DSXchange

The Furbeowulf project is achieving massively parallel processing design and implementation with commodity products. Their use of a "watchdog" processor to manage both process allocation and network traffic is particularly clever.

Please post the detail of the error message, which includes information about what stage it occurred in. You also need to tell us what kind of stage was involved, and an overview of the job design would help also. Ultimately we need to establish what was trying to be opened, so that we can narrow do...

The Oconv() function supports range checking. Oconv(value,"Rn,m") returns the value itself if it is between n and m inclusive, or "" otherwise. Multiple ranges can be tested with a single Oconv() function, for example Oconv(value,"R1,10;101,110;201,210") returns t...

"Too big" might be too much data for a transaction. What is your commit rate (rows per transaction) setting? You could perhaps try committing more frequently rather than splitting the source file.

Even simpler might be to use a before-job or after-job subroutine that invokes ExecSH. The Input Value field contains the UNIX command line needed to execute your script.

A more graceful method would be to use a job sequence containing an Execute Command activity.

DataStage does not support stand-alone BASIC programs - in DataStage you must only create routines that can be invoked by other DataStage components. Yes, you do this in the Routines branch or in the Job Control tab of job properties.

Search the forum for information about the UNLOCK command. You must have Administrator access to use this command.

The question might even be about ulimit.

Sasikiran, DataStage is a complex product - you really do need to be more specific with your questions. Welcome aboard, by the way. :D

I have one of those, but it uses SAP prescience (VE vill tell you how to run your business!).

That's a permissions problem. You will probably need your DB2 DBA to look at it. It's not a DataStage file.

You MUST have access to the before Data Set - it's physically impossible to determine which rows have been deleted by inspecting the after Data Set.

Use a Reject link to trap the line that does not correspond to the metadata described in the schema.

My guess is that you're writing to a Sequential File stage, and there is a node pool called export in the configuration file but none of the nodes in this pool is the conductor node (that is, also belongs to the default node pool).

It's just a warning signifying that a time request from a system call returned a value that is earlier than the start time for the operator. By 0.353 seconds. It may be that the start time is obtained via a different system call, it may be that something odd happened with your system clock(s), it ma...

From The Data Warehouse Lifecycle Toolkit by Kimball et al (pages 212-216 in the 1998 edition): Factless fact tables are the preferred method for recording events in a data warehouse where there is no natural numeric measurement associated with an event. Factless fact tables also are used to guarant...