IBM Mainframe Forum

sergeyken wrote:[I wonder how the last V record is fit into the end of U physical block? Either the first part of record is placed at the very end of one block, and the tail of the same record moved to the beginning of the next block? Or the whole record is moved to the beginning of the next block when it does not fit the end of previous one?

The only time you have data that corresponds to sergeyken's description is with VBS data. This is handled in the RDW; the RDW for the last segment in the physical record is flagged as either the first or middle segment, and the first RDW in the next phyisical record is flagged as either the last segment or another middle segment. Note, too, that IBM refers to these RDWs as SDWs (Segment Desriptor Words).

Regular V format data has complete logical records in every physical record

steve-myers wrote:

sergeyken wrote:[I wonder how the last V record is fit into the end of U physical block? Either the first part of record is placed at the very end of one block, and the tail of the same record moved to the beginning of the next block? Or the whole record is moved to the beginning of the next block when it does not fit the end of previous one?

The only time you have data that corresponds to sergeyken's description is with VBS data. This is handled in the RDW; the RDW for the last segment in the physical record is flagged as either the first or middle segment, and the first RDW in the next phyisical record is flagged as either the last segment or another middle segment. Note, too, that IBM refers to these RDWs as SDWs (Segment Desriptor Words).

Regular V format data has complete logical records in every physical record

The actual question is not about the STANDARD processing of VB, VBS records, but: how the STANDARD utilities do handle the NONSTANDARD format U, which they initially were not supposed to handle without some tricks?

As a side point, we finally found a solution to our underlying problem (keep or display the RDW for binary transfers outside of z/OS). Originally we were using a utility called Universal Data Mover (UDM); we figured it has an option to be added to the 'attrib' command: