| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-03-06 | |||
| 14:39:13 | johnthetubaguy | dansmith: I seem to be stuck in a pile of post PTG admin at the moment, I feel like quitting that and doing some reviewing, that looks like a good place to start | |
| 14:39:43 | dansmith | johnthetubaguy: heh, well, I think she still has to address some feedback from you, or so she said on friday | |
| 14:40:38 | johnthetubaguy | dansmith: ah, OK, at a first glance its looking much closer now | |
| 14:41:11 | openstackgerrit | Alex Xu proposed openstack/nova-specs master: The ResourceProviderTraits API - Manage traits with ResourceProviders https://review.openstack.org/345138 | |
| 14:42:05 | dansmith | johnthetubaguy: cool, I've not gone through it in detail, but was just testing it underneath my cells one as it resolved several problems for me | |
| 14:42:20 | johnthetubaguy | cool | |
| 14:50:33 | openstackgerrit | John L. Villalovos proposed openstack/nova master: Use flake8-import-order https://review.openstack.org/435575 | |
| 14:58:23 | openstackgerrit | Lee Yarwood proposed openstack/nova master: libvirt: Check if domain is persistent before detaching volume https://review.openstack.org/441204 | |
| 15:06:40 | jaypipes_ | cdent: morning. got in after midnight last night. just getting started this morning. can you recap the scheduler meeting for me please? anything urgent for me to address other than fixing up the few things mriedem and you pointed out on the Ironic get_inventory() patch? | |
| 15:07:40 | edleafe | jaypipes_: does http://eavesdrop.openstack.org/meetings/nova_scheduler/2017/nova_scheduler.2017-03-06-14.00.log.html help? :) | |
| 15:07:51 | openstackgerrit | Alin Balutoiu proposed openstack/os-vif master: vif_plug_ovs: Skip setting MTU on Windows when plugging devices https://review.openstack.org/441855 | |
| 15:07:56 | jaypipes_ | edleafe: don't have time to read the full thing, really.. | |
| 15:08:17 | edleafe | The only thing was it wasn't clear why PUT/DELETE was considered so dangerous | |
| 15:08:33 | artom | dansmith, oh hey, you're up :) As main reviewer for device tagging last time around, can I ask you to start checking out https://review.openstack.org/#/c/376021/ ? | |
| 15:08:38 | artom | mriedem, same to you ^^ | |
| 15:08:39 | edleafe | especially since they would both fail with 409 if any of the traits were in use | |
| 15:08:57 | cdent | jaypipes_: yeah, as edleafe points out the main discussion was around traits api, and we reached some sound conclusions | |
| 15:09:16 | dansmith | artom: yeah, I've got it in the queue | |
| 15:09:25 | jaypipes_ | cdent: k. guess I'll read back through the IRC log. | |
| 15:09:54 | artom | dansmith, wootness | |
| 15:10:15 | cdent | jaypipes_: nothing pressing, so I wouldn't stress it | |
| 15:11:11 | jaypipes_ | cdent: meh, I'd love to make some good forward progress on traits this week, possibly merging the spec and the first few in the code series (after more reviews of course) | |
| 15:11:57 | edleafe | jaypipes_: well, we agreed to go with the simpler approach, with the understanding that we could always add to it later if necessary | |
| 15:11:58 | cdent | jaypipes_: well, it's under control, I would guess we'll have a spec merge in the next 24-48 hours | |
| 15:12:15 | jaypipes_ | cool, that' | |
| 15:12:23 | jaypipes_ | s great news guys, thanks cdent and edleafe | |
| 15:12:36 | jaypipes_ | and sorry for missing the meeting. | |
| 15:21:07 | arne_r | mriedem: Hey Matt, could you take a look at this? https://review.openstack.org/#/c/436868/3 Without it, nova is still broken on s390 in newton | |
| 15:21:15 | openstackgerrit | Evgeny Antyshev proposed openstack/nova master: get_model method missing for Ploop image https://review.openstack.org/441989 | |
| 15:31:08 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Short circuit notifiations when not enabled https://review.openstack.org/428260 | |
| 15:55:40 | openstackgerrit | Lee Yarwood proposed openstack/nova master: compute: Move detach logic from manager into driver BDM https://review.openstack.org/439520 | |
| 15:55:41 | openstackgerrit | Lee Yarwood proposed openstack/nova master: objects: Add attachment_id to BlockDeviceMapping https://review.openstack.org/437665 | |
| 15:55:41 | openstackgerrit | Lee Yarwood proposed openstack/nova master: compute: Only destroy BDMs after successful detach call https://review.openstack.org/440693 | |
| 15:55:42 | openstackgerrit | Lee Yarwood proposed openstack/nova master: db: Add attachment_id to block_device_mapping https://review.openstack.org/437597 | |
| 15:55:43 | openstackgerrit | Lee Yarwood proposed openstack/nova master: Make BlockDeviceMapping object support uuid https://review.openstack.org/242603 | |
| 15:55:44 | openstackgerrit | Lee Yarwood proposed openstack/nova master: Add uuid column to BlockDeviceMapping https://review.openstack.org/242602 | |
| 16:04:04 | cfriesen | can someone explain to me the logic behind https://github.com/openstack/nova/blob/master/nova/virt/hardware.py#L1109 ? I thought that if the flavor policy was "prefer" then we would use the image policy, if one was specified. | |
| 16:04:55 | cfriesen | sfinucane: ^ | |
| 16:05:24 | johnthetubaguy | sfinucan: ^ | |
| 16:08:52 | cfriesen | sfinucan: I think Nikola's commit 24997343 changed the logic of this code block, but I don't think it was intentional based on the commit message | |
| 16:09:28 | johnthetubaguy | mriedem: could you remove the -2 on this one now please https://review.openstack.org/#/c/418025/ | |
| 16:23:45 | smcginnis | I don't see any placement-api setup in the Ocata install guide. | |
| 16:23:54 | smcginnis | Is that missing, or am I overlooking it. https://docs.openstack.org/ocata/install-guide-ubuntu/nova-compute-install.html | |
| 16:30:25 | mriedem | smcginnis: https://review.openstack.org/#/c/438328/ | |
| 16:30:49 | mriedem | smcginnis: and https://docs.openstack.org/developer/nova/placement.html | |
| 16:30:54 | smcginnis | mriedem: Thanks | |
| 16:35:48 | mriedem | dansmith: nova.conf on the controller doesn't need access to the main nova [database] right? | |
| 16:36:02 | mriedem | i'm looking at http://docs-draft.openstack.org/28/438328/3/check/gate-openstack-manuals-tox-doc-publish-checkbuild/319104b//publish-docs/draft/install-guide-ubuntu/nova-controller-install.html | |
| 16:36:12 | dansmith | mriedem: it does today | |
| 16:36:23 | dansmith | mriedem: going forward it will point at cell0 | |
| 16:36:34 | mriedem | oh plus compute nodes | |
| 16:36:37 | mriedem | for the scheduler fallback | |
| 16:38:42 | mriedem | johnthetubaguy: would be nice to get these done https://review.openstack.org/#/q/status:open+project:openstack/nova-specs+branch:master+topic:mova-implemented-ocata-specs | |
| 16:38:50 | mriedem | marks the implemented specs for ocata | |
| 16:42:32 | mriedem | cdent: the placement_database config option is not used at all in code right now, right? | |
| 16:43:27 | cdent | mriedem: ah, you make a very good point | |
| 16:44:19 | mriedem | cdent: you know i'm reviewing that install guide now right? :) | |
| 16:44:21 | cdent | mriedem: are you refering to the docs in progress in the install guides? | |
| 16:44:24 | mriedem | bingo | |
| 16:44:25 | cdent | jinx | |
| 16:45:27 | cdent | thanks for catching that. I was reading through that and thinking "this should all be second nature, but it's not" | |
| 16:47:24 | mriedem | https://bugs.launchpad.net/nova/+bug/1670419 | |
| 16:47:24 | openstack | Launchpad bug 1670419 in OpenStack Compute (nova) "placement_database config option help is wrong" [Low,Triaged] | |
| 16:47:42 | mriedem | i wonder if we should just remove those options for now since they have some scary help text about saying you should use something separate to avoid messy data migrations | |
| 16:47:45 | mriedem | but we don't actually use the thing | |
| 16:48:01 | mriedem | they could be added back in when https://blueprints.launchpad.net/nova/+spec/optional-placement-database is done | |
| 16:50:42 | cdent | Sigh. I thought when we decided not to use it we had removed all of it, but apparently not. | |
| 16:50:52 | mriedem | i'll remove | |
| 16:50:53 | mriedem | and backport | |
| 16:51:15 | cdent | The code at https://review.openstack.org/#/c/362766/ continues to work but the efforts to have some discussion about how to move it forward keep stalling. | |
| 16:52:28 | openstackgerrit | Lee Yarwood proposed openstack/nova master: libvirt: Check if domain is persistent before detaching volume https://review.openstack.org/441204 | |
| 16:52:48 | johnthetubaguy | mriedem: yeah, I can try hit those, just snowed under in 1 million things right now, or feels that way | |
| 16:53:38 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Remove unused placement_database config options https://review.openstack.org/442035 | |
| 16:53:46 | mriedem | johnthetubaguy: remember how when we were in ATL and kept telling you to stop signing up for things? | |
| 16:54:04 | johnthetubaguy | mriedem: I wish it was those things that were distracting me, but yeah | |
| 16:54:09 | efried_ | Snow. That must suck. 74 degrees in Austin ;-) | |
| 16:54:45 | johnthetubaguy | mriedem: most of those are dispatched to other folks now though, at least I think so | |
| 17:02:09 | openstackgerrit | Artom Lifshitz proposed openstack/nova-specs master: SPICE native client support https://review.openstack.org/442040 | |
| 17:02:53 | artom | sgordon, I added you to ^^^, maybe you can help with the use cases. | |
| 17:06:41 | sgordon | artom, ack | |
| 17:07:07 | artom | sgordon, many thanks | |
| 17:11:28 | mriedem | cdent: heh, also, the guide doesn't actually tell you to run placement under wsgi http://docs-draft.openstack.org/28/438328/3/check/gate-openstack-manuals-tox-doc-publish-checkbuild/319104b//publish-docs/draft/install-guide-ubuntu/nova-controller-install.html#install-and-configure-components | |
| 17:11:32 | mriedem | that section is commented out | |
| 17:11:44 | mriedem | see how that jumps from step 7 to step 1 | |
| 17:12:11 | cdent | mriedem: that's part of why I left the bit about <Directory> | |
| 17:12:42 | cdent | I'm assuming that the deb or rpm packages do the configuration of apache, but are getting it somewhat wrong | |
| 17:12:48 | mriedem | oh | |
| 17:12:56 | mriedem | i forgot about package config magic | |
| 17:13:54 | openstackgerrit | Artom Lifshitz proposed openstack/nova-specs master: SPICE native client support https://review.openstack.org/442040 | |
| 17:16:10 | mriedem | wow they have placement_database config on the compute nodes too | |
| 17:16:14 | mriedem | that's way off | |
| 17:17:59 | cdent | clearly I need to stop reviewing while on company calls | |
| 17:20:40 | mriedem | spotz: i've gone through https://review.openstack.org/#/c/438328/ | |
| 17:20:45 | mriedem | spotz: there are some clear issues in there | |
| 17:21:35 | spotz | mriedem: I saw your comments thanks! Hopefully we can give it another run through with a better outcome:) | |
| 17:21:47 | mriedem | it's mostly easy stuff, | |
| 17:21:51 | mriedem | and removing quite a few unused things | |
| 17:22:22 | spotz | Stuff that would cause the No valid hosts? I saw another comment that's not just me | |
| 17:23:59 | mriedem | sdague: no more _LE?! | |
| 17:24:11 | sdague | mriedem: well, that's the question | |