z/OS DFSMShsm Implementation and Customization Guide
Previous topic | Next topic | Contents | Contact z/OS | Library | PDF


Changing the amount of time ABACKUP waits for an ML2 volume to become available

z/OS DFSMShsm Implementation and Customization Guide
SC23-6869-01

When ABACKUP processes a data set on ML2 tape, it updates the primary and migration volume record (MCV) of the volume being processed to indicate that it is in use by ABACKUP. If the data set spans volumes, the MCVs of the first and last volumes are updated. If a tape volume is already in use by another function, ABACKUP waits up to 30 minutes for the volume to become available as the default. If after waiting 30 minutes, the volume still remains unavailable, message ARC6254A prompts the operator to continue the wait or to cancel the operation. The end of this wait means that the MCV has become unmarked as being in use by the other function; no physical tape allocation is attempted until after the MCV has been updated.

You can alter the default wait time of 30 minutes by entering the following PATCH command:
 PATCH .ABRCB.+4A2 X'nnnn' VERIFY(.ABRCB.+4A2 X'005A')
Note: X'nnnn' is the hexadecimal representation of the number of 20-second intervals to be tried. The default decimal value is 90, which produces a 30-minute wait.

If ABACKUP finds, while backing up migrated data sets, that an ML2 tape is in use by another ABACKUP task, it temporarily skips backing up from that tape and continues backing up from any other needed ML2 tapes that are not in use.

After all ML2 tapes have been processed or found to be in use, another pass can occur if any ML2 tapes were skipped. After a tape was skipped, and if no other data sets were backed up from tape in the current pass, ABACKUP delays for five minutes waiting for the other ABACKUP commands to complete using the ML2 tapes in contention. ABACKUP then retries accessing the data sets on the ML2 tapes that were skipped during the previous pass. If at least one ML2 tape is still in use, ABACKUP retries again, starting another pass.

ABACKUP retries a maximum of nine times, for a possible total of ten attempts.

If at the end of nine retries there is still at least one ML2 tape still in use by another ABACKUP command, ABACKUP issues an ARC6254A message for each such volume. If any response is N (No), ABACKUP processing fails with an ARC6259E message without any additional retries. If all the responses are Y (Yes), ABACKUP writes one ARC6255I message and retries a maximum of another nine times. If at the end of the second nine retries, there is still at least one ML2 tape in use, the ABACKUP processing fails with an ARC6261E message.

To patch the between-pass delay value, issue the following command:
PATCH .ABRCB.+2A X'0008' /* Delay by ABACKUP for 8 minutes */
                         /* between passes for another     */
                         /* ABACKUP to finish with an ML2  */
                         /* needed by both                 */
To patch the number of retry loops, issue the following command:
PATCH .ABRCB.+38 X'04'   /* Make 4 retry loops for ABACKUP */
                         /* needing ML2 tape(s) in use by  */
                         /* other ABACKUP(s)               */

Go to the previous page Go to the next page




Copyright IBM Corporation 1990, 2014