| *** erlon has joined #openstack-meeting-alt | 01:15 | |
| *** yikun has joined #openstack-meeting-alt | 01:16 | |
| *** yamamoto has quit IRC | 01:29 | |
| *** yamamoto has joined #openstack-meeting-alt | 01:29 | |
| *** hongbin has joined #openstack-meeting-alt | 01:48 | |
| *** yamamoto has quit IRC | 02:01 | |
| *** yamamoto has joined #openstack-meeting-alt | 02:02 | |
| *** yamamoto has quit IRC | 02:06 | |
| *** dave-mccowan has quit IRC | 02:30 | |
| *** yamamoto has joined #openstack-meeting-alt | 02:43 | |
| *** yamamoto has quit IRC | 02:57 | |
| *** yamamoto has joined #openstack-meeting-alt | 02:58 | |
| *** yamamoto has quit IRC | 03:02 | |
| *** yamamoto has joined #openstack-meeting-alt | 03:46 | |
| *** kaisers has quit IRC | 03:56 | |
| *** kaisers has joined #openstack-meeting-alt | 04:05 | |
| *** kaisers has quit IRC | 04:12 | |
| *** kaisers has joined #openstack-meeting-alt | 04:19 | |
| *** tetsuro has joined #openstack-meeting-alt | 04:25 | |
| *** yamamoto has quit IRC | 04:29 | |
| *** hongbin has quit IRC | 04:32 | |
| *** tetsuro has quit IRC | 04:37 | |
| *** tetsuro has joined #openstack-meeting-alt | 04:38 | |
| *** janki has joined #openstack-meeting-alt | 04:54 | |
| *** yamamoto has joined #openstack-meeting-alt | 04:58 | |
| *** apetrich has joined #openstack-meeting-alt | 05:06 | |
| *** sheel has joined #openstack-meeting-alt | 05:11 | |
| *** jtomasek has joined #openstack-meeting-alt | 05:17 | |
| *** yamamoto has quit IRC | 05:45 | |
| *** e0ne has joined #openstack-meeting-alt | 06:21 | |
| *** tetsuro has quit IRC | 06:22 | |
| *** tetsuro has joined #openstack-meeting-alt | 06:25 | |
| *** yamamoto has joined #openstack-meeting-alt | 06:44 | |
| *** giblet is now known as gibi | 06:50 | |
| *** kopecmartin|off is now known as kopecmartin|ruck | 07:02 | |
| *** slaweq_ has joined #openstack-meeting-alt | 07:12 | |
| *** jtomasek has quit IRC | 07:16 | |
| *** jtomasek has joined #openstack-meeting-alt | 07:17 | |
| *** rcernin has quit IRC | 07:21 | |
| *** ttsiouts has joined #openstack-meeting-alt | 07:23 | |
| *** rdopiera has joined #openstack-meeting-alt | 07:37 | |
| *** ttsiouts has quit IRC | 07:39 | |
| *** yamamoto has quit IRC | 07:55 | |
| *** ttsiouts has joined #openstack-meeting-alt | 07:57 | |
| *** tssurya has joined #openstack-meeting-alt | 08:15 | |
| *** finucannot is now known as stephenfin | 08:15 | |
| *** belmoreira has joined #openstack-meeting-alt | 08:17 | |
| *** yamamoto has joined #openstack-meeting-alt | 08:53 | |
| *** alexchadin has joined #openstack-meeting-alt | 09:03 | |
| *** alexchadin has quit IRC | 09:15 | |
| *** tetsuro has quit IRC | 09:29 | |
| *** bhavikdbavishi has joined #openstack-meeting-alt | 09:47 | |
| *** yamamoto has quit IRC | 09:49 | |
| *** bhavikdbavishi has quit IRC | 09:52 | |
| *** ttsiouts has quit IRC | 10:01 | |
| *** jtomasek has quit IRC | 10:09 | |
| *** janki has quit IRC | 10:12 | |
| *** janki has joined #openstack-meeting-alt | 10:13 | |
| *** jtomasek has joined #openstack-meeting-alt | 10:28 | |
| *** rossella_s has joined #openstack-meeting-alt | 10:33 | |
| *** rcernin has joined #openstack-meeting-alt | 10:36 | |
| *** dpawlik has joined #openstack-meeting-alt | 10:37 | |
| *** slaweq_ is now known as slaweq | 10:44 | |
| *** slaweq is now known as dpawlik_ | 10:45 | |
| *** dpawlik_ is now known as slaweq | 10:45 | |
| *** dave-mccowan has joined #openstack-meeting-alt | 10:47 | |
| *** yamamoto has joined #openstack-meeting-alt | 10:48 | |
| *** alexchadin has joined #openstack-meeting-alt | 10:53 | |
| *** rossella_s has quit IRC | 10:55 | |
| *** rossella_s has joined #openstack-meeting-alt | 10:55 | |
| *** yamamoto has quit IRC | 11:00 | |
| *** yamamoto has joined #openstack-meeting-alt | 11:00 | |
| *** rossella_s has quit IRC | 11:04 | |
| *** rossella_s has joined #openstack-meeting-alt | 11:05 | |
| *** yamamoto has quit IRC | 11:07 | |
| *** erlon has quit IRC | 11:08 | |
| *** rcernin has quit IRC | 11:17 | |
| *** rossella_s has quit IRC | 11:19 | |
| *** rossella_s has joined #openstack-meeting-alt | 11:19 | |
| *** rossella_s has quit IRC | 11:30 | |
| *** rossella_s has joined #openstack-meeting-alt | 11:31 | |
| *** rossella_s has quit IRC | 11:39 | |
| *** rossella_s has joined #openstack-meeting-alt | 11:39 | |
| *** yamamoto has joined #openstack-meeting-alt | 11:46 | |
| *** belmorei_ has joined #openstack-meeting-alt | 11:48 | |
| *** rossella_s has quit IRC | 11:50 | |
| *** belmoreira has quit IRC | 11:51 | |
| *** rossella_s has joined #openstack-meeting-alt | 11:54 | |
| *** alexchadin has quit IRC | 11:56 | |
| *** yamamoto has quit IRC | 11:59 | |
| *** yamamoto has joined #openstack-meeting-alt | 11:59 | |
| *** yamamoto has quit IRC | 12:01 | |
| *** yamamoto has joined #openstack-meeting-alt | 12:01 | |
| *** raildo has joined #openstack-meeting-alt | 12:02 | |
| *** e0ne has quit IRC | 12:04 | |
| *** yamamoto has quit IRC | 12:06 | |
| *** helenafm has joined #openstack-meeting-alt | 12:07 | |
| *** yamamoto has joined #openstack-meeting-alt | 12:08 | |
| *** rossella_s has quit IRC | 12:09 | |
| *** rossella_s has joined #openstack-meeting-alt | 12:11 | |
| *** yamamoto has quit IRC | 12:12 | |
| *** e0ne has joined #openstack-meeting-alt | 12:13 | |
| *** ttsiouts has joined #openstack-meeting-alt | 12:14 | |
| *** fried_rice is now known as efried | 12:49 | |
| *** yamamoto has joined #openstack-meeting-alt | 12:51 | |
| *** lbragstad has joined #openstack-meeting-alt | 12:54 | |
| *** iyamahat has joined #openstack-meeting-alt | 12:54 | |
| *** jgrosso has joined #openstack-meeting-alt | 12:55 | |
| *** ijw has joined #openstack-meeting-alt | 12:56 | |
| *** ijw has quit IRC | 12:56 | |
| *** ijw has joined #openstack-meeting-alt | 12:57 | |
| *** ijw has quit IRC | 12:59 | |
| *** ijw_ has joined #openstack-meeting-alt | 13:00 | |
| *** tetsuro has joined #openstack-meeting-alt | 13:04 | |
| *** e0ne has quit IRC | 13:05 | |
| *** rossella_s has quit IRC | 13:07 | |
| *** rossella_s has joined #openstack-meeting-alt | 13:07 | |
| *** dustins has joined #openstack-meeting-alt | 13:13 | |
| *** lbragstad has quit IRC | 13:24 | |
| *** rossella_s has quit IRC | 13:28 | |
| *** rossella_s has joined #openstack-meeting-alt | 13:29 | |
| *** erlon has joined #openstack-meeting-alt | 13:30 | |
| *** erlon_ has joined #openstack-meeting-alt | 13:31 | |
| *** cloudrancher has quit IRC | 13:31 | |
| *** cloudrancher has joined #openstack-meeting-alt | 13:32 | |
| *** lbragstad has joined #openstack-meeting-alt | 13:32 | |
| *** erlon has quit IRC | 13:35 | |
| *** rossella_s has quit IRC | 13:42 | |
| *** rossella_s has joined #openstack-meeting-alt | 13:43 | |
| *** sheel has quit IRC | 13:50 | |
| *** cloudrancher has quit IRC | 13:51 | |
| *** janki has quit IRC | 13:51 | |
| *** janki has joined #openstack-meeting-alt | 13:52 | |
| *** cloudrancher has joined #openstack-meeting-alt | 13:52 | |
| *** jaypipes has joined #openstack-meeting-alt | 13:53 | |
| *** takashin has joined #openstack-meeting-alt | 13:54 | |
| efried | #startmeeting nova_scheduler | 14:00 |
|---|---|---|
| openstack | Meeting started Mon Oct 8 14:00:19 2018 UTC and is due to finish in 60 minutes. The chair is efried. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:00 |
| openstack | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:00 |
| *** openstack changes topic to " (Meeting topic: nova_scheduler)" | 14:00 | |
| openstack | The meeting name has been set to 'nova_scheduler' | 14:00 |
| takashin | o/ | 14:00 |
| jaypipes | o/ | 14:00 |
| * gibi cannot join this time | 14:01 | |
| * efried strikes gibi from agenda | 14:01 | |
| efried | Bueller? Bueller? | 14:02 |
| efried | guess we'll get started and let people wander in. | 14:02 |
| efried | #link agenda https://wiki.openstack.org/wiki/Meetings/NovaScheduler#Agenda_for_next_meeting | 14:02 |
| efried | #topic last meeting | 14:03 |
| *** openstack changes topic to "last meeting (Meeting topic: nova_scheduler)" | 14:03 | |
| efried | #link last minutes: http://eavesdrop.openstack.org/meetings/nova_scheduler/2018/nova_scheduler.2018-10-01-14.00.html | 14:03 |
| efried | Any old business? | 14:03 |
| *** mriedem has joined #openstack-meeting-alt | 14:03 | |
| mriedem | o/ | 14:03 |
| efried | #topic specs and review | 14:03 |
| efried | #link latest pupdate: http://lists.openstack.org/pipermail/openstack-dev/2018-October/135475.html | 14:03 |
| *** openstack changes topic to "specs and review (Meeting topic: nova_scheduler)" | 14:03 | |
| * bauzas waves a bit late | 14:03 | |
| efried | Anything to call out from the pupdate? (Will talk about extraction a bit later) | 14:04 |
| efried | #link Consumer generation & nrp use in nova: Series now starting at https://review.openstack.org/#/c/605785/ | 14:04 |
| efried | No longer in runway. Was gonna ask gibi the status of the series, but he's not attending today. | 14:04 |
| efried | Bottom patch has some minor fixups required | 14:05 |
| *** ganso has joined #openstack-meeting-alt | 14:05 | |
| efried | an interesting issue raised by tetsuro | 14:05 |
| mriedem | i need to look at that one, | 14:05 |
| mriedem | since i talked with gibi about the design before he wrote it | 14:06 |
| efried | which is that we can't tell whether the destination is nested in many cases until after we've already decided to schedule to it. | 14:06 |
| bauzas | also some concern about how the filters could check the computes in case allocation candidates are only about nested RPs | 14:06 |
| efried | which means we don't know whether we need to run the scheduler until... we've run the scheduler. | 14:06 |
| mriedem | i have the same problem with the resize to same host bug | 14:06 |
| efried | well, if you know the host already, you can go query that host to see if he's nested. And if so, you have to run the scheduler. | 14:07 |
| bauzas | I had a concern on changing the behaviour in https://review.openstack.org/#/c/605785/9/nova/compute/api.py@4375 | 14:07 |
| *** ganso has left #openstack-meeting-alt | 14:07 | |
| *** munimeha1 has joined #openstack-meeting-alt | 14:07 | |
| bauzas | if we want to call the scheduler anyway, we should have a new microversion IMHO | 14:07 |
| mriedem | bauzas: gibi and i had talked about the behavioral changes, but we didn't think a new microversion would be needed here, | 14:08 |
| mriedem | but it's messy i agree, | 14:08 |
| bauzas | heh | 14:08 |
| bauzas | huh* even | 14:08 |
| mriedem | we already broke the force behavior in pike when we made sure we could claim allocations for vcpu/disk/ram on force | 14:08 |
| mriedem | here we're breaking that if nested | 14:08 |
| mriedem | the more we depend on claims in the scheduler, the less we can honor force | 14:09 |
| bauzas | if we want to stop forcing a target (wrt I'm fine with), I just think we should still signal it for operators | 14:09 |
| efried | Can we add a column to the hosts table caching whether the host uses nested/sharing? | 14:09 |
| bauzas | like, you wanna still not call the scheduler ? fair enough, just don't ask for 2.XX microversion | 14:09 |
| bauzas | >2.XX even | 14:10 |
| jaypipes | why does it matter if we go from a non-nested host to a nested host? I mean, if the nested host supports the original requested resources and traits, who cares? | 14:10 |
| mriedem | i don't think we want to allow people to opt into breaking themselves | 14:10 |
| efried | bauzas: But if we don't call the scheduler, we literally *can't* schedule to a nested host | 14:10 |
| bauzas | efried: how can I target a nested resource provider ? | 14:10 |
| bauzas | could someone give me examples ? | 14:10 |
| efried | jaypipes: a) How would you know if it does? b) if any of the resources are in child providers, you need GET /a_c to give you a proper allocation request. | 14:10 |
| bauzas | operators target compute services | 14:11 |
| jaypipes | efried: and? | 14:11 |
| efried | and that (calling GET /a_c rather than just cloning the alloc onto the dest) is a behavior change. | 14:11 |
| mriedem | we should probably table this until gibi is around to talk about it, because i know he and i talked about a bit of this before he started this code | 14:11 |
| jaypipes | efried: if the scheduler returns a destination, we use it. who cares if the resources ended up being provided by child providers or not. | 14:11 |
| efried | that's the point. The scheduler returns a destination if we call the scheduler. | 14:12 |
| efried | We're talking about a code path where previously we *didn't* call the scheduler. | 14:12 |
| efried | IIUC. | 14:12 |
| mriedem | jaypipes: the question is when you force and bypass the scheduler | 14:12 |
| jaypipes | ah... force_host rears its ugly-ass head yet again. | 14:12 |
| mriedem | yes | 14:12 |
| bauzas | not force_hosts | 14:12 |
| mriedem | same idea | 14:12 |
| bauzas | force_hosts is only for boot | 14:12 |
| mriedem | i think we should table until gibi is around | 14:12 |
| bauzas | but it's calling the scheduler | 14:12 |
| efried | yeah. | 14:12 |
| mriedem | i could try to dig up our irc conversation but it'd be hard probably | 14:13 |
| efried | or we could just proceed, and make a big decision that affects his whole world for the next six months. | 14:13 |
| bauzas | compared to livemigrate/evacuate where you litterally can bypass scheduler | 14:13 |
| jaypipes | I guess I still don't see why we care. If the destination host (forced or not) supports the original request, why do we care? | 14:13 |
| efried | chicken/egg. We don't know if it supports the original request unless we call the scheduler algo to find that out. | 14:13 |
| *** SteelyDan is now known as dansmith | 14:13 | |
| mriedem | well, we claim outside of the scheduler | 14:14 |
| efried | I'm not sure to what extent ops expect "force" to mean "don't call the scheduler" though. | 14:14 |
| mriedem | today | 14:14 |
| bauzas | I still don't get why we're concerned by nested resource providers being targets | 14:14 |
| jaypipes | efried: why can't we ask the destination host in pre-live-migrate? | 14:14 |
| mriedem | like i said, we already sort of broke the live migration 'force' parameter in pike, | 14:14 |
| mriedem | when conductor started claiming | 14:14 |
| bauzas | efried: since live-migrate API is existing AFAIK | 14:14 |
| bauzas | mriedem: shit, I missed that then | 14:14 |
| efried | bauzas: If any of the resources that we need come from nested providers, we must use GET /a_c to come up with a proper allocation request. | 14:14 |
| bauzas | efried: isn't that a bit related to the concern I had about candidates be only on nested resource providers ? | 14:15 |
| bauzas | we somehow need to know which root RP we're talking about | 14:15 |
| mriedem | bauzas: see https://review.openstack.org/#/c/605785/9/nova/conductor/tasks/live_migrate.py@132 and scheduler_utils.claim_resources_on_destination for history | 14:16 |
| *** rossella_s has quit IRC | 14:16 | |
| efried | so, tabling until we can involve gibi. Moving on. | 14:16 |
| mriedem | +1 | 14:16 |
| efried | Extraction | 14:16 |
| efried | Info in the pupdate ---^ | 14:16 |
| efried | cdent out this week. edleafe, mriedem, status? | 14:16 |
| efried | Oh, Ed isn't around either. It's all on you mriedem | 14:17 |
| *** e0ne has joined #openstack-meeting-alt | 14:17 | |
| mriedem | umm | 14:17 |
| * mriedem looks | 14:17 | |
| mriedem | https://review.openstack.org/#/c/604454/ is the grenade patch which is passing, | 14:17 |
| mriedem | cdent updated that with the proper code to create the uwsgi placement-api config | 14:18 |
| efried | #link https://review.openstack.org/#/c/604454/ is the grenade patch which is passing | 14:18 |
| mriedem | the devstack change that depends on it is still failing though https://review.openstack.org/#/c/600162/ | 14:18 |
| efried | #link the devstack change that depends on it https://review.openstack.org/#/c/600162/ | 14:18 |
| efried | This is the $PROJECTS issue? | 14:18 |
| mriedem | tl;dr there are other jobs that devstack runs which aren't cloning the placement repo yet, | 14:18 |
| mriedem | i have patches up for that, but they aren't passing and i haven't dug into why yet | 14:19 |
| mriedem | yeah https://review.openstack.org/#/c/606853/ and https://review.openstack.org/#/c/608266/ | 14:19 |
| bauzas | I have good news for extraction | 14:19 |
| bauzas | https://review.openstack.org/#/c/599208/ has been tested and works on a physical machine with pGPUs | 14:20 |
| mriedem | efried: looks like my d-g patch for updating $PROJECTS passed, just failed one test in tempest | 14:20 |
| mriedem | so just rechecks it looks like | 14:20 |
| bauzas | next step will be to write some functional test mocking this ^ | 14:20 |
| efried | nice | 14:20 |
| jaypipes | bauzas: nice. | 14:20 |
| efried | bauzas: That's more of a reshape nugget than extraction, though? | 14:21 |
| jaypipes | efried: we agreed that that was a requirement for extraction. | 14:21 |
| bauzas | efried: well, I thought we agreed on this being a priority for the extraction :) | 14:21 |
| efried | oh, I guess we said we were going to want ... yeah | 14:21 |
| efried | I forget why, actually. | 14:21 |
| bauzas | anyway | 14:22 |
| jaypipes | let's not rehash that. | 14:22 |
| efried | oh, right, it was a requirement for the governance split | 14:22 |
| efried | not for getting extracted placement working. | 14:22 |
| efried | cool cool | 14:22 |
| jaypipes | I have a spec topic... | 14:22 |
| *** beekneemech is now known as bnemec | 14:22 | |
| efried | anything else on extraction? | 14:22 |
| mriedem | tl;dr it's close | 14:22 |
| efried | sweet | 14:22 |
| mriedem | for the grenade/devstack ci/infra bits | 14:22 |
| bauzas | I need to disappear, taxi driving my kids from school | 14:23 |
| efried | jaypipes: Want to go now or after the other spec/review topics? | 14:23 |
| jaypipes | so I have repeatedly stated I am not remotely interested in pursuing either https://review.openstack.org/#/c/544683/ or https://review.openstack.org/#/c/552105/. I was under the impression that someone (Yikun maybe?) who *was* interested in continuing that work was going to get https://review.openstack.org/#/c/552105/ into a state where people agreed on it (good luck with that), but as of now, I've seen little action on it other than | 14:24 |
| jaypipes | negative reviews. | 14:24 |
| * efried click click click | 14:24 | |
| mriedem | jaypipes: yeah yikun has been busy with some internal stuff after a re-org, | 14:24 |
| jaypipes | so my question is should I just abandon both of the specs and force the issue? | 14:24 |
| mriedem | i can send an email to see what's going on and if we still care about those | 14:24 |
| jaypipes | k, thx | 14:24 |
| *** rossella_s has joined #openstack-meeting-alt | 14:25 | |
| efried | This could relate to the next-next topic on the agenda actually. | 14:25 |
| *** dtrainor has joined #openstack-meeting-alt | 14:26 | |
| efried | we were talking about using the file format proposal embedded in the | 14:26 |
| efried | #link device passthrough spec https://review.openstack.org/#/c/591037/ | 14:26 |
| efried | as a mechanism to customize provider attributes (prompted by the belmoreira min_unit discussion) | 14:26 |
| efried | jaypipes agreed to review ^ with that in mind | 14:26 |
| jaypipes | efried: yes. | 14:27 |
| jaypipes | efried: I have found it very difficult to review. will give it another go this morning. | 14:27 |
| efried | The "initial defaults" thing is still weird. | 14:27 |
| efried | and not addressed in there (yet) | 14:27 |
| efried | bauzas suggested to split out the part of the spec that talks about the file format, and do the device passthrough aspect on its own. | 14:28 |
| efried | Which sounds like a good idea to me, considering the various ways we've talked about using it. | 14:28 |
| efried | okay, moving on. | 14:30 |
| efried | last week, the | 14:30 |
| efried | #link HPET discussion http://lists.openstack.org/pipermail/openstack-dev/2018-October/135446.html | 14:30 |
| efried | led to an interesting precedent on using traits for config | 14:30 |
| jaypipes | another spec ... I pushed a new rev on https://review.openstack.org/#/c/555081/ | 14:30 |
| jaypipes | (cpu resource tracking) | 14:30 |
| efried | #link CPU resource tracking spec https://review.openstack.org/#/c/555081/ | 14:30 |
| *** lpetrut has joined #openstack-meeting-alt | 14:32 | |
| efried | any discussion on traits-for-config or CPU resource tracking? | 14:33 |
| efried | any other specs or reviews to discuss? | 14:33 |
| mriedem | i personally hope that cpu resource tracking is not something we pursue for stein | 14:33 |
| mriedem | while we're still trying to land reshaper et al | 14:33 |
| mriedem | reshaping all instances on all compute nodes is going to be rough during upgrade | 14:34 |
| mriedem | unless we can do that offline | 14:34 |
| jaypipes | mriedem: so let's hold off getting new clean functionality so that upgrades can be prolonged even longer until end of 2019? | 14:35 |
| mriedem | yes? | 14:36 |
| dansmith | I feel like we've been putting off numa topo in placement a while now | 14:36 |
| mriedem | i think getting reshaper and bw-aware scheduling and all that stuff has been around long enough that we need to get those done first | 14:36 |
| *** rossella_s has quit IRC | 14:36 | |
| dansmith | so I don't disagree that it's going to be a big reshape, but.. dang, we've been working towards it for a while now and.. | 14:36 |
| jaypipes | mriedem: I don't get the argument that adding another data migration (reshape action) makes upgrades harder than having one to do in a release cycle. | 14:37 |
| mriedem | i would just like fewer things to worry about | 14:37 |
| dansmith | if we end up with something for gpus that requires compute nodes to be online, | 14:37 |
| dansmith | it'd be massively better for FFU to have both of those in the same release | 14:37 |
| dansmith | vs. two different (especially back-to-back) releases | 14:37 |
| mriedem | do we need the computes online for the cpu resource tracking upgrade? | 14:38 |
| dansmith | yes | 14:38 |
| dansmith | they have to do it themselves, I think, because only they know where and what the topology is | 14:38 |
| jaypipes | dansmith: right, unless we go with a real inventory/provider descriptor file format. | 14:39 |
| dansmith | jaypipes: well, that just pushes the problem elsewhere.. you still have to collect that info from somewhere | 14:39 |
| *** rossella_s has joined #openstack-meeting-alt | 14:39 | |
| jaypipes | dansmith: it's already in the vast majority of inventory management systems. | 14:39 |
| efried | waitwait, the *admin* is going to be responsible for describing NUMA topology? It's not something the driver can discover? | 14:39 |
| dansmith | efried: we should have the driver do it for sure | 14:40 |
| jaypipes | efried: the operator is ultimately responsible for *whether* a compute node should expose providers as a tree. | 14:40 |
| efried | phew | 14:40 |
| dansmith | jaypipes: but we can't just requre the operator to have that and build such a mapping, IMHO | 14:40 |
| dansmith | but even still, | 14:40 |
| mriedem | why would operators care how we model things internally? | 14:40 |
| jaypipes | efried: a lot of operators don't want or need to deal with NUMA. they just have needs for dedicated CPU and shared CPU resources and don't care about NUMA. | 14:41 |
| efried | Yeah, I can live with a "use numa or not" switch. | 14:41 |
| dansmith | the driver is the only one that can decide how existing allocations map to that information, IMHO, so unless you want to run the driver against the DB from a central node... even still, there are numa pinnings that the driver has done already we need to know about | 14:41 |
| efried | I was just afraid you were talking about requiring the op to say "and this CPU is in NUMA node 1, and this CPU is in NUMA node 2 and..." | 14:41 |
| dansmith | mriedem: they don't, that's why making them write a topo description for each (type of) compute node to do this migration would be mega-suck | 14:41 |
| dansmith | efried: I think that's what jaypipes is saying | 14:41 |
| dansmith | and I think that's not reasonable | 14:42 |
| dansmith | efried: I don't think a "numa or not" switch is reasonable either, FWIW | 14:42 |
| jaypipes | dansmith: ops *already* have that. they all have hardware profiles which describe the different types of hardware they provide to guests. | 14:42 |
| dansmith | they just want it to work | 14:42 |
| dansmith | jaypipes: all ops do not have that | 14:42 |
| dansmith | jaypipes: but even still, they don't have the information about what numa allocations we've already done for existing instances | 14:42 |
| jaypipes | dansmith: agreed completely with that last statement. | 14:43 |
| efried | With a generic inventory/provider descriptor file, you could allow the op to override/customize. But I would think we would want the default to be automatic detection/configuration resulting in at least a sane setup. | 14:43 |
| jaypipes | it's a shame the guest NUMA topology and CPU pinning were implemented as such a tightly coupled blobject mess. | 14:43 |
| mriedem | while i agree it would be best if we can do all the reshapes we know we need to do in the same release to ease the pain, i just wanted to state that i'm worried about trying to bite this off in stein with everything else that's going on | 14:43 |
| dansmith | mriedem: there's risk there for sure, no doubt | 14:44 |
| efried | We also still can't do generic affinity without a placement API change, just to bring that up again. | 14:44 |
| dansmith | I'm not saying it's critical, I'm just saying writing it off now seems like a bad idea to me | 14:44 |
| *** rossella_s has quit IRC | 14:44 | |
| mriedem | i'll admit the only part of that spec i've read is the upgrade impact | 14:45 |
| mriedem | then i had to go change my drawers | 14:45 |
| dansmith | mriedem: I guess I'm not sure why that's a surprise at this point, | 14:45 |
| mriedem | will artom's stuff depend on this? | 14:45 |
| dansmith | but maybe I have just done more thinking about it | 14:45 |
| mriedem | artom's stuff = numa aware live migration | 14:45 |
| dansmith | mriedem: artom's stuff kinda conflicts with this.. if this was done his stuff would be easier I think | 14:45 |
| mriedem | dansmith: yeah i've just avoided thinking about this | 14:45 |
| *** rossella_s has joined #openstack-meeting-alt | 14:46 | |
| mriedem | ok, i need to get updated on what he plans to do with that as well | 14:46 |
| mriedem | anyway, i'll be quiet now | 14:46 |
| efried | dansmith, jaypipes: any last words? | 14:48 |
| dansmith | no. | 14:48 |
| efried | Home stretch | 14:49 |
| efried | #topic bugs | 14:49 |
| efried | #link Placement bugs https://bugs.launchpad.net/nova/+bugs?field.tag=placement | 14:49 |
| *** openstack changes topic to "bugs (Meeting topic: nova_scheduler)" | 14:49 | |
| efried | any bugs to highlight? | 14:49 |
| jaypipes | efried: go Browns? | 14:49 |
| mriedem | ugliest FG ever | 14:50 |
| efried | Horns to 5-1 by a toenail. Khabib face cranks Connor to a tap, then attacks his training team. Derek Lewis's balls are hot. Other weekend sports news? | 14:51 |
| *** edleafe has joined #openstack-meeting-alt | 14:51 | |
| efried | I guess we're really in | 14:51 |
| efried | #topic opens | 14:51 |
| *** openstack changes topic to "opens (Meeting topic: nova_scheduler)" | 14:51 | |
| dansmith | is that real sports news? | 14:51 |
| mriedem | yes | 14:52 |
| dansmith | hah. okay, sounded made up | 14:52 |
| mriedem | https://deadspin.com/khabib-nurmagomedov-taps-out-conor-mcgregor-attacks-co-1829580622 | 14:52 |
| dansmith | shows how much I know | 14:52 |
| * bauzas waves again | 14:52 | |
| efried | and https://www.youtube.com/watch?v=F_E6jXHMPs4 | 14:53 |
| efried | okay, anything else? | 14:53 |
| * edleafe arrives super-late | 14:53 | |
| *** gcb_ has joined #openstack-meeting-alt | 14:53 | |
| efried | edleafe: Anything to bring up before we close? | 14:53 |
| *** e0ne has quit IRC | 14:54 | |
| bauzas | we also had https://www.youtube.com/watch?v=KgwmPhAu0tc | 14:54 |
| efried | Thanks y'all | 14:55 |
| efried | #endmeeting | 14:55 |
| *** openstack changes topic to "OpenStack Meetings || https://wiki.openstack.org/wiki/Meetings/" | 14:55 | |
| openstack | Meeting ended Mon Oct 8 14:55:31 2018 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:55 |
| openstack | Minutes: http://eavesdrop.openstack.org/meetings/nova_scheduler/2018/nova_scheduler.2018-10-08-14.00.html | 14:55 |
| openstack | Minutes (text): http://eavesdrop.openstack.org/meetings/nova_scheduler/2018/nova_scheduler.2018-10-08-14.00.txt | 14:55 |
| openstack | Log: http://eavesdrop.openstack.org/meetings/nova_scheduler/2018/nova_scheduler.2018-10-08-14.00.log.html | 14:55 |
| *** helenafm has left #openstack-meeting-alt | 14:55 | |
| *** mriedem has left #openstack-meeting-alt | 14:55 | |
| *** tetsuro has quit IRC | 14:56 | |
| *** lpetrut has quit IRC | 14:57 | |
| *** ijw_ has quit IRC | 14:58 | |
| *** ijw has joined #openstack-meeting-alt | 14:58 | |
| *** takashin has left #openstack-meeting-alt | 14:59 | |
| *** cloudrancher has quit IRC | 15:01 | |
| *** cloudrancher has joined #openstack-meeting-alt | 15:02 | |
| *** dklyle has joined #openstack-meeting-alt | 15:04 | |
| *** cloudrancher has quit IRC | 15:07 | |
| *** cloudrancher has joined #openstack-meeting-alt | 15:08 | |
| *** e0ne has joined #openstack-meeting-alt | 15:11 | |
| *** janki has quit IRC | 15:20 | |
| *** panda is now known as panda|off | 15:23 | |
| *** rossella_s has quit IRC | 15:35 | |
| *** rossella_s has joined #openstack-meeting-alt | 15:37 | |
| *** e0ne has quit IRC | 15:40 | |
| *** e0ne has joined #openstack-meeting-alt | 15:42 | |
| *** tssurya has quit IRC | 15:47 | |
| *** e0ne has quit IRC | 15:49 | |
| *** rossella_s has quit IRC | 15:51 | |
| *** rossella_s has joined #openstack-meeting-alt | 15:51 | |
| *** rossella_s has quit IRC | 15:56 | |
| *** cloudrancher has quit IRC | 15:59 | |
| *** cloudrancher has joined #openstack-meeting-alt | 15:59 | |
| *** rossella_s has joined #openstack-meeting-alt | 16:00 | |
| *** gcb_ has quit IRC | 16:03 | |
| *** cloudrancher has quit IRC | 16:06 | |
| *** cloudrancher has joined #openstack-meeting-alt | 16:08 | |
| *** belmorei_ has quit IRC | 16:11 | |
| *** rdopiera has quit IRC | 16:21 | |
| *** ttsiouts has quit IRC | 16:21 | |
| *** panda|off has quit IRC | 16:22 | |
| *** ttsiouts has joined #openstack-meeting-alt | 16:22 | |
| *** panda has joined #openstack-meeting-alt | 16:23 | |
| *** ttsiouts has quit IRC | 16:26 | |
| *** efried has quit IRC | 16:30 | |
| *** rossella_s has quit IRC | 16:31 | |
| *** efried has joined #openstack-meeting-alt | 16:37 | |
| *** gyee has joined #openstack-meeting-alt | 16:48 | |
| *** jgrosso has quit IRC | 16:57 | |
| *** apetrich has quit IRC | 17:11 | |
| *** iyamahat has quit IRC | 17:15 | |
| *** e0ne has joined #openstack-meeting-alt | 17:24 | |
| *** iyamahat has joined #openstack-meeting-alt | 17:28 | |
| *** e0ne has quit IRC | 17:29 | |
| *** iyamahat_ has joined #openstack-meeting-alt | 17:32 | |
| *** iyamahat has quit IRC | 17:36 | |
| *** apetrich has joined #openstack-meeting-alt | 17:47 | |
| *** e0ne has joined #openstack-meeting-alt | 17:54 | |
| *** Swami has joined #openstack-meeting-alt | 18:13 | |
| *** panda has quit IRC | 18:26 | |
| *** panda has joined #openstack-meeting-alt | 18:28 | |
| *** kopecmartin|ruck is now known as kopecmartin|off | 18:45 | |
| *** hongbin has joined #openstack-meeting-alt | 18:50 | |
| *** slagle has joined #openstack-meeting-alt | 18:52 | |
| *** tssurya has joined #openstack-meeting-alt | 18:53 | |
| *** _pewp_ has quit IRC | 18:54 | |
| *** _pewp_ has joined #openstack-meeting-alt | 18:55 | |
| *** diablo_rojo has joined #openstack-meeting-alt | 19:04 | |
| *** sambetts|afk has quit IRC | 19:07 | |
| *** sambetts_ has joined #openstack-meeting-alt | 19:10 | |
| *** armstrong has joined #openstack-meeting-alt | 19:11 | |
| *** slagle has quit IRC | 19:40 | |
| *** diablo_rojo has quit IRC | 19:42 | |
| *** cloudrancher has quit IRC | 19:43 | |
| *** cloudrancher has joined #openstack-meeting-alt | 19:44 | |
| *** jtomasek has quit IRC | 20:13 | |
| *** e0ne has quit IRC | 20:19 | |
| *** slagle has joined #openstack-meeting-alt | 20:20 | |
| *** e0ne has joined #openstack-meeting-alt | 20:20 | |
| *** tssurya has quit IRC | 20:28 | |
| *** lbragstad has quit IRC | 20:37 | |
| *** slagle has quit IRC | 20:37 | |
| *** lbragstad has joined #openstack-meeting-alt | 20:37 | |
| *** e0ne has quit IRC | 20:48 | |
| *** raildo has quit IRC | 20:57 | |
| *** iyamahat__ has joined #openstack-meeting-alt | 21:52 | |
| *** erlon_ has quit IRC | 21:54 | |
| *** iyamahat_ has quit IRC | 21:55 | |
| *** slaweq has quit IRC | 22:01 | |
| *** dustins has quit IRC | 22:09 | |
| *** armstrong has quit IRC | 22:12 | |
| *** ijw has quit IRC | 22:19 | |
| *** hongbin has quit IRC | 22:38 | |
| *** dave-mccowan has quit IRC | 22:49 | |
| *** munimeha1 has quit IRC | 22:50 | |
| *** rcernin has joined #openstack-meeting-alt | 22:50 | |
| *** dave-mccowan has joined #openstack-meeting-alt | 22:51 | |
| *** pbourke has quit IRC | 23:08 | |
| *** pbourke has joined #openstack-meeting-alt | 23:08 | |
| *** lbragstad has quit IRC | 23:16 | |
| *** cloudrancher has quit IRC | 23:24 | |
| *** cloudrancher has joined #openstack-meeting-alt | 23:24 | |
| *** Swami has quit IRC | 23:37 | |
| *** tpsilva has quit IRC | 23:41 | |
| *** gyee has quit IRC | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!