| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-08-13 | |||
| 09:58:04 | slaweq | efried: I just opened bug https://bugs.launchpad.net/nova/+bug/1839961 as I see that this test if failing quite often | |
| 09:58:05 | openstack | Launchpad bug 1839961 in OpenStack Compute (nova) "Test tempest.api.compute.servers.test_novnc.NoVNCConsoleTestJSON.test_novnc failing often" [Undecided,New] | |
| 09:58:49 | slaweq | efried: please take a look at it if You will have few minutes, I opened it for nova but maybe it's e.g. tempest issue - I'm not 100% sure | |
| 10:30:08 | shilpasd | stephenfin: thanks for review 'isolate_aggregate' pre-request filter patches | |
| 10:30:34 | shilpasd | stephenfin: just small query on https://review.opendev.org/#/c/671075/6/nova/scheduler/request_filter.py@68 and @72 | |
| 10:30:46 | shilpasd | if we have only one ResourceRequest, then we can omit union, right? | |
| 10:31:37 | stephenfin | shilpasd: Yeah, I think so. That's what we do in nova/scheduler/utils.py https://github.com/openstack/nova/blob/master/nova/scheduler/utils.py#L461-L483 | |
| 10:32:05 | stephenfin | Ignore the numbered_groups_from_flavor piece and I think you have exactly what you want | |
| 10:33:01 | stephenfin | Also, there's a check there to see if 'Flavor.extra_specs' exists. However, I assumed it always would so I don't know why that's there. Maybe just tests? | |
| 10:35:55 | shilpasd | thanks for reference, will address accordingly, and in my changes will check do we need this check 'if 'Flavor.extra_specs' exists', then only add | |
| 10:36:29 | shilpasd | stephenfin: and for https://review.opendev.org/#/c/671074/5/nova/objects/aggregate.py@541 | |
| 10:36:34 | stephenfin | Makes sense. To say everything is passing so far, I imagine you don't, but it would be no harm to trace the paths to see if the field is ever _not_ set | |
| 10:36:37 | shilpasd | will edit doc message | |
| 10:37:55 | shilpasd | sure, will check if field is ever _not_ set, is any causing issue | |
| 11:01:19 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Make evacuation respects anti-affinity rule https://review.opendev.org/649963 | |
| 11:03:09 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Make evacuation respects anti-affinity rule https://review.opendev.org/649963 | |
| 11:08:27 | openstackgerrit | Boxiang Zhu proposed openstack/nova master: Fix live migration break group policy simultaneously https://review.opendev.org/651969 | |
| 11:44:52 | openstackgerrit | ya.wang proposed openstack/nova master: Add property check when create/rebuild/resize. https://review.opendev.org/674523 | |
| 11:44:52 | openstackgerrit | ya.wang proposed openstack/nova master: Support report driver migrate feature https://review.opendev.org/674522 | |
| 11:44:53 | openstackgerrit | ya.wang proposed openstack/nova master: Change method: _live_migration_flags https://review.opendev.org/674525 | |
| 11:44:53 | openstackgerrit | ya.wang proposed openstack/nova master: Scheduler translate properties to traits https://review.opendev.org/674524 | |
| 11:44:54 | openstackgerrit | ya.wang proposed openstack/nova master: Add releasenote and update docs. https://review.opendev.org/674526 | |
| 11:59:05 | mriedem | stephenfin: replied in https://review.opendev.org/#/c/627892/ | |
| 12:02:22 | openstackgerrit | Peter Penchev proposed openstack/nova master: libvirt: use native AIO mode for StorPool Cinder volumes. https://review.opendev.org/676172 | |
| 12:26:11 | mriedem | tssurya: i've informally reserved 2.76 for you | |
| 12:37:49 | tssurya | mriedem: just saw the ML, thanks a lot Matt! :) | |
| 12:40:24 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Prevent init_host test to interfere with other tests https://review.opendev.org/675553 | |
| 12:56:11 | yonglihe | melwitt: I came up with some comment to the code you concerned, hope it address your problem. glad to have you take a | |
| 12:56:12 | yonglihe | look: https://review.opendev.org/#/c/627765/36/nova/compute/manager.py | |
| 12:58:43 | yonglihe | sean-k-mooney: I bet you will be here again -:) , so could you please take another round of review for https://review.opendev.org/#/c/627765. | |
| 12:59:58 | sean-k-mooney | hehe i am, ill try an take a look again later today | |
| 13:01:43 | yonglihe | Thanks | |
| 13:02:18 | yonglihe | Have a good review day, Sean, -:) . | |
| 13:36:21 | tssurya | dansmith, mriedem: I opened this bug a few days ago : https://bugs.launchpad.net/nova/+bug/1839009 , would be could if someone could confirm it since I am not exactly sure if its a bug or if its intended to be that way. | |
| 13:36:22 | openstack | Launchpad bug 1839009 in OpenStack Compute (nova) "os-server-external-events does not behave correctly for failed single events" [Undecided,New] | |
| 13:39:09 | dansmith | tssurya: yeah, looks like maybe the assumption around "if accepted events" meaning "no instances" is too naive | |
| 13:39:46 | tssurya | dansmith: yea | |
| 13:42:44 | dansmith | I dunno what the api recommendation is here, | |
| 13:42:56 | dansmith | but I would think that just removing the else: raise bit there would be fine, | |
| 13:43:04 | dansmith | since we calculate the overall and per-event result regardless | |
| 13:43:35 | dansmith | I think that else is left from my original code that didn't try to do per-event status | |
| 13:45:25 | tssurya | dansmith: yea we could, but wouldn't it change the existing api behaviour? | |
| 13:47:17 | dansmith | well, obviously it would, but I think it falls under the "people don't write clients to depend on obviously broken behavior" | |
| 13:47:57 | dansmith | we could specifically check for "they provided a legit single-event but the instance just wasn't found, so do this special 404" I guess | |
| 13:49:30 | tssurya | dansmith: hmm yea I didn't do anything to the bug since I wasn't sure what was the right way out | |
| 13:49:53 | tssurya | but like you said as least for ironic the client assumes everything is done correctly on server side | |
| 14:03:22 | redkrieg | lyarwood: thanks for getting back to me. if there's anything I can do to help with stable device rescue, please let me know. it would solve some big pain points for me and my clients. | |
| 14:07:38 | openstack | Launchpad bug 1669468 in devstack "duplicate for #1839961 tempest.api.compute.servers.test_novnc.NoVNCConsoleTestJSON.test_novnc fails intermittently in neutron multinode nv job" [Medium,Fix released] - Assigned to melanie witt (melwitt) | |
| 14:07:38 | efried | slaweq: I've seen https://bugs.launchpad.net/nova/+bug/1839961 quite a bit as well. mriedem is our resident bugmeister, though, perhaps he has insight. | |
| 14:08:00 | slaweq | efried: thx, yes, mriedem already marked it as duplicate of other bug | |
| 14:08:03 | slaweq | sorry for noise than :) | |
| 14:08:09 | mriedem | the devstack patch was merged yesterday | |
| 14:08:18 | slaweq | and thx for fix | |
| 14:08:24 | mriedem | http://status.openstack.org/elastic-recheck/#1669468 | |
| 14:08:33 | mriedem | if it's still hitting we might have other problems | |
| 14:08:42 | mriedem | https://review.opendev.org/#/c/675721/ | |
| 14:09:29 | efried | now the list is down to 167 | |
| 14:16:42 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Support reverting migration / resize with bandwidth https://review.opendev.org/676140 | |
| 14:17:21 | mriedem | tssurya: when are you on vacation? | |
| 14:26:31 | Tianhao_Hu | About this issue, there are files left after cold migration every time. But the directory left is empty and has no effect on cold migration. I'm wondering that if we can think this is not a bug? Can anyone give me some help on this? | |
| 14:26:35 | openstack | Launchpad bug 1824858 in StarlingX "nova instance remnant left behind after cold migration completes" [Low,Confirmed] - Assigned to hutianhao27 (hutianhao) | |
| 14:26:35 | Tianhao_Hu | https://bugs.launchpad.net/starlingx/+bug/1824858 | |
| 14:30:49 | mriedem | tssurya: dansmith: efried: https://review.opendev.org/#/c/645611/ | |
| 14:32:05 | mriedem | i didn't compare to my last set of comments (PS7) because the diff is too large | |
| 14:32:22 | mriedem | tssurya: if you're going to be on vacation soon, like leaving today/tomorrow, then i can take this over | |
| 14:33:09 | mriedem | i think my main concern is whether or not operators can disable ironic sending the event until nova is completely upgraded | |
| 14:33:28 | efried | They can, there's a conf option on the ironic side. | |
| 14:33:48 | efried | https://review.opendev.org/#/c/664842/19/ironic/conf/nova.py | |
| 14:33:50 | mriedem | efried: you mean just to disalbe nova? | |
| 14:33:53 | dansmith | and unless ironic really cares, it shouldn't be an issue | |
| 14:34:02 | mriedem | ah send_power_notifications | |
| 14:34:06 | mriedem | yeah we should mentio nthat in the nova release note | |
| 14:34:10 | dansmith | I expect ironic is sending these to be informative, but not depending on the result | |
| 14:34:16 | mriedem | dansmith: that's not the issue, | |
| 14:34:50 | mriedem | the issue is if you're doing rolling upgrades of your computes, you've upgraded ironic before nova, then nova control plane, and you can be getting the events, the api sets the task_state on the instance and casts to an old compute that doesn't know how to handle this event and then your instance is stuck with a task_state that won't go back to None w/o admin intervention | |
| 14:35:20 | dansmith | okay yeah, good point | |
| 14:35:23 | openstackgerrit | Stephen Finucane proposed openstack/nova master: mypy: Add type annotations to 'nova.pci' https://review.opendev.org/676209 | |
| 14:35:23 | openstackgerrit | Stephen Finucane proposed openstack/nova master: WIP! tox: Add mypy target https://review.opendev.org/676208 | |
| 14:35:24 | openstackgerrit | Stephen Finucane proposed openstack/nova master: zuul: Add 'mypy' job https://review.opendev.org/676210 | |
| 14:35:33 | dansmith | even if we can disable the ironic side, the nova side needs to not do that | |
| 14:36:25 | dansmith | tbh, the api setting the task state has kinda rubbed me the wrong way the whole time anyway, | |
| 14:36:39 | dansmith | as it seems like we should just forward this down to the compute and let it do its thing if needed | |
| 14:36:57 | mriedem | i mentioned in ealier review that the API could check the compute service version, but because volume-extended was merged w/o doing that and has the same issue, it was going to slide here as well | |
| 14:37:19 | dansmith | setting task state is to help with races to take action on events, and if we can't do the power state short-circuit because something else has started, it's fine to fail/log late | |
| 14:37:32 | stephenfin | mriedem: Question on https://review.opendev.org/#/c/631581/44/nova/conductor/tasks/cross_cell_migrate.py | |
| 14:37:57 | mriedem | dansmith: that was another thing i pointed out, the API code isn't handling UnexpectedTaskStateError if we lose a race so the API would 500 and fail for all events | |
| 14:38:10 | mriedem | *all events in that request | |
| 14:39:02 | dansmith | mriedem: yeah, so I think I'd use this as an argument to not make this something more complex on the api side | |
| 14:41:37 | mriedem | stephenfin: replied | |
| 14:45:05 | mriedem | stephenfin: i realized late that the cinder 3.44 check in there can be removed now https://review.opendev.org/#/c/649759/ | |
| 14:45:15 | mriedem | i wrote the cross-cell patch before we bumped the min required cinder, | |
| 14:45:18 | mriedem | so i'll remove that in a follow up | |
| 14:45:33 | stephenfin | I also missed that. Yeah, follow-up would be good (y) | |
| 15:09:56 | dansmith | mriedem: I commented on tssurya's patch about the api task_state involvement, and also thought of another problem with that | |
| 15:26:18 | mriedem | dansmith: i don't think the libvirt thing is an issue if you're compute is train+ | |
| 15:26:37 | mriedem | b/c the driver will raise NotImplementedError and @reverts_task_state will set the task_state to None | |
| 15:26:37 | dansmith | no? | |
| 15:26:54 | dansmith | is that a call? | |
| 15:26:56 | mriedem | i commented on that in one of the unit tests - as in that scenario is not asserted | |
| 15:27:09 | mriedem | @reverts_task_state is a decorator on the compute managermethod | |