| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-10-22 | |||
| 10:37:30 | adrianc | sean-k-mooney: Hi, id like to collaborate on https://blueprints.launchpad.net/nova/+spec/libvirt-neutron-sriov-livemigration can we have a blue-jeans meeting to discuss ? | |
| 10:38:49 | sean-k-mooney | adrianc: sure we could that said it is best to keep as much of these discussion on irc/ml/gerrit when it comes to the design side | |
| 10:41:08 | openstackgerrit | Adam Spiers proposed openstack/nova-specs master: Add spec for libvirt driver launching AMD SEV-encrypted instances https://review.openstack.org/609779 | |
| 10:41:20 | sean-k-mooney | adrianc: have you reviewed the spec i have published yet? | |
| 10:41:53 | sean-k-mooney | ah i see you have | |
| 10:41:56 | Dinesh_Bhor | Hi All, Is there a recommended/official API rate limiter for Nova? | |
| 10:42:14 | adrianc | sean-k-mooney: agree(also commented on your spec), I have also a WIP POC, myself and moshele would like to have a short sync on the topic | |
| 10:44:06 | sean-k-mooney | adrianc: did you have a time in mind. i would like th discuss the spec/blueprint at the next nova team meeting to see if we can at least get the blueprint approved | |
| 10:44:22 | sean-k-mooney | regarding the poc how far have you gotten | |
| 10:44:40 | sean-k-mooney | i had started on the same but i have been pulled into too other activities also | |
| 10:45:26 | sean-k-mooney | i had hoped to work on adding pci address claiming this week/next | |
| 10:45:54 | adrianc | sean-k-mooney: for indirect SRIOV with macvtap its working apart of the fact that no multiple port binding implementation in this mech driver | |
| 10:46:49 | adrianc | sean-k-mooney: including the pci resource claim for the migrated instance | |
| 10:47:16 | sean-k-mooney | adrianc: well wiht what i have published where im not doing the claim macvtap works | |
| 10:48:00 | sean-k-mooney | https://review.openstack.org/#/c/607365/ crrects the multiple port binding issue | |
| 10:48:21 | sean-k-mooney | then when i comment out this exception https://review.openstack.org/#/c/608788/1/nova/network/neutronv2/api.py | |
| 10:48:43 | sean-k-mooney | the migration succeeds. i was planning on trying to get the claim to work. | |
| 10:48:50 | sean-k-mooney | later this week | |
| 10:49:06 | adrianc | sean-k-mooney: was the neutron port in ACTIVE state ? | |
| 10:49:11 | adrianc | after migration that is | |
| 10:49:36 | sean-k-mooney | i did not check but i could ssh to the vm. | |
| 10:50:15 | sean-k-mooney | i belive it should be but we may need to modify the ml2 driver on the neutron side. | |
| 10:50:44 | sean-k-mooney | it has been almost 2 weeks since i worked on this since i was on training most of last week | |
| 10:52:40 | sean-k-mooney | adrianc: do you have any review pushed to gerrit | |
| 10:54:06 | adrianc | sean-k-mooney: not yet, its on local branch. | |
| 11:04:20 | sean-k-mooney | adrianc: did you have a time in mind to chat about this futher. | |
| 11:06:08 | adrianc | sean-k-mooney: im on PTO tomorrow but Wednesday would be great | |
| 11:08:47 | sean-k-mooney | sure that should work i have one or two meetsing around 4pm IST(UTC+1) | |
| 11:13:32 | adrianc | sean-k-mooney: great, can you set something at around 11AM (UTC+1) | |
| 11:14:36 | sean-k-mooney | that works for me. im not sure if there are any US based folks that should join but assumeing not then yes | |
| 11:16:42 | adrianc | great, no US folks from my side. | |
| 11:17:44 | sean-k-mooney | cool we may want to check in with artom who will not be online at that time but i can sync with him later | |
| 11:18:30 | artom | I'm here, barely | |
| 11:19:11 | sean-k-mooney | artom: its 6am for you now right? | |
| 11:19:17 | artom | 7:20 | |
| 11:19:28 | aspiers | ouch | |
| 11:20:04 | sean-k-mooney | artom: why? why would you be awake and on irc at this time? | |
| 11:21:00 | artom | Heh, I woke up on my own, 6-7 hours of sleep is usually enough for me | |
| 11:21:33 | artom | And I wanted to get a downstream escalation out of the way, hence laptop and IRC | |
| 11:22:07 | sean-k-mooney | artom: i was just talking to adrianc about having a call about sriov livemigtion at 11:00 utc+1 on wednesday. is that something you want to join i was sugessting it might be a bit too early for you but not sure you needed to attend either | |
| 11:22:19 | artom | Also, I need to literally go buy milk, otherwise 1. kids will complain that I'm a bad parent and 2. I'll have nothing to put in my coffee | |
| 11:22:58 | artom | sean-k-mooney, err, can you give me a time delta for the meeting? Like, "in 3 hours"? | |
| 11:22:59 | sean-k-mooney | artom: hehe no worres coff.. childern are importent | |
| 11:23:38 | sean-k-mooney | artom: it 12:22 now so it would be your 6 am on wednesday | |
| 11:23:49 | artom | sean-k-mooney, hah, no. | |
| 11:23:51 | artom | :) | |
| 11:34:17 | jangutter | sean-k-mooney: is https://review.openstack.org/#/c/608693/ now more or less in line with your idea of the "generic representor" thingy? If so, I'll update the spec again to reflect this. | |
| 11:46:42 | openstackgerrit | Jie Li proposed openstack/nova-specs master: Support volume-backed server rebuild https://review.openstack.org/532407 | |
| 11:50:27 | artom | sean-k-mooney, thanks for thinking of including me, but 6:00 is too damn early for a meeting | |
| 11:50:46 | artom | I'll happily review the specs/code in the normal fashion | |
| 12:03:03 | sean-k-mooney | artom: no worries ill can give you the clip notes if it impacts the numa aware migration in anyway | |
| 12:08:51 | sean-k-mooney | jangutter: sorry spinning a few plates currently ill comment on the review later today | |
| 12:09:34 | jangutter | sean-k-mooney: no rush at all, crockery comes first | |
| 12:09:58 | sean-k-mooney | jangutter: if i remember correctly it did but ill have to check it later to confirm | |
| 12:29:37 | jaosorior | hey folks, when configuring a vendordata plugin, from where is the vendordata plugin called? is it from nova-metadata? or from nova-api? or the compute itself? | |
| 12:38:31 | jaypipes | aspiers: ok, sorry for delay. https://review.openstack.org/#/c/609779/ is now reviewed. | |
| 12:38:44 | aspiers | thanks! | |
| 12:38:47 | jaypipes | np! :) | |
| 12:39:01 | jaypipes | aspiers: sorry, was out at a $work thing all last week :( | |
| 12:39:10 | aspiers | np at all | |
| 12:39:17 | aspiers | I had enough other feedback to keep me busy ;-) | |
| 12:39:23 | jaypipes | rock on :) | |
| 12:40:05 | sapd1 | Hi everyone, I get this errror when live migrate instance. http://paste.openstack.org/show/732581/ | |
| 12:40:18 | sapd1 | I upgraded from queens to rocky | |
| 12:40:46 | sapd1 | libvirt: 3.6 and qemu 2.11 | |
| 13:08:45 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Transform compute_task notifications https://review.openstack.org/482629 | |
| 13:29:23 | openstackgerrit | Takashi NATSUME proposed openstack/nova stable/rocky: Remove unnecessary redirect https://review.openstack.org/607400 | |
| 13:29:56 | openstackgerrit | Takashi NATSUME proposed openstack/nova stable/queens: Remove unnecessary redirect https://review.openstack.org/607402 | |
| 13:30:51 | openstackgerrit | Adam Spiers proposed openstack/nova-specs master: Add spec for libvirt driver launching AMD SEV-encrypted instances https://review.openstack.org/609779 | |
| 13:31:00 | aspiers | jaypipes: ^^^ | |
| 13:39:02 | openstackgerrit | Matthew Edmonds proposed openstack/nova master: Use tempfile for powervm config drive https://review.openstack.org/610174 | |
| 13:40:05 | jaypipes | aspiers: rock on. | |
| 13:43:19 | openstackgerrit | Markus Hentsch proposed openstack/nova-specs master: Spec for the Nova part of Image Encryption https://review.openstack.org/608696 | |
| 13:55:56 | efried | n-sch meeting in 5 minutes in #openstack-meeting-alt | |
| 14:03:06 | mriedem | alex_xu: you should probably take a look at this https://review.openstack.org/#/c/609591/ | |
| 14:03:55 | stephenfin | mriedem: Could you take a look at this bug fix series this week? https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bug/1784353 | |
| 14:04:10 | alex_xu | mriedem: got it, will take a look | |
| 14:04:44 | mriedem | stephenfin: yeah | |
| 14:04:49 | stephenfin | ta | |
| 14:05:44 | sapd1 | Does anyone get this problem when upgrade from queens to rocky? | |
| 14:10:42 | s10 | sapd1: looks like this issue is related to the upgrade to the qemu 2.11, not nova. | |
| 14:11:43 | sapd1 | s10: I have some compute nodes which are running qemu 2.11 and it gets same issue. | |
| 14:12:43 | sapd1 | Does nova-compute require higher qemu/libvirt version/ | |
| 14:13:09 | sapd1 | sorry s/2.11/2.10/ | |
| 14:13:52 | mriedem | johnthetubaguy: you might care about this https://review.openstack.org/#/c/609043/ | |
| 14:26:46 | mriedem | stephenfin: these are pretty easy https://review.openstack.org/#/q/topic:bug/1763051+(status:open+OR+status:merged) - first is a refactor, second is a try/catch/log but don't blow up sort of thing | |
| 14:27:33 | mriedem | oh sorry, latter is try/catch/rollback | |
| 14:27:45 | mriedem | was mixing that up with a similar live migration error handling patch | |
| 14:27:58 | stephenfin | mriedem: Sounds good | |
| 14:28:15 | stephenfin | mriedem: Also, I took a look at those debug logs you reworked for the issue awaugama and I are seeing. +2'd the first but do you think the second one should be merged now? Comments are confusing/misleading | |
| 14:28:26 | mriedem | i believe your employer cares about those live migration ones | |
| 14:29:23 | stephenfin | This one (efried got to it too) https://review.openstack.org/#/c/597560/ | |
| 14:29:33 | mriedem | what is confusing/misleading? | |
| 14:29:47 | mriedem | i also mostly restored/cleaned that up for awaugama to recreate the problem | |
| 14:29:59 | mriedem | if it's not helpful then i don't care if we don't use it | |
| 14:30:02 | stephenfin | "I don't really intend these overly verbose logs to get merged." doesn't apply anymore, right? | |
| 14:30:39 | mriedem | compare PS7 to PS1 | |
| 14:30:46 | mriedem | it's a much slimmer version | |
| 14:31:29 | stephenfin | ack | |
| 14:31:38 | efried | mriedem: See comments, I still think it's going to be excessive. | |