DSXchange

Hi Regu; It is not as easy as you think. Use the Reporting Assistant in the Manager and Move all Jobs to the documentation Tool. Start the Tool with the -key pressed. You are then in Access. Then you have to look into the right table to find the details. If you have used parameters for your hashfile...

Is it possible to get a list of hashed files used by all jobs in a particular project. I expect I can do this in DS.TOOLS some how, but I also get the following error when I try to run it through telnet. exec(): 0509-036 Cannot load program bin/uv because of the following errors: 0509-150 Dependent ...

Hi Sylvain, this depends also from the DataStage Release. Early DataStage 5.1 Releases open for each link an own session unless you enable transaction grouping. Later Releases open for each OCI Stage only one session but it is more complicated if the links comes from different Transformers. Regards,...

Bonjour Sylvain, two links to one stage is nothing special and should not be the problem. For to solve the problem you have to understand a bit the transaction system from oracle. If you have to much commits to a (reading) table during the read process the snapshot of the (reading) data from a certa...

Hi everyone, Does anyone know how DataStage handles two links to the same OCI ? Is a connection opened for each of them ? Because we have two links, Update and Insert, to the same OCI and the jobs aborts with Oracle error "snapshot segment too old" and the only reason to explain this would be a conf...

Hi Tom, There are two ways to DS.TOOLS: (1) In administrator choose a Project and Excecute and write "DS.TOOLS" as command. As Ray told already the Menu doesnt look so nice because the line feed and carrige return are not handled as they should, but it works. (2) You telnet to the machine, where Dat...

DS.TOOLS is a green-screen menu that is executed within the DataStage environment on the server. It *can* be executed via the Command window in the Administrator client, but the formatting of the menu is lost. Instead, you need to establish a telnet session to your UNIX server, cd to the project dir...

This is a topic for an orphaned message.

what is DS.TOOLS ? When i check in the administrator module (on the NT client) i have a tunetables menu - but it only includes read and write cache size for hashed file stages (999MB in each) and Active-to-Active link properties. I cant find anything about the memory manegement of the Aggregator sta...

In the DS.TOOLS menu is an option (Option 6 - Adjust job tunable properties) for tuning the memory management of the Aggregator stage. You can set the initial size, the threshold at which it increases in size, the multiplier for this increase, a second threshold and increment. -- ___________________...

Hi All, In my job, there are 2 links write to a DB2 stage. I have set transaction size to zero for both links. When I run the job, the "inserts Or updates" committed at the link level. Meaning, if the insertion fails during second link, whatever inserted by first link is stands committed. Just wonde...

Aggregator cant sort and sum large amount of data. There are two options that we can do. 1) Putting in a sort stage: You need to put in a sort stage in front of Aggregator to sort your data in advance then specify the sort order in aggregator stage. This would take of the problem but it slows your j...

Check to make sure your data is sorted prior to aggregation and that you have the sort and sort order columns defined in the input tab of the aggregator stage. If you do not define the sort order to the aggregator stage it will try and sort the input in memory and when you have a large amount of dat...

Hi Tom, Have you sorted the record on the field that you use for Aggregate?? This could solve your problem. Could you pls let us know if is like that?? Hope this help. Best regards Giovanni Boccia Fromtom.peter.nielsen@nordea.com Todatastage-users@oliver.com Cc DateMon, 17 Feb 2003 16:45:56 +0100 Su...

Instead of Using Agg-Stage try to write the data into a Hashfile with an UVStage and read them with the group by Option, this way Aggregation could also be done. Problem is known since 3.x-Times. Wolfgang > -----Original Message----- > From: tom.peter.nielsen@nordea.com > [mailto:tom.peter.nielsen@n...