DSXchange

Welcome aboard. :D
Instead of using an OLEDB stage, try using an ODBC stage. When you import the "table definition" from Excel, make sure to check the box that says "include system tables": Excel ODBC driver presents worksheets as system tables.

The pronoun you're looking for is "you". We have a user registered called "U" who might take offence.

Administrator client

Select project

Properties

Unprotect Project

"Clear Status File" does not touch any job designs.

Whatever deleted your sub-job earlier was not "Clear Status File".

Of course it is, you have not changed anything. The from previous run" message will include additional diagnostic information about the cause of the problem.

No. Use two Lookup stages. Run the reject links from them through a Funnel stage to accumulate the rejected rows. Or make the lookup failed rule on the first Lookup stage Continue, and perform the second lookup conditionally on one of the first lookup columns not being NULL.

Surely both these functions require the date to be converted as an additional argument?

Every stage is written to the DataStage Plug-in API. Plug-in really only means "optionally installed". With parallel jobs you will find that, by default, every free stage is automatically installed - the only parallel plug-in stages are the ones for which you have to pay extra.

Look at the score. The import operator is the Sequential File stage. Are you sure the field is defined as a string, not a date, there?

The process had a problem reading from a socket. Sockets are used between player processes during repartitioning in a multi-machine environment. You need to do a little more detective work - which stage generated the error? You may need to disable operator combination to narrow the focus of your inv...

No you don't. Just specify the 20 in the data format string. There are examples in the manuals and on-line help.

The \r indicates that the file originated on a Windows system. Change the record delimiter string to \r\n

WAT is a Buddhist temple.

The word you are looking for is WHAT.

Finicky things, computers.
LIST DICT RT_LOG to display the available column names that you might use in a query. FULL.TEXT is among them, FULL.NAME is not.

Sequential file is fastest, and does not remove duplicates.