A VOLSEARCH job asks for the first full tape to be mounted three times. Why?

Products

• VM:Backup for z/VM
• VM:Backup High-Speed Disaster Recovery Option (HiDRO)
• Mainframe VM Product Manager
• VM:Manager Suite for Linux on Mainframe
• VM:Manager Suite for z/VM
• VM SUITE

Issue/Introduction

Why does VOLSEARCH job ask for the first full tape to be mounted three times?

Resolution

A VOLSEARCH restore jobs consists of two main phases:

1. Volser scan phase; builds JIV and JRB control blocks
   
   
2. Read input tape phase; reads the tapes for restoring

The read input tape phase is then broken into two phases based on how the incremental backup instances should be applied;

"phase one" - oldest to most recent (e.g. REPLACE, CLEAR, etc. options), and
"phase two" - most recent to oldest (e.g. NEWFILE option)

Note: "phase one" is really any non-NEWFILE option

NEWFILE is the default restore type for CMS data. (RESTYPE record)

CLEAR is the default restore type for SFS file space data. (RESSFS record)

When both CMS data and SFS file space data are in the restore template and if OPTIONS RESTYPE NEWFILE are specified, then three mounts will occur:

1. Volser scan phase
   
   
2. Read input tape phase one for CLEAR data (SFS file space data)
   
   
3. Read input tape phase two for NEWFILE data (CMS data)

When both CMS data and SFS file space data are in the restore template and if OPTIONS RESTYPE REPLACE is specified, then two mounts will occur:

1. Volser scan phase
   
   
2. Read input tape phase one for CLEAR and REPLACE data
   (SFS filespace data and CMS data)