Migrating Datastage Jobs to Production

Post questions here relative to DataStage Server Edition for such areas as Server job design, DS Basic, Routines, Job Sequences, etc.

Moderators: chulett, rschirm, roy

Post Reply
Lotus26
Premium Member
Premium Member
Posts: 48
Joined: Tue Jul 13, 2004 2:09 pm

Migrating Datastage Jobs to Production

Post by Lotus26 »

Hi


Can any one explain the procedure and steps to move datastage job from Development environment to Production environment.
Regards
Lotus26
datastage
Participant
Posts: 229
Joined: Wed Oct 23, 2002 10:10 am
Location: Omaha

Post by datastage »

It's probably considered a 'best practice' these days to make use of the Version Control client. Initialize components into the VC repository and then promote them to a production enviroment, ideally making them read-only and doing a mass compile at the same time.

Otherwise, there is still the old standby of using DataStage Manager to export componts to a .dsx file and then import them into a Prod environment
Byron Paul
WARNING: DO NOT OPERATE DATASTAGE WITHOUT ADULT SUPERVISION.

"Strange things are afoot in the reject links" - from Bill & Ted's DataStage Adventure
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

The vital piece is to make sure that you've got everything you need.

If you've been diligent about recording dependencies (this is why the Dependencies tab is there), then Version Control can pick up the dependent objects.

There may be certain things external to DataStage that you also need to promote, such as custom buildops (in PX) or scripts. The secret here is to create a custom folder for Version Control. VC insists that the custom folder be within the project directory, but I've found that it's happy to work with a symbolic link (UNIX) or shortcut (Windows).

In a server environment, you may also need to create any needed hashed files in the production project. This can most readily be accomplished by validating the jobs in which there exist Hashed File stages (or possible UV stages) with input links.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

ray.wurlod wrote:If you've been diligent about recording dependencies (this is why the Dependencies tab is there), then Version Control can pick up the dependent objects.
Actually, unless something has changed and I haven't noticed, VC does not need the Dependencies tab to find dependant objects. I'm not saying it doesn't actually use it, but it is perfectly happy inspecting all of the code being promoted and finding all of the jobs, routines, etc that it thinks need to go along for the ride. I'm not advocating everyone leave dependencies blank, just pointing out that VC isn't something that absolutely needs it.
And then ray.wurlod wrote:There may be certain things external to DataStage that you also need to promote, such as custom buildops (in PX) or scripts. The secret here is to create a custom folder for Version Control. VC insists that the custom folder be within the project directory, but I've found that it's happy to work with a symbolic link (UNIX) or shortcut (Windows).
I found this little trick way back when it didn't really work at that well. Imagine if you will, a man happily setting up a symbolic link in his project pointing back to his custom scripts directory. Then, picture in your mind that same man just as happily importing all of this scripts into VC. Lastly, contemplate his confusement when all of his scripts ceased to function. A quick check found that VC had written his import comments (much like happens in a job's long description) into all of his scripts. :roll: Needless to say, this little faux paus was quickly corrected when reported.
And lastly ray.wurlod wrote:In a server environment, you may also need to create any needed hashed files in the production project. This can most readily be accomplished by validating the jobs in which there exist Hashed File stages (or possible UV stages) with input links.
So that's what that does! :wink: Back In The Day, I validated everything all the time. Not only was it the only way to create hash files in jobs, but it was supposed to check your OCI syntax and other items of that ilk as well. It mostly worked. Nowadays, 'Validate' is probably something my little charges have never clicked on, in spite of my efforts to point out situations just like this where it still helps. What happens when you don't do this when a job is promoted to production for the first time? If you are monitoring job statuses, your first production run 'fails' when the creation of the hash file during the run generates a 'warning' that the hash file was created and the whole process grinds to a halt. :lol:


Bottom line, and back to the OPs question, much like I had already posted when they asked this at ADN... Version Control is the short answer. Read the docs. Play with it and then come back and ask specific questions. You can always drop and rebuild the VERSION project once you get the hang of it.


Edit: Wow, a whopping 4 frequent flier miles for this post. :? So much for that trip to Australia I was planning...
Last edited by chulett on Mon Jul 19, 2004 6:38 am, edited 1 time in total.
-craig

"You can never have too many knives" -- Logan Nine Fingers
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Yes, there seems to be an inordinate penalty for using a quote. Walter?
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Post Reply