| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-04-09 | |||
| 14:14:34 | openstackgerrit | Jay Pipes proposed openstack/nova-specs master: mirror nova host aggregates to placement API https://review.openstack.org/545057 | |
| 14:14:48 | jaypipes | mriedem: ^ done. | |
| 14:20:28 | openstackgerrit | Matt Riedemann proposed openstack/nova-specs master: mirror nova host aggregates to placement API https://review.openstack.org/545057 | |
| 14:20:54 | mriedem | thanks | |
| 14:20:54 | mriedem | jaypipes: tweaked the alternative section, but +2 | |
| 14:23:40 | jaypipes | mriedem: thx | |
| 14:29:07 | openstackgerrit | Dan Smith proposed openstack/nova master: Avoid dumping stack on BuildAbortException https://review.openstack.org/559751 | |
| 14:29:16 | dansmith | mriedem: ^ | |
| 14:29:47 | openstackgerrit | Chris Dent proposed openstack/nova master: Isolate placement database config https://review.openstack.org/541435 | |
| 14:29:47 | openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | |
| 14:29:48 | openstackgerrit | Chris Dent proposed openstack/nova master: WIP: Ensure that os-traits sync is attempted only at start of process https://review.openstack.org/553857 | |
| 14:32:10 | kashyap | alex_xu: Hi, when you're around -- maybe you'd know the answer to my question on `enabled_pef_events`: | |
| 14:32:38 | kashyap | alex_xu: As it stands, seems like Nova accepts far beyond three events (cmt, mbmt, mbml). Isn't it? | |
| 14:33:07 | kashyap | alex_xu: Because the source for that config attributes shows it accepts a string list -- so all values that are supported by libvirt are accepted by Nova | |
| 14:33:13 | kashyap | (At least that's what the doc says) | |
| 14:36:42 | kashyap | alex_xu: Disregard me, answered my question -- yes, it supports broader set of events besides those tree | |
| 14:36:45 | kashyap | s/tree/three/ | |
| 14:37:47 | mriedem | esberglu: why is https://review.openstack.org/#/c/554688/ in the runways etherpad? | |
| 14:37:57 | mriedem | it's not tied to a blueprint, but is just powervm driver specific, | |
| 14:38:13 | mriedem | if it's not a bp, please just include it in the powervm driver section of the other priority review etherpad https://etherpad.openstack.org/p/rocky-nova-priorities-tracking | |
| 14:38:26 | esberglu | mriedem: Will do | |
| 14:38:35 | mriedem | thanks | |
| 14:45:20 | openstackgerrit | Chris Dent proposed openstack/nova master: Use nova.db.api directly https://review.openstack.org/543262 | |
| 14:46:52 | mriedem | sahid: bauzas: this spec tracking is really odd https://review.openstack.org/#/c/511188/ - that's for rocky, for a blueprint which is closed | |
| 14:47:29 | mriedem | why isn't there a "overhead-pin-set" spec for rocky? | |
| 14:48:03 | bauzas | mriedem: shit, you're right | |
| 14:48:09 | bauzas | mriedem: the URL is wrong | |
| 14:49:01 | mriedem | and the name of the spec is wrong | |
| 14:49:05 | openstackgerrit | Eric Berglund proposed openstack/nova master: PowerVM Driver: Localdisk https://review.openstack.org/549300 | |
| 14:49:06 | bauzas | I haven't looked at the related BP, my bad | |
| 14:49:13 | mriedem | sahid: can you fix that up please? | |
| 14:49:46 | bauzas | I think sahid didn't changed the spec when we asked him to create that BP | |
| 14:50:12 | mriedem | finucannot: you can remove the -2 on https://review.openstack.org/#/c/510897/ - the spec is approved | |
| 14:50:18 | bauzas | mriedem: thanks for that btw. | |
| 14:50:26 | bauzas | honestly, I should also look at the BP | |
| 14:58:18 | sahid | mriedem: at the beginning it was just an update of the already approved spec, that is why we did not have really noticed the issue | |
| 14:59:25 | bauzas | sahid: ping me when you're done with the new change and I'll +2 | |
| 14:59:33 | bauzas | for renaming I mean | |
| 14:59:58 | sahid | yes i will | |
| 15:11:41 | mriedem | dansmith: replied in https://review.openstack.org/#/c/558059/ | |
| 15:17:37 | jaypipes | cdent: ok, everyone gone | |
| 15:17:55 | cdent | roger | |
| 15:18:24 | openstackgerrit | Merged openstack/nova master: Support extending attached ScaleIO volumes https://review.openstack.org/554679 | |
| 15:20:20 | mriedem | jaypipes: efried: -1 on tetsuro's spec https://review.openstack.org/#/c/559466/ - the bp link is wrong for one, but i'm also wondering if this is just to enable out of tree filters/weighters, or if he actually wants to add a vGPU weighter to nova | |
| 15:20:24 | mriedem | i'd be supportive of that if so | |
| 15:20:52 | mriedem | question inline about the slippery slope of including provider trees in the response also | |
| 15:21:32 | mriedem | bauzas: you would probably be interested in that idea ^ | |
| 15:24:22 | mriedem | jgwentworth: dansmith: i guess we were supposed to be actively brainstorming for the forum https://etherpad.openstack.org/p/YVR-nova-brainstorming | |
| 15:25:51 | mriedem | efried: jaypipes: cdent: edleafe: i'm sure there are placement things that could be at the forum | |
| 15:26:02 | mriedem | e.g. cinder talking about maybe leveraging it in their scheduler | |
| 15:26:15 | cdent | mriedem: yes, it's on y list | |
| 15:26:30 | cdent | got a couple things, including the cinder thing | |
| 15:26:35 | mriedem | efried: design sessions at the summit | |
| 15:26:51 | bauzas | mriedem: ack, I'll look | |
| 15:27:18 | bauzas | -ETOOMANYPLACEMENTSPECS | |
| 15:27:51 | bauzas | should we stop feeding the placement mogwaï after midnight ? | |
| 15:28:11 | openstackgerrit | Ed Leafe proposed openstack/nova-specs master: Add Generation to Consumers https://review.openstack.org/556971 | |
| 15:29:14 | efried | bauzas: It's water that makes 'em multiply. Feeding after midnight turns 'em evil. | |
| 15:29:19 | edleafe | efried: ^^ addressed your concerns | |
| 15:30:50 | bauzas | efried: ha, right. So, people, stop hydrating your specs | |
| 15:31:14 | jaypipes | mriedem: it's not to enable anything in particular. it's just to remove a restriction in the returned results of allocation candidates' provider_summaries section that wasn't necessary to restrict. | |
| 15:31:25 | bauzas | and don't give'em a review after midnight or they'll turn into something bad | |
| 15:31:54 | bauzas | efried: technically, growing mogwais is good | |
| 15:32:11 | bauzas | if you respect the 3 rules | |
| 15:33:15 | bauzas | jaypipes: what's up with the placement channel ? can I rejoin it ? | |
| 15:33:24 | cdent | bauzas: yes, it's ready | |
| 15:33:34 | bauzas | :'( | |
| 15:34:09 | jaypipes | bauzas: I warned you to leave the channel :) | |
| 15:34:49 | bauzas | jaypipes: well, you kick after the first warning, I hope your pugs behave correctly :p | |
| 15:39:12 | openstackgerrit | Eric Fried proposed openstack/nova-specs master: Return all resources in provider_summaries spec https://review.openstack.org/559466 | |
| 15:39:20 | efried | mriedem, jaypipes: I went ahead and made those edits ^ | |
| 15:39:26 | efried | tetsuro ^ | |
| 15:42:24 | bauzas | efried: see my comment on the spec | |
| 15:42:33 | bauzas | efried: about interop | |
| 15:43:23 | bauzas | efried: if people wanna add some query related to inventories, they should add it to Placement, not by a filter | |
| 15:43:33 | efried | bauzas: Responding. | |
| 15:43:48 | bauzas | efried: because the Placement logic needs to be interoperable for the same query | |
| 15:44:49 | mriedem | i just don't want to spend time making this change if nothing is going to leverage it in-tree right now | |
| 15:45:08 | bauzas | if we accept to pass back the inventories, then you'll be sure that one day or later, people will just use placement for basic checks, and will do python filtering for their specific calls | |
| 15:45:14 | mriedem | if out of tree filters/weighters want to get full provider tree and inventory information, they can query placement directly | |
| 15:45:17 | bauzas | and then, we're done with claims | |
| 15:45:26 | bauzas | what mriedem said | |
| 15:48:18 | jaypipes | what? | |
| 15:49:31 | efried | jaypipes: Help me out here. What's the reason we want to include non-requested resources and other-RPs-in-same-tree in provider summaries? | |
| 15:49:56 | edleafe | filters/weighers? | |
| 15:50:20 | jaypipes | efried: so that filters like the NUMATopologyFilter and ComputeCapabilitiesFilter can use the inventory usage and trait information. | |
| 15:50:33 | mriedem | if you need to filter on a resource class, that should be part of the requested resource in the query which puts it in the response, right? | |
| 15:51:02 | mriedem | this spec doesn't say anything about returning a full view of the resource provider traits either, only what was requested | |
| 15:51:11 | mriedem | or do you already get the full set of traits back in provider summaries? | |
| 15:51:12 | efried | mriedem: As I tried to explain, we have no syntax to talk about resource classes we're *not* requesting. | |
| 15:51:14 | jaypipes | mriedem: not if you want to know that a sibling NUMA node is "more packed" than another. | |
| 15:51:28 | efried | yeah, and that. | |
| 15:52:44 | mriedem | sure i get the weighing thing here | |
| 15:53:02 | mriedem | i'm basically -1 until something in-tree is going to leverage this | |
| 15:53:26 | mriedem | because provider 'summary' is not going to be a summary anymore, it's going to be a full blown view of the provider, it sounds like anyway | |
| 15:53:53 | efried | mriedem: That's really what was intended from the start, I believe. It was an artifact of the impl that it wound up being a subset. | |
| 15:54:21 | jaypipes | mriedem: we already return all the rp traits, not just the ones requested. https://github.com/openstack/nova/blob/master/nova/api/openstack/placement/objects/resource_provider.py#L3426 | |
| 15:54:25 | efried | mriedem: Surely you're not suggesting that "something in-tree is going to leverage this" is necessary in the Rocky timeframe? | |
| 15:54:42 | mriedem | efried: why would that be shocking? | |
| 15:54:46 | jaypipes | mriedem: summary only referred to the aggregation of usage counts. | |