DSXchange

Even via datastage you can get this done but we need logic. I bet even the stored procedure executes some sort of logic. If you can get us that logic maybe someone can help you design you, or even walk you through the entire job.

The issue is certainly at the database end. It allows a certain amount of bytes but more is being inserted. Its just a warning and i believe the data is getting inserted. Check the table and see at what point it chops off the rest of the data.

I like to do everything via job parameters. Not like to mess with environment variables. No special reason, just a person choice.

Is that all that your script contains, i bet not. I hope you are sourcing the dsenv file within your script. Also, is this sequence multiinstance? And what exactly do you mean that the control is not being returned back to the script. How many jobs are there in the sequence, do they execute properly...

Rows into columns huh, that sounds like a job for the pivot stage. Look into the pivot stage along with the row merger.

So you have different delimiters. Then i suppose you will be reading both the columns as a single column, right? In the transformer use the INDEX() function to detect a pipe. If present send it through the link which eventually gets loaded in the database. If no pipe is found, send it down the rejec...

You said this is happening in production. Was the same experienced in development or even system test environments?
What are the after sqls? If they are deletes, and the table size is huge, then that might be taking a long time. And during the execution of after sqls.

You can also get DMBSCODE of rejected records. But even then it will just be a code. You can have a small table with dbms codes and what it means. Once you have the dbms codes, you can do the lookup and get the message. You can also write a bat script and get all the warning messages from a particul...

YOu can check the log files in the director. But if the message file shows that the data got loaded then make sure you are in the same environment, same database and same table.

Another piece of advice. Split up your job. Modularize it. It might not speed up things but will sure make the jobs more readable, restartablity will be easier too. Plus you will get a big help in the unix sort.

Search this forum for more detailed descriptions on how to generate surrogate keys. WHich is what you want, a running sequential number. You will even find ways to get the max from the table and then add one to it for your further inserts. Search is your friend.

I know right, thats Ray's signature question. And i just love his initial reply, "Nothing is not the correct answer", really cool. Coming back to the OP query, check if the table you are loading changed. Check with the DBA if he put any triggers on it. If a trigger fires on every insert, that tends ...

Did you try it without the keyword 'NONRECOVERABLE' ???

Well, your load replace syntax seems to be correct but db2 is reading something more than that. Thats why its giving you SQL0104N which just means its a syntax error. Even the SQL0104N help is not very helpful as it just says, hey, your syntax is messed up, fix it. Close that window, open a new one ...

I like the heading, heh, 'same problem'. Wow, Even the RMM stage cannot tackle this one. It just gets better and better here doesnt it