| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-02-28 | |||
| 10:22:56 | gibi | bauzas: ^^ | |
| 10:23:15 | stephenfin | bauzas, gibi: Both refer to config options that workaround bugs in older libvirt versions that we no longer support. In the former, I just dropped it. In the latter, I did the full deprecation cycle. Thoughts on which is preferable? | |
| 10:23:42 | bauzas | stephenfin: I'm in favor of being gentle with our ops | |
| 10:25:21 | bauzas | stephenfin: what's the libvirt min version we support ? | |
| 10:25:24 | bauzas | kashyap: ^? | |
| 10:25:49 | gibi | stephenfin: the config driver format config still works even if it is not needed to set to vfat | |
| 10:25:49 | stephenfin | bauzas: 3.0 | |
| 10:25:58 | gibi | stephenfin: so there keeping the config is meaningful | |
| 10:26:32 | gibi | stephenfin: would disable_libvirt_livesnapshot still disable the snapshot in Stein if we keep it? | |
| 10:26:44 | gibi | if yes, then I would just deprecate that config too | |
| 10:26:55 | stephenfin | gibi: Yeah, it would. I'll deprecate that instead so | |
| 10:26:59 | bauzas | stephenfin: the problem is that we provided a flag which, even if tied to a specific version, allows a specific behaviour | |
| 10:27:17 | bauzas | stephenfin: so people can use it even if they have newer libvirt | |
| 10:37:34 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Deprecate 'disable_libvirt_livesnapshot' option https://review.openstack.org/626932 | |
| 10:37:59 | stephenfin | bauzas, gibi: Updated to deprecate instead of remove outright, if you fancy hitting that too ^ | |
| 10:39:25 | openstackgerrit | Yongli He proposed openstack/nova master: Adds the server group info into show server detail API. https://review.openstack.org/621474 | |
| 10:42:05 | kashyap | bauzas: Hi | |
| 10:42:18 | kashyap | bauzas: Was briefly AFK; it's documented in the driver.py :-) | |
| 10:42:20 | openstackgerrit | Yongli He proposed openstack/nova master: Adds the server group info into show server detail API. https://review.openstack.org/621474 | |
| 10:43:26 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Dedupe controller install guides https://review.openstack.org/638715 | |
| 10:43:27 | openstackgerrit | Stephen Finucane proposed openstack/nova master: docs: Dedupe compute install guides https://review.openstack.org/638716 | |
| 10:43:51 | kashyap | bauzas: Currently these: | |
| 10:43:51 | kashyap | MIN_LIBVIRT_VERSION = (3, 0, 0) | |
| 10:43:52 | kashyap | MIN_QEMU_VERSION = (2, 8, 0) | |
| 10:44:43 | openstackgerrit | Stephen Finucane proposed openstack/nova master: conf: Call out where pci.alias should be set https://review.openstack.org/624445 | |
| 11:46:02 | mkarpiarz | Hi. Can someone please take a look at https://bugs.launchpad.net/nova/+bug/1780225/comments/3? | |
| 11:46:04 | openstack | Launchpad bug 1780225 in OpenStack Compute (nova) "Libvirt error when using --max > 1 with vGPU" [High,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | |
| 11:46:45 | mkarpiarz | I can see 2 ways of fixing this | |
| 11:48:02 | mkarpiarz | 1. Store the list of available mediated devices in some global structure. | |
| 11:49:54 | mkarpiarz | 2. Let the instance fail in launch and return back to spawn(), so it can check available devices again. | |
| 11:51:52 | mkarpiarz | Any other ideas? | |
| 11:54:18 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP context: Ensure templated connection URLs are rendered before use https://review.openstack.org/639607 | |
| 12:12:44 | openstackgerrit | Merged openstack/nova master: conf: Deprecated 'config_drive_format' https://review.openstack.org/608407 | |
| 12:32:00 | openstackgerrit | Artom Lifshitz proposed openstack/nova master: CONF.enable_numa_live_migration is not needed >= Stein https://review.openstack.org/640021 | |
| 12:35:06 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add remove_resources_from_instance_allocation to report client https://review.openstack.org/639653 | |
| 12:36:09 | gibi | efried: fixed nits and replied about the retry in ^^ | |
| 12:43:01 | openstackgerrit | sean mooney proposed openstack/nova master: Add free for claimed, allocated devices https://review.openstack.org/616120 | |
| 12:43:01 | openstackgerrit | sean mooney proposed openstack/nova master: Allow per-port modification of vnic_type and profile https://review.openstack.org/607365 | |
| 12:43:02 | openstackgerrit | sean mooney proposed openstack/nova master: Add get_instance_pci_request_from_vif https://review.openstack.org/619929 | |
| 12:43:02 | openstackgerrit | sean mooney proposed openstack/nova master: SR-IOV Live migration indirect port support https://review.openstack.org/620115 | |
| 12:43:03 | openstackgerrit | sean mooney proposed openstack/nova master: libvirt: auto detach/attach sriov ports on migration https://review.openstack.org/629589 | |
| 12:43:55 | sean-k-mooney | resolved minor merge conflict and rebased to master otherwise no other changes in that series | |
| 12:47:47 | tacco | hi there.. i have a issue with a redeployed hypervisor.. after deployment i see a "ResourceProviderCreationFailed: Failed to create resource provider" error when trying to start nova or migrating a instance to this hypervisor https://pastebin.com/zMHUUYEW | |
| 12:55:49 | yaawang | hi, could anyone please take a review on https://review.openstack.org/#/c/637834/, this a patch about libvirt cpu models select :-) | |
| 13:31:57 | openstackgerrit | Lee Yarwood proposed openstack/nova master: WIP context: Ensure templated connection URLs are rendered before use https://review.openstack.org/639607 | |
| 13:33:10 | artom | sean-k-mooney, wait, we can have nested virt in nodepool now? mnaser, is vexxhost doing that now? | |
| 13:33:48 | mnaser | artom: we’re doing it but | |
| 13:33:53 | sean-k-mooney | artom: technically. the ci jobs will still try to disable it but we can renable it | |
| 13:33:59 | mnaser | I’m trying not to broadcast it too much | |
| 13:34:15 | artom | mnaser, oops, sorry :) | |
| 13:34:22 | mnaser | I want the jobs that use it are jobs that need to use it, not.. just do it can be faster | |
| 13:34:41 | mnaser | So Magnum is a case where it was useful and that’s why we added it | |
| 13:34:41 | artom | mnaser, well, NUMA live migration *needs* to use it :) | |
| 13:34:44 | sean-k-mooney | mnaser: i brought it up in the context of testing nuam aware live migration | |
| 13:34:55 | mnaser | Yeah then I’d be supportive of it | |
| 13:35:00 | artom | mnaser, sweet! | |
| 13:35:14 | sean-k-mooney | artom: as i said it would have to be non voting however | |
| 13:35:28 | yonglihe | mriedem: alex_xu: https://review.openstack.org/621474 updated per all comments. I have to have rest. have a good one everyone. | |
| 13:35:29 | artom | sean-k-mooney, yeah that's fine, we have to start somewhere | |
| 13:35:43 | artom | Even just being able to trigger a job manually that posts publicly would be a great first step | |
| 13:36:20 | sean-k-mooney | artom: if ovh/packet reenable it in the future we might be able to make it voting eventually | |
| 13:36:30 | artom | sean-k-mooney, baby steps :) | |
| 13:36:32 | sean-k-mooney | artom: well you could start by adding it to the experimental pipeline | |
| 13:36:52 | sean-k-mooney | that requires a comment ot trigger it | |
| 13:36:58 | artom | sean-k-mooney, wouldn't I have to start by moving whitebox to the openstack namespace? | |
| 13:37:08 | sean-k-mooney | yes | |
| 13:37:15 | sean-k-mooney | i ment after that | |
| 13:37:27 | sean-k-mooney | and you know after we write all the tests | |
| 13:37:41 | artom | Hey, I have some written already :) | |
| 13:38:04 | sean-k-mooney | did you look at the list i gave in the email | |
| 13:38:17 | artom | Yeah | |
| 13:38:19 | sean-k-mooney | feel free to automate those while i test them | |
| 13:38:52 | artom | sean-k-mooney, that'll kinda depend on whether the absence of func tests is a blocker for mriedem | |
| 13:39:04 | artom | If it is, those take priority before FF | |
| 13:39:15 | sean-k-mooney | yep | |
| 13:39:16 | artom | Other than that timing issue, I'll happily hax those up | |
| 13:43:03 | sean-k-mooney | artom: once you get the first functional test working i suspect the rest will be fairly staight forward | |
| 13:44:05 | sean-k-mooney | if you can add the functional test as a follow up that will reduce churn and make my testing eaisier | |
| 13:44:42 | artom | sean-k-mooney, yep, good idea. I can start with the tests in a new patch on top, and we can squash later | |
| 13:45:04 | sean-k-mooney | yep or even just keep it on top | |
| 13:45:46 | sean-k-mooney | squashing them to keep them with the feature code is nice but i dont think its required | |
| 14:00:24 | sean-k-mooney | mnaser: by they way do you use kvm on ubuntu 18.04 for vexhost? | |
| 14:00:33 | mnaser | sean-k-mooney: centos 7 | |
| 14:00:38 | sean-k-mooney | ah ok | |
| 14:00:46 | mnaser | but we've managed to get the perfect combo for functional nested virt | |
| 14:00:51 | mnaser | thanks a lot to the kata containers folks | |
| 14:00:57 | sean-k-mooney | there is a bug in the ubunut 18.04 kernel that breaks nested virt | |
| 14:01:18 | sean-k-mooney | i had to replace mine with the mainline 4.20 kernel to fix issues | |
| 14:01:19 | mnaser | yeah our ci is worked out pretty often both by magnum and kata and their nested virt has been super stable | |
| 14:02:46 | sean-k-mooney | ya i went with ubuntu 18.04 because kolla-ansibel source deploys work better on ubuntu then centos in my previos experience | |
| 14:03:02 | sean-k-mooney | mnaser: do ye use openstack ansible for your production installation | |
| 14:03:07 | mnaser | sean-k-mooney: yeah | |
| 14:03:56 | sean-k-mooney | i must get around to checking that out at some point but since its working im not going to touch it untill i need to upgrade :) | |
| 14:10:04 | mdbooth | bauzas: Do you recall why rebuild is forced to be on the same host? Is it just because the way rebuild is written doesn't allow for a migration, or is it fundamental to the intent of rebuild? Or some other reason? | |
| 14:12:10 | sean-k-mooney | mdbooth: why would a rebuild ever need to migrate if the image does not change | |
| 14:12:21 | sean-k-mooney | if the image does change we hit the schulder | |
| 14:12:25 | mdbooth | bauzas: I ask because I was looking at the attach/detach root volume patches. They look fine, but shelve/unshelve is just distasteful, so having it as the only available option seems undesirable. Reading the original spec, it sounds to me like they want to rebuild on a different host... | |
| 14:12:30 | mdbooth | Which is an evacuate# | |
| 14:12:42 | mdbooth | Evacuate with a different root volume. | |
| 14:12:43 | mdbooth | Hmm | |