| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2018-03-29 | |||
| 13:47:57 | jaypipes | It is that latter point that I am afraid of, since it means we essentially throw out all the scheduling heuristics that are now made in placement (since we would no longer be able to filter on providers having a trait but would need to add some mechanism to detect providers that *could have* a trait on them, if certain conditions apply) | |
| 13:48:08 | openstackgerrit | Sylvain Bauza proposed openstack/nova-specs master: Proposes Multiple GPU types https://review.openstack.org/557065 | |
| 13:48:08 | efried | jaypipes: #2 is a bit sticky, though, innit? The consumer requests an instance with FPGA with function X, cyborg finds a previously-unfettered device suitable for use and flashes it with function X, then sets a trait saying it's got function X on it now. Is that "consumer requests causing traits to be set"? | |
| 13:48:12 | dansmith | jaypipes: yup | |
| 13:48:29 | dansmith | efried: it has given that device capability | |
| 13:48:48 | bauzas | jaypipes: dansmith: very simple spec about providing a config option for defining a list of pGPUs a specific type would be supported https://review.openstack.org/557065 | |
| 13:48:51 | dansmith | it has changed the device in some meaningful way that means it can do things | |
| 13:48:58 | efried | dansmith: tbc, *I* am fine with that scenario. I'm asking if it violates jaypipes' #2 | |
| 13:49:12 | jaypipes | efried: it would, yes. | |
| 13:49:16 | bauzas | dansmith: jaypipes: if you can review it quickly, that would unblock me from using n-r-p for libvirt vGPUs | |
| 13:49:27 | dansmith | jaypipes: o.O | |
| 13:50:02 | bauzas | mriedem: dansmith: I'll sponsor https://review.openstack.org/#/c/544730/ and above | |
| 13:50:14 | dansmith | jaypipes: now I'm confused :) | |
| 13:50:16 | jaypipes | efried: if cyborg configures its inventory/traits dynamically *at the time and due to a specific request from a consumer*, I have an issue with that | |
| 13:50:35 | dansmith | jaypipes: heh, that's what you just said you were okay with | |
| 13:50:45 | mriedem | alex_xu_: interesting idea to deprecate the backup API in https://review.openstack.org/#/c/511825/ - you should probably start a ML thread on that if you want wider input | |
| 13:50:46 | jaypipes | dansmith: it's a difference in timing. | |
| 13:51:06 | dansmith | jaypipes: no, I said above "sets the trait when programming" | |
| 13:51:19 | alex_xu_ | mriedem: will do that later | |
| 13:51:22 | jaypipes | dansmith: when is the programming done. | |
| 13:51:44 | jaypipes | dansmith: that was a question, sorry.. when is the programming of the slots done? | |
| 13:52:03 | jaypipes | dansmith: if it's done after GET /allocation_candidates, then that won't work. | |
| 13:52:13 | dansmith | jaypipes: sure it will | |
| 13:52:23 | jaypipes | dansmith: since clearly the trait won't be useful in filtering out destination providers. | |
| 13:52:29 | alex_xu_ | jaypipes: dansmith note, for programmable fpga, it inventory always 1 | |
| 13:52:32 | dansmith | jaypipes: it would be preferred | |
| 13:52:56 | alex_xu_ | there is no race like we tweak traits in the fly for gpu | |
| 13:53:07 | dansmith | jaypipes: hangout? | |
| 13:53:11 | jaypipes | dansmith: sdure | |
| 13:53:12 | jaypipes | sure | |
| 13:53:20 | jaypipes | set it up. I will coffee up. | |
| 13:53:35 | jaypipes | let me grab a pug for emotional support. | |
| 13:53:39 | dansmith | https://hangouts.google.com/call/vBep88lVuQCIThVzSqUZAAEE | |
| 13:54:06 | mriedem | powervm interface attach/detach bp patch needs another +2 https://review.openstack.org/#/c/546813/ | |
| 13:54:15 | dansmith | e'yone's invited of course | |
| 13:56:43 | openstackgerrit | Claudiu Belu proposed openstack/nova master: tests: autospecs all the mock.patch usages https://review.openstack.org/470775 | |
| 13:59:06 | bauzas | dansmith: just for sanity, left some comments on https://review.openstack.org/#/c/544730/16 but +W'd it | |
| 14:10:26 | openstackgerrit | Claudiu Belu proposed openstack/nova master: tests: autospecs all the mock.patch usages https://review.openstack.org/470775 | |
| 14:10:27 | openstackgerrit | Claudiu Belu proposed openstack/nova master: WIP: replace spec with autospec https://review.openstack.org/557299 | |
| 14:11:28 | mriedem | bhagyashris: at least one more update in here i think https://review.openstack.org/#/c/511825/ | |
| 14:17:26 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Fix bad management of _TRAITS_SYNCED flag https://review.openstack.org/557722 | |
| 14:17:33 | cdent | mriedem: that ^ fixes it in my tests | |
| 14:18:34 | mriedem | let's recheck it a few times | |
| 14:20:02 | bhagyashris | mriedem: ok thank you for review. you mean to depricate the create backup api and newly introduce deleteBackup api will delete the backup using the rotation parameter | |
| 14:21:16 | mriedem | claudiub: can you update this quick? https://review.openstack.org/#/c/557377/ | |
| 14:21:46 | mriedem | bhagyashris: no, the alternative is to simply deprecate the createBackup API, and not introduce a new deleteBackups API | |
| 14:22:07 | mriedem | since createBackup can all be done using orchestration of existing nova/glance APIs | |
| 14:22:33 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Filter resource providers by forbidden traits in db https://review.openstack.org/556472 | |
| 14:22:34 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Filter allocation candidates by forbidden traits in db https://review.openstack.org/556660 | |
| 14:22:34 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Parse forbidden traits in query strings https://review.openstack.org/556819 | |
| 14:22:35 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Support forbidden traits in API https://review.openstack.org/556820 | |
| 14:24:15 | jaypipes | mriedem: amen, brother. | |
| 14:25:01 | mriedem | makes me curious when/why createBackup was added | |
| 14:25:12 | mriedem | but was likely so long ago that any blueprint for it has little to no detail | |
| 14:25:22 | jaypipes | mriedem: correct. | |
| 14:26:00 | openstackgerrit | Claudiu Belu proposed openstack/nova master: vmware: Fixes _detach_instance_volumes method https://review.openstack.org/557377 | |
| 14:27:15 | bhagyashris | mriedem: ok i am not getting this: cleanup the backups using a rotation value controlled client-side | |
| 14:27:49 | mriedem | bhagyashris: createBackup can be totally done using existing compute and image APIs today, without the actual createBackup API, | |
| 14:28:02 | mriedem | i can snapshot an instance with specific metadata to indicate it's a backup | |
| 14:28:12 | dansmith | alex_xu_: if that "what are you saying" was a joke, that was excellent :P | |
| 14:28:16 | mriedem | and then i can, as a tenant user, delete older backup snapshots in glance | |
| 14:28:57 | mriedem | bhagyashris: my point being, you as the tenant (or script, or external service, whatever) can do all of the createBackup stuff without the compute API (nova) doing it for you | |
| 14:29:17 | alex_xu_ | dansmith: I'm good at pretend to not understand what you guys saying | |
| 14:29:24 | dansmith | haha | |
| 14:29:35 | bauzas | dansmith: again, one comment on https://review.openstack.org/#/c/544729/16 but not holding me to +W | |
| 14:29:53 | bauzas | dansmith: tl;dr: I guess we're talking of aggregate UUIDs, not names | |
| 14:30:11 | openstackgerrit | Chris Dent proposed openstack/nova master: [placement] Support forbidden traits in API https://review.openstack.org/556820 | |
| 14:30:37 | openstackgerrit | Surya Seetharaman proposed openstack/nova master: Allow scheduling only to enabled cells (Filter Scheduler) https://review.openstack.org/550527 | |
| 14:30:38 | openstackgerrit | Surya Seetharaman proposed openstack/nova master: Add --enable and --disable options to nova-manage update_cell https://review.openstack.org/555416 | |
| 14:30:38 | openstackgerrit | Surya Seetharaman proposed openstack/nova master: Update the cells FAQs and scheduler maintenance docs. https://review.openstack.org/556459 | |
| 14:30:59 | mriedem | bauzas: good point | |
| 14:31:17 | bauzas | I don't want to throw the whole series | |
| 14:31:23 | bauzas | since the code will still work | |
| 14:31:31 | bauzas | at least until that point | |
| 14:31:36 | mriedem | if we wanted to use a ListOfUUIDsField, we'd want to do that now | |
| 14:31:45 | dansmith | bauzas: list of uuids is not enough | |
| 14:31:48 | bauzas | mriedem: not really | |
| 14:31:56 | dansmith | bauzas: which is why I didn't create one | |
| 14:31:57 | bauzas | mriedem: because of the AND OR | |
| 14:32:03 | bauzas | dansmith: yeah, I know | |
| 14:32:03 | dansmith | we need a list of list of uuids, | |
| 14:32:09 | bauzas | for me, it's just a docstring correction to make | |
| 14:32:12 | dansmith | which I think is kinda nuts | |
| 14:32:16 | mriedem | oh right, the comma | |
| 14:32:22 | bauzas | I'm fine with keeping a list of strings as of noxw | |
| 14:32:37 | dansmith | so these are "aggregate requests" and depend on the placement schema (which is strict) to handle it | |
| 14:32:39 | bauzas | but the fact that we will have strings made of UUIDs and commas | |
| 14:32:58 | bauzas | so, I just want to clarify the docstring, that's it | |
| 14:34:03 | dansmith | bauzas: yeah, fair neough | |
| 14:34:20 | dansmith | bauzas: the names make it easier to reason about the example, but I can make a point of explaining what they are | |
| 14:38:29 | bhagyashris | mriedem: But giving provision through nova API to createBackup of server backup is more user friendly | |
| 14:39:07 | mriedem | bhagyashris: well, it is because that's what is available today :) | |
| 14:39:32 | mriedem | bhagyashris: to be clear, i'm not -2 on your spec, i'm just saying that the alternative to deprecate should be mentioned in there | |
| 14:42:44 | mriedem | claudiub: can you also propose a stable/queens backport for https://review.openstack.org/#/c/557377/ ? | |
| 14:42:54 | mriedem | i've got a queens 17.0.2 release patch waiting and can hold it up for this | |
| 14:43:02 | mriedem | since it was a regression in queens 17.0.1 | |
| 14:43:43 | openstackgerrit | Claudiu Belu proposed openstack/nova stable/queens: vmware: Fixes _detach_instance_volumes method https://review.openstack.org/557727 | |
| 14:44:45 | bhagyashris | mriedem: yeah... still waiting for others opinion but deprecation will lose the user friendly and that most imp IMO. | |
| 14:47:01 | bhagyashris | johnthetubaguy: Hi waiting for opinion on same ^ (https://review.openstack.org/#/c/511825/) | |
| 14:47:57 | johnthetubaguy | bhagyashris: I like the idea of documenting the alternative, even if we decide against it | |