DSXchange

Our latest approach is: DRS (1) --> xfm --> DRS (2) --> hashed_file (2) ^ | DRS (2) -->hashed_file (1) This: 1) Ensures hashed_file (1) always exists 2) Is always in sync with DRS (2) (the target table) 3) Ensures new SKs are derived correctly Pros: Ensures the hashed file is always in sync with th...

Our latest approach is: DRS (1) --> xfm --> DRS (2) --> hashed_file (2) ^ | DRS (2) -->hashed_file (1) This: 1) Ensures hashed_file (1) always exists 2) Is always in sync with DRS (2) (the target table) 3) Ensures new SKs are derived correctly Pros: Ensures the hashed file is always in sync with th...

I am extracing data from an oracle database and i am getting out of process memory error from oracle.This is a production backup database where i will not find a way to change any settings. Are there some thing that we could do from the Oracle stage in the server ORA-04030: out of process memory wh...

I have found some previos threads about the same issue:

64 bit install of datastage on Windows?

Installing DS8 on 64-bit WINDOWS

The answer, basically, is that it is possible but not supported.

Thank you.

Nowhere in the documentation, nor at IBM' site I could find information about DataStage 8.0.1 Server running over Windows 2003 Enterprise Server 64bit.

Does anyone here has that configuration running succesfully?

Thanks in advance,

I thought the FILELOCK function had an additional parameter to wait "automatically" a given time when the file was locked by another user/process.

I will use the code you provided, thank you.

Hello all, I am using the FILELOCK statement to ask for an exclusive access to a hashed file and it is working very well. TargetName = "TargetFile" Open TargetName TO HashedTargetFile Then FILELOCK HashedTargetFile ... But, in case the file is locked by another user I would like my routine...

I am not sure what it was.
I double checked my query and it was fine. Restarted the server and still the problem persisted.

So, I exported, deleted and imported the job and the problem of the "extra" stages reported dissapeared.

Thank you.

Your were right, I ran a test for field 8 and this is the result: LinkName.. Transformer_3 StageType. CTransformerStage Field8.... LinkName.. InterProcess_7 StageType. CCustomStage Field8.... CInterProcess LinkName.. Aggregator_9 StageType. CCustomStage Field8.... AGGREGATOR LinkName.. Oracle_OCI_27...

Hello all, I have a new question. :oops: I am trying to identify from the Repository for each of my Jobs the total number of stages of these types it uses: Link_Partitioner Transformer Sort Aggregator Although I have been trying (experimenting) with different queries, the objective is to have someth...

I am trying to determine at the beggining of a sequence if it was started by another sequence, a parent. (I hope you'll understand what I am trying to tell). I believe it is possible to do it using a parameter, right? Apart from that, Is there a different solution available? Using, for example, a Ro...

Ok. We discussed here your answers (not those involving The Simpsons :) ) and decided to give a 'weight' to each job based only in our knowledge of the job design and the tests and stats we have collected so far. Now, Craig mentioned... Ken Bland offers a set of Job Control Utilities free from his w...

Hello,

that's very interesting and I have taken my notes regarding methrics. I would like to know more specificly what methods you used to "throttle" the jobs.

Regards

Hello, By the end of the next month we are going to move a new completed Phase to production environment. This phase, a collection of jobs and job sequences, will be running over an IBM xSeries with 2 quad-core CPU and 4GB RAM. We want to control the system workload specifically related to job proce...

No, not the Repository :wink: Right, but look beyond the subject to the reason why they are 'connecting to UniData6 from Designer' Yes, thank you. It is explained in the main post the reason why we are trying to connect to UniVerse: "to query all the job names inside that sequence and its subse...