| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-12-21 | |||
| 19:12:01 | edleafe | maybe mriedem will | |
| 19:12:35 | mriedem | jaypipes: replied | |
| 19:13:03 | mriedem | oomichi: no one ever brought that up for queens, | |
| 19:13:13 | mriedem | in the 2-3 months they could have brought it up | |
| 19:13:46 | oomichi | mriedem: ok, let's skip it | |
| 19:14:17 | mriedem | given how trivial this is we could possibly make an exception | |
| 19:14:36 | mriedem | i just don't generally like to make exceptions because then you do it for one blueprint and everyone else expects it | |
| 19:16:57 | melwitt | there's a pretty simple libvirt driver refactor needing a +W that's en route to native luks encryption if anyone can review https://review.openstack.org/#/c/460243 | |
| 19:17:01 | jaypipes | mriedem: k, both bottom patches +Wd | |
| 19:17:07 | mriedem | oomichi: replied | |
| 19:17:09 | mriedem | jaypipes: danka | |
| 19:17:16 | jaypipes | mriedem: bitte | |
| 19:18:57 | edleafe | jaypipes: thx. Guess we'll wait til Jan for the final patch | |
| 19:19:07 | edleafe | (unless you're up for it now!) | |
| 19:19:16 | mriedem | i haven't gone through the final patch yet | |
| 19:19:21 | jaypipes | edleafe: the one that needs a rebase? | |
| 19:19:23 | mriedem | so yeah figured january at this rate | |
| 19:19:25 | mriedem | plus that | |
| 19:19:26 | mriedem | :) | |
| 19:19:40 | jaypipes | mriedem: I'm reading your tc IRC log... | |
| 19:20:05 | mriedem | edleafe: would be cool if you could find an existing functional test for resize + reschedule that we could do the same caching scheduler extension test like i added this morning | |
| 19:20:09 | mriedem | jaypipes: you like that? | |
| 19:20:17 | mriedem | better get comfortable | |
| 19:20:27 | edleafe | mriedem: I'll look | |
| 19:20:31 | jaypipes | mriedem: it's "interesting" so far. | |
| 19:20:35 | melwitt | I read it yesterday | |
| 19:20:44 | mriedem | which tantrum are we talking about? | |
| 19:20:48 | mriedem | several days ago, or yesterday? | |
| 19:21:00 | jaypipes | mriedem: I mean, I already knew everyone hated nova (cores), but .. | |
| 19:21:04 | jaypipes | mriedem: yesterday | |
| 19:21:22 | mriedem | turns out they specifically hated john for one specific thing brought up years ago | |
| 19:21:28 | melwitt | yeah, it was ... illuminating | |
| 19:21:29 | mriedem | but that now includes me and everyone else | |
| 19:22:30 | edleafe | you're all the same | |
| 19:24:29 | mriedem | especially Vek | |
| 19:24:36 | mriedem | he's offendor #1 | |
| 19:24:48 | mriedem | *offender | |
| 19:25:33 | openstackgerrit | Ed Leafe proposed openstack/nova master: Make conductor pass and use host_lists https://review.openstack.org/511358 | |
| 19:25:33 | openstackgerrit | Ed Leafe proposed openstack/nova master: Don't try to delete build request during a reschedule https://review.openstack.org/528835 | |
| 19:25:34 | openstackgerrit | Ed Leafe proposed openstack/nova master: Change compute RPC to use alternates for resize https://review.openstack.org/526436 | |
| 19:25:52 | edleafe | mriedem: jaypipes: ^^ the rebase got 'em all | |
| 19:27:38 | mriedem | oomichi: i think "If you need help debugging" is the correct english | |
| 19:27:56 | mriedem | otherwise "If you need help with debugging"? | |
| 19:28:21 | mriedem | granted, i think most people who have english has a second language are probably more knowledgeable of the actual grammar rules :) | |
| 19:28:22 | oomichi | mriedem: "If you need help with debugging" is more easy for me | |
| 19:28:56 | jroll | another option is "if you need help to debug" | |
| 19:28:58 | mriedem | oomichi: 私はそれを変更する | |
| 19:29:00 | jroll | I like with | |
| 19:29:15 | oomichi | mriedem: ありがとう | |
| 19:29:43 | mriedem | oomichi: as for the "API extension" part, | |
| 19:29:47 | mriedem | i know like 'resource' | |
| 19:29:57 | mriedem | but not sure what to use beside extension. handler? controller? just API? | |
| 19:30:03 | mriedem | "route"? | |
| 19:35:07 | melwitt | mriedem: would it not be a REST API "resource"? that's what ppl told me to call it when I added /usages to placement API for example | |
| 19:35:22 | oomichi | mriedem: I think API is better in them. handler, controller and route are just implementation side | |
| 19:36:09 | mriedem | melwitt: but there are several API extensions under nova/api/openstack/compute/ which work on the same resource | |
| 19:36:11 | mriedem | like servers | |
| 19:36:25 | mriedem | consider the user_data, block_device_mapping_v2, etc extensions | |
| 19:36:26 | melwitt | ah, right | |
| 19:36:33 | mriedem | we have a similar modeling in our api samples directories | |
| 19:36:37 | mriedem | which is why i'm trying to show it's a 1:1 | |
| 19:36:45 | melwitt | yeah. I was thinking of the os-blah ones that are separate | |
| 19:37:16 | oomichi | API becomes non-extensible, API means a single API, and we don't need to have separated tests for each extension | |
| 19:37:46 | melwitt | there aren't any separate directories under doc/api_samples/servers/ fwiw. just microversion directories | |
| 19:38:38 | melwitt | it's nice to have them separated for organizing and finding them though, IMHO | |
| 19:39:35 | mriedem | oomichi: then what do we call this? https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/user_data.py | |
| 19:39:43 | mriedem | it's an extension of the servers api | |
| 19:40:04 | mriedem | eventually, one day, these should all be collapsed and removed, which sdague started and gmann was working on | |
| 19:40:11 | mriedem | but i'm not sure what to call those except 'extensions' | |
| 19:40:11 | oomichi | mriedem: hehe, that is a good sample | |
| 19:40:53 | oomichi | mriedem: IIUC alex_xu wants to merge these code into main servers.py with json-schema | |
| 19:41:03 | oomichi | to remove these extension code | |
| 19:41:31 | mriedem | yeah i know | |
| 19:41:39 | mriedem | but until then what do we call these | |
| 19:41:59 | oomichi | at this time, it is difficult to find another word except extension | |
| 19:42:30 | mriedem | ha | |
| 19:42:33 | mriedem | exactly! | |
| 19:43:07 | mriedem | i could put a note about how "extension" doesn't actually mean we support extending the API, and link to the docs about that, but it seems like that would get messy | |
| 19:43:20 | melwitt | the user_data example doesn't have its own directory though, from what I see | |
| 19:43:34 | mriedem | you know what melanie | |
| 19:44:00 | mriedem | ok but https://github.com/openstack/nova/tree/master/doc/api_samples/server-metadata | |
| 19:44:01 | mriedem | does | |
| 19:44:12 | oomichi | mriedem: please don;t put such note, I am fine with current one which is better | |
| 19:44:12 | mriedem | and whatever the hell this thing is https://github.com/openstack/nova/tree/master/doc/api_samples/os-quota-sets-noop | |
| 19:44:13 | mriedem | :) | |
| 19:44:36 | mriedem | oomichi: ok so i'll change the 'debug' part | |
| 19:44:37 | mriedem | and leave the rest | |
| 19:44:46 | oomichi | hehe | |
| 19:45:02 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Document testing guide for new API contributions https://review.openstack.org/529618 | |
| 19:46:14 | mriedem | https://docs.openstack.org/nova/latest/contributor/api.html#modularity | |
| 19:46:19 | mriedem | "The Nova REST API is separated into different controllers in the directory ‘nova/api/openstack/compute/’" | |
| 19:46:29 | openstackgerrit | Merged openstack/nova stable/pike: [placement] Fix an error message in API validation https://review.openstack.org/526465 | |
| 19:46:30 | mriedem | since the doc already says "controllers" maybe we should just be consistent? | |
| 19:50:14 | melwitt | I was originally going to put the noop quotas directory under os-quota-sets/ but thought maybe the top level one was less confusing. I can try to move it under os-quota-sets/ | |
| 19:52:32 | melwitt | another way I tried was putting the noop samples at the same level in os-quota-sets/ without their own directory and have the noop tests look for a noop prefix, but thought maybe that was worse ¯\_(ツ)_/¯ | |
| 20:00:21 | mriedem | i was just giving you shit about that | |
| 20:03:51 | melwitt | k :) | |
| 20:39:07 | openstackgerrit | Merged openstack/nova master: Pass RequestSpec to ConductorTaskAPI.build_instances https://review.openstack.org/515495 | |
| 20:41:12 | cfriesen | mriedem: do you know why we're calling self._update() at https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L590 ? Having that means that every time we call_update_available_resource() we'll save invalid data, then overwrite it a bit later with valid data. | |
| 20:42:30 | mriedem | cfriesen: oh boy | |