My site is running an MVS server at level 3.1.2.0. We just went through a
disaster recovery test. We used the recovery plan file produced by DRM to
rebuild our server and then started doing restores from 3590 copy pool tapes.
At one point we had three restore processes competing for two tape drives. A
'query mount' command showed that volume 600058 was mounted and in use, and
inspecting the tape drives showed that 600058 was in fact mounted. A series of
'qeury session ... f=d' commands showed that one of the restore processes was
waiting for 600058 to become available, but that neither of the other restore
processes was using it. The situation resolved itself when a mount wait
timeout occured after an hour. At that point 600058 was dismounted, the job
that had been waiting for it switched to waiting for a mount point, and a
different tape was mounted for another restore process. Eventually a mount
point became available, 600058 was mounted, and the restore process that had
been waiting for it read the desired data from it with no further trouble. Is
this a known problem?