Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-05-16
13:53:54 dansmith I feel like it's a fix to a piece of logic that was wrong
13:54:11 dansmith sure, nova could continue to do the thing it does today to work around it, but
13:54:32 dansmith anyway, whatever, I'd rather just get it done regardless
13:55:59 mriedem do what y'all want, i will step aside on this one
13:57:33 dansmith no, it's not that big of a deal, we've got a spec, let's just roll
13:57:37 dansmith getting it working is the important thing
14:00:34 openstackgerrit Vladyslav Drok proposed openstack/nova-specs master: Allow having placement inventories with reserved value equal to total https://review.openstack.org/568613
14:01:30 BobBall mriedem / naichuans: Naichuan is on PRC time but yes, the XenAPI bits of the vgpu-rocky series are definitely ready for a runway slot.
14:05:25 dansmith vdrok: see my latest feedback on your spec? if it's not clear, I can just tweak the wording myself so we can land that
14:05:46 mriedem BobBall: yeah it's in the runway now,
14:05:54 vdrok dansmith: yeah just noticed that, I'll add a couple sentences there, thank you
14:05:56 mriedem i know they are in china, just dropping pings here for later reading
14:06:03 mriedem BobBall: but thanks for following up
14:06:08 dansmith vdrok: cool
14:07:12 BobBall mriedem: Thanks :) I was in a nice 2 hour meeting so didn't see the message until late
14:09:00 openstackgerrit Vladyslav Drok proposed openstack/nova-specs master: Allow having placement inventories with reserved value equal to total https://review.openstack.org/568613
14:12:50 dansmith mriedem: so, how is the virt driver going to know if it should return no inventory or reserved=total?
14:13:10 dansmith or I guess we'll just require the new microversion and fail if not
14:13:13 dansmith so nevermind
14:13:40 mriedem yar
14:16:25 dansmith mriedem: jaypipes: let's get this over with: https://review.openstack.org/#/c/568613/7
14:17:31 jaypipes dansmith: I'll let mriedem do that. I'm supportive of it.
14:18:49 mriedem will look in a bit
14:21:32 mriedem random question,
14:22:25 mriedem on the rare days that the sun comes up here, at 6am or so, i find myself waking up early, and not wanting to get in the shower right away because everyone else is still sleeping, so i go unshowered until like mid morning, is that a normal work-from-home trajectory when summer starts?
14:22:29 stephenfin efried: No idea what's going on here, I'm afraid. Might be worth reaching out to infra about that https://review.openstack.org/#/c/568065/
14:23:01 efried mriedem: How often do you shave?
14:23:05 mriedem hells bells,
14:23:08 mriedem you know i'm a baby face right?
14:23:16 mriedem twice per week tops
14:23:17 efried Well, I didn't want to be impolite.
14:23:41 mriedem i will shave once the teenage mustache is pronounced
14:24:03 efried Having surprisingly little trouble picturing you with peachfuzz moustache.
14:24:10 mriedem and to keep up appearances with the mom that's at the same bus stop
14:24:50 stephenfin Anyone wanna fix an annoying warning/broken test? https://review.openstack.org/#/c/568532/
14:25:10 mriedem efried: i guess that's the pattern, whether or not my wife is at work in the morning and i have to go to the bus stop
14:25:10 efried mriedem: My perspective on showering is different. I usually only shower after I train. Which is around 8x/week these days. But some days I don't train until evening, so showering first thing in the morning isn't a thing for me.
14:25:11 mriedem problem solved
14:25:35 stephenfin efried: 8 times??
14:26:15 efried stephenfin: Yeah, 2x T & Th, Saturday is my day off.
14:29:58 mriedem dansmith: vdrok: jaypipes: +W
14:30:11 jaypipes thx
14:30:29 dansmith where's the code patch for it?
14:30:47 vdrok Thank you folks for reviewing this so fast!
14:31:12 mriedem https://review.openstack.org/#/c/564838/
14:36:13 bauzas mriedem: if you wanna know my life, I take shower before kids wake up, or then I'm done
14:36:39 bauzas mriedem: and given I need to taxi them to the school every morning, I guess other parents wouldn't like that
14:36:55 bauzas smelling is not an option
14:45:35 openstackgerrit Merged openstack/nova master: Move image conversion to privsep. https://review.openstack.org/554437
14:45:47 openstackgerrit Merged openstack/nova master: We don't need utils.trycmd any more. https://review.openstack.org/554439
14:48:38 mriedem i forgot about this until last night, but shall we deprecate running api services with eventlet? https://review.openstack.org/#/c/549510/
14:52:23 openstackgerrit Merged openstack/nova-specs master: Allow having placement inventories with reserved value equal to total https://review.openstack.org/568613
14:56:29 openstackgerrit Stephen Finucane proposed openstack/nova-specs master: Rename spec to match blueprint https://review.openstack.org/568870
15:00:50 bauzas mriedem: please, yes
15:00:51 bauzas +2
15:01:11 bauzas we already signaled since pike, so let's send another batsignal now
15:05:20 openstackgerrit Stephen Finucane proposed openstack/nova master: Simplify instance name generation https://review.openstack.org/516573
15:15:46 openstackgerrit Eric Fried proposed openstack/nova master: Debug logs for allocation_candidates filters https://review.openstack.org/568712
15:16:36 jroll efried: what do you train for? running or?
15:17:02 efried jroll: I own & run a Brazilian jiu jitsu school
15:17:09 jroll ah, neat :)
15:18:09 efried 7 of those 8 are me leading class (and sparring with my students). The last one is me going to my mothership to receive instruction.
15:29:24 openstackgerrit Matt Riedemann proposed openstack/nova master: Remove deprecated monkey_patch config options https://review.openstack.org/568880
15:40:58 openstackgerrit Matt Riedemann proposed openstack/nova master: Remove deprecated monkey_patch config options https://review.openstack.org/568880
15:47:19 openstackgerrit Merged openstack/nova master: PowerVM Driver: Snapshot https://review.openstack.org/543023
15:49:51 mriedem efried: maybe you have input on this? https://review.openstack.org/#/c/549053/
15:50:27 efried mriedem: I was going to let esberglu answer. If that doesn't happen fairly soon, I'll jump in.
15:50:37 openstackgerrit Zack Cornelius proposed openstack/nova master: Implement file backed memory for instances in libvirt https://review.openstack.org/567876
15:51:19 efried mriedem: Big picture, we do use some of those capability bits to set the respective ones on the driver itself - they're different depending on which disk driver you're using - and others internally to figure out whether we can do certain things.
15:51:52 openstackgerrit Zack Cornelius proposed openstack/nova-specs master: Libvirt file backed memory https://review.openstack.org/563704
15:53:10 efried mriedem: But I haven't dug in to see whether that logic truly exists yet in the in-tree incarnation.
15:53:54 mriedem i just figured if you have an abc parent class with methods/properties that by default raise NotImplementedError, those should be marked as abstract methods/properties
15:54:26 mriedem i.e. i think the point is that the subclass won't instantiate unless it implements those methods
15:56:21 openstackgerrit Balazs Gibizer proposed openstack/nova master: Add request_spec.RequestGroup versioned object https://review.openstack.org/568840
15:56:22 openstackgerrit Balazs Gibizer proposed openstack/nova master: Add requested_resources field to RequestSpec https://review.openstack.org/567267
15:56:23 openstackgerrit Balazs Gibizer proposed openstack/nova master: Transfer port.resource_request to the scheduler https://review.openstack.org/567268
15:56:48 mriedem lbragstad: policy question for you on https://review.openstack.org/#/c/568706/4/nova/api/openstack/placement/handlers/usage.py@93 - this api is defined as https://developer.openstack.org/api-ref/placement/#list-usages
15:56:55 mriedem and takes a required project_id query string parameter
15:57:13 mriedem right now our default rule uses check string role:admin
15:57:42 mriedem but thinking about admin_or_owner, would it make more sense if the target to the policy check is the project_id from the query param?
15:58:03 mriedem note placement doesn't have an admin_or_owner rule defined because most resources in placement don't have projects/users associated with them
15:58:29 lbragstad mriedem: so - it's safe to assume placement is mostly made up of system level resources?
15:58:30 mriedem or, i haven't gotten to the ones yet that do, like https://developer.openstack.org/api-ref/placement/#list-allocations
15:58:43 mriedem well, usages and allocations are per-consumer
15:58:56 mriedem and a consumer for now has a project/user (instance records, migrations, etc)
15:59:12 lbragstad that makes sense
15:59:15 mriedem i think a volume in cinder at some point would be a resource provider consumer
15:59:47 lbragstad but usages and allocations are per project in some cases and per user in others, right?
16:00:08 lbragstad (e.g. a user can have 5 ssh keys)
16:00:11 mriedem well, we always create them using the instance.project_id and instance.user_id i think
16:00:31 lbragstad ah
16:00:31 mriedem except i'm not sure if we pass user_id in all cases, sec
16:00:58 efried mriedem: oh, the abstract business, okay. Yeah, I understand, and yeah, abc is kind of a retrofit in python, not really any different behavior-wise from raising NotImplementedError. But we can probably clean that up if it's a thing for you.
16:01:04 lbragstad ok - i always thought that project_id was safe to assume in all cases (since pretty much everything requires a project scoped token today)
16:02:09 mriedem yes we always pass project_id,
16:02:10 gibi melwitt: I cannot make it to the tomorrow's nova team meeting so I cannot chair it.
16:02:26 mriedem and during scheduling we always pass user_id too, i just knew we had something different about how nova does that https://github.com/openstack/nova/blob/master/nova/scheduler/utils.py#L842
16:02:31 melwitt gibi: k, np. thanks for the heads up
16:03:02 mriedem lbragstad: yeah so allocations will have project_id and user_id, and /usages response is built based on the allocations data i believe
16:03:30 lbragstad mriedem: so - based on the fact that project_id is always required when calling list-usages, i think it's safe to say that is a project-specific API

Earlier   Later