On 10/06/2025 07:48, Harsh Ailani wrote:
Hello core reviewers,
*The following reviews are very important in the upcoming cinder release for IBM SVf Cinder Driver :*
925450: [IBM SVf driver] Adding support for temporary volumegroup | https://review.opendev.org/c/openstack/cinder/+/925450
926844: [IBM SVf Driver] Added support for volumegroup-snapshot for temporary volumegroup | https://review.opendev.org/c/openstack/cinder/+/926844
916722: [IBM SVf Driver] Added support for restorefromsnapshot CLI for volumegroups | https://review.opendev.org/c/openstack/cinder/+/916722
938771: [IBM SVf Driver] Added support of clone operation for volumegroup | https://review.opendev.org/c/openstack/cinder/+/938771
929906: Doc: Documentation updates for ibm-svf-volumegroup feature. | https://review.opendev.org/c/openstack/cinder/+/929906
They are dependent on each other and our upcoming features are all dependent on these reviews.
Zuul has passed on these and our internal customer reviews are also completed.
This is not really how opensource works. this email is more of a demand then a request specially given the subject line. "Re: [CINDER] Dependent reviews need to be merged" its perfectly fine to inform the community that you think your code is ready for review or even say there has been some internal review but openstack practices open develoment https://docs.openstack.org/project-team-guide/open-development.html Which means we do all code review in public and generally discussage doing extensive internal review as that context is not evident in the review. doign excessive internal review can also be problematic as internal style or code preference can sometimes lead to designs that do not fit with upstream so i woudl stongly encurage any company to do all the develoment and design _UPSTREAM-FIRST_ and engage in code review the openstack way https://docs.openstack.org/project-team-guide/review-the-openstack-way.html i am not a member of the cidner core team so i cannot speak on there behalf with regards to polilcy but i will not the blueprint you are uisng https://blueprints.launchpad.net/cinder/+spec/ibm-svf-volumegroup was approved for the Zed release. it is nto targeted at teh current release. looking at the gerrit topic https://review.opendev.org/q/topic:%22pbha_merge_vgs%22 i also do not see a cidner-spec while we all parctice open design https://docs.openstack.org/project-team-guide/open-design.html each project handels the paper work a littel differnecly so if you have not already spoken to the cider core team on irc, mailing list or at the PTG to ensure they are happy to proceed with this enhancement using the old zed i would recommend doing that before going much futher. most project require designed to be approved again at the start of each cycle and the curent blueprint as last approved for zed so it is 7 cycles out of date. if this was in nova the lack of an approved blueprint or spec would recive an automatic -1 form reviewer explainable winy this is not procduerly valid and asking you to first discuss with the team if it needed a spec or blueprint. that would elevate to a procedural -2 if the original -1 was ignored with a more direct message stating that if you do not engage with the design approval part of the 4 opens and follow the community process for new feature that your proposal cannot proceed. one thing that cinder used to enforce is third party ci for storage backends. quickly jumping to the last patch in the serice i am not directly seeing a any ibm ci posting result on those changs. they may exist and i just missed them but if you dont have proper end to end testing of this that may also be a problem for moving this series forward. again i suggest talking to the cinder core team about there expecations and process before going much further with the implementations.
Requesting you to please review them and help us merge them in the upcoming release.
Thanks and Regards,
Harsh