DSXchange

Does that cover XML? If so, it can be published on www.datastagexchange.com (a sister site to this one).

You need to supply at least one query parameter from the row being processed. Set it from the job parameter in an upstream Transformer stage.

Does the update strategy "replace existing rows" (which equates to delete then insert) work for you?

You must connect the reference data via either ODBC stage or UV stage - these are the only stage types in server jobs that support multi-row return.

Is there any possibility that a job which has been run with unlimited warning can corrput the "dsrpcd" file? None whatesoever. This executable is not even in a directory accessed by DataStage processes. dsrpcd is started, typically, by the init process. DSEsettag.sh does override entries in the uvc...

Does either of your sources contain newline characters in the data? If so, have you adjusted the table definition to indicate that this is the case?

So the GUI is connecting but dsjob is not. What does that suggest to you? Try specifying authentication information (-host, -user and -password options) in your dsjob command line.

The message about large objects suggests you're trying to return a CLOB or a BLOB. This is not supported.

Can you post the code of your after-job subroutine, maybe we can find your glitch. hJob = DSJ.ME Message = "Link row counts:" StageList = Convert(",", @FM, DSGetJobInfo(hJob, DSJ.STAGELIST)) Loop Remove StageName from StageList Setting MoreStages LinkList = Conver...

By designing (a) so that they do finish and/or (b) that hang conditions are detected and managed.

1) No.
2) No. It's not necessary or even desirable.
3) No. There's no such thing as an OLEDB driver. Anywhere.

It's beginning to look like the software (whether DataStage or other) itself might be killing processes with SIGKILL signal, which is never recommended practice. Do you run from a job sequence that includes a Terminator activity? Is the machine heavily loaded when this occurs and, if so, does the op...

Sounds like getting a briefing on the basics from your Oracle DBA would not be a bad idea either!

There are many ways. Possibly the easiest is to follow the Row Generator stage with a Copy stage whose second output sends rows to a file.

Not true. Join stage supports more than two inputs. They are called Left, Intermediate and Right. I believe that it still executes pairwise joins into intermediate result sets, however, the same as most database engines do.

The Lookup stage is a composite operator; it creates two operators - one to load the reference (virtual) Data Set, the other to perform the actual lookup operation. You might benefit by increasing the buffer sizes on the inputs to Lkup3 and possibly Lkup4.