| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2020-09-09 | |||
| 10:59:46 | brinzhang_ | sean-k-mooney, gibi: the change of https://review.opendev.org/#/c/729563/12/nova/conductor/manager.py@1273 was restored, and I left comments in this patch | |
| 12:06:19 | openstackgerrit | Hemanth Nakkina proposed openstack/nova master: Update pci stat pools based on PCI device changes https://review.opendev.org/749175 | |
| 12:49:52 | pawan-gupta | Hi, I have setup a Compute Cluster and I am trying to push a small script using vendor data, It is working good in most of the distributions except CentOS-7. when I do cloud-id, it gives me ci-b64:b3B..... instead of openstack, I am not sure, what I am missing here. Cloud-Init version is 18.5. | |
| 12:55:22 | sean-k-mooney | brinzhang_: yep just saw it. ya your right we delete the arqs in shelve offload so in unshelve we just need to create them not delete and create. | |
| 12:56:02 | brinzhang_ | sean-k-monney: yeah | |
| 12:56:20 | brinzhang_ | sean-k-mooney: Ask a questions | |
| 12:57:31 | brinzhang_ | gibi, sean-k-mooney: If we used Nova in Rocky release, and we want to make placement project join the openstack Rocky release, is there any risk? | |
| 12:57:53 | brinzhang_ | In rocky release, the placement still in Nova | |
| 13:00:10 | sean-k-mooney | you mean what use standalone placement fomr a later release instead of in tree placemnt on rocky | |
| 13:00:36 | sean-k-mooney | that should work fine unless we messed up the placement microversions in some way | |
| 13:00:41 | brinzhang_ | We used docker to deploy the OpenStack env, if pull the placement-Ussuri join, it will be a independent docker | |
| 13:01:10 | sean-k-mooney | brinzhang_: ah instead of hosted form the nova api? | |
| 13:01:26 | sean-k-mooney | brinzhang_: i tought it was always a seperate container | |
| 13:01:38 | brinzhang_ | sean-k-mooney: yeah | |
| 13:01:56 | sean-k-mooney | as in i though kolla for example always ran placment in its own container even when it was in tree | |
| 13:02:18 | brinzhang_ | if so I think it work ok, the Ussuri placement works in the OpenStack Rocky env | |
| 13:02:33 | sean-k-mooney | ya it should | |
| 13:02:49 | brinzhang_ | ack thanks sean-k-mooney | |
| 13:02:58 | sean-k-mooney | we havent really tested that but that what microverions are for and they have pretty good microverion testing in placment | |
| 13:03:31 | brinzhang_ | I know the placement microversion, Iwill see it later | |
| 13:04:12 | yingjisun | gibi: here is the latest result of patch | |
| 13:04:24 | yingjisun | http://207.189.188.190/logs/07/742407/6/check-vote/ext-nova-zuul/664e0fe/ | |
| 13:04:25 | gibi | yingjisun: yes I saw it thanks. It looks good | |
| 13:04:42 | yingjisun | :) | |
| 13:04:47 | gibi | yingjisun: I will approve the undeprecate patch on Thursday | |
| 13:05:02 | yingjisun | I see. Thank you. | |
| 13:05:07 | lyarwood | gmann: https://review.opendev.org/#/c/744883/ - I'm trying to wrap the existing gate script for evacuation in an ansible role as you have done for the parts of grenade | |
| 13:05:31 | lyarwood | gmann: any idea what I'm doing wrong here as zuul isn't able to find the role I've added in the same change | |
| 13:05:48 | gmann | lyarwood: checking.. | |
| 13:06:34 | gmann | 1 min, fixing policy patch pep8 | |
| 13:06:46 | tosky | lyarwood: - run-nova-evacuate-hook but the role is roles/run-evacuate-hook/README.rst | |
| 13:06:57 | lyarwood | argh! | |
| 13:07:04 | tosky | :) | |
| 13:07:12 | lyarwood | tosky: thanks :) | |
| 13:09:24 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP zuul: Introduce nova-multinode-evacuate https://review.opendev.org/744883 | |
| 13:10:21 | openstackgerrit | Ghanshyam Mann proposed openstack/nova master: Migrate default policy file from JSON to YAML https://review.opendev.org/748059 | |
| 13:12:16 | gmann | stephenfin: updated^^, sorry about that, it did not hit me locally due to type annotation checks | |
| 13:13:11 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP zuul: Introduce nova-multinode-evacuate https://review.opendev.org/744883 | |
| 13:25:18 | openstackgerrit | Federico Ressi proposed openstack/nova master: Log stack trace when fails launching a defined domain https://review.opendev.org/750652 | |
| 14:04:34 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP zuul: Introduce nova-multinode-evacuate https://review.opendev.org/744883 | |
| 14:06:02 | sean-k-mooney | lyarwood: is that a job ^ | |
| 14:06:14 | sean-k-mooney | i assume its not some new api action right | |
| 14:06:25 | sean-k-mooney | it sjust the job where we will test evacuate | |
| 14:06:46 | tosky | it's part of the removal of the remaining live-migration job | |
| 14:06:58 | sean-k-mooney | ya that is what i was assuming | |
| 14:07:38 | sean-k-mooney | i know lee has been workign on that for a while and we discussed that in the past but have not been following this too closely | |
| 14:17:26 | openstackgerrit | Artom Lifshitz proposed openstack/nova stable/train: post live migration: don't call Neutron needlessly https://review.opendev.org/750670 | |
| 14:21:25 | bauzas | gibi: fwiw, looking at your change, no question yet | |
| 14:21:32 | bauzas | stephenfin: have you seen my point ? | |
| 14:21:39 | gibi | bauzas: cool, thanks | |
| 14:24:10 | stephenfin | bauzas: oh, sorry, I missed that | |
| 14:24:22 | stephenfin | Lemme refresh that real quick | |
| 14:24:28 | bauzas | np, have you time for respinning it ? | |
| 14:24:32 | bauzas | cool ta | |
| 14:24:47 | stephenfin | sean-k-mooney: See, ta is catching on :P ^ | |
| 14:25:14 | sean-k-mooney | hehe it is a very irish thing, slowly we will take over | |
| 14:27:00 | openstackgerrit | Artom Lifshitz proposed openstack/nova stable/rocky: post live migration: don't call Neutron needlessly https://review.opendev.org/750673 | |
| 14:27:43 | artom | Oh wait, I f'ed that up, it's r s t | |
| 14:27:50 | artom | So stein before rocky | |
| 14:28:52 | openstackgerrit | Artom Lifshitz proposed openstack/nova stable/stein: post live migration: don't call Neutron needlessly https://review.opendev.org/750674 | |
| 14:30:51 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Remove resize, cold migration from vTPM limitations https://review.opendev.org/750675 | |
| 14:30:55 | stephenfin | bauzas: ^ | |
| 14:30:58 | openstackgerrit | Artom Lifshitz proposed openstack/nova stable/stein: post live migration: don't call Neutron needlessly https://review.opendev.org/750674 | |
| 14:31:44 | stephenfin | I didn't add anything to the release note since we already decided not to list specific operations in there. See Ie470af3e738327c6f2800f386dbe43319f896222 | |
| 14:31:58 | bauzas | stephenfin: err, I'd have loved to have the doc in the same patch, but kk | |
| 14:32:47 | bauzas | ah you're right and I'm old | |
| 14:32:51 | bauzas | nevermind this then | |
| 14:35:53 | openstackgerrit | Artom Lifshitz proposed openstack/nova stable/rocky: post live migration: don't call Neutron needlessly https://review.opendev.org/750673 | |
| 15:03:29 | stephenfin | bauzas: What's the chances of looking at https://review.opendev.org/#/q/topic:sriov-functional-tests+status:open too? | |
| 15:03:54 | bauzas | stephenfin: atm looking at the hairy patch on SRIOV attach/detach from gibi | |
| 15:03:58 | bauzas | maybe next one ? | |
| 15:04:19 | bauzas | stephenfin: do the above also checks sriov attach/detach ? | |
| 15:04:43 | gibi | bauzas: nope, but my patch has functional test for it | |
| 15:04:50 | stephenfin | No, not needed for that | |
| 15:04:53 | bauzas | kk | |
| 15:04:57 | stephenfin | and SR-IOV attach/detach is higher priority, for sure | |
| 15:04:58 | bauzas | so it's independent | |
| 15:05:07 | gibi | it is | |
| 15:05:10 | bauzas | k, continuing to review gibi's homework | |
| 15:05:35 | gibi | yeah I working from home so the code I push is basically a homework :D | |
| 15:05:38 | bauzas | and fwiw, the above series is functional tests adds, right? if so, we don't need to merge them before tomorrow's FF | |
| 15:05:51 | stephenfin | bauzas: correct | |
| 15:06:10 | bauzas | kk all good, back to gerrit and github | |
| 15:06:54 | stephenfin | gibi: I assume XenAPI-removal will have to be punted come M3? :-\ | |
| 15:07:14 | stephenfin | I'm waiting on the Tempest fix to merge but gate failures are killing me :-( | |
| 15:07:49 | gibi | stephenfin: if nobody else showing up reviewing that series then I'm affraid so | |
| 15:08:13 | gibi | ohh the tempest part is still open that does not help either | |
| 15:09:21 | lyarwood | sorry been stuck downstream for a while but let me know if I can help with anything | |
| 15:09:45 | bauzas | stephenfin: well, if that's not a Facebook poke, you can surely do it | |
| 15:09:58 | bauzas | but I'm already married fwiw | |
| 15:10:59 | stephenfin | bauzas: keep your smut off IRC, plz | |
| 15:11:07 | stephenfin | ;) | |
| 15:12:42 | stephenfin | lyarwood: It's clearly not the highest priority, but I have two series that I'd like to close off: the XenAPI-removal and the SR-IOV tests. Former is feature'ish so that's more important rn and starts here https://review.opendev.org/#/c/749300/ | |
| 15:13:25 | stephenfin | You won't be able to look at anything above the 'remove os-agents' change until the Tempest change merges but the first half should be good to go | |
| 15:26:10 | openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Remove unused 'affect_auto_assigned' parameter https://review.opendev.org/744871 | |
| 15:26:11 | openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Add type hints for various allocate functions https://review.opendev.org/744870 | |
| 15:31:16 | bauzas | gibi: couple of questions in https://review.opendev.org/#/c/740995/17/nova/compute/manager.py | |
| 15:31:25 | gibi | looking | |
| 15:31:41 | bauzas | gibi: mostly about exception handling in order to make sure we have correct error behaviours | |
| 15:32:01 | bauzas | the overall logic looked to me good | |