DSXchange

Huh... I haven't seen a LONG since the 7.x release of Oracle, anything nowadays should use a CLOB. Typically. Some kind of Large Object anywho. You may be able to leverage the DBMS_LOB package, specifically the substr function there, to pull out just what you need - if you know the max data size in ...

You've definitely got a problem on your hands. Not only are Oracle LONG fields way obsolete and a PITA to deal with in any tool, they are not supported in DataStage.

What is the version of the Oracle database you are accessing?

Two ways.

1) Pass in the value to use as a Job Parameter and then use that parameter in the output filename.

2) Create a fixed filename in the job and then 'after job' rename / timestamp the file via your O/S, either directly or scripted.

Right. That or use a tool like Version Control for promoting changes as it doesn't nuke everything in the process.

They are gone as import deletes the old job, including the log.

If you've got it working fine now... I certainly wouldn't worry about changing it. Just keep that in mind as a possibility next time.

No, talking Server here. Did you really mean 7.5.1 for the version?

You certain you are using the ODBC stage for a reference lookup and that's the stage (on the left-hand side of the editor) you clicked on?

Most of the time, yes, which is why it is an option.

Once he finds the right place, a single click will get him exactly what he wants so no need to redesign the job.

Ok, let's start over. The Execute Command stage is a Sequence job stage. My post was for Server jobs. It would be best if you fully explained exactly what it is you are trying to accomplish. The more details you give, the more relevant the answer will be. Looping in a Sequence job is accomplished us...

Sorry, didn't mean you to do that on the canvas, you need to click on it from inside the Transformer editor then select 'Link Properties'.

Right-click on the reference link itself, the option to enable multiple rows is at the bottom of the dialogue box from what I recall.

Only the UV and ODBC stages can return a 'multi-row result set'. Can your 'reference table' be accessed via ODBC? You haven't stated what database you are using.

There is a reference Link property you'd need to enable after you set that up.

That's because the SP2 patch was not a 'crash' fix but rather a compatibility fix for Service Pack 2 which caused certain stages to not open. You need to get the specific 'Designer crashes when editing Sequence jobs' patch.

That information really isn't public knowledge. You are meant to retrieve what you are looking for from the job logs, either via the various 'DSGet' functions provided for use in a job or by using the equivalent options of 'dsjob' scripted from the command line. For example, I would think a search f...

And you get them from... ?