DSXchange

we did drop all the indexes.. the main culprit could be the referential integrity, check, primary key and not null constraints on the target columns though.. which we cannot drop

yes, i did work with bulk loader before and never used that ever again, coz i thought it was useless, in the sense it was so slow. .. from all your suggestions i thought that i might've been wrong and wanted to learn more about its options and features to improve performance... anyway.. yeah, actual...

yes, i do agree that TOAD may have misled me (my query was the exact view defintion that i copied), but in the Datastage job data starts flowing in from the oraoci stage that reads from that view pretty soon, i mean in couple minutes and not hours.. its the processing thats taking so long. About the...

record length is about 300bytes average pretty small actually.. but you are right about array size.. though in this case it seems okay. Generally, even i try to use 500 for medium sized files and 1000 when i think its a small record..byte wise.. but apart from these 2 things.. i read about buffer si...

yeah.. actually there is not much happening on the database when the job is run ..usually coz its after hours..same reason why network traffic might not play much part.. but about replacing oraoci with orabulk.. let me ask this.. i am doing a lot of user-defined sql on target end.. can i still do it...

Guys, I havent worked with jobs that populate more than a couple hundred thousands records in the past, so very naively i want to ask this question.. how long should a simple server job that reads from a view (view joins 4 tables.. but when i run this view in Toad, it gives results in 7 mins) return...

I have to capture the stage or the link last err into a reject table/file. I used to use DSLinkLastErr or DSStageLastErr in a server job.. i was looking if anybody would know about similar macros or routines which i could use to get the same kind of details.

thanks in advance for the help.

actually, i have the correct the summary that you have assumed In Transformer, you have a constraint field for the first Oracle table. You want "rejects" from that constraint to go to the second Oracle table. Then you have a second constraint to the third Oracle table. I do not have a spec...

wow Teej,
you are really a geek..works perfectly well.. and i do have administrator and access to it as well

Thanks a ton

error message is

Link perfHdrRejectORAIn, Constraint: perfHdrORAIn.REJECTED
Variable 'perfHdrORAIn.REJECTED' not defined

Guys, I need to be able to set the commit frequency on the target Oracle Enterprise stage to be able to avoid problems with rollback segments. would like suggestions from anyone who has done that before. I know there was a Transaction Handling feature in server jobs that enabled us to specify 'rows ...

i have a source stage (oracle), a transformer and 3 target oracle stages. My requirement is that i have to capture rejects of first target ONLY into second oracle table. I tried using otherwise in transformer (with the link ordering as first target, second target reject table, third target),but it d...

hey guys, back with another question.. i have a source stage (oracle), a transformer and 3 target oracle stages. My requirement is that i have to capture rejects of first target ONLY into second oracle table. I tried using otherwise in transformer (with the link ordering as first target, second targ...

unfortunately no.., atleast according to the manuals too.. the transformer can accept only one input (reference/stream irrespective) in a parallel job.. such a big constraint on development of parallel jobs...

Yeah, actually thats what i tried meanwhile, i wrote a user-defined sql on source oracle stage, with just the sequence name as one column. It worked fine. I am sure we can do it at the target end also. The lookup part .. Since my design has one source, one target oracle stage and one transformer, if...