DSXchange

Be patient. DataStage is capturing these details automatically. Once the job completes use a routine, the dsjob command or even another DataStage job to extract the information either from the job log or using the DataStage API.

Does the file exist? How was is created? For this to work, it must - must - have been created with a Data Set stage; it is a common mistake to believe that one can create a Data Set with a Sequential File stage.

Check with your UNIX administrator to determine whether the file unit table is becoming full.

Depending on what stage type you are using, and how you are sending the rows to the table (upsert vs write), there may be an additional column called sqlcode on the reject link.

However, capture of all warnings is probably easiest done by post-processing the job log.

Xpert, you really are confused. The original poster did specify Windows, but it's 7.5x2 (not 7.5.2) that supports parallel job compilation and execution on Windows. And please don't use SMS-style abbreviations here; English is not everyone's first language, so good English is difficult enough. dstag...

Welcome aboard. :D This is highly unusual. What symptoms lead you to this conclusion? Possibly the best solution is to restore from a recent system backup, but don't do that unless DataStage is shut down. However, don't rush in - your diagnosis may not be correct. Answer the question and we may be a...

There you go! I don't use $PROJDEF or $ENV in development until the very very end. But I do switch it before testing, so there's one more thing with which to annoy the testers: "but that's how it will be in production!".

No advantage whatsoever. A Copy stage would be much less impact as a placeholder, and also give you the ability to suck out a few rows (for audit/sampling) to a second output link. Indeed, if it's a straight through copy, the Copy stage is likely to be "optimized out" when the job is compiled.

Not bad - use Code tags to get a grid. The DataStage roles do not confer any rights at all. There is no Administrator role; the DataStage administrator is the user dsadm or the superuser on UNIX, or a member of the Administrators group on Windows. The other three roles are gates. For example, if you...

Get your DBA to purge or enlarge the log.

Learn not to rush, grasshopper. :D

Look at the comments in the generated source code (RT_BP2245/JOB.1923857383.DT.1413534682.TRANS1) to find which link is referred to by Pin%%V0S5P2.

There is no circumstance under which the routine result ($ReturnCode) will be overridden by the master sequence, unless you explicitly code for that to happen in the expression that delivers the value to the controlled job's parameter.

It would be much easier to solve if you could find some mechanism for deriving this rogue column somewhere in the main stream. There is no way to accomplish that you ask, except for a second job that updates the target via the Dynamic RDBMS stage.

A Hashed File is simply another mechanism for implementing a database table. So imagine that you're inserting new rows into a database table. Create the hashed file with a UV stage but edit the DDL so that you get auto-generated key values. CREATE TABLE ErrorTable ( SurrKey INTEGER NOT NULL DEFA...