Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2025-02-13
15:53:11 bauzas gibi: sean-k-mooney's approach mimicing the metricsweigher seems a reasonable trade-off as a follow-up patch
15:53:38 dansmith gibi: yeah, all goes to how strongly you expect it to work. I expected weak optimization because it needs to be fully automatic
15:53:55 dansmith or intended to be fully automatic I mean
15:54:01 bauzas basically, it makes the weigh multiplier less easy to configure but that gives you a possibility to weigh per prop
15:54:14 sean-k-mooney for hard affinity we have the "isolating aggrates with placement" feature
15:54:22 dansmith I pretty much assume any weigher implementation is just trying to make better-than-random decisions, but not anything you'd bet on
15:54:49 dansmith bauzas: you mean my suggestion of a dict of properties and weights?
15:55:26 dansmith I think that gets closer to gibi's goal, but I also offered that suggestion along with a good example of why it's too simplistic
15:55:27 sean-k-mooney the weigher multipler is for seting the relitve weight between weighers when nomalising
15:55:50 sean-k-mooney the per property weight is internal
15:56:03 sean-k-mooney to how we calulate a weight for a given host before we normalise
15:56:13 sean-k-mooney so they have 2 distinct usecases
15:57:13 sean-k-mooney anyway shall we pick this up in an hour
15:58:51 bauzas yup, let's see ourselves in 1 hour
15:58:55 bauzas I'll ping back here
15:59:05 bauzas others can chime if they want
17:02:01 gibi need 2 minutes and I can join
17:02:32 bauzas okay, I'm starting a gmeet
17:03:52 bauzas https://meet.google.com/vye-tjns-bai <= for discussing about https://review.opendev.org/c/openstack/nova/+/940642/
17:42:13 sean-k-mooney bauzas: gibi dansmith melwitt on an unrelated note. neutron removed linuxbridge supports weeks ago
17:42:25 sean-k-mooney os-vif and nova have linux bridge jobs
17:42:45 sean-k-mooney the nova one only tirggers on specific files
17:43:17 sean-k-mooney tehre was a mailing list post about this a while ago but i just want to get poeopel opions on how to move forward
17:43:34 sean-k-mooney first ill put up a patch to drop hte job form nova's check pipeline
17:44:09 sean-k-mooney beyond that there is a patch to remove the job form os-vif but i think we should also remove the linux brindge os-vif plugin
17:44:43 sean-k-mooney os-vif's policy has always been we only have pluggins for in tree neutron backends
17:45:16 sean-k-mooney so question 1 is do i defer the plugin removal to 2025.2 or do it in 2025.1/epoxy
17:45:51 sean-k-mooney second question is there is linux bridge speciric code in nova. wehn shoudl we remove that?
17:46:07 sean-k-mooney my propsoal is to formally deprecate the linux bridge support in nova this cycle
17:46:18 sean-k-mooney and remove the linux brdige code in nova in 2025.2
17:47:22 sean-k-mooney you would only be able to user linux brdige in nvoa as part fo 2025.1 if you had an older neutron but since we never deprecated support you could argure that that is the best path forward
17:47:34 sean-k-mooney we coudl also just be more agressive and delete the code now
17:48:28 sean-k-mooney bauzas: gibi dansmith melwitt ^ we dont have to decied now but let me know what ye woudl perfer
17:49:14 dansmith idk, deprecate and lag on removal I guess
17:50:43 sean-k-mooney if we go with that approch ill change my review on https://review.opendev.org/c/openstack/os-vif/+/941567
17:58:57 sean-k-mooney ill prepare patches to do the formal dpercations and unblock ci this evening and we can review in the irc meeting on tusday. the non clint lib freeze is next week so we need to do an os-vif release wiht the deprecation but i that thats ok
18:36:05 opendevreview sean mooney proposed openstack/os-vif master: Deprecate linux bridge supprot https://review.opendev.org/c/openstack/os-vif/+/941585
18:36:06 opendevreview sean mooney proposed openstack/os-vif master: remove linux bridge plugin https://review.opendev.org/c/openstack/os-vif/+/941586
18:37:08 opendevreview sean mooney proposed openstack/os-vif master: Deprecate linux bridge plugin https://review.opendev.org/c/openstack/os-vif/+/941585
18:37:09 opendevreview sean mooney proposed openstack/os-vif master: remove linux bridge plugin https://review.opendev.org/c/openstack/os-vif/+/941586
19:02:27 bauzas sean-k-mooney: still there ?
19:03:38 bauzas sean-k-mooney: just a question, given the logic of the conf options for the metricsweigher sound reasonable to me, I wonder whether I should duplicate those options for my specific weigher or reuse those metrics options by adding a doc modification saying that the ImageMetaWeigher could also use them ?
19:04:15 sean-k-mooney yep
19:04:40 sean-k-mooney you need to duplicate them
19:04:40 bauzas yep about which alternative ? dup or reuse ?
19:04:44 sean-k-mooney yep im here
19:04:44 bauzas ok
19:05:27 bauzas OK then let's go for specifying them
19:05:31 bauzas thanks
19:06:01 sean-k-mooney the other reason they have to be seperated is
19:06:32 sean-k-mooney with the new weither if the new config option is not defied we effectivly weigh each option as 1
19:06:35 sean-k-mooney i.e. evenly
19:07:03 sean-k-mooney but when its defiend we woudl only apply the weigts to the specified values and ignore the rest
19:07:30 sean-k-mooney where as for the metric weigher the cofnig option is required for it to work
19:07:40 sean-k-mooney instead of being optional
19:07:46 sean-k-mooney so we dont want to share it
19:26:01 mikal Morning
19:29:09 mikal It seems like its been fairly quiet on the spice patches overnight, apart from gmann approving the temptest test (thanks gmann!). Are there any actions items for me right now that I should know about?
19:29:20 bauzas sean-k-mooney: fine, then I'll just provide *one* additional config option, by default we'll stick to evenly weigh the props
19:30:25 sean-k-mooney +1
19:30:50 sean-k-mooney mikal: i honestly didnt have time to look at the current state
19:31:03 sean-k-mooney but you got it to the point of workign and testign in a ci job
19:31:19 sean-k-mooney so i think we could move forward with merging some/all of these?
19:31:24 mikal sean-k-mooney: yeah, no worries.
19:31:41 sean-k-mooney alos we should move forward iwth os-trait patch if not allready done?
19:31:46 mikal sean-k-mooney: yes. The tempest test passes in the ovs-plug job, assuming that the entire rube goldberg machine of changes is landed
19:31:52 gmann mikal: pinged frickler for devstack change 2nd review review in qa channel
19:32:14 mikal sean-k-mooney: yeah, once the spice-direct series is landed the os-traits change will be the main thing blocking me. It blocks the sound / usb changes.
19:32:18 mikal gmann: thank you.
19:32:46 sean-k-mooney we need a second +2 on https://review.opendev.org/c/openstack/os-traits/+/940418 and we need to a release before the non clinet lib freeze next week
19:32:54 mikal sean-k-mooney: I am a little unsure if we want to compress your zuul changes into a single change with the spice API changes or keep it like this. I definitely don't want to take credit for your work by stealing your changes.
19:33:21 sean-k-mooney steal all the things
19:33:27 mikal sean-k-mooney: that freeze is a week before feature freeze, or at the same time?
19:33:29 sean-k-mooney im fine with squashing them if you want
19:33:38 gmann bauzas: not sure if you see my message early this week. but let me know what I need to do to get this spec merged? I added it in Nova meeting agenda for spec exception approval (but i would not be able to attend meeting due to time conflict). I started code change in parallel if that help, manager role is almost ready and will work on service role change https://review.opendev.org/c/openstack/nova-specs/+/937650
19:33:49 sean-k-mooney mikal: the non client freeze is a week before feature freeze
19:33:52 mikal sean-k-mooney: I'd prefer to not squash them unless there's a really solid reason why we should. I think the history is useful.
19:34:16 sean-k-mooney then i think we can jsut try and move them forward as is
19:34:22 bauzas gmann: this cycle, I unfortunately don't have a lot of time upstream
19:34:26 bauzas so I saw your ping
19:35:01 sean-k-mooney gmann: is the code written and tested?
19:35:13 sean-k-mooney becasue otherwise i think this need to go to next cycle
19:35:14 bauzas yeah, that's my question
19:35:43 sean-k-mooney https://review.opendev.org/c/openstack/nova/+/941056
19:35:48 sean-k-mooney https://review.opendev.org/c/openstack/nova/+/941347
19:35:49 gmann sean-k-mooney: I have started that, manager role changes is almost ready (fixing some unit and tempest test)
19:35:51 sean-k-mooney are teh nova patches
19:36:02 gmann service role change I will work soon
19:36:03 gmann https://review.opendev.org/q/topic:%22bp/policy-service-and-manager-role-default%22
19:36:18 gmann ^^ this is all series where service role changes i need to rebase/fix
19:36:37 sean-k-mooney oh there are others ok
19:37:21 sean-k-mooney we kind of have a lot of patches that need review so i kind of feel like this should move to next cycle
19:37:34 sean-k-mooney on the ohter had i dont really want to be the one to punt it so..
19:39:06 gmann well, thing is I am not sure if I can spend more time on upstream activities in next cycle (depend on various factor) so I thought of getting it done in nova this cycle
19:40:50 gmann anyways I added it in meeting agenda and will be ok based on how team decide. unfortunately I will not be able to attend meeting but you knows the current situation on this so I will check meeting log/decision.
19:44:37 bauzas gmann: noted, we'll discuss this on next tues
19:44:45 gmann bauzas: thanks
19:45:58 bauzas sean-k-mooney: one last question about the weigher, do we agree on the fact that props not set on the pre-prop setting won't be weighed ?
19:46:21 bauzas we'll just skip them if so
19:46:58 bauzas ie. default of non-set props will be 0.0

Earlier   Later