DSXchange

Write them to a hashed file using the relevant keys. Last one in wins. The Hashed File stage implements destructive overwrite of keys.

Or use a Transformer stage with stage variables to detect change (or lack of) in sorted input.

Not an environment variable, but you can add more directories (from other file systems) as scratch disk in your configuration file. If the space is being demanded by sorting (necessary for the join stage, ,and inserted by the Framework if you neglect to do so), then you can force the sort to use mor...

I hvae mapped Column3 in the output of the change capture stage.I think that it is not the problem of mapping or not mapping the column (Column3) in the output of change capture stage. Well, Mister Expert, if it's not that what is it? Putting a Column3 on the input link might help, but only if it h...

WHAT "sh file"? We're talking about a DataStage routine here.

Because you have not posted that code, we can not determine where the return value of 1 is coming from. Therefore, in light of that unwillingness, YOU have to be the detective.

The only possibility is unintentional or intentional sabotage. There is nothing AT ALL in this product that would clear out everything in the DSEngine directory. Indeed, because certain files (such as SYS.MESSAGE) are open whenever any DataStage process is running, it would be almost impossible to a...

Your "the only way for a "all or nothing" strategy is to have a common Oracle Stage to load data with Rows per Transaction = 0" would not work for me - I am using DB2. Read again what Craig and I have said. It's robust and reliable. Then get yourself some heavy duty protection because your DBA is go...

... or a job sequence that runs parallel jobs.

There's nowhere for Column3 to get its value from, so DataStage is alerting you to the fact that it is being given its default value.

For that u can use (Head -2) or else go before routine head -2 gives the first two rows - it does not eliminate them. Some shells allow the form head +2 which will eliminate the first two lines. Otherwise use @INROWNUM > 2 as a constraint expression in a Transformer stage downstream of the Sequenti...

Your ExecuteScript routine is returning the value 1. Presumably somewhere in the routine the statement Ans = 1 exists, and is being executed in this case. Conventionally a non-zero exit status in UNIX/Windows indicates other than a successful completion. DataStage extends this convention to the valu...

APT_DUMP_SCORE won't help you with that - it merely posts the score into the job log. No, you may need to adapt your job to include debugging statements (such as echo in a before-job ExecSH subroutine).

Let me clarify: The source stage metadata needs to be supplied at runtime as well. Has anyone actually verified that? On one class that I attended it was asserted that RCP enabled yielded the same result as "SELECT * FROM tablename" at least when using the Enterprise stage types. This was given as ...

Probably you have (correctly) de-duplicated the Update inputs, but have duplicates in the Master input. As each match is made in a Merge stage the update row is consumed, so it's not there when the duplicate comes in from the Master input (and therefore you have an unmatched Master).

I am getting error.Please suggest this error asap.... My suggestion is that the error is "Could not load dsstpora.so" Terms like "urgent" and "asap" generate negative karma, and demotivate us from responding. DSXchange is an all-volunteer site, whose members post as and when they can, if they have ...

When the job starts the score is composed on the conductor node and then has to be distributed to each of the nodes mentioned in the configuration file. This step (distribution) has failed. You probably need to look at how you have rsh set up.