| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-03-30 | |||
| 11:00:51 | kashyap | stephenfin: Looks fine to me; just looked | |
| 11:25:07 | kashyap | stephenfin: I'm somewhat ambivalent about the folding in here: https://review.opendev.org/c/openstack/nova/+/782411 | |
| 11:26:06 | kashyap | I'm not against it; but just that sometimes having the separate doc w/ a single focus can be less distracting. One can link to the other, generic CPU modelss doc, though. | |
| 11:31:48 | sean-k-mooney | kashyap: too much linking can make reading them in git/github or pdf form so its a trade off but ya topic docs can be useful. i dont partically like the extream we take it to downstream | |
| 12:01:55 | sean-k-mooney | stephenfin: do you have a partical wording you would like me to add for the release note in the cpu_max_* patch | |
| 12:28:13 | gibi | stephenfin: I guess it is OK that these docs are only merged for Xena now | |
| 12:33:32 | sean-k-mooney | stephenfin: https://review.opendev.org/c/openstack/nova/+/769614/2//COMMIT_MSG#18 is that better? | |
| 12:34:32 | kashyap | sean-k-mooney: Yeah; I hear ya | |
| 12:50:39 | stephenfin | gibi: I think so, yeah | |
| 12:51:17 | gibi | stephenfin: OK, cool, I'm working through it | |
| 12:54:38 | stephenfin | sean-k-mooney: Yeah, that makes sense. So iiuc what we want to say is that previously we would attempt to best map the guest NUMA topology of pinned instances to the host NUMA topology, but that was buggy and wasn't updated e.g. during live migration so it's been removed? | |
| 12:55:24 | sean-k-mooney | well i could say that or i could try an implemente it ussing toplogy sorting | |
| 12:55:31 | sean-k-mooney | rahter then filtering | |
| 12:55:44 | sean-k-mooney | i feel like that should be a sperate patch however | |
| 12:58:25 | sean-k-mooney | if i proceed with removal of the code then yes ill update the release as you sugesst noting we tired but it was buggy and now removed | |
| 13:04:27 | stephenfin | makes sense | |
| 13:20:51 | sean-k-mooney | stephenfin: what would you like me to do note we remove the behavior and try and add the sorting in a follow up, add the sortign first in a prevous patch or just remove the behavior and note it | |
| 13:21:16 | stephenfin | I think the sorting should probably be separate so a follow-up? | |
| 13:26:36 | sean-k-mooney | ok ill adress the other comments so and then ill push and start working on the follow up | |
| 13:28:05 | sean-k-mooney | thanks for persiting with this i was exclusivly thinking baout this form the point of view of unpinned guests with numa toplogies | |
| 13:39:44 | bauzas | gibi: I only see one bugfix with https://bugs.launchpad.net/nova/+bugs?field.tag=wallaby-rc-potential | |
| 13:40:08 | bauzas | gibi: stephenfin: what about the cpu_max patch ? | |
| 13:40:36 | stephenfin | I don't think so. It's latent | |
| 13:41:10 | bauzas | https://review.opendev.org/q/project:openstack/nova+branch:stable/wallaby we only merged lyarwood's patch for os-brick | |
| 13:41:20 | bauzas | ok, so for the moment, only one patch for RC2 | |
| 13:41:31 | bauzas | gibi: when are you planning the RC2 release ? | |
| 13:48:43 | sean-k-mooney | bauzas: ya the cpu max patch is not an rc candiate | |
| 13:48:54 | bauzas | ack | |
| 13:49:11 | sean-k-mooney | ill backport it to stable/wallaby as normal once we do the release. | |
| 13:53:18 | bauzas | all good | |
| 13:56:27 | bauzas | folks, I think we have a consensus about nova shouldn't supporting rabbit MQs be unstable, right? | |
| 13:56:32 | bauzas | context : https://bugs.launchpad.net/nova/+bug/1921804 | |
| 13:56:35 | openstack | Launchpad bug 1921804 in OpenStack Compute (nova) "leftover bdm when rabbitmq unstable" [Undecided,New] | |
| 13:57:06 | dansmith | bauzas: I think we have always said, and can only say, that we expect rabbit to work, yes | |
| 13:57:26 | bauzas | yup | |
| 13:57:50 | bauzas | I'll move the bug status to Wontfix | |
| 13:58:07 | dansmith | if there's some particularly nasty bug we create for ourselves if rabbit breaks at some specific time, then mitigating that would be good, but I think in general we should expect rabbit to work | |
| 14:00:43 | gibi | bauzas: agreed about the leftover bdm, until it does not create a failure later just leaks some db resources then I'm OK to move it to wontfix | |
| 14:00:48 | bauzas | the bug is about the fact that we create a bdm even the volume is not attached | |
| 14:08:24 | bauzas | mmm, maybe we should document this in https://docs.openstack.org/nova/latest/contributor/project-scope.html | |
| 14:11:35 | gibi | bauzas: about the RC2, release should be cheap, but we don't have to rush. It is totally OK to have the RC proposed early next week. | |
| 14:11:44 | bauzas | yup | |
| 14:11:55 | bauzas | my question was more about the time we need to look at existing bugs | |
| 14:12:05 | gibi | looking now is good | |
| 14:12:07 | bauzas | as we could find regressions | |
| 14:12:27 | bauzas | I don't want to create a RC2 this week if we need another RC3 eventually | |
| 14:12:38 | gibi | yupp | |
| 14:47:15 | gibi | stephenfin: FYI there is a rendering issue in https://review.opendev.org/c/openstack/nova/+/781438/2/doc/source/admin/cpu-models.rst#60 but not serious so I'm OK if it is fixed in a fup | |
| 14:47:56 | stephenfin | damn, ack - will do a follow-up | |
| 15:01:07 | sean-k-mooney | stephenfin: so the max cpu error. it looks like that only happens when pinning is enabled | |
| 15:01:28 | sean-k-mooney | when i remove hw:cpu_policy=dedicated it goes away | |
| 15:01:31 | stephenfin | yeah, that would make sense based on my comments | |
| 15:01:59 | sean-k-mooney | ya so ill leave that in the extra specs with hw:mem_page_size=small | |
| 15:04:34 | sean-k-mooney | actully ill use hw:mem_page_size='any' its technically not required but its incorrect to not set it | |
| #openstack-nova - 2021-03-31 | |||
| 00:17:34 | openstackgerrit | sean mooney proposed openstack/nova master: Test numa and vcpu topologies bug: #1910466 https://review.opendev.org/c/openstack/nova/+/769601 | |
| 00:17:36 | openstackgerrit | sean mooney proposed openstack/nova master: Fix max cpu topologies with numa affinity https://review.opendev.org/c/openstack/nova/+/769614 | |
| 00:17:36 | openstack | bug 1910466 in OpenStack Compute (nova) "NUMA instance spawn fails on get_best_cpu_topology when there is no 'threads' preference" [Medium,In progress] https://launchpad.net/bugs/1910466 - Assigned to sean mooney (sean-k-mooney) | |
| 01:11:13 | openstackgerrit | Brin Zhang proposed openstack/nova-specs master: Re-propose Remove tenant_id https://review.opendev.org/c/openstack/nova-specs/+/783783 | |
| 07:17:19 | sorin-mihai | is it possible to avoid getting .novalocal appended to the hostname of the instances? | |
| 07:24:20 | openstackgerrit | Merged openstack/nova master: docs: Clarify value for 'hw:cpu_realtime' extra spec https://review.opendev.org/c/openstack/nova/+/782708 | |
| 07:32:36 | openstackgerrit | Balazs Gibizer proposed openstack/placement master: Add support for RP re- and un-parenting https://review.opendev.org/c/openstack/placement/+/784020 | |
| 08:03:10 | openstackgerrit | Balazs Gibizer proposed openstack/osc-placement master: Mark microversion 1.37 support https://review.opendev.org/c/openstack/osc-placement/+/784023 | |
| 08:03:37 | openstackgerrit | Balazs Gibizer proposed openstack/osc-placement master: Mark microversion 1.37 supported https://review.opendev.org/c/openstack/osc-placement/+/784023 | |
| 08:23:53 | bauzas | just in case sorin_mihal goes back : https://docs.openstack.org/nova/latest/configuration/config.html#api.dhcp_domain | |
| 09:43:02 | openstackgerrit | Merged openstack/nova master: docs: Use proper headers in nova-manage man page https://review.opendev.org/c/openstack/nova/+/726191 | |
| 09:52:25 | gibi_away | I will be away mostly this afternoon | |
| 10:11:50 | openstackgerrit | Merged openstack/nova stable/victoria: api: Log os-resetState as an instance action https://review.opendev.org/c/openstack/nova/+/772480 | |
| 10:31:36 | openstackgerrit | Stephen Finucane proposed openstack/nova master: doc: Further tweaks to the CPU models document https://review.opendev.org/c/openstack/nova/+/784066 | |
| 10:32:02 | stephenfin | sean-k-mooney: ^ | |
| 10:33:39 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Further tweaks to the CPU models document https://review.opendev.org/c/openstack/nova/+/784066 | |
| 10:38:09 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Correct typos in 'nova-manage db' docs https://review.opendev.org/c/openstack/nova/+/784067 | |
| 10:39:31 | openstackgerrit | Merged openstack/nova stable/ussuri: Make _rebase_with_qemu_img() generic https://review.opendev.org/c/openstack/nova/+/780780 | |
| 10:40:09 | openstackgerrit | Merged openstack/nova stable/ussuri: Use absolute path during qemu img rebase https://review.opendev.org/c/openstack/nova/+/757083 | |
| 10:40:28 | openstackgerrit | Merged openstack/nova stable/ussuri: libvirt: Use specific user when probing encrypted rbd disks during extend https://review.opendev.org/c/openstack/nova/+/773444 | |
| 10:55:34 | openstackgerrit | Claudiu Belu proposed openstack/nova stable/train: WIP: live migration: Pass limits onto scheduler-chosen destination https://review.opendev.org/c/openstack/nova/+/784071 | |
| 11:02:13 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Correct typos in 'nova-manage cells_v2' docs https://review.opendev.org/c/openstack/nova/+/784073 | |
| 12:45:12 | openstackgerrit | Merged openstack/nova stable/train: Test for disabling greendns https://review.opendev.org/c/openstack/nova/+/761763 | |
| 13:09:30 | openstackgerrit | Stephen Finucane proposed openstack/nova-specs master: trivial: Fix spelling, formatting of vDPA spec https://review.opendev.org/c/openstack/nova-specs/+/781260 | |
| 14:43:59 | kashyap | stephenfin: Are you sure the default for AArch64 is coming from libvirt, instead of QEMU? | |
| 14:44:05 | kashyap | (For 'host-passthrough', i.e) | |
| 14:44:16 | stephenfin | nope | |
| 14:44:52 | kashyap | I'll confirm on the change; let me look in the code | |
| 14:45:03 | kashyap | And also ask an AArch64 dev, for good measure | |
| 14:46:27 | kashyap | I discussed this on an upstream docs change ... /me goes to find the link | |
| 14:47:19 | kashyap | stephenfin: Okay, from my own review comment: | |
| 14:47:25 | kashyap | For AArch64, we now set the ``max`` as the default CPU model; it does the right thing in context of both QEMU TCG (plain emulation) and for KVM (hardware acceleration). | |
| 14:47:25 | kashyap | [quote] | |
| 14:47:29 | kashyap | [/quote] | |
| 14:47:59 | kashyap | stephenfin: So I don't know where you got that 'host-passthrough' bit for AArch64; I'll comment on the change for the record | |
| 14:48:21 | stephenfin | I think that was based on comments from sean-k-mooney | |
| 14:48:47 | kashyap | That's incorrect, I'm afraid | |
| 14:51:45 | kashyap | stephenfin: Unlike x86, AArch64 does not even _have_ a default CPU model. We discussed it at length in this patch here: https://review.opendev.org/c/openstack/nova/+/709494 | |
| 14:56:11 | sean-k-mooney | kashyap: not accoundign to the code | |
| 14:56:33 | kashyap | sean-k-mooney: Very well it is; according to the code | |
| 14:56:44 | kashyap | https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L4753 | |
| 14:56:53 | kashyap | sean-k-mooney: And see this release note: https://review.opendev.org/c/openstack/nova/+/709494/21/releasenotes/notes/bug-1864588-737c29560effd16e.yaml | |
| 14:57:20 | kashyap | Which I recall distinctly asking to word it that way (because it's truth) :) | |
| 15:05:44 | sean-k-mooney | ill be starting the call soon by the way | |
| 15:05:50 | sean-k-mooney | just fisnihgn a differnt call | |
| 15:06:06 | sean-k-mooney | oh upstream yeah :) | |