DSXchange

Then please mark the topic as Resolved... unless you want to talk about how you horked up the log file in the first place.

Ok, so actually no input examples, just output with the lead function. Fine. This is just bog standard 'group change detection' done with stage variables. As noted, you can't look ahead but you can easily look 'behind' and stash values from previous records. So you'll need variables to save the curr...

Be restarted yes, but not magically restart on its own.

NO. From the Adminstrator client, connect to your project and navigate to the Command option. Once there, issue CLEAR.FILE RT_LOG635 which will empty the log. Then via the Directory re-establish any Auto Purge settings you had in that job's log. A 'blink' is a 'backwards link' and getting one means ...

'Exactly similar'?

Please take a moment and show us, based on the input data you posted earlier, what your output data needs to look like after the transformations are performed.

Parallel Job Developer's Guide, Chapter 28, Modify Stage.

Welcome! The use of a dblink would basically be transparent to DataStage and requires adding "@linkname" to the table name in the from clause. At least that's how it works in Oracle, assume others are similar if not the same.

In my experience, a dblink is neither faster nor a silver bullet.

You could also do a 'LIST.DICT DS_AUDIT' from the Administrator client. Search the forums for COMO to find ways to get that output in a file.

It's not in the public domain, but you can do a 'LIST.DICT' on it or just a 'select * from' and see what you get.

Well, that sucks. Anyone know if 'correcting' this is anywhere in the DataStage roadmap?

I knew what you were talking about. The sequential file stage can neither parse XML nor display the complete contents of a LongVarchar. You could, however, view it if your incoming XML was pretty printed but as one very long data stream? Nope.

Not really a 'double negative', just an opportunity to disable at the job level an option at the Sequence level. Use it in a Restartable sequence for any job that should always run each time, regardless of the current checkpoint status.

How do you capture the counts in 'after-job' in Job properties.The procedures listed there do not contain 'DSGetLinkInfo'. If this is another lovely 'not in the v8 online help' thing, then use your pdf documentation files - the Advanced PX, Server Developer and BASIC pdf all mention those functions...

ExecSH, dsjob -report, redirect output physically as a file!!!

You have an 'employee', 'department' and 'error' table and they all have the same metadata? I find that hard to believe. Seems like you'll need to build X jobs and then have an intelligent process that reads your list of tables and determines which job to run based on the name found. Perhaps using s...