Earlier  
Posted Nick Remark
#openstack-nova - 2021-12-14
16:09:48 sean-k-mooney well not all project have a m2 spec freeze
16:09:59 sean-k-mooney so it may be less impactful on others
16:10:05 sean-k-mooney not form me
16:10:19 gmann yeah, its project to project
16:10:20 bauzas sean-k-mooney: you're right, this is maybe us who haven't considered it correctly when we thought about Yoga-2 as a spec freeze
16:10:34 bauzas anyway, not a problmem
16:10:48 bauzas we'll see how things go after new year
16:10:57 bauzas #topic Review priorities
16:11:02 bauzas #link https://review.opendev.org/q/status:open+(project:openstack/nova+OR+project:openstack/placement)+label:Review-Priority%252B1
16:11:07 bauzas #link https://review.opendev.org/c/openstack/nova/+/816861 bauzas proposing a documentation change for helping contributors to ask for reviews, needs a second +2
16:11:13 bauzas thanks gibi btw.
16:11:29 gibi thanks bauzas
16:11:49 gibi the current list of review prio looks OK to me
16:11:49 gmann I am ok with that, just 1 comment i added to make it clear on non-core adding +1
16:11:51 bauzas about the unified limits series, I was confused with what to review but this is fixed
16:12:00 gmann but that is ok to add in followup also, not blocker
16:12:08 bauzas after discussing it with melwitt internally
16:12:23 bauzas gmann: ack, will look
16:12:48 bauzas I really want us to merge the unified limits series, this has been overdue for too long
16:12:58 sean-k-mooney i agree
16:13:17 bauzas so I'll continue to review, but given melwitt's on PTO, I guess the progress would be better after next year
16:13:18 sean-k-mooney unified limist assumign its readay i think would be a big win and release highlight
16:14:02 gibi there is a list of patches on the yoga-python-testing that we need to look at I will add them to priorities
16:14:13 bauzas any series or patches that people would want to highlight to the team for reviews ?
16:14:14 sean-k-mooney gmann: on a related note do we have a nova spec for the chagne in RBAC direction and what that means for nova
16:14:22 bauzas heh, jinxed by gibi
16:14:27 bauzas gibi: patches ?
16:14:32 gibi yep
16:14:37 gibi gmann has a list of them
16:14:40 bauzas gibi: I mean, links ?
16:15:09 gmann yeah, this is list for auditing all policy #link https://wiki.openstack.org/wiki/Nova/rbac
16:15:11 gibi https://review.opendev.org/c/openstack/python-novaclient/+/819208
16:15:15 gibi https://review.opendev.org/c/openstack/os-traits/+/819205
16:15:18 gibi https://review.opendev.org/c/openstack/osc-placement/+/819203
16:15:21 gibi https://review.opendev.org/c/openstack/nova/+/821636
16:15:25 gibi https://review.opendev.org/c/openstack/placement/+/819206
16:15:30 gmann and we can do audit in wiki page in one shot or in code changes also.
16:15:41 gmann gibi: thanks
16:16:25 bauzas gibi: I marked the nova and placement ones with the R-P flag
16:16:33 gmann sean-k-mooney: that was for RBAC one, and we have BP also for that #link https://blueprints.launchpad.net/nova/+spec/policy-defaults-refresh-2
16:16:50 bauzas sean-k-mooney: we said on a meeting yes for a specless BP approval
16:16:54 gmann yeah
16:16:56 bauzas (for the policy thing)
16:17:01 sean-k-mooney bauzas: ok
16:17:16 sean-k-mooney normally i think we shoudl treath policy like an api change
16:17:31 sean-k-mooney and use a spec for it but we can proceed without one for now
16:17:46 bauzas this is the defaults, but meh
16:17:47 gmann I have pushed SYSTEM reader->system admin on top of dansmith server policy changes. and I will be pushing more changes this week
16:17:58 sean-k-mooney bauzas: well to me policy change are interop issues
16:18:09 bauzas sean-k-mooney: I don't disagree
16:18:13 sean-k-mooney so change default is a breaking change but lest not repoen that
16:18:27 bauzas but for users, those are transparent if defaults correctly scoped
16:18:36 gmann well policy change does not require spec always. we have fixed (new or default change) many as bug also
16:19:09 gmann yeah, as it is overall change in RBAC, and as described in goal we have way that users will be broken by default
16:19:23 gmann and migration path and plan for them to new policy in next release
16:19:45 gmann *user will not be broken
16:19:53 gmann sorry for typo. :)
16:20:11 sean-k-mooney just to not consume all the time on this is the plan this cycle to walk back the work we had done and focus on identifying the work need with most of the new work happing next cycle
16:20:20 sean-k-mooney i will review the wiki you provided offline
16:20:40 gmann yeah, specially we are isloating the system form project resource and no system reader in yoga
16:20:52 sean-k-mooney ack
16:21:11 gmann sean-k-mooney: thanks, wiki audit will be helpful
16:21:48 bauzas I think we also said that the TC would document this
16:22:02 bauzas the policy changes, I mean
16:22:14 sean-k-mooney its in the tc spec to some degree
16:22:14 bauzas I even wrote this in the BP when I approved
16:22:20 gmann basically this is what we will do (phase1 ) in nova in yoga #link https://governance.openstack.org/tc/goals/selected/consistent-and-secure-rbac.html#phase-1
16:22:22 bauzas that's my point
16:22:24 sean-k-mooney i would like to see a nova game plan for how to aling to that
16:22:35 sean-k-mooney which is why i ask about a nova spec
16:22:57 sean-k-mooney but if most of that work will happen next cycle im fine with the wiki and specless blueprint for this cycle
16:22:59 bauzas that's why we said we can start to digest and see whether we need to document more the changes in the spec
16:23:03 bauzas right
16:23:08 bauzas moving on ?
16:23:12 sean-k-mooney +1
16:23:16 gmann yeah, its there in TC goal
16:23:41 bauzas #topic Stable Branches
16:23:46 bauzas stable gates are not blocked
16:23:49 bauzas shit
16:23:56 bauzas elodilles: it's you who should tell it
16:24:06 elodilles :)
16:24:12 elodilles so they are not blocked
16:24:16 gibi :D
16:24:18 elodilles (as far as I've seen it before the meeting with a quick double-check)
16:24:28 gmann i thought it is very blocked and bauzas typo :)
16:24:30 elodilles though on some branches (wallaby, victoria?) patches need many rechecks to be able to merge
16:24:58 bauzas interesting dystopîa
16:25:05 elodilles gmann: hopefully not o:)
16:25:05 bauzas between gmann and elodilles
16:25:16 gmann \o/
16:25:42 elodilles or are they? :-o
16:27:50 bauzas so, what to say ? I guess we need to double check
16:27:51 bauzas :)
16:28:32 elodilles i haven't seen any blocking issue, but i'm now confused :)
16:28:58 elodilles gmann: did i miss something?
16:29:16 gmann ah no, I have not seen too. I think recheck is more needed that is uit
16:29:18 gmann it
16:29:44 gmann I saw in other project stable/train but not nova
16:30:17 elodilles oh, thanks
16:30:43 elodilles cool for nova then

Earlier   Later