DSXchange

I would imagine that those "Internal errors" in their C code would need to go back to the vendor for clarification / issue resolution. Check with your official support provider, perhaps there's a patch available.

As I noted, I don't kill them but rather use DS.TOOLS to 'log them out'. Hopefully there's a difference there... and perhaps even a handling of the LOCK table as well.

Yes, you do know the meaning because you read what Arnd and I posted... right? And if you read them, you'll know what needs doing to get more help.

Please start your own topic.

You'll need to set it to Manual. Then either use your own (precreated) ctl rather than the one it generates or tweak it on the fly (sed, perhaps) each time. Then you run the sqlldr command.

:idea: People, please wrap your ascii art with 'code' tags to preserve all of the lovely work you've done lining crap up. That and liberal use of the 'Preview' option. That being said, I have no idea what you mean by this: The First HASHFILE1 is looked up with TRANS! and Second HASHFILE2 is looked u...

Mmmm.... shiny.

Specifically, one could use ExecSH before or after job to run a shell script (or any other operating system command) that is on the DataStage server. There are other ways, of course.

An exact search for CLOB here should help answer your question, seeing as how you're not the first person to ask.

Right... in a Server job that message is logged by the active stage which usually means the Transformer is named in it but that doesn't mean the problem lies there. And as Arnd noted, all it means by itself is that the job aborted, something you already knew. There should be other messages logged wi...

Nothing jumps out at me as being incorrect but then I've never had a need to connect to Informix from DataStage. Have you tried using that connection in an ODBC stage in a job yet? Something simple that doesn't require much metadata be manually input into the stage? Wondering if the issue is specifi...

I've done the same thing when a project gets 'locked' like this in versions prior to 8, I don't see why the same game would not be approriate here. I get a list of defunct / old connections and start logging them out via the DS.TOOLS menu until people can get back in. Any session that doesn't close ...

No, this option is specific to the stage so it does not affect any other aspect of the job. Not according to our testing. We recreated the deadlock with the check boxes in both states and what should be a fatal error is a warning in both cases. :? Not sure how your testing relates to what I wrote a...

You (no pun intended) mean something already written and willing to be shared? No to both on my part, unfortunately. :( Assuming you haven't done something like ETLStats to have your run stats stored off elsewhere, I would imagine you'd need to work through all job logs and query them individually t...

No, this option is specific to the stage so it does not affect any other aspect of the job.