DSXchange

If you follow best practice logs ARE stored in RT_LOGnnn files in version 8. IBM realized that storing them in the common metadata repository imposed to great an overhead, and now provide RTLogging and ORLogging settings (per project) and recommend that you use RTLogging - that is, logging in local ...

You could remove it from the dictionary file for the rule set or change its default width in that dictionary, assuming you're using an editable rule set.

raghav_ds wrote:I have searched the documentation, but could not find any information on XML Processing instructions.

Anybody worked on constructing the XML Processing Instructions in DataStage.

Did you look in the XML Pack and XML Transformations manuals?

Use any format you like, provided it's the format expected by whatever application is going to consume the .dat file.

Key change is set up as an output column from the Sort stage. This column is transmitted to the Transformer stage on its input link.

The magic word is probably "unbounded".

Check environment variables TMPDIR and, possibly, TEMP.

I doubt that this is anything at all to do with XMETA - it's more likely to be the number of subdirectories in the project directory.

Just make the column wider on the output link Columns tab.

Are you running the 29 jobs consecutively? Running them simultaneously (or in bunches, if you don't have the capacity for all 29) should cut some processing time.

Learn about partitioning. Parallel stages that are not constrained to execute in a node pool will executed in every node in the default node pool. The score will show which node handles each sequential stage. Partitioning controls which rows of data go to which nodes, not the nodes on which a partic...

Your first requirement could be met by a Remove Duplicates stage preserving the First record from each group, assuming the data are sorted correctly. You specified removal of the asterisk, but the output in your example preserved the asterisk. Removal is most easily effected by a Convert() function....

No, because sorted data is the only efficient way to perform comparison.

What's your source? Have 0xFF characters been used as fillers somewhere?

Is there an NLS button in the General tab of your Administrator client?

This almost certainly suggests that the C++ compiler has not been correctly installed or that, if it has, it (or its environment variables) has not been correctly configured. Do you get compile-time errors with these jobs? Does the Transformer stage have any flags set in the Build tab of stage prope...