DSXchange

Code in the production environment must not be altered.

Your actions also imply that you are not sure of the change. So doing that in production adds more risk.

You may be better-off doing the change in some test env and ensuring the change works before migrating to prod.

Did you check the permissions for that user?

Did you try using any other db? You may be able to analyze the performance with help of the DBA. What is your disk and memory performance like? You can do a selective load into the uv by joining the source for uv with the source for your job. More information on the data may assist others in giving ...

As ArndW suggested, you need to identify the most occurring period using sort or agg or any external method.

I will suggest using that as a parameter in your next job or use Unix command(s) to prefix to each record.

Write the link from the source stage to a file.

There can be lots of reasons for having less rows.

Your assumptions are correct.

Check the space in the tmp and project dir.

You also need to analyze the reason and business logic for this variation.

Best will be to write into a seq file order by the keys and do a diff.

Did you try using the DB stage?

First check will be with stage variables.

Did you verify that the ODBC points to the correct SQL Server DSN?

Simplest will be to bounce DS.

The records are of varying columns.

The records are of varying columns.

Job sequence loop is available from v7.5 and above.

The value you have can directly be written into the db.

Do you get any error message?

Note - Please post in appropriate forum.