Deployment and Version Control

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
Munish
Participant
Posts: 89
Joined: Sun Nov 19, 2006 10:34 pm

Deployment and Version Control

Post by Munish »

Hi All,

Our project is in deployment stage.
There are a few doubts I want to clear up,
I tried using DSX search and as usual it helped me a lot, however I am left with a few doubts.

1. Can we compile from the command line in a Unix (Solaris) environment?
>> I think one can not compile from command line.
viewtopic.php?t=106597&highlight=compile+command+line


2. When jobs are exported separately as executables, how are dependencies managed? Is each file autonomous, and once each job is imported we can expect it all to hang together correctly, or is there a notion of libraries, similar to in Java?

3. If you import a job with the same name does it simply overwrite?

4. Do we need to import jobs in a specific order in a project depending upon there dependency?

5. How will we be running our sequence from the command line - do we call a specific job, or the project ...?

6. How do we "lock" the jobs in production?

Thanks and regards,
Munish
MK
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

  • 1. No.

    2. Not relevant, since "executable" is not at the O/S level. A dependent object is called by finding it in the Repository.

    3. Yes, but it can prompt first. Uncheck "Overwrite Without Query".

    4. No. See answer to Q2.

    5. It's up to you and how you have encapsulated the dependencies. If they're in a job sequence, it's the job sequence that you run.

    6. Version Control can make them read-only. You can make your production project a protected project also.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Munish
Participant
Posts: 89
Joined: Sun Nov 19, 2006 10:34 pm

Post by Munish »

Thanks Ray,
Could I please request you to put some more light on Point 2.

In application,
if I need to run individual job, I have to take care of dependencies.

How does this happen in DS,
when I import individual job does it takes dependencies like param etc. with it.

If not, what else should be imported prior to import the job.

Thanks,
Munish
MK
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

The Dependencies tab in job properties and Routine properties serves as a check list for other objects that need to be promoted. Version Control can detect that objects are recorded there, and cause the other objects to be promoted (after prompting for permission to do so).

Otherwise it's the responsibility of the promoter to ensure that all dependent objects either already exist or are promoted.

(Amplification of Point 2)
DataStage components execute in a virtual machine, either the DataStage Run Machine (server jobs and job sequences) or the Orchestrate shell (parallel jobs). What is generated can not be executed directly from an operating system shell.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
bmsq
Premium Member
Premium Member
Posts: 32
Joined: Mon Oct 30, 2006 9:19 pm

Post by bmsq »

Thanks Ray,
Highly Appreciated.
Closing the thread.
Regards,
Munish
Post Reply