Running CA Disk HSMGEN2 job, during IBM DFSMShsm to CA Disk conversion, it can happen to get messages: ADSHC142 4512 UNABLE TO COMPLETE DFHSM TO CA DISK CONVERSION OF DATA SET dsn - ADSHC142 DATA SET BYPASSED

Document ID : KB000007609
Last Modified Date : 14/02/2018
Show Technical Document Details
Issue:

In CA Disk 12.5 Installation Guide – Appendix G we document the procedure to automatically convert from IBM DFSMShsm to CA Disk and in the target library CCUWJCL we provide sample jobs for this conversion.

Jobs HSMGEN1 and HSMGEN2 are part of this conversion as they manage the creation of the correct commands to promote the data movement from IBM DFSMShsm to CA Disk.

 

It can happen that job HSMGEN2 gets the following message for more than one dataset:

ADSHC142 4512 UNABLE TO COMPLETE DFHSM TO CA DISK CONVERSION OF DATA SET dsn 

ADSHC142 DATA SET BYPASSED 

 

The job ends RC=05, the dataset is not moved from IBM DFSMShsm to CA Disk and the conversion cannot progress.

This can occur when the conversion job HSMGEN2 from DFSMShsm to CA Disk  is managing the ML2 (Migration Level 2) datasets , which should be moved from disk to tape.

How to correct this situation, allowing the conversion to progress? 

Environment:
CA Disk – CA Datacom – IBM DFSMSHsm
Cause:

Here the explanation we provide in the CA Disk 12.5 Backup and Restore Message Reference Guide for the error message:

 

4512

UNABLE TO COMPLETE DFHSM TO CA DISK CONVERSION OF DATA SET ddd DATA SET BYPASSED

Reason:

A matching DSNINDEX record, as created by ADSHC002 or ADSHC003, was not found.

The data set is bypassed due to the missing record necessary to reconstruct the original information, like some of the dates and SMS attributes.

Action:

Verify that the record has not expired by checking the original MCDS / BCDS.

If further assistance is required, contact CA Technical Support.

 

 

Resolution:

The error:

 

ADSHC142 4512 UNABLE TO COMPLETE DFHSM TO CA DISK CONVERSION OF DATA SET dsn 

ADSHC142 DATA SET BYPASSED 

 

usually occurs during the HSMGEN2 conversion of the ML2 datasets from IBM DFSMShsm to CA Disk, trying to move them from dasd to tape.

The dataset is not moved under CA Disk and the conversion doesn’t progress to the end correctly.

If this is the case, it is necessary to check the CA Disk Sysparm value set for option SYNCDEVS.

The default is SYNCDEVSN99 and it could cause the problem.

 

To bypass the error, it is necessary to set the option as:

 SYNCDEVSN01

Here the explanation we provide in the CA Disk 12.5 Backup and Restore System Guide:

 

SYNCDEVSn99

Valid for ARCHIVE and BACKUP

This sysparm determines whether or not the SYNCDEV CCW is to be issued during Backup or Archive after every data set or if the disposition processing of data sets will be deferred until the final data block of the data set is moved from the control unit to the media. The default value of N99 causes CA Disk to defer the disposition processing by queuing up to '99' data sets. After the data set(s) that are in the buffer that is in the tape control unit has been written to the media, disposition processing will be performed for all queued data sets that are now complete.

A SYNCDEV command will be issued to the device if the number of queued data sets exceeds the number specified. A value of 'Y' or 'N01' or 'N00' will cause CA Disk to issue the SYNCDEV during cartridge processing after every data set.

While SYNCDEVSn99 will improve throughput when small data sets are involved during backup or archive on high density tapes, additional recovery steps like Tape Analysis, Archvol Rebuild, and so on, are needed if a tape I/O error or other exceptional condition preventing CA Disk from properly closing the tape media is detected during Backup or Archive processing.

When SYNCDEVS is specified to allow data set disposition queuing for Backup and Archive, the ENQ on each data set queued is retained until disposition processing is performed for that data set so that the Backup or Archive copy retains its integrity. A data set that had been archived could be read or written if the ENQ were released prior to disposition processing. The length of time the ENQ is held after the Archive or Backup for a data set is complete is unpredictable and depends on when the next data set completes the Archive or Backup process, or, at the latest, the completion of the job.

If you have jobs that are waiting for a data set to be free, you may want to trade the backup time reduction for data set availability. You can use SYNCDEVS specified with a Y to work. Changing the number from 99 to 80 or 75 may not change much depending on the control unit buffer size and the average data set size. Seeting it to 5 or 10 will allow some streaming to take place but more SYNCDEVs will be issued, slowing the Backup job.

The DSNINDEX field DSNRATIO reflects the actual IDRC compression ratio ONLY if SYNCDEVS is set to 'Y' or N01 or 'N00'. DSNRATIO will be set to 100 (no compression) if data set disposition queuing is performed and the data set is smaller than 20MB. It will be set to the value calculated if the data set is larger than 20MB but may not accurately reflect the actual compression for the data set. It will more accurately reflect the data set's actual compression on the media as the size of the data set increases. It will also be affected by the value of IDRCSIZE sysparm in that data sets below the threshold specified in IDRCSIZE will not have the IDRC statistics retrieved and thus DSNRATIO will not reflect the actual IDRC compression ratio.

This sysparm is effective only if the output media includes a cartridge tape drive of some kind. If the output media is disk storage, the disposition processing is performed immediately.

 

If, after this suggestion, the abend occurs again, please open a Case with CA Support.