| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-07-06 | |||
| 14:19:41 | mriedem | assuming we have any actions with >1 event | |
| 14:19:47 | dansmith | because we don't declare all our events ahead of time | |
| 14:19:51 | dansmith | would be nice if we did, but.. | |
| 14:19:57 | mriedem | sure, that's tasks api | |
| 14:19:58 | mriedem | i guess | |
| 14:20:00 | dansmith | right | |
| 14:20:05 | mriedem | watch your footing | |
| 14:20:18 | dansmith | heh | |
| 14:20:35 | mriedem | so i'm thinking "expose the events but hide the traceback for non-admins by default (in a microversion)" is the winner | |
| 14:20:47 | dansmith | I guess so | |
| 14:20:52 | dansmith | I like #1, but... yeah :) | |
| 14:20:53 | mriedem | "winner" | |
| 14:21:01 | mriedem | do quotes help? | |
| 14:21:04 | dansmith | yeah, hard to feel like you won, when that's the solution | |
| 14:21:05 | dansmith | heh | |
| 14:21:41 | mriedem | i also don't feel like working this into the same microversion, since it will clutter up an already large change, | |
| 14:21:56 | mriedem | but also not sure i want to create a new separate microversion just for that | |
| 14:22:38 | mriedem | alright, anyway, this was helpful | |
| 14:22:41 | dansmith | yeah, I dunno.. would be nice to save a microversion, but reading the history, it'll be like "this volume thing and oh also we exposed this other random thing, kthxbai" | |
| 14:22:46 | sean-k-mooney | stephenfin: by the way if you have not seen https://review.openstack.org/#/c/390513/ and https://review.openstack.org/#/c/390512/ they are releated to the issue your trying to solve | |
| 14:23:27 | stephenfin | sean-k-mooney: I think I saw one but not the other. I'll take a look now :) | |
| 14:24:08 | stephenfin | sean-k-mooney: meanwhile, I _think_ this is correct, but I'm not sure if it's even necessary given your comments about not using a fqdn for hostname https://review.openstack.org/#/c/480676/ | |
| 14:24:15 | mriedem | dansmith: that's basically how the services/hypervisors uuid api changes read out in the rest api version history and release notes :) | |
| 14:24:22 | dansmith | mriedem: heh | |
| 14:24:26 | mriedem | "take and return uuid instead of id, plus a whole bunch of other shit" | |
| 14:24:27 | sean-k-mooney | stephenfin: the first one is proposing a single api endpoint to return all the info to populate the metadat. the second is the nova/neutont negociation sepc where i would like to use os vif as the serialisation format | |
| 14:24:45 | mriedem | 500 microversions also sucks | |
| 14:24:59 | dansmith | yeah | |
| 14:26:38 | sean-k-mooney | stephenfin: lookin at it now. did you add the dns_domain field in the previous patches | |
| 14:27:12 | stephenfin | sean-k-mooney: I don't think it's necessary - that model does weird stuff with a 'meta' field that's mostly autopopulated | |
| 14:27:22 | stephenfin | sean-k-mooney: Look at the patch that added 'mtu' support | |
| 14:30:01 | openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: fix memory locking limit https://review.openstack.org/472633 | |
| 14:30:06 | jaypipes | dansmith, ildikov: :) | |
| 14:31:41 | mriedem | mgagne: fyi in case you haven't been following along https://review.openstack.org/#/c/454322/27/nova/compute/api.py | |
| 14:31:57 | mriedem | mgagne: let's talk about how you'd like to proceed - if you want me to work on the os-instance-actions api change in there | |
| 14:32:38 | openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add BDM to InstancePayload https://review.openstack.org/448779 | |
| 14:40:56 | foutatoro | hi all, I'm trying to create KVM instance with ubuntu cloud images used in Openstack with virt-install but I always get the error "No bootable device found". | |
| 14:41:06 | foutatoro | can someone help me to solve that please ? | |
| 14:42:15 | openstackgerrit | Merged openstack/os-vif master: Rehome unit tests to ``tests\unit`` folder https://review.openstack.org/476623 | |
| 14:42:27 | openstackgerrit | Merged openstack/os-vif master: Use ``assert_has_calls`` to check function calls https://review.openstack.org/477469 | |
| 14:57:08 | jaypipes | while `tox -epy27,pep8`; do process_emails.sh; done | |
| 14:58:10 | mriedem | ildikov: so the cinder v3 api changes are stacked such that swap volume should get tested in the top patch which does the new style attach, right? | |
| 14:58:50 | mriedem | ildikov: because there is a swap volume test in tempest, but i don't think it runs in the normal tempest dsvm jobs because it's marked as 'slow' | |
| 14:59:19 | ildikov | mriedem: it should run as it was failing until I added the swap patch in the chain | |
| 14:59:43 | mriedem | ok i'll take a look | |
| 15:00:33 | ildikov | mriedem: you can take a thorough look at both the dependencies and on the attach patch | |
| 15:00:43 | ildikov | just to keep everyone busy :) | |
| 15:06:59 | jaypipes | mriedem: ralonsoh is looking for this (yet unmerged) patch: https://review.openstack.org/#/c/466040/ | |
| 15:07:20 | jaypipes | mriedem: but you're right to say that 0.3.2 doesn't include that (since obvi it's not merged yet) | |
| 15:07:57 | jaypipes | I'm approving that now. it needed rebasing after the unit test re-org | |
| 15:07:59 | openstackgerrit | sahid proposed openstack/nova master: libvirt: fix memory locking limit https://review.openstack.org/472633 | |
| 15:08:08 | mriedem | jaypipes: ok | |
| 15:08:14 | mriedem | something didn't smell right | |
| 15:08:31 | jaypipes | mriedem: ack | |
| 15:08:46 | jaypipes | mriedem: there's also an os-vif thing too :) | |
| 15:09:05 | mriedem | plus this monster https://review.openstack.org/#/c/449257/ | |
| 15:09:25 | jaypipes | mriedem: well, yeah, but that depends on all the others :) | |
| 15:09:36 | mriedem | right | |
| 15:09:39 | mriedem | i see a few os-vif changes, | |
| 15:09:44 | mriedem | not sure which are related, they are all marked as bug fixes | |
| 15:16:41 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for RP traits https://review.openstack.org/474550 | |
| 15:16:42 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for traits https://review.openstack.org/474186 | |
| 15:16:42 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for allocations https://review.openstack.org/470933 | |
| 15:16:42 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for usages https://review.openstack.org/480563 | |
| 15:16:43 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Make placement_api_docs.py failing https://review.openstack.org/480924 | |
| 15:16:44 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for RP usages https://review.openstack.org/450105 | |
| 15:16:44 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for RP allocations https://review.openstack.org/450550 | |
| 15:16:44 | openstackgerrit | Andrey Volkov proposed openstack/nova master: [placement] Add api-ref for allocation_candidates https://review.openstack.org/481112 | |
| 15:22:38 | openstackgerrit | Jay Pipes proposed openstack/nova master: placement: scheduler uses allocation candidates https://review.openstack.org/476631 | |
| 15:22:39 | openstackgerrit | Jay Pipes proposed openstack/nova master: WIP - placement: claim resources in scheduler https://review.openstack.org/476632 | |
| 15:24:28 | jaypipes | dansmith, mriedem, efried, bauzas: addressed all of your comments on https://review.openstack.org/476631 | |
| 15:24:42 | efried | just pulling it up, thanks. | |
| 15:25:05 | openstackgerrit | Sylvain Bauza proposed openstack/nova master: Pass the req_dest fielf of ReqSpec in the filter props https://review.openstack.org/481116 | |
| 15:25:44 | jaypipes | efried: I changed the code style to put every word and paren on its own line, just for you and dansmith. | |
| 15:25:52 | efried | Love it. | |
| 15:25:56 | jaypipes | heh | |
| 15:27:12 | dansmith | jaypipes: (I hope you are kidding | |
| 15:27:12 | dansmith | ) | |
| 15:27:27 | jaypipes | dansmith: no, | |
| 15:27:28 | jaypipes | not, | |
| 15:27:29 | jaypipes | kidding, | |
| 15:27:31 | jaypipes | ) | |
| 15:27:37 | jaypipes | dansmith: j | |
| 15:27:38 | jaypipes | k | |
| 15:28:06 | jaypipes | boom. | |
| 15:28:10 | jaypipes | mission accomplished. | |
| 15:29:17 | mriedem | some day someone will have to explain the kerfuffle from last week about ( | |
| 15:29:18 | mriedem | this | |
| 15:29:18 | mriedem | ) | |
| 15:30:49 | jaypipes | mriedem: basically, it makes dansmith and efried want to do this to my head: http://bit.ly/2tl79kj | |
| 15:31:11 | jaypipes | and edleafe :) | |
| 15:32:01 | bauzas | mriedem: dansmith: not sure you remembered our convo yesterday about https://bugs.launchpad.net/nova/+bug/1702454 but I just provided a fix https://review.openstack.org/481116 | |
| 15:32:01 | openstack | Launchpad bug 1702454 in OpenStack Compute (nova) "Transforming the RequestSpec object into legacy dicts doesn't support the requested_destination field" [High,In progress] - Assigned to Sylvain Bauza (sylvain-bauza) | |
| 15:32:13 | bauzas | mriedem: dansmith: targeted for backport to ocata and newton as well | |
| 15:32:48 | mriedem | jaypipes: well, just keep an eye out for when one of them do something similar with a dict or list and then you can drop the gd mfing hammer | |
| 15:33:24 | jaypipes | mriedem: I'm a pacifist. | |
| 15:33:24 | mriedem | bauzas: what is a "fielf"? | |
| 15:33:37 | efried | mriedem TL;DR: jaypipes has been indoctrinating himself in golang, where this kind of wasted vspace is acceptable, even encouraged. Which would be one thing if it matched the style of the code anywhere else in nova, but it doesn't. For dicts, yes. For lists, less so, but sometimes. For method calls, never. | |
| 15:33:38 | bauzas | mriedem: a covfefe | |
| 15:33:39 | jaypipes | mriedem: it's a field that dresses up at Christmas. | |