| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-06-12 | |||
| 19:39:00 | oomichi | mriedem: sdague: ^^^ | |
| 19:39:53 | sdague | oomichi: nice | |
| 19:44:35 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/472814 | |
| 19:46:44 | openstackgerrit | Dan Smith proposed openstack/nova master: DNM: Canary https://review.openstack.org/473565 | |
| 19:48:14 | openstackgerrit | Merged openstack/nova master: Use plain routes list for block_device_mapping instead of stevedore https://review.openstack.org/462180 | |
| 19:48:57 | openstackgerrit | Merged openstack/nova master: Remove unused variable https://review.openstack.org/473210 | |
| 19:55:54 | mriedem | thanks oomichi | |
| 19:57:59 | mriedem | oomichi: question in there about using strict=True | |
| 20:02:20 | mriedem | sdague: i think we might want to hold off on https://review.openstack.org/#/c/473564/ | |
| 20:02:49 | oomichi | mriedem: I think strict parameter is not so matter here because the input is already validated with jsonschema before | |
| 20:02:50 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/472814 | |
| 20:03:32 | mriedem | oomichi: hmm, good point - how about a negative test to show that though? | |
| 20:03:42 | mriedem | by passing forced_down='invalid' | |
| 20:04:31 | oomichi | mriedem: that is for coparing a response with expectation. The response should be also boolean instead of string | |
| 20:05:01 | mriedem | oomichi: i don't understand what you mean | |
| 20:05:10 | oomichi | mriedem: ah do you want to add 'invalid' as a negative test | |
| 20:05:13 | mriedem | yes | |
| 20:05:15 | oomichi | mriedem: I can do that | |
| 20:05:18 | mriedem | assert it raises ValidationError | |
| 20:06:07 | oomichi | yeah, making | |
| 20:16:47 | openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova master: Add bool_from_string for force-down action https://review.openstack.org/473564 | |
| 20:17:23 | mriedem | oomichi: looks good, just waiting for zuul | |
| 20:18:02 | oomichi | mriedem: no problem, thanks for taking care :) | |
| 20:18:06 | mriedem | you could remove the is_legacy_v2() thing | |
| 21:00:17 | openstackgerrit | Jay Pipes proposed openstack/nova master: placement: produce set of alternative allocations https://review.openstack.org/473377 | |
| 21:01:06 | jaypipes | dansmith, cdent, edleafe, mriedem: now with moar functional testing. ^ | |
| 21:01:26 | cdent | but does it slice, dice and julienne? | |
| 21:05:58 | dansmith | jaypipes: are you working on the api exposure bit next? | |
| 21:06:12 | jaypipes | dansmith: yes | |
| 21:07:12 | dansmith | melwitt: I assume you're working on those comments on your upper placement usages thing and will fix the rebase need when you push/ | |
| 21:08:15 | melwitt | dansmith: yes. currently working on test coverage for recheck_quotas cases as I rebase the quotas series | |
| 21:08:31 | dansmith | melwitt: talking about this: https://review.openstack.org/#/c/469635/12 | |
| 21:08:59 | melwitt | dansmith: yeah, I know. trying to say that's next after what I'm currently doing since I've got -1s on two series | |
| 21:09:08 | dansmith | gotcha | |
| 21:18:27 | dansmith | melwitt: I can fix up the 404 comment and rebase I think | |
| 21:18:32 | dansmith | if you want | |
| 21:22:48 | melwitt | dansmith: I'll do it real quick. oomichi had one other note too | |
| 21:22:57 | dansmith | melwitt: already partway through | |
| 21:23:15 | dansmith | melwitt: I was mostly asking if you had pending local changes so as not to mess those up | |
| 21:23:20 | melwitt | hm, okay | |
| 21:23:31 | dansmith | melwitt: if you want me not to push I shan't :) | |
| 21:24:06 | melwitt | y'all are itching to review these huh :P | |
| 21:24:35 | dansmith | well, they need get-in'ing right? | |
| 21:24:43 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add bool_from_string for force-down action https://review.openstack.org/473564 | |
| 21:25:15 | melwitt | yup. gimme a minute | |
| 21:26:18 | melwitt | dansmith: I'll push this in a minute, if that's okay | |
| 21:26:23 | dansmith | okay | |
| 21:26:49 | melwitt | thanks | |
| 21:29:26 | openstackgerrit | melanie witt proposed openstack/nova master: placement project_id, user_id in PUT /allocations https://review.openstack.org/469634 | |
| 21:29:26 | openstackgerrit | melanie witt proposed openstack/nova master: placement: Add GET /usages to placement API https://review.openstack.org/469635 | |
| 21:39:26 | jaypipes | bauzas: would appreciate a review on https://review.openstack.org/#/c/469634/ if you can. I can't +2 since I wrote a chunk of it. | |
| 21:39:44 | jaypipes | alex_xu: you too :) ^^ | |
| 21:46:34 | openstackgerrit | Merged openstack/nova master: libvirt: handle missing rbd_secret_uuid from old connection info https://review.openstack.org/472266 | |
| 21:51:35 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Handle uuid in HostAPI.compute_node_get https://review.openstack.org/461532 | |
| 21:51:35 | openstackgerrit | Matt Riedemann proposed openstack/nova master: WIP: Handle uuids in os-hypervisors API https://review.openstack.org/471899 | |
| 21:51:36 | openstackgerrit | Matt Riedemann proposed openstack/nova master: WIP: Use uuid for id in Services API https://review.openstack.org/464280 | |
| 21:51:37 | mriedem | dtp: here are the updates ^ can you look over the PUT /os-services/{service_id} changes in there? I've got the tests done. | |
| 21:51:46 | mriedem | dtp: just need to work on the api-ref updates, probably tomorrow | |
| 21:53:26 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Add bool_from_string for force-down action https://review.openstack.org/473564 | |
| 21:54:29 | dtp | mriedem sure | |
| 21:56:42 | mriedem | it's going to *suck* squashing in the os-hypervisors change when that happens | |
| 21:56:51 | mriedem | but that's another battle | |
| 22:31:10 | cfriesen | if anyone's on....do we create a migration record for a "rebuild", and if so where is that in the code? | |
| 22:31:47 | openstackgerrit | OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/472814 | |
| 22:54:11 | openstackgerrit | Merged openstack/nova master: Add service_token for nova-glance interaction https://review.openstack.org/419662 | |
| 22:59:03 | mriedem | efried: does ^ close the service token bp? | |
| 22:59:15 | mriedem | the rest looked like dnm testing patches | |
| 23:07:41 | openstackgerrit | Takashi NATSUME proposed openstack/nova master: Avoid forcing translation on logging calls https://review.openstack.org/413876 | |
| 23:12:32 | mriedem | good god has anyone read the "Strawman Queens PTG week slicing" thread yet and can summarize? | |
| 23:12:37 | mriedem | because i don't want to start it | |
| 23:37:49 | openstackgerrit | Ed Leafe proposed openstack/nova master: Extract custom resource classes from flavors https://review.openstack.org/473627 | |
| 23:41:59 | mriedem | dansmith: i think i just realized something we broke with the os-services API being cellsified | |
| 23:42:13 | mriedem | broke/made more obviously not dumb | |
| 23:42:33 | mriedem | for things like enable/disable, today you pass in the host/binary to lookup the service | |
| 23:42:45 | mriedem | with multi-cells, we now lookup the host mapping by the host | |
| 23:42:51 | mriedem | host mapping will only exist for compute services | |
| 23:42:54 | openstackgerrit | John Griffith proposed openstack/nova master: WIP/POC Implement new attach Cinder flow https://review.openstack.org/330285 | |
| 23:42:59 | mriedem | but i think before you could technically disable nova-scheduler services | |
| 23:43:03 | mriedem | which doesn't mean anything, but you could do it | |
| 23:50:00 | openstackgerrit | Ed Leafe proposed openstack/nova master: Extract custom resource classes from flavors https://review.openstack.org/473627 | |
| 23:53:16 | mriedem | Kevin_Zheng: pretty easy bug to fix https://bugs.launchpad.net/nova/+bug/1697563 | |
| 23:53:16 | openstack | Launchpad bug 1697563 in OpenStack Compute (nova) "api-ref: GET /os-services does not describe host/binary request query filter parameters" [Medium,Triaged] | |
| 23:58:43 | dansmith | mriedem: yeah, so someone was recently asking me why disable on nova-scheduler didn't actually stop those services at the OS level | |
| 23:59:05 | dansmith | mriedem: we should probably return "40x: You can't do that thing" if they try | |
| 23:59:08 | mriedem | the api-ref docs all say "compute" service too | |
| 23:59:10 | mriedem | which is how i found this | |
| 23:59:13 | dansmith | okay | |
| 23:59:30 | mriedem | so that's the question i have in https://review.openstack.org/#/c/464280/9/nova/api/openstack/compute/services.py@281 | |
| 23:59:35 | mriedem | which is, do we now make this a 400 | |
| 23:59:44 | mriedem | before that new microversion, it's going to be a 404 now for non-compute services | |
| #openstack-nova - 2017-06-13 | |||
| 00:00:01 | mriedem | with the microversion, we could get non-compute things again, or just close that weird gap | |
| 00:00:40 | mriedem | probably better to restrict to nova-compute services for things like enable/disable/forced_down | |
| 00:08:04 | dansmith | mriedem: you mean after we made the change it's 404 now instead of 200 right? | |
| 00:12:23 | mriedem | yes | |
| 00:12:41 | mriedem | we could get it back to a 200, but it seems better to just fix the weird behavior we used to allow | |
| 00:13:08 | mriedem | this would be good to remember for when someone shows up in a year asking why their weird script no longer disables nova-scheduler services... | |
| 00:13:15 | dansmith | agreed | |
| 00:13:19 | dansmith | not sure 404 is the right thing though | |
| 00:13:47 | dansmith | since we expose the service, they did a PUT on it.. probably needs to be 499: Dumbass or something | |
| 00:14:22 | mriedem | or a 400 | |