| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2019-02-13 | |||
| 14:46:08 | sean-k-mooney | i have done ^ 3 times in the last week for this issue on different systems | |
| 14:46:28 | sean-k-mooney | but ill look at actully just makeing devstack fix it properly next week | |
| 14:46:36 | bauzas | cdent: edleafe: efried: somehow, we changed from nova-placement-api.conf to placement-api.conf without thinking of folks who upgrade | |
| 14:46:38 | bauzas | :) | |
| 14:46:48 | bauzas | words of wisdom, people do upgrade :) | |
| 14:47:01 | cdent | of devstack? | |
| 14:47:01 | sean-k-mooney | bauzas: yep but its also devstack soo.... | |
| 14:47:22 | cdent | i thought the three letters on the front was a clue? | |
| 14:47:43 | sean-k-mooney | i assume grenade is just nuking the configs and replaceing them with new ones | |
| 14:48:12 | sean-k-mooney | cdent: it still shows up on the openstack survay in prod ocationally | |
| 14:48:51 | sean-k-mooney | cdent: honestly with the ansible rules we use in the gate for deploying it and the swap to systemd for all the things its actully a lot less dumb then it used to be | |
| 14:49:02 | sean-k-mooney | cdent: it now support host reboots again | |
| 14:49:03 | bauzas | cdent: you don't imagine how much devstack is insightful for deployers :) | |
| 14:49:17 | cdent | I'm not disputing that devstack should clean up apache configs better. Just trying to point out that this is an old and not related to placement problem and not something that is super urgent. it's just a bug that needs to be fixed | |
| 14:49:17 | bauzas | but anyway, not a big deal | |
| 14:49:21 | cdent | anyone can fix it | |
| 14:50:17 | sean-k-mooney | fixed and then backported as this only happens if you swap branches | |
| 14:50:51 | jroll | I'm surprised clean.sh doesn't remove apache configs, seems like an easy fix | |
| 14:50:55 | sean-k-mooney | which developer do a lot but ci systems and normal peole do not | |
| 14:51:44 | cdent | jroll: I really think it simply comes down to "nobody has got around to it yet or been burned enough to bother". sean-k-mooney is nearly there. | |
| 14:51:47 | sean-k-mooney | jroll: it also does not remove apache. i mean that is the super simple fix but not makeing the config generate condtional woudl also have worked | |
| 14:51:51 | cdent | sounds like bauzas is already there but doesn't have time | |
| 14:53:26 | sean-k-mooney | im takeing 2 days off this week and need to get other stuff done before i do which is the only reason i havent done it yet. | |
| 14:58:22 | kashyap | As promised earlier: http://lists.openstack.org/pipermail/openstack-discuss/2019-February/002665.html -- [ops][nova] Heads-up: Upcoming version bump for libvirt and QEMU in 'Stein' | |
| 14:59:41 | openstackgerrit | Kashyap Chamarthy proposed openstack/nova master: libvirt: Bump MIN_{LIBVIRT,QEMU}_VERSION for "Stein" https://review.openstack.org/632507 | |
| 14:59:42 | openstackgerrit | Kashyap Chamarthy proposed openstack/nova master: libvirt: Drop MIN_LIBVIRT_PARALLELS_SET_ADMIN_PASSWD https://review.openstack.org/632514 | |
| 15:00:28 | kashyap | gibi: stephenfin: Just a commit message update on the main patch ^ | |
| 15:02:35 | openstackgerrit | Alexandra Settle proposed openstack/nova master: Adding cross refs for config options in scheduler filter guide https://review.openstack.org/636635 | |
| 15:06:35 | bauzas | gibi: so, I uploaded my change | |
| 15:06:46 | bauzas | gibi: but now the func test you wrote is failing | |
| 15:06:50 | bauzas | weird, heh ? | |
| 15:06:52 | gibi | bauzas: sorry, I haven't looked it yet | |
| 15:07:05 | bauzas | no worries, just saying | |
| 15:08:21 | bauzas | gibi: I don't get why but the root RP has a VGPU inventory, hence the mismatch | |
| 15:09:30 | bauzas | gibi: http://logs.openstack.org/59/631559/5/check/nova-tox-functional/100cfb6/job-output.txt.gz#_2019-02-13_11_19_37_217415 | |
| 15:09:40 | bauzas | gibi: any reason why you don't provide the VGPU inventory first ? | |
| 15:19:41 | openstackgerrit | Kashyap Chamarthy proposed openstack/nova master: libvirt: Drop MIN_LIBVIRT_PARALLELS_SET_ADMIN_PASSWD https://review.openstack.org/632514 | |
| 15:39:14 | gibi | bauzas: sorry I was dragged away from the keyboard | |
| 15:39:24 | bauzas | gibi: heh, we all are | |
| 15:39:38 | bauzas | or, rather, always dragged from upstream | |
| 15:42:56 | gibi | bauzas: if https://review.openstack.org/#/c/631559/5/nova/tests/functional/libvirt/test_reshape.py@130 fails then it means that the reshape does not moved the VGPU resource away from the compute RP to the GPU RP | |
| 15:43:15 | gibi | bauzas: https://review.openstack.org/#/c/631559/5/nova/tests/functional/libvirt/test_reshape.py@123 should trigger the reshape | |
| 15:43:41 | bauzas | I'll dig into pdbing it | |
| 15:43:58 | bauzas | in the meantime, lemme mark my series as -2 | |
| 15:54:04 | openstackgerrit | Alexandra Settle proposed openstack/nova master: Adding cross refs for config options in scheduler filter guide https://review.openstack.org/636635 | |
| 15:59:07 | gibi | bauzas: http://logs.openstack.org/59/631559/5/check/nova-tox-functional/100cfb6/job-output.txt.gz#_2019-02-13_11_19_37_215302 here is when the compute is restarted and should do the reshape | |
| 16:02:00 | openstackgerrit | Jack Ding proposed openstack/nova master: Flavor extra spec and image properties validation https://review.openstack.org/620706 | |
| 16:24:19 | melwitt | o/ | |
| 16:50:45 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Use math.gcd starting with python 3.5 https://review.openstack.org/636669 | |
| 16:51:40 | melwitt | mriedem: there's a patch proposed to add basic info about projects to the openstack-map, if you're interested in reviewing the nova info https://review.openstack.org/636604 | |
| 17:01:46 | openstackgerrit | Vlad Gusev proposed openstack/nova stable/queens: Handle unicode characters in migration params https://review.openstack.org/636678 | |
| 17:09:09 | mriedem | melwitt: commented | |
| 17:09:13 | mriedem | dansmith: replied on https://review.openstack.org/#/c/469675/ | |
| 17:09:34 | mriedem | dansmith: for sanity, i think we likely just want to stop using check_availability_zone in both the api and compute | |
| 17:09:39 | mriedem | and write an api-specific version of that | |
| 17:10:26 | mriedem | from the api, gather all the precreated volumes and their azs, | |
| 17:10:30 | mriedem | if any are different, it's an error | |
| 17:10:38 | mriedem | if any are different from the requested server az, it's an error | |
| 17:10:58 | mriedem | if the server is created without a specific az, and the volumes all have the same az, then put that on the request spec | |
| 17:12:32 | dansmith | mriedem: yeah, collect the data and make the call in the compute api | |
| 17:12:55 | dansmith | mriedem: like I said, if the "user requested an az and it doesn't match the volume" case still needs to be handled deep for cleanup reasons, then okay, | |
| 17:13:34 | dansmith | but (a) default_az means something totally different to me when reading the compute/api code and (b) I don't expect either of these mismatch checks to happen so deep down in that file | |
| 17:14:03 | dansmith | so if you don't want to refactor this at all, I'd at least ask that you change the name of that parameter with commentage up in compute/api aboutit | |
| 17:15:28 | mriedem | "the "user requested an az and it doesn't match the volume" case" is what would be in nova-compute after a volume is created | |
| 17:15:33 | mriedem | so yeah i think that remains | |
| 17:16:01 | dansmith | wait | |
| 17:16:23 | dansmith | oh right, yeah | |
| 17:17:00 | mriedem | which reminds me of https://bugs.launchpad.net/nova/+bug/1497253 | |
| 17:17:02 | openstack | Launchpad bug 1497253 in OpenStack Compute (nova) "different availability zone for nova and cinder when AZ is not explicitly given" [Low,Fix released] - Assigned to Dan Smith (danms) | |
| 17:19:21 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Remove get_config_vhostuser https://review.openstack.org/565471 | |
| 17:19:32 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Validate bandwidth configuration for other VIF types https://review.openstack.org/636383 | |
| 17:19:44 | openstackgerrit | Stephen Finucane proposed openstack/nova master: Further de-dupe os-vif VIF tests https://review.openstack.org/636384 | |
| 17:21:12 | stephenfin | gibi: Trivial doc fix here, if you fancy taking a look https://review.openstack.org/#/c/636635/ | |
| 17:26:09 | mriedem | dansmith: right so i'm thinking leaving the nova-compute behavior the same, and probably move that check_availability_zone method to nova.virt.block_device where it's actually used | |
| 17:26:16 | mriedem | and writing something new in the API | |
| 17:26:22 | dansmith | makes sense | |
| 17:26:25 | mriedem | that has the user-requested (or not) AZ context | |
| 17:26:59 | mriedem | having said that, it's pot-pie-o-clock and i've got a physical this afternoon (first in 10+ years) so my attendance will be spotty | |
| 17:27:14 | mriedem | oh and (don't tell laura) i have to get a valentine | |
| 17:27:44 | dansmith | mriedem: you might be approaching that age where the doctor starts wanting to take your relationship to the next level | |
| 17:27:58 | mriedem | i'm holding out for my 40s | |
| 17:28:03 | dansmith | heh | |
| 17:34:32 | tzumainn | hi! quick question about the expected behavior with nova and ironic; I have four baremetal nodes, and I've created one server that runs on one of the nodes, but if I run "nova hypervisor-servers" against each node, that server shows up for every single node | |
| 17:40:36 | melwitt | tzumainn: are you running one nova-compute service for all four baremetal nodes? if so, I think that might be why it shows up for every node | |
| 17:41:08 | melwitt | I'd have to look into the code to find how the hostname => server lookup is done to confirm | |
| 17:41:30 | melwitt | jroll ^ | |
| 17:41:43 | tzumainn | melwitt, ah, yeah, I am | |
| 17:42:44 | jroll | I'm not sure what "nova hypervisor-servers" is | |
| 17:42:56 | jroll | but yeah, probably only goes by compute service hostname, not node name | |
| 17:43:26 | melwitt | it's a command to get a list of servers given a hostname | |
| 17:46:29 | bauzas | dansmith: stop me 1 sec, do we now need to set the host mappings when deploying with devstack ? | |
| 17:46:56 | dansmith | bauzas: devstack should do that for you | |
| 17:47:06 | jroll | melwitt: ah, sounds like a bug in that API response | |
| 17:47:26 | bauzas | dansmith: that was my assumption but I got a Host 'XXX' is not mapped to any cell | |
| 17:47:54 | dansmith | bauzas: on a vanilla devstack setup? shouldn't happen afaik | |
| 17:48:30 | bauzas | technically, a reinstalled devstack but with a ./clean.sh and a reclone=true | |
| 17:48:38 | melwitt | jroll: I get what you mean now, might be returning host instead of node | |
| 17:48:41 | bauzas | weirdo | |
| 17:49:02 | jroll | melwitt: yeah | |
| 17:49:34 | melwitt | thanks | |