| opendevreview | Seyeong Kim proposed openstack/nova master: libvirt: Support boot_index for multiple block devices https://review.opendev.org/c/openstack/nova/+/963665 | 02:17 |
|---|---|---|
| *** mhen_ is now known as mhen | 02:27 | |
| opendevreview | Will Szumski proposed openstack/nova master: Randomise best hosts https://review.opendev.org/c/openstack/nova/+/968544 | 09:25 |
| opendevreview | Johannes Kulik proposed openstack/nova-specs master: Spec for asynchronous volume-attach API https://review.opendev.org/c/openstack/nova-specs/+/958900 | 10:38 |
| opendevreview | Johannes Kulik proposed openstack/nova-specs master: Spec for asynchronous volume-attach API https://review.opendev.org/c/openstack/nova-specs/+/958900 | 11:15 |
| r-taketn | Hello everyone. I have uploaded the SEV/SEV-ES refactoring codes. Could you please review it and provide some feedback? | 11:33 |
| r-taketn | #link https://blueprints.launchpad.net/nova/+spec/generalize-sev-code | 11:33 |
| r-taketn | It was discussed at PTG (https://etherpad.opendev.org/p/nova-2026.1-ptg#L687) and decided to make this specless. And, I have run SEV/SEV-ES tests (both manually and using tempest) with uploaded codes. | 11:37 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/nova-specs master: [spec] Add Cross-AZ scheduling blueprint https://review.opendev.org/c/openstack/nova-specs/+/900296 | 11:49 |
| masahito | hi nova team, please review the bug fix for 500 Internal Server error in the lock API https://review.opendev.org/c/openstack/nova/+/946223 . Thanks. | 14:34 |
| noonedeadpunk | sean-k-mooney: sorry for bothering you again... I'm trying to process reviews on the spec, and got confused myself as well. Given that weights are run per hypervisor, then basically we can't use heapq basically? | 17:16 |
| noonedeadpunk | there's no reason to build AZ to Host mapping, as we just need to get current hypervisor aggregates, and fetch AZs from it. And then fetch azs foreach request_spec.instance_group.members | 17:17 |
| noonedeadpunk | and then like count how many instances are in the same az, and use that as a negative multiplier? | 17:18 |
| sean-k-mooney | we can the heap woudl need to be be in the request spec object | 17:19 |
| sean-k-mooney | althernitively you can impelent the wigher by overriding the follow fucntion | 17:19 |
| sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/weights.py#L100-L120 | 17:19 |
| sean-k-mooney | that will allow it to process all hosts at once and you jsut return the list of wights | 17:20 |
| sean-k-mooney | noonedeadpunk so the point i was orgianly trying to make is you rare not ment to fetch data in the weighers | 17:20 |
| sean-k-mooney | only use the data form the host state or request spec | 17:21 |
| sean-k-mooney | we do not have any weigher that operate on aggreats today btu we do have filters that do | 17:22 |
| sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/scheduler/filters/aggregate_instance_extra_specs.py#L50 | 17:22 |
| noonedeadpunk | ok, I was just looking at existing weighters, but seems none of them is actually using weigh_objects | 17:23 |
| sean-k-mooney | but even then you lotice that we never do lookups on object or db queries | 17:23 |
| sean-k-mooney | noonedeadpunk: correct non of the intree ones weigh all the hosts at once | 17:23 |
| sean-k-mooney | https://github.com/openstack/nova/blob/master/nova/scheduler/weights/affinity.py | 17:23 |
| sean-k-mooney | is the best refence for what you currently want to do | 17:24 |
| noonedeadpunk | yeah, was checking that, and it seems that it should have everything needed | 17:24 |
| noonedeadpunk | except not doing it at once | 17:24 |
| sean-k-mooney | so i ugess it does indrectly | 17:25 |
| sean-k-mooney | actully no | 17:25 |
| sean-k-mooney | it does not have the count of isntace in the curren group in the a given az | 17:25 |
| sean-k-mooney | if you were to weigh all host you shoudl have the info you need however | 17:26 |
| noonedeadpunk | but then if weigh_objects is used - does that still run for each hypervisor? | 17:26 |
| sean-k-mooney | since you can gorup the host_state obejct by AZ then count the number of instnace form the isntnace grou pin each group | 17:26 |
| sean-k-mooney | noonedeadpunk: no it will run once | 17:26 |
| noonedeadpunk | so.... then also doing a query is not that bad? | 17:27 |
| sean-k-mooney | you don tneed to do a query | 17:27 |
| sean-k-mooney | the info shoudl already be there | 17:27 |
| sean-k-mooney | if you look at the base impoealtion of weigh objects https://github.com/openstack/nova/blob/master/nova/weights.py#L100-L120 | 17:27 |
| sean-k-mooney | that the forlop that runs the weigher over all host one by one | 17:28 |
| sean-k-mooney | weighed_obj_list is the list of HostState objects | 17:28 |
| noonedeadpunk | so weighed_obj_list is list of HostState? | 17:28 |
| sean-k-mooney | weight_properties is the request spec | 17:28 |
| noonedeadpunk | ok | 17:28 |
| noonedeadpunk | yes | 17:28 |
| sean-k-mooney | we never got a round to fixing the fact they are called diffent thing in diffeent palces | 17:29 |
| * noonedeadpunk will add description there.... | 17:29 | |
| noonedeadpunk | yeah, ok, makes sense | 17:29 |
| noonedeadpunk | thanks for clarifying it | 17:29 |
| noonedeadpunk | as I got a little bit confused | 17:30 |
| sean-k-mooney | ya i proably could have explaid this better | 17:30 |
| sean-k-mooney | it only clicked now that we shoudl ahve all the data we need already | 17:30 |
| sean-k-mooney | but that to implement this weigher we shoudl jsut overried weigh_objects | 17:30 |
| noonedeadpunk | yeah | 17:31 |
| noonedeadpunk | I had no idea what these parameters are tbh | 17:31 |
| noonedeadpunk | *arguments | 17:31 |
| sean-k-mooney | ya so in newer filtere/weigher we name them what they are | 17:31 |
| sean-k-mooney | but in th older once we never fixt them to make sense | 17:32 |
| noonedeadpunk | ++ | 17:32 |
| noonedeadpunk | now it looks *way* simpler then I originally anticipated | 17:33 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/nova-specs master: [spec] Add Cross-AZ scheduling blueprint https://review.opendev.org/c/openstack/nova-specs/+/900296 | 17:52 |
| sean-k-mooney | noonedeadpunk: same | 18:09 |
| noonedeadpunk | um? | 18:19 |
| noonedeadpunk | sean-k-mooney: sorry, did not get what you meant | 18:19 |
| sean-k-mooney | its simpler then i was expecting also | 18:20 |
| noonedeadpunk | ah | 18:20 |
| noonedeadpunk | :D | 18:20 |
| noonedeadpunk | I just thought it was comment to spec update :D | 18:23 |
| opendevreview | Sean Mooney proposed openstack/nova-specs master: add spec for resouce notifications https://review.opendev.org/c/openstack/nova-specs/+/967712 | 19:02 |
| noonedeadpunk | fwiw, we've found a bug with parallel migrations on ubuntu 24.04 with libvirt 10.0. It's a libvirt bug actually, and I think it's fixed in 10.2 | 19:56 |
| noonedeadpunk | but parallel live migrations fail with `libvirt.libvirtError: internal error: unable to execute QEMU command 'blockdev-add': Certificate does not match the hostname` | 19:56 |
| noonedeadpunk | so /o\ | 19:56 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/nova-specs master: [spec] Add Cross-AZ scheduling blueprint https://review.opendev.org/c/openstack/nova-specs/+/900296 | 19:59 |
| noonedeadpunk | ah. and that is the case only for local storage, so when `--block-migration` is used | 19:59 |
| noonedeadpunk | sorry, fixed in 10.3, not 10.2 | 20:01 |
| noonedeadpunk | https://libvirt.org/news.html#v10-3-0-2024-05-02 | 20:01 |
| noonedeadpunk | https://github.com/libvirt/libvirt/commit/5d48c5d215071526383b8fc50d81ecde62e4111b | 20:20 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!