Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2025-07-01
16:46:29 sean-k-mooney i belive we do
16:46:36 dansmith that surprises me
16:46:53 sean-k-mooney we have it form the compute node object i belive
16:47:01 dansmith sure, but in hoststate?
16:47:38 sean-k-mooney so this is the poc https://review.opendev.org/c/openstack/nova/+/947503/9/nova/scheduler/weights/compute_created.py
16:48:27 dansmith we have an object, but I didn't think it was queried from the db and thus not complete, but maybe that's not right
16:49:08 sean-k-mooney so yes https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L176
16:49:26 sean-k-mooney you could be correct about it being a partial object
16:49:26 gibi based on the poc we have HostState.service.created_at as created_at does not expect to change it is fair to assume it is up to date
16:49:45 gibi isn't expected to change
16:50:34 sean-k-mooney so it looks like this is only populated via update its not populated iva __init__
16:51:02 sean-k-mooney so im not sure if its there if you disatble the track_isntace_state notifications stuff
16:51:17 dansmith okay we are doing a scatter to gather nodes and hosts from cells
16:51:18 sean-k-mooney we woudl have to confirm that its always present in both confiugrations
16:51:30 dansmith I was thinking we did an optimized construction from a list there
16:51:44 dansmith sean-k-mooney: yeah, good point
16:52:11 sean-k-mooney i think that config only affect the instnace object and how those are populated
16:52:16 dansmith because this could also be triggering massive lazy loads, which would be bad
16:52:19 sean-k-mooney but assuming we have that alwasy
16:52:43 sean-k-mooney i think it would be ok to use it in the weigher provide ya we are not lazy loading this
16:53:04 dansmith we need to make sure this is run in CI at least once so we can examine logs for lazy loads
16:56:32 sean-k-mooney ya we can turn this on in one fo the jobs
16:57:08 sean-k-mooney it wont really have much of an effect with only 1-2 nodes
16:57:10 dansmith we don't have to permanently, but just a DNM on top of this to get a run and make sure we don't see "lazy loading service.created_at" :)
16:58:10 sean-k-mooney we are wrappign the sercie in a readonly dict class that i have not really seen before. i would assume that shoudl also block lazyloading?
16:58:24 sean-k-mooney goign back on topic
16:58:34 sean-k-mooney are we ok with appvoing this as a specless blueprint
16:58:38 sean-k-mooney and continuing to review
16:59:16 dansmith idk
16:59:53 Uggla :+1 for me it looks an interesting weigher.
17:00:20 Uggla gibi, ok for you ?
17:00:24 gibi I'm fine
17:00:46 sean-k-mooney i dont nessiarly have an objection just note that if there are lazy loadign issues they will need to be resolved ebfore we can merge it
17:00:56 dansmith same
17:00:58 sean-k-mooney so it might not be ready before FF
17:01:17 sean-k-mooney but im ok to move to the review and take a look a ci logs ectra
17:01:30 Uggla jonnyb is that ok for you ?
17:01:35 jonnyb thats fine with me, i wasnt aware that lazy loading could be an issue. but good point
17:02:04 sean-k-mooney you defaulted to 1.0 so in theroy its enabled by default
17:02:15 sean-k-mooney so we should b eable to look at any of the current ci logs
17:02:28 Uggla ok last topic:
17:02:28 Uggla #topic (gibi) Workaround for cpython3.13 GC bug needs review https://review.opendev.org/c/openstack/nova/+/952966
17:02:36 dansmith oh, definitely seems like it should be disabled by default
17:02:53 dansmith that's merged?
17:03:05 sean-k-mooney well we can defintly change that. in the review and ya the gc thing is merged
17:03:26 gibi yepp it is merged
17:03:29 Uggla dansmith yes it seems
17:03:48 gibi I proposed the backport to Epoxy as I know debian want it there
17:04:23 gibi (the cpython bug fixing effort is progressing in the background)
17:05:14 gibi so other than the backport there is nothing else left here
17:05:18 gibi we can move on
17:05:41 sean-k-mooney jonnyb: so without diging deeply https://zuul.opendev.org/t/openstack/build/8a9b0a0404034a79a95621b18e16534b/log/controller/logs/screen-n-sch.txt#1017 you schduler was enabeld and i dont see lazy loadign but we need to look at that properly after the meeting
17:05:44 Uggla ok cardoe, wants to discuss about a patch
17:06:03 Uggla cardoe please go ahead, and then we will close.
17:06:07 cardoe https://review.opendev.org/c/openstack/nova-specs/+/471815
17:06:36 cardoe It is a proposed spec change to add trunk ports with their vlan tags to the network_data.json
17:06:47 sean-k-mooney ya im +2 on that gibi is +1 because they wont have review bandwith i belvie
17:07:04 gibi I'm +2-1=+1 signalling review bandwidth yes
17:07:16 cardoe It's something written originally by someone not involved in the effort currently.
17:07:23 cardoe But neutron supports defining ports like that.
17:07:55 cardoe There have been a few patches floating around from different operators that are using that functionality as a downstream patch.
17:07:56 dansmith sean-k-mooney: jonnyb: agree, I see the weigher being weighed and no lazy loads in that log, so seems fine
17:08:06 gibi having +1 from rubasov (Bence) is a good sign, he was involved in the original work of trunk
17:08:15 cardoe I've been working to consolidate the downstream operators around a single implementation and wanting to get it upstreamed.
17:08:20 sean-k-mooney yep
17:08:48 cardoe We've got temptest tests written and an implementation.
17:08:52 sean-k-mooney there is code for noav and tempest test too
17:09:04 sean-k-mooney ya https://review.opendev.org/q/topic:%22bp/expose-vlan-trunking%22
17:09:20 sean-k-mooney this is useful for both vms adn for ironic
17:09:24 cardoe This is just part of my TC work to try and get downstream operators to push back to the community.
17:10:29 sean-k-mooney so the requwst is bascily for another core to sponsor/review this before spec freeze on thruday
17:10:57 sean-k-mooney and then to review the actul pactchs before m3
17:11:54 sean-k-mooney the one thing i will say is i think we only have turnks in one job
17:12:12 sean-k-mooney so we will need to confirm that this si properly tested in our ci and that that job passes
17:13:07 sean-k-mooney the logs have rotated and the one nova patch that is requried https://review.opendev.org/c/openstack/nova/+/941227 is in merge conflict if i ty an rebase
17:13:17 cardoe I'll work to get folks to update things based on feedback.
17:13:25 sean-k-mooney oh its just missign isnged off by
17:14:50 Uggla cardoe something else you want to ask ?
17:15:16 cardoe No my only question is how can I get this on the roadmap
17:18:12 Uggla We are overtime, so we should close.
17:18:19 Uggla Thanks all
17:18:25 Uggla #endmeeting
17:18:25 opendevmeet Meeting ended Tue Jul 1 17:18:25 2025 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)
17:18:25 opendevmeet Minutes: https://meetings.opendev.org/meetings/nova/2025/nova.2025-07-01-16.00.html
17:18:25 opendevmeet Minutes (text): https://meetings.opendev.org/meetings/nova/2025/nova.2025-07-01-16.00.txt
17:18:25 opendevmeet Log: https://meetings.opendev.org/meetings/nova/2025/nova.2025-07-01-16.00.log.html
17:21:53 sean-k-mooney cardoe: if we cant proceed with it this cycle i woudl say its still wroth rebaseign for next cycle
17:22:03 sean-k-mooney over all i think geting this in to upstream is a good thing
17:22:38 sean-k-mooney and we shoud encurrage upstream operator change like this when they are in scope of nova ot address
17:23:18 Uggla sean-k-mooney, if you have a min, what do you think of this bug: https://bugs.launchpad.net/nova/+bug/2115116 to my mind it is out of our scope but I'd like your feelings.
17:24:02 sean-k-mooney so i have been kind fo frustrated with some of the install guid work that has been sumited lately
17:24:11 cardoe It's this constant catch-22. The core teams have their own workloads and aren't there to respond to the whims of the operators. So the operators need to contribute more from a review standpoint. But then the operators are all counting spend and wanting to equate their return on investment.
17:25:27 sean-k-mooney Uggla: i see where they are coming form but personlly i dont think the mysql info should be in the install guide either
17:25:52 gibi the core team are basically employed by a company which is also counting profit, so operators could do the same, i.e. paying contributors to become cores
17:25:54 sean-k-mooney beyound a very high level over view of how to use teh config option ectra
17:26:23 Uggla sean-k-mooney, btw there is so many ways to install it.
17:26:44 sean-k-mooney Uggla: i personaly woudl mark it opipion
17:27:13 sean-k-mooney its not invalid i just think we would be better limiting the scope of the doc
17:28:31 sean-k-mooney we do not incldue it in the ubuntu ones either for what its worht
17:28:37 sean-k-mooney so it not a disto specific gap

Earlier   Later