

Found this with a quick search, specifically wanted to point out this bit:

Note: If you encounter a problem when running a job which uses a server shared container in a parallel job, you could try increasing the value of the DSIPC_OPEN_TIMEOUT environment variable in the Parallel ► Operator specific category of the environment variable dialog box in the InfoSphere DataStage Administrator.

Worth a shot.

I have increased from 30 (default) to 500 - no improvement, still the same error

Well, is your environment the same as it was before? (stand alone host vs grid or cluster)

Are any other jobs broken or just this one?

Can you show the code that actually loads the other project code?

Sorry for the late response - I wasn't in office.

Well, is your environment the same as it was before?

We also changed the environment:
Version 8.5 was a Standalone system
Version 11.5 works with Master and 2 Engines.

Are any other jobs broken or just this one?

Only problems with this job. But at the moment this is the only job using the Server Container.

Can you show the code that actually loads the other project code?

Not sure what you mean - the OSH Code?

In your shared container, there must be some code (basic?) that is making a call to open a project/job.

Can you show us what you are doing there?

Hi,

the container is very simple:

Input > Transformer (using string-functions and other format routines) -> Text File Output

sometimes during a migration things hiccup and it's sometimes best to simply make a new container from scratch. I would rename the existing container. Create a new container (not a copy of process), and drag new stages onto the canvas and repolulate with the desired logic. Recompile container, (force) recompile the calling job. test.

Additional Information:
I can use the Container with Server Jobs without any problems. So I dont think rebuild will fix the problem ... but i will try.

Have you involved your official support provider yet? And are you aware of the limitations on the use of Server Shared Containers in a Parallel job? Things like no MPP or Cluster systems? And I would put any thoughts of it working on a really old release and a different architecture out of your head while trying to get it to work on the latest release.

I'll have to defer to others here who may actually be using them but I have vague memories of another requirement of using them... the need for the Server "engine" to be installed on any node you run the PX job on? Or is that another false memory implanted by my therapist? Haven't had any luck searching so thought I would throw that out there.