ZFS Backup error ADSDF102 4106 00000081 CODE EF17623E
Article ID: 259395
Updated On:
Products
Issue/Introduction
It can happen that the Disk Backup job is executed in one Lpar to backup several SMS pools that contain zFS from different Lpars, so some of them are mounted in z/OS where job runs while others are mounted in their respective Lpars .
When backup runs and tries to access the so-called "remote" zFS , the below error message is issued :
ADSDF102 4106 ADR980E (001)-DTDSC(01), THE BPX1PCT PROGRAM FAILED DURING
QUIESCE PROCESSING FOR DATA SET datasetname WITH RETURN CODE
ADSDF102 4106 00000081 AND REASON CODE EF17623E
Is it possible with Disk to bypass this error and enable these zFS backups too
Environment
Release : All
Resolution
The error message is the following:
...
ADR980E (001)-DTDSC(01), THE BPX1PCT PROGRAM FAILED DURING QUIESCE PROCESSING FOR DATA SET data set name WITH RETURN CODE 00000081 AND REASON CODE EF17623E
...
and this is an IBM DFSMSdss message just because this backup is done using this product.
Here DFSMSdss is trying to quiesce the zFS file before performing the dump and this is just because, starting from DISK 12.5 PTFs SO07283 and SO10594, we demanded to DFSMSdss the complete serialization, quiesce and backup of the zFS files.
From the IBM Message Guide :
Explanation
If the operation is QUIESCE, the source data set could not be serialized. This message can be received if a source z/OS® file system data set is unmounted while DFSMSdss is attempting to serialize the data set.
If the operation is UNQUIESCE or RESET, serialization for the data set was lost while the data set was being processed....
-----------
checking the Return Code and Reason Code:
...
129 0081 ENOENT No such file, directory, or IPC member exists.
...
...
Action
The aggregate specified cannot be found. Correct the aggregate name and try again.
Parent topic:
EFxxrrrr reason codes
....
so this is something related to IBM DFSMSdss behavior. In addition another observation coming from the below IBM suggestions:
...
Back up mounted zFS data sets with logical data set copy or dump. This will provide the quiesce serialization mechanism (BPX1PCT) to ensure data integrity.
For non-shared file systems, you must run the DFSMSdss copy or dump job on the same system where the data set is mounted for update; the quiesce is only effective on the system it is issued from. For shared file systems, you can run the DFSMSdss job from any system in the sysplex, because the quiesce is handled globally across all systems. TOL(ENQF) is not honored during logical dump of zFS and is not honored for a source zFS data set during logical copy.
...
In this case Disk is not responsible for this behavior but IBM DFSMSdss needs to manage in this way the zFS in order to avoid any corruption of the file. It is a DFSMSdss requirement.
Feedback
