DSXchange

Ken, I don't have a real elaborate version, but basically I use the os.system() module on the Windows Server called directly from the unix shell. This still requires SFU to be installed as it provides the "gateway". Python makes it platform independent. here is the essence of the code (uns...

If you have a UNIX Datastage Server I would start them under UNIX itself. If you have a Windows DataStage Server why bother starting jobs via UNIX. We have a Windows DataStage server and client(s). We currently *DO NOT* have at this client a Unix license. Unix is controlling the process of running ...

remsh and rsh are OS programs for running commands on remote systems. From your Unix box, use remsh to execute dsjob on the Windows DS server. First you need SFU installed on the Windows Server (Unix Windows Services). Yes, it looks like we're forced to go this route. Apparently there isn't a clien...

We want to setup a unix cron process that kicks of a DataStage job on windows. Apparently the dsjob executable is not part of the DataStage client so we could not just request a Unix client for this purpose. The other method is to write a dsjob interface (DOS Batch) to start the DataStage batch and ...

You never assigned HashFileName to it's input parameter. DataStage does not automatically make parameters available.

Add this line to the top of your batch code:

I found that sometimes it's easier to write a short python script to pre process the file.

Ascential support has entered an enhancement request for the ORAOCI9 plugin to support long datatypes.

ray, good point. This topic should be "moved" to the DataStage forum since 5.2 is almost legacy now.

You may want to open a case with support to see if there are patches that need to be applied.

Which specific release of DataStage 5 are you on?
I remember there being memory leaks in the aggregator and ended up getting a patch for it.

Topic: Oracle LONG DataTypes not supported in DataStage OCI plug-in. Product: DataStage Release: 2.2 Date Entered: 10/31/1997 Fixed in Release: Date Closed: Full Description: Oracle LONG DataTypes not supported in DataStage OCI plug-in stage. Resolution: None given. Workaound Provided: None given. ...

has anyone reported this as a bug?
It's a problem with both ORAOCI8 and ORAOCI9I plugins.

Since DS/OCI Stage does not support LongVarChar, I specified a Varchar with a length of 9,999,999. Will the CLOB be truncated if the length is longer? Is it possible to specify an <unlimited> size?

sounds like you should enroll in the basic DS Training.

We are having a problem telling the Aggregator Stage that the input into the Aggregator Stage. We have only 1 Key we wish to aggregate on. On the Key on the "Input" tab we said "1" in the Sort Column and "ascending" on Sort Order. but we blew up doing this. Is there so...

the Aggregator just pulls the "Last" value from the input link. I may try to instead replace it with a constraint but for now it works. If I go the constraint route I may have to put in a dummy row at then end as ray suggests.