DSXchange

That's right. Char(253) - also known as @VM - is a "value mark" which delimits values in a multi-valued field, as you've discovered.

Your structural metadata (on the Format tab) are not correct for the file. In particular the record delimiter (or record delimiter string) property is incorrect.

Enter the column name and data type, then double click on the row number of the Columns grid. You can also right click on the row itself in the Columns grid, and choose Edit Row from the menu.

Unless the rows in the files are exceedingly wide, I don't see why this error is being thrown. The approach is certainly valid. Maybe larger buffers would help.

It's an Oracle error, not a DataStage error. One (at least) of the objects referred to in your SQL no longer exists or, perhaps, you're not using the correct schema name.

StartLoopActivityName.$Counter

Why not just use the Manager client to perform a Where Used analysis on the table definition?

@ITERATION

Not necessarily. It will be wherever the DB2 database is.

If you're thinking of using a sequence in XMETA, there's still no requirement for the database (metadata repository tier) to be co-resident with the engine tier.

Then all you need to do is to use stage variables to compare the current row with the previous row (assuming that they're appropriately sorted and partitioned) and set an indicator column value on the output.

There isn't, but server jobs ARE supported in version 8.x.

Note the missing trailing quote character. It would appear that you're getting a "mark character" into the key of a hashed file. This is impermissible.

The word "perforce" is an adverb meaning "necessarily". I claim only that getting a block of keys out of a local state file will be quicker than sending a request over a network to invoke a database and issue a query against a sequence in that database then wait for the response ...

As far as I am aware the design of DSODB is not documented in the public domain. But you could open it in the database-specific tool (for example DB2 Control Center) if you want to look at it.

ORCHESTRATE is an object (much like SQLCA is an object). In this case ORCHESTRATE has the same properties as your input link, so that you can refer to column names as properties of the object, for example ORCHESTRATE.CUSTOMER for the CUSTOMER column.