Cause:
Backups on the NFS backend shared the same "backup" prefix.
Consequence:
If two different backups shared the same container, the backups overwrite each other's data.
Fix:
Generate the backup prefix based on the backup UUID and the volume ID.
Result:
Backup data will no longer get overwritten.
The fix was verified on RHOS10 system
The relevant rpm packages ware upgraded manually on RHOS10 system
tests ware performed as described in the bug description
3 backup made and each backup been restored and the volume
content was check to be as is at the time of the backup created.
Bug was successfully verified according to test case RHELOSP-27027
Setup Cinder backend LVM and backup backend is NFS
rpm version:
[root@controller-0 ~]# rpm -q openstack-cinder
openstack-cinder-9.1.4-6.el7ost.noarch
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.
For information on the advisory, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.
https://access.redhat.com/errata/RHBA-2017:2660