DSXchange

Moderator: please remove this duplicate post

Others: please respond to other post

You are getting whatever the ODBC driver reports during import. The only way to change that is to edit it afterwards (or perhaps to get a different ODBC driver).

Hello all, I am getting an error in a server job as out of memory space.what might be the problem? Thanks The problem is that you're out of memory space. Something on your system is eating memory (actually, everything is). You need to apply some diagnostic tools, such as vmstat, to identify the cul...

Not sure that this can be automated; but then every site I've worked at elect to use moderate publication, so perhaps there's something that I've not encountered.

Yes. And RT_STATUS*** as well.

These are hashed files, and therefore directory/*30 structures, rather than operating system files.

They have corresponding VOC entries.

There are other VOC entries containing last modified information.

The longer I remain in this business the crazier are the business "requirements" that I encounter!

Welcome aboard. :D When you run a DataStage job from Director there is a Job Run Options dialog that has a Limits tab. It is here that the limit of 50 warnings was set. Investigate the job log to find out what is causing these warnings, and correct the cause. The job should then run satisfactorily.

Ask the vendor about a class called Programming with DataStage BASIC which used to be on the books.

If they are no longer offering it, ask whether DSXchange (through its Strategies2Learn partner) might be able to provide such training.

None whatsoever. You can store anything into a job's user status area, including a dynamic array of arbitrary structure. Whether the activity $UserStatus accurately retrieves this, or truncates it to field #1 only, is something I've never tested. It ought not to - so if you can reproduce it and need...

Lose the illegal outer-join part of your query.

ProfileStage can generate a target database specification and mapping document from source data (analyzed), and also generate DSX files. So clearly it's possible.

But, again, why do you want to re-invent the wheel?

You can do this by asking for the "interim status" - the status that the job would have if its after-job subroutine weren't keeping it running.

The server Merge stage requires two File Name properties. It reads those files, and places the join specified between them onto its output link.

It does not support input links of any kind.

<irony>
Also "friendly fire". All complaints to billg@microsoft.com. Who will probably suggest replacing your ETL tool with the infinitely superior Microsoft product (believing everything that Microsoft marketing says). </irony>

The reason for that is that every "A" will generate the same hashvalue.

It's the same as in SQL - if you group by a column that contains only one distinct value you will end up with one group.

Prefer Round Robin or Random, or partition on a different key.