Since recycle can occur concurrently on different DFSMShsm hosts, it is possible for a user on one host to request recycle of a tape that is currently in use by recycle on another host. In this case the request fails, regardless of whether the request was for use of the tape for recycle input or recycle output.
Table 1 is a summary of possible volume contention in a multiple DFSMShsm-host environment.
Function That Has Volume | Function in Other Host That Needs Volume | Result |
---|---|---|
Migration | Migration | No contention. |
Recall | Migration | Migration selects a new target volume. |
Recycle (source) | Migration | Migration selects a new target volume. |
Recycle (target) | Migration | No contention. |
Migration, |
Recall | Recall initiates a tape takeaway, and periodically retries the recall. When migration or recycle processing is complete for the data set being processed, the allocated volume is released and a new volume is selected for migration or recycle processing. |
Recall | Recall | A specified period of time after an ML2 tape is mounted, and when a recall completes, the tape is taken away if a request exists on another host that is of higher priority than the request that normally would have been performed on the recalling host from the already mounted tape. |
ABACKUP | Recall | Recall periodically retries accessing the data set on the tape. If Recall is a WAIT request, after a delay, it initiates a tape takeaway, which causes ABACKUP to skip any remaining data sets that are needed by ABACKUP on the tape and process from other migration and user tapes. ABACKUP later tries to process those data sets that were skipped earlier. |
ARECOVER | Recall | Recalls are retried until the tape is available or a retry limit is reached. |
TAPECOPY (source) | Recall | If tapecopy takeaway is enabled, Recall retries for 15 minutes before requesting tape takeaway. If TAPECOPY has not completed its copy within one minute of noting the tape takeaway request, the copy fails. |
Migration | Recycle (source) | Recycle of this volume fails. Recycle of other eligible volumes continues. |
Recall | Recycle (source) | Recycle of this volume fails. Recycle of other eligible volumes continues. |
Recycle (source) | Recycle (source) | Recycle skips the volume because another host is recycling it. |
Recycle (target) | Recycle (source) | Recycle fails. |
Migration | Recycle (target) | No contention. |
Recall | Recycle (target) | Recycle selects a new target volume. |
Recycle (source) | Recycle (target) | Recycle selects a new target volume. |
Recycle (target) | Recycle (target) | No contention. |
Migration, |
ABACKUP | ABACKUP initiates a tape takeaway and periodically retries the data set. When migration or recycle processing is complete for the data set being processed, the allocated volumes are released. A new volume is selected for migration or recycle output contention, or a new connected set is selected if there is recycle source contention. |
Recall | ABACKUP | ABACKUP is retried until the tape is available or a retry limit is reached. |
ABACKUP | ABACKUP | ABACKUP skips over the tape in use and processes from other migration tapes. It reattempts to process the skipped tape later. |
TAPECOPY (source) | All except Recall and ABACKUP | Function fails. |
TAPECOPY (source) | ABACKUP | If tapecopy takeaway is enabled, ABACKUP retries for 15 minutes before initiating a tape takeaway. If TAPECOPY has not completed its copy within one minute of noting the takeaway request, the copy fails. |