Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-12-10
06:40:26 openstackgerrit xulei proposed openstack/nova master: Optimization for nova-api _checks_for_create_and_rebuild https://review.opendev.org/698125
09:00:06 openstackgerrit Kashyap Chamarthy proposed openstack/nova master: libvirt: Remove MIN_QEMU_FILE_BACKED_DISCARD_VERSION https://review.opendev.org/695915
09:00:07 openstackgerrit Kashyap Chamarthy proposed openstack/nova master: libvirt: Remove MIN_{LIBVIRT,QEMU}_FILE_BACKED_VERSION https://review.opendev.org/695942
09:32:50 luyao efried, stephenfin : I invite you to look at spec for live migration with vpmem if you get time. https://review.opendev.org/#/c/695863
10:00:23 shilpasd bauzas: thanks for review https://review.opendev.org/#/c/650188 >> NFS spec
10:05:31 bauzas shilpasd: I'll try to look more
10:05:43 bauzas still working downstream unfortunately
10:07:01 shilpasd bauzas: ok, thank you
10:07:14 shilpasd attending nova weekly to have more review from nova core
10:08:07 shilpasd this just for FYI
10:12:38 openstackgerrit Eric Xie proposed openstack/nova master: Report trait 'COMPUTE_IMAGE_TYPE_PLOOP' https://review.opendev.org/698132
11:01:36 openstackgerrit Stephen Finucane proposed openstack/nova master: Switch to uses_virtio to enable iommu driver for AMD SEV https://review.opendev.org/696697
11:01:37 openstackgerrit Stephen Finucane proposed openstack/nova master: Create a controller for qga when SEV is used https://review.opendev.org/693072
11:01:37 openstackgerrit Stephen Finucane proposed openstack/nova master: Also enable iommu for virtio controllers and video in libvirt https://review.opendev.org/684825
11:07:02 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Convert remaining API tests to use neutron https://review.opendev.org/696509
11:07:02 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Drop nova-network-base security group tests https://review.opendev.org/696508
11:07:03 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove 'is_neutron_security_groups' function https://review.opendev.org/696511
11:07:03 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Convert remaining unit tests to neutron https://review.opendev.org/696510
11:07:04 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Correct some broken VIF tests https://review.opendev.org/696513
11:07:04 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove nova-network security group driver https://review.opendev.org/696512
11:07:05 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Copy shared utils from nova-net module https://review.opendev.org/696515
11:07:05 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove firewall support https://review.opendev.org/696514
11:07:06 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove 'MetadataManager' https://review.opendev.org/696748
11:07:06 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove final references to nova-network https://review.opendev.org/696516
11:07:07 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove dependency on nova-net from fake cache https://review.opendev.org/696749
11:07:07 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Add TODOs to remove security group-related objects https://review.opendev.org/696517
11:07:08 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Rename 'nova.network.neutronv2' -> 'nova.network' https://review.opendev.org/696745
11:07:08 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Kill it https://review.opendev.org/696518
11:07:09 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove unnecessary 'neutronv2' prefixes https://review.opendev.org/696776
11:07:09 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Make the security group API as module https://review.opendev.org/696746
11:07:10 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove db methods for ProviderMethod https://review.opendev.org/697150
11:07:10 openstackgerrit Stephen Finucane proposed openstack/nova master: nova-net: Remove unused exceptions https://review.opendev.org/697149
11:58:52 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add unified '_(build|create)_flavor' helper functions https://review.opendev.org/697538
11:58:52 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add unified '_build_server' helper function https://review.opendev.org/697537
11:58:53 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add '_create_server' helper https://review.opendev.org/697540
11:58:53 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add '_delete_server' to 'InstanceHelperMixin' https://review.opendev.org/697539
12:58:43 openstackgerrit Merged openstack/nova master: Nix os-server-external-events 404 condition https://review.opendev.org/698037
12:58:49 openstackgerrit Merged openstack/nova master: Disable NUMATopologyFilter on rebuild https://review.opendev.org/689861
14:22:55 efried gmann, mriedem, dansmith: How do we feel about backporting the external events fix?
14:23:28 dansmith efried: find on principle, but unless there's something that is going to use it, I'm not sure there's a point
14:23:37 dansmith efried: cyborg master tests against nova master yeah?
14:23:57 efried Yeah, tight dependency, several things they're relying on (including your event cancel)
14:24:45 efried I can buy "wait until needed". It should be isolated enough that it would be easy to do at any point.
14:25:15 dansmith sure, but seems about like any other cross-service dependency,
14:25:49 dansmith unless they're going to run against the older service, there's no reason to backport something like that, and they clearly aren't going to get all of nova's cyborg integration stuff backported
14:28:27 openstackgerrit sean mooney proposed openstack/nova master: FUP for in-place numa rebuild https://review.opendev.org/698260
14:29:13 sean-k-mooney efried: ^ that should adress the comments on the first two patches. i have a checkup at my doctors in 15 mins so ill look at the final patch when i get back
14:29:31 mriedem i agree with what dan said
14:30:23 efried dansmith: right, I wasn't thinking of cyborg, I was thinking of some other service that needs to be able to operate against different levels of nova. If they have to change code for the fix, it might be easier for them if the fix was backported rather than them having to code for either possibility.
14:30:41 dansmith well,
14:30:42 efried I guess "other service" is just neutron, ironic, cinder
14:30:50 dansmith I thought we said we didn't expect anyone else to notice,
14:30:58 efried famous last words
14:31:05 dansmith so I'd want to know what actual service would care in what actual situation
14:31:05 sean-k-mooney neutron is the only one one that i think would send events maybe cinder and i dont think neutron ever checks the responce code
14:31:19 dansmith not doubting that it's possible, just saying.. backports like that need reasons, IMHO
14:31:26 efried ironic does too now for the power state thing
14:31:34 efried Fine by me, just talking it through.
14:31:35 dansmith efried: if they're going to notice, then backporting something they'll notice is really much worse
14:31:37 sean-k-mooney efried: but does it actully check the responce code
14:32:00 sean-k-mooney most of the external events usage just fired the event and forgot about it
14:32:05 efried dansmith: yeah, that makes sense, because we can't count on the backport being applied in a given field situation, right?
14:32:37 dansmith efried: backports need to be things that can be applied (or not) in any order to a running deployment..
14:32:40 sean-k-mooney also in the ironic case the only failure mode we know of would be for shelved instaces. is that a thing with ironic?
14:32:53 sean-k-mooney it could be form bfv i guess but in general i woudl guess not?
14:33:03 sean-k-mooney anyway got to run
14:33:16 dansmith efried: if we backport it, you should expect someone would have yum upgraded one machine at a time, in any order, potentially without upgrading neutron or anything else, before or after
14:33:53 dansmith efried: so if we introduce some different behavior, there are lots of ways that could be much worse than a coordinated upgrade, which would generally be smoked out ahead of time, rolled to prod in a maintenance window, etc
14:34:11 efried sean-k-mooney: It just uses the code for logging purposes https://review.opendev.org/#/c/664842/19/ironic/common/nova.py
14:34:22 efried so the behavior would change, but the result (error or not) would be the same.
14:37:02 efried oh, I lied.
14:37:11 efried it will return True in the 207/4xx case.
14:39:40 efried ...which is only ever checked in UT. Phew.
14:39:54 openstackgerrit Merged openstack/nova stable/rocky: Do not update root_device_name during guest config https://review.opendev.org/696353
14:43:02 openstackgerrit Matt Riedemann proposed openstack/nova master: WIP: Add NodeOwnerFilter https://review.opendev.org/697331
15:11:02 stephenfin efried: Think I could entice you to burn through a few more of those nova-net patches today?
15:11:38 efried stephenfin: Totally. When do you bugger off til 2020?
15:12:25 stephenfin I'm around until the 27th, minus the 25th/26th
15:12:40 stephenfin gone the new years week
15:12:46 stephenfin *week of new years
15:13:33 efried stephenfin: cool. So, can I trade you nova-net for vTPM (not today, I've got some more updates to make)?
15:14:36 efried dun dun dahhh
15:15:16 stephenfin I'll allow it.
15:15:22 efried Thanks.
15:15:33 efried Is gibi_off off for the remainder?
15:24:55 aarents kashyap: thks for the review https://review.opendev.org/#/c/696084/ I added some remarks about if go on rebase versus flatten
15:25:26 kashyap aarents: No problem; will look. I'm running around like a head-less chicken doing last-minute stuff
15:27:30 aarents k no prob
15:28:56 efried dansmith, mriedem, sean-k-mooney: https://review.opendev.org/698273 (ironic refactor for the external events fix)
15:30:09 kashyap efried: If you're curious about yesterday's "competition", I won ... the "loser's final", if that makes sense
15:30:22 sean-k-mooney i have not seen us use the ddt stuff much i should look into that more closely sometime
15:30:25 kashyap (There were 10 folks; two groups of 5 each.)
15:30:38 efried As a veteran of tournaments, both single elimination and round robin, it makes perfect sense. How many in the bracket?
15:30:50 efried disregard, race condition.
15:30:59 kashyap Hehe
15:31:12 efried sean-k-mooney: I actually stole that from dustinc, whom I now consider to be our ddt expert.
15:32:17 efried kashyap: congratulations. There are many reasons competition is a positive experience, and winning is only a small (and IMO not close to the most important) aspect.
15:32:32 sean-k-mooney thanks for adding the comment on the ddt elements, it makes it easier to follow without having to figure it out
15:32:54 kashyap efried: Yeah, I barely practised the last few Mondays; and went in there just to see how _different_ players play in a game vs. routine practise
15:33:07 efried kashyap: On that note: http://www.taylorbjj.com/why-i-dont-compete/

Earlier   Later