Description of problem: Setup: OSP13z10 HP3Par Storage backend Goal: Do backup of volumes Operation: #1 Create a snapshot of a volume #2 Create a volume of that snapshot On the HP3Par storage we see 3 operations: #1 Create the snapshot #2 Create a separate thick volumes (createvvcopy in HP terms) of the original volume that is never used after #3 Create a vlun to attach to the compute node That extra step (#2) is not needed or wanted. We discovered why this is happening. Its because of the HP variable called 'convert_to_base'. On OSP13 even with the latest z release (z14) the HP driver included is 4.0.8. In that version convert_to_base is set to True. There is no way to force it to false. That is why in version 4.0.11 an extra spec has been created to allow openstack to interact with that variable and they also made that variable false by default. In OSP16.1 We do have this new version and it works as advertise and it solved the problem I am describing above. The request here is to backport 4.0.11 HP driver version to OSP13. Version-Release number of selected component (if applicable): OSP13 (any release) HP Driver 4.0.8 How reproducible: 100% Steps to Reproduce: 1. Create snapshot of volume 2. Create volume of that snapshot 3. Wait several hours to have volumes created. Actual results: Unnecessary step of creating a volume that is not used or needed. Expected results: Removed that extra step. Additional info:
@Pablo, can you take this?
Exception flag granted
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 (Red Hat OpenStack Platform 13.0 bug fix and enhancement 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-2021:2385