DSXchange

lstsaur wrote:If you have licensed with the Federation server, then JDBC is used for every connectivity.

This may be what the OP's interlocutor was referring to.

You have to set it to True. Have you brought the environment variable into your job as a job parameter? Its default value is usually False, though you may have changed that to $PROJDEF. When you run the job, make sure it is set to True. Check in the second event logged for your job that it is set th...

I was told that when accessing a UNIX source from a UNIX ETL server, it's not really ODBC and JDBC would work better. Whoever told you that has no idea what they are talking about. What gives? Are the ODBC connections we set up in UNIX 'real' ODBC connections? Yes, they are "real" ODBC connections....

I'd be looking more closely at how Espresso gets the job status back and makes its decisions.

Why should you give them any tools at all? They already have some perfectly good tools for monitoring the system. You can, of course, identify the processes associated with DataStage jobs, which will allow them to target their monitoring. In server jobs the process ID of each active stage can be got...

Welcome aboard.
You need a simple script.
Step 1. dsjob -ljobs projectname
Step 2. Iterate through this list using dsjob -jobinfo projectname jobname to get the start/finish times. Compare these to current date and you have the desired solution.

That and "unable to insert record into table" is occurring on both your processing nodes, so it's either a data problem or the table space being full in the database server (your DBA can advise on the latter, the complete log message may have additional information).

If anything it would be a VarBinary or LongVarBinary but, as Craig noted, BLOBs are not supported.

Had you bothered to search you would have found that the Project menu has a Print option and, in the dialog that opens, you have the opportunity to select a check box that allows you to print selected events to file, either in summary or detail view.

The sparse lookup is a last resort, when you can't find any other way.

Try the "bucket" technique that I suggested - you may be able to use a normal lookup then.

You asked the same question in the General forum; I answered it there.

Yes, here is probably the more appropriate forum, but what's done is done.

Think about how a sparse lookup works. It will necessarily be slow. Alternatives include doing the join the the database, if both data sources are in the same database instance, or using a Join stage. Are all 500 million rows needed? You might create a "bucket table" containing only the keys in the ...

It's a run-time error. The problem occurred on the node identified in your configuration file as node1 which, from the [0,2], I am guessing is the first node name in your configuration file. You now need to identify which process it is. Are there any other warnings logged? Enable environment variabl...

Welcome aboard. First and foremost, requirements will be driven by the business needs. Generically you need to determine what they need to know, how it is to be measured (the "facts") and along what dimensions (for example time, geography, product, and so on) the analysis is to be performed. This in...

That statement is complete rubbish. And 100% wrong.

It's always "hashed" file.