DSXchange

Many Thanks !!!

I was wondering how can the last value in an array be retrived through sql. Here is the SQL I am using SELECT @ID, "KEY", DTC, CREATOR, PREVDTD, PREVDELETOR, DTM, MODIFIER FROM DS_AUDIT WHERE ("KEY" = '2\' || '#JOBNAME#'); Some of the fields like MODIFIER , DTM are multi valued. ...

I saw from the server side tracing that a datastage subroutine DSR_AUDIT is being called. Saw the object code for this is DSG_BP.O directory. How can I call this routine in a server routine. This is what I saw in the tracing files 2007-04-05 16:52:00: DSR_AUDIT IN = Arg1=5 Arg2=2 Arg3=RateFileCreate...

Ray, We might not be changing our version any time soon. People dont seem to jump onto latest versions that fast here :) The server side tracing idea was great. It did give me enough clues on trying to execute existing inbuilt basic functions to retreive what I want. Needless to say the code you, ki...

I wonder if its possible to capture the sql strings datastage executes when a log of a job is selected or a category is selected in manager that brings up the list. I stumbled on this when some of our job log hashed files physically got deleted somehow. When I clicked the job log in the director, it...

But that would make the job read only for the person who wants to work on it. Then the developer will have to switch it to NRO everytime he has to work on it and make it back to RO every time he logs off the machine, or takes a break or chooses to work on something else. Instead if he can hold a loc...

Unfortunately, there is a lot of exporting and importing done all the time. As requirements change, we import dsx copies from production back into development to be in sync. Is it possible to lock a job from being changed by locking/changing permissions on any of the internal files of a job, like RT...

Ok, I was reading thru my post again, and my punctuation might have given out the wrong impression. I meant ("put an annotation in red stupid", kind of an answer). Just didnt want anyone who take the time to help folks like me to get offended

For some reason I like to do things the complex way :) Heres a thought. Make the developers invoke the routine to make jobs read only. At the same time the routine writes the time and the developer user name to a universe file . Do the same for the reverse process. Now we have a trail of who is work...

Sounds like a much simpler plan than trying to lock jobs. I think the red color should work Have to see how it works out

In fairness to the developers, we have been hit by some insane deadlines, and multiple requirements are flowing in often that end up touching the same job. Rather than checking with all the developers if any of them is using a job, which can take time , It becomes a little easy to be able to know th...

Read Only sounds good. All I want is for others to know that someone is working on a job before they start modifying it. The responsibility then falls on that person to identify the individual who is working on that job to figure out when they can start using it. A job locked by a user sounds a bett...

Hi, Considering DataStage does not have a check in check out facilty, how can that be implemented by other means ? I would like to work on a job, and at the end of the day lock it. The next day, I come back and unlock it. Are there any universe commands that would lock a job ? I know I dont have the...

But again, the problem is the same code with the same data runs perfectly fine after recompiling the job and re-running it. I would think all things being equal, the job should abort again, but runs successfully instead.

But that doesnt explain why this job runs in development. It also runs well in production for two days before it starts aborting and we need to re compile it. Then we get a two run ticket !!