DSXchange

Isn't that basically how any subscription based CDC process works? I haven't been on the receiving end of every one on the planet but that's what I recall from the ones I've seen.

Yes, the Key columns go into the where clause of the delete so it will be based on it. However, as Arnd notes only values you send in will be deleted. You'd need to clarify the relationship between what you are trying to delete versus the "random subset" left behind. For example: 1234 1234...

peep wrote:i m sorry i cant read all of your post.
it was truncated .
can you please inbox?

No. That's not how it works here. Premium posts require a Premium Membership to read.

Start by actually reading the posts you can see rather than making people repeat the same thing over and over.

Like.... what? Which part of your example is the "sequential file column value"?

As noted, unless you pass the value as a parameter you'd need an external table to use the file as if it were a table in a SQL query.

The ones you're using when you said "This can be done with the help of 2 or 3 stage variables" - they're the ones in there that start with an "@".

So what happens when you just do this: $DSHOME>../PXEngine/bin/orchadmin rm /IIS/....node1/datasets/jobname.ds :? FYI: I can't tell if you are including everything from the comma onwards in the command when you run it or if that is just meant to be informational. And I'm assuming you will replace jo...

OK... for the umpeenth time, you are not deleting 'descriptor files' you are deleting datasets which the descriptor file is pointing to and... describing. Yes, it will also be deleted during the process but the goal of the command is to remove the dataset associated with it. For the "rm" c...

One of the reasons I very rarely used them darn IPC stages.

What aggregation method are you using, hash or sort? Hash is the default but since the data is already sorted you might try 'Sort' instead. It depends on the number of groups you end up processing...

rajneesh75 wrote:I am curious if server is on another machine(windows or unix) would the same command run/not run the job on that server given that i am giving the hostname:port no in the command

Yes. As noted.

PhilHibbs wrote:He asked about running them on the workstation, not on the server.

From the workstation, not on.

Why not just do it in your source SQL? I'm assuming a database table, of course. Otherwise use the Aggregator to do the exact same thing: group by A & B, maximum(C).

First: I found 19 matches when I did an exact search for "This directory is not set up for DataStage" so we've talked about this before. Secondly I hoped you answered that question with a big "N" for no. As a search would have shown, you should only be invoking the DataStage shel...

Welcome. Yes, you can seeing as how the jobs can only run from the DataStage server. In the past, we had to get our hands on the dsjob executable from a Windows server release plus the supporting ddls in order to do that but from what I recall it is now automatically installed on each client regardl...

In 'older' versions setting the default would only effect new jobs. My understanding is that behaviour changed in a more recent release such that jobs that are still using the default (i.e. have not been overridden) would use the new default automagically. Can someone confirm / deny?