When an RC/Migrator analysis is actually executed there are often utilities being used for unloading and loading data. When the execution is finished without any problems you don't want to have these files left behind to be an annoyance to your dasd management. How do we delete them effectively ?
RC/Migrator has two Analysis Options that will delete unload datasets and temporary datasets for you at the end of a successful execution.
DATASET DELETE OPTIONS UNLOAD ==> Y TEMPORARY ==> Y
The first is used to indicate whether you want to delete the datasets RC/Migrator creates to hold the unloaded data for the Unload/Reorg utilities. These datasets are dynamically allocated as part of the Unload utility process. If they are not automatically deleted, you will need to delete them manually later.
The second is used to indicate whether you want to delete the temporary datasets used by the DB2 utilities. If they are not automatically deleted, you will need to delete them manually later.
What do these options do in practice?
At the end of the load for each table you will see code like this:
.SYNC 40 'LOAD TABLE USERID.TBROLE' .CALL LOADDEL .DATA DSN(userid.DBCORP.TID37.DATA) <---------if you said Y to number one "UNLOAD ==> Y" you get this line generated from your unload. DSN(userid.DBCORP.TID37.DATA.D) <------the rest of these are to do with the loads. DSN(userid.DBCORP.TID37.DATA.E) DSN(userid.DBCORP.TID37.DATA.M) DSN(userid.DBCORP.TID37.DATA.U) DSN(userid.DBCORP.TID37.DATA.O) .ENDDATA
The above block is generated for each table. The number of work datasets will vary depending on your utility. CA Fast Load and IBM Load use different numbers of work datasets. The example above is for an IBM load Utility. Some sites have sometimes added the RLSE parm to the unload dataset allocation however during a restart operation this can cause trouble when the unload dataset has already been released and Batch processor does not know that this has happened and tried to possibly delete it before starting again.