DSXchange

Are you really having 80 open connections?

Not even that. Just add the word DISTINCT in the Derivation cell of the first column in the Columns grid. You don't need user-defned SQL.

Go to Status view.
Disable display of Categories in the View menu.
Optionally sort by Status. Or sort by job name.
Print your report.
Re-enable display of Categories.

Having the DataStage deadock daemon ( dsdlockd ) running is my preferred way. You can also run it in probe mode (dsdlockd -p). There is a file called dsdlockd.config in the DataStage Engine directory where you can configure the deadlock daemon to auto-start. There is a menu called DEADLOCK.MENU with...

Examine all the square-bracketed pieces in the first message. [DataStage][SQL Client][ODBC][Microsoft][Controlador ODBC Excel]Precision value not valid The message is being generated by the Microsoft ODBC driver for Excel. Therefore the limitation is in that driver, not in DataStage. Get in touch wi...

When's the interview?

What kind of job? What stage types are involved? The answer will change depending on these factors.

What is the data type of the input column? It needs to be a string of some kind for Maveric's suggestion to work; if it's a date or timestamp you need to apply appropriate conversion functions. Given that you have one- or two-digit days and months adds complexity.

It's free space while the job is running, not before or afterwards, that is at issue here.

One of your file systems has filled during the running of the job. Probably scratch disk, but there's not enough information to say so for sure.

murari83 wrote:I am planing to implemented in user define stages(custom,build,wrapped).

Why?
What specifically do you want to do that the myriad stages supplied with the product can't do?
Are you even aware of how the three user-defined stage types differ from each other?

Easy. Get the NLS settings right. In DataStage, this means the TIME category of the locale.

Why do you specify "without using Director"? It's by far the easiest way.

The status of each and every job is in a separate table in the Repository.

You use dslictool AFTER logging out the process.
You log out the process using the Cleanup Resources utility in Director or one of the Logout options from the DS.TOOLS menu - you NEVER kill a DataStage process with kill -9.

Search can certainly help. Just leave out the numeric part of the name when searching for the error message.

Using the installer that ships with IBM Information Server. You will also need the licensing XML file that IBM provides. You will probably also need more hardware.

Yes. No. It's horses for courses. Control-M might be preferable in some circumstances, because it's an enterprise scheduler - you can schedule other, non-DataStage, activities with it as well as DataStage activities. There's probably a good compromise, where you might do everything in job sequences ...