DSXchange

You can add a second Transformer stage to your job design to filter the backdated records onto a separate stream before progressing with whatever processing is currently in your job. The backdated records could be processed in the same job, or written to a staging file for processing by a subsequent...

I have no intention of doing so in this case. The DataStage API remains, and can be used reliably to interrogate the log. The approach using a job suffers from the need to determine the log table name. An approach using the DataStage API does not.

The repeats are because the warning is generated on each processing node mention in your configuration file; the ",0" indicates partition number 0, for example. It is a warning not because anything is wrong, but because something unusual or unexpected is happening at run time; in this case you are w...

Manuals from IBM training classes and Orchestrate Operators manual page 7-8. The latter lists all the recognized regular expression characters and "%" is not among them. IBM Center of Excellence claims that the Transformer stage is now faster than the Filter stage because the latter is interpreted; ...

But could be on a different machine. The service-oriented architecture of version 8.0 requires that you access metadata through the Metadata Delivery service; the actual location and structure are thereby hidden. And it (the job log) may not be one table any more - none of the supported databases fo...

You seem to have two IN operators in one clause: WHERE CTR_POL_NUM IN (CTR_SOURCE_SYSTEM IN ('HC' , 'HF') The "did not finish OK" message is a strange artifact of Routine activity; it uses the same convention as Execute Command, regarding the return value as an "exit status"; if it's non...

Indeed, all of those and, depending on circumstances, one or two others as well. You will probably have to convince your DBA that SELECT privilege does not allow anyone to do any damage.

The Filter stage is a strange hybrid. With LIKE it expects regular expressions rather than SQL wildcards. Therefore what you need is

It does this to ascertain how many tables meet your filter criteria. It then (if the count is non zero) performs another SELECT from DBA_TABLES to return a list of tables so you can choose one to import. Therefore, yes, you do need SELECT privilege to DBA_TABLES, and about eight other system tables ...

Not might, will cease to exist. However, I am sure there will be (is) some equivalent where you can invoke the Metadata Delivery service.

Your input record has nine bytes but your metadata (record schema) defines only seven. Can only guess at the second one, but it looks like you have specified - or implied through "preserve partitioning" - an explicit partitioning algorithm, so DataStage is warning you that its default (using Entire ...

You might try Vincent's suggestion, but note that the default value clause needs to be enclosed in curly braces and it and the "table" in square brackets.

The *.fs files are plain text. You can view them with the cat command, for example.

The log records are in a database table within the Repository. The name of the table is RT_LOGnnnn, where nnnn is the job number associated with the job. You can search the forum or use LIST DICT RT_LOG to get the column names available to query the log; you could create a DataStage server job to do...

Where did you find the "$" syntax for a CALL statement? This normally means a GCI call, which means that a reference to the function must exist within the General Call Interface mechanism. How are you invoking your server routine in a parallel job? Is it within a server shared container, or within a...