| *** tetsuro has joined #openstack-placement | 00:27 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: placement: Use INNER JOIN for requied traits https://review.openstack.org/573113 | 00:48 |
|---|---|---|
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Test alloc_cands with indirectly sharing RPs https://review.openstack.org/519601 | 01:05 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Fix nits in nested provider allocation candidates(2) https://review.openstack.org/573475 | 01:20 |
| *** mriedem_afk is now known as mriedem | 01:23 | |
| *** mriedem is now known as mriedem_afk | 01:37 | |
| *** tetsuro has quit IRC | 01:53 | |
| *** tetsuro has joined #openstack-placement | 01:53 | |
| *** mriedem_afk is now known as mriedem | 01:54 | |
| *** mriedem has quit IRC | 02:10 | |
| openstackgerrit | Shuo Liu proposed openstack/nova-specs master: fix spelling mistake https://review.openstack.org/573479 | 02:20 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova master: Remove mox in unit/network/test_neutronv2.py (2) https://review.openstack.org/573481 | 02:22 |
| openstackgerrit | Shuo Liu proposed openstack/nova-specs master: fix spelling mistake in rst file https://review.openstack.org/573487 | 02:49 |
| openstackgerrit | Steve Kowalik proposed openstack/nova master: Switch to oslo_messaging.ConfFixture.transport_url https://review.openstack.org/573046 | 03:26 |
| *** tetsuro has quit IRC | 04:48 | |
| *** tetsuro has joined #openstack-placement | 04:49 | |
| *** bhagyashri_s has quit IRC | 05:01 | |
| *** bhagyashri_s has joined #openstack-placement | 05:05 | |
| -openstackstatus- NOTICE: Zuul stopped receiving gerrit events around 04:00UTC; any changes submitted between then and now will probably require a "recheck" comment to be requeued. Thanks! | 06:01 | |
| *** e0ne has joined #openstack-placement | 06:09 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return all resources in provider_summaries https://review.openstack.org/558045 | 06:25 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return all providers - multiple providers https://review.openstack.org/559480 | 06:25 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Return all providers - single provider https://review.openstack.org/573545 | 06:25 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Fix nits in nested provider allocation candidates(2) https://review.openstack.org/573475 | 06:37 |
| *** e0ne has quit IRC | 06:44 | |
| openstackgerrit | Naichuan Sun proposed openstack/nova master: xenapi(N-R-P): Add API to support vgpu resource provider create https://review.openstack.org/520313 | 06:45 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs master: Create specs directory for Stein https://review.openstack.org/573602 | 07:06 |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Add microversion for nested allocation candidate https://review.openstack.org/565487 | 07:15 |
| *** e0ne has joined #openstack-placement | 07:26 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: Fix nits in placement-return-all-resources series https://review.openstack.org/573604 | 07:34 |
| *** bauzas is now known as PapaOurs | 07:38 | |
| openstackgerrit | Tetsuro Nakamura proposed openstack/nova master: WIP: Optimize member_of check for nested providers https://review.openstack.org/573203 | 07:38 |
| *** tetsuro has quit IRC | 07:43 | |
| *** takashin has left #openstack-placement | 08:03 | |
| *** ttsiouts has joined #openstack-placement | 08:12 | |
| *** gibi is now known as giblet | 08:18 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: API: add support to abort queued live migration in microversion 2.63 https://review.openstack.org/573136 | 08:31 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: libvirt: Don't report DISK_GB if sharing https://review.openstack.org/560459 | 08:45 |
| giblet | jaypipes, efried: I had to update ^^ to fix the functional test failure and added to extra assert as well. I think the 3 of us can send this in now. | 08:48 |
| giblet | s/to/two/ | 08:55 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add policy to InstanceGroup object and api models. https://review.openstack.org/563375 | 08:57 |
| openstackgerrit | Yikun Jiang (Kero) proposed openstack/nova master: Add InstanceGroupPolicy object https://review.openstack.org/573628 | 08:57 |
| *** ttsiouts has quit IRC | 09:16 | |
| *** ttsiouts has joined #openstack-placement | 09:16 | |
| *** ttsiouts has quit IRC | 09:20 | |
| *** ttsiouts has joined #openstack-placement | 09:21 | |
| *** edmondsw has joined #openstack-placement | 09:26 | |
| *** edmondsw has quit IRC | 09:30 | |
| openstackgerrit | Shuo Liu proposed openstack/nova master: fix keytone to keystone in releasenotes.po https://review.openstack.org/573640 | 09:30 |
| openstackgerrit | jiang wei proposed openstack/nova master: Add action initiator attribute to the instance info https://review.openstack.org/536243 | 09:33 |
| *** ttsiouts has quit IRC | 09:34 | |
| *** ttsiouts has joined #openstack-placement | 09:36 | |
| *** ttsiouts has quit IRC | 10:06 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova master: API: add support to abort queued live migration in microversion 2.63 https://review.openstack.org/573136 | 10:08 |
| openstackgerrit | Bhagyashri Shewale proposed openstack/nova master: libvirt: Don't report DISK_GB if sharing https://review.openstack.org/560459 | 10:27 |
| *** stephenfin is now known as finucannot | 10:44 | |
| *** ttsiouts has joined #openstack-placement | 11:00 | |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Test boot with more ports with bandwidth request https://review.openstack.org/573317 | 11:31 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add request_spec.RequestGroup versioned object https://review.openstack.org/568840 | 12:02 |
| *** nicolasbock has joined #openstack-placement | 12:13 | |
| *** ttsiouts has quit IRC | 12:32 | |
| jaypipes | giblet: +2 from me. | 12:36 |
| *** ttsiouts has joined #openstack-placement | 12:37 | |
| *** edleafe is now known as figleaf | 12:37 | |
| *** bhagyashris_ has joined #openstack-placement | 12:44 | |
| *** ttsiouts has quit IRC | 12:45 | |
| *** bhagyashri_s has quit IRC | 12:48 | |
| *** edmondsw has joined #openstack-placement | 13:02 | |
| *** ttsiouts has joined #openstack-placement | 13:04 | |
| giblet | jaypipes: I can send it to the gate, do we want to wait for efried to take a look too? | 13:06 |
| *** edmondsw has quit IRC | 13:06 | |
| jaypipes | giblet: I'm cool with you sending it, yes. | 13:14 |
| jaypipes | giblet: efried wrote a chunk of it so didn't want to +W it. | 13:14 |
| giblet | jaypipes: Ohh, OK. sending it through... | 13:14 |
| bhagyashris_ | jaypipes, giblet: Thanks for review :) | 13:18 |
| giblet | bhagyashris_: thanks for working on fixing the bug | 13:19 |
| bhagyashris_ | :) | 13:21 |
| jaypipes | bhagyashris_: no worries, thanks for your patience with us. | 13:21 |
| *** efried is now known as fried_rice | 13:21 | |
| jaypipes | ooooh. | 13:21 |
| *** jaypipes is now known as leakypipes | 13:21 | |
| bhagyashris_ | jaypipes: :) no issues | 13:28 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: libvirt: Start populating NUMACell.network_info field https://review.openstack.org/564441 | 13:30 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: objects: Add InstanceNUMANetworkInfo object https://review.openstack.org/564442 | 13:30 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Unchain _get_phynet_info from _get_port_vnic_info https://review.openstack.org/564443 | 13:30 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Add 'create_resource_requests' to network API https://review.openstack.org/564444 | 13:30 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Add '_get_network_tunnel_status' https://review.openstack.org/564445 | 13:30 |
| *** mriedem has joined #openstack-placement | 13:32 | |
| fried_rice | giblet, leakypipes, bhagyashris_: \ō/ | 13:35 |
| giblet | fried_rice: happy Friday! | 13:35 |
| leakypipes | fried_rice: is that a halo? | 13:35 |
| bhagyashris_ | \o/ | 13:36 |
| fried_rice | leakypipes: It only really looks like that with the double-arm 'yay' going on. It's my crew cut. | 13:38 |
| leakypipes | :) | 13:38 |
| -openstackstatus- NOTICE: A misapplied distro security package update caused many jobs to fail with a MERGER_FAILURE error between ~06:30-12:30 UTC; these can be safely rechecked now that the problem has been addressed | 13:47 | |
| *** edmondsw has joined #openstack-placement | 13:54 | |
| openstackgerrit | Eric Fried proposed openstack/nova master: fix tox python3 overrides https://review.openstack.org/572974 | 13:58 |
| PapaOurs | fried_rice: you know that I like diagrams, don't you ? :) | 13:59 |
| fried_rice | PapaOurs: I was just teasin, daddy bear. | 13:59 |
| * PapaOurs chilling | 14:00 | |
| PapaOurs | fried_rice: honestly, using segdiag is fine with me | 14:00 |
| PapaOurs | fried_rice: it's just that you wrote something I think that's just implementation stuff | 14:00 |
| fried_rice | PapaOurs: I agree it wouldn't have been sufficient on its own - but the necessary details are called out in the sections below, yes? Is there anything missing? | 14:00 |
| fried_rice | oh. Yeah, I like to write specs in such a way that anyone could go and implement them, even if the original authors all go down in a plane crash. I must have learned that in school. | 14:01 |
| PapaOurs | :) | 14:02 |
| fried_rice | PapaOurs: Seriously though, thinking through the impl raised some design questions it's better to address now. Like whether we need the list of affected providers. Which I still need to think through some more. | 14:04 |
| PapaOurs | fried_rice: sure, but the concern I have is that people could be -1 because of some implementation discussion, while they could miss the main reason why we provide this spec, ie. the new placement API resource | 14:05 |
| fried_rice | PapaOurs: Well, we definitely need the spec to include more than just the placement API piece. | 14:07 |
| fried_rice | PapaOurs: That's only part of the picture. | 14:07 |
| PapaOurs | fried_rice: sure, of course | 14:07 |
| PapaOurs | anyway, see this comment I made as a just nit | 14:08 |
| PapaOurs | fried_rice: if people discuss about the main one, fine with me | 14:08 |
| fried_rice | PapaOurs: Your point is taken, though. If we start to get into too much of a battle there, I can remove some impl details. | 14:08 |
| fried_rice | PapaOurs: Hoping it doesn't come to that. | 14:08 |
| PapaOurs | fried_rice: cool then | 14:08 |
| PapaOurs | fried_rice: yeah, hopefully too ! | 14:08 |
| fried_rice | PapaOurs: btw, unrelated subject, the DISK_GB/libvirt thing is in the gate. Did you want to have a look at that before it disappears? | 14:09 |
| PapaOurs | well, if it's in the gate.. :) | 14:09 |
| PapaOurs | I was more looking at specs | 14:09 |
| fried_rice | PapaOurs: Okay. I thought you were keeping an eye on that for your own purposes. But maybe you only cared about the prior one that implemented update_provider_tree. Anyway https://review.openstack.org/560459 if interested. | 14:10 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: placement: always create consumer records https://review.openstack.org/567678 | 14:12 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: add consumers generation field https://review.openstack.org/557958 | 14:12 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: placement: Allocation.consumer field https://review.openstack.org/565405 | 14:12 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: rework allocation handler _allocations_dict() https://review.openstack.org/565407 | 14:12 |
| openstackgerrit | Jay Pipes proposed openstack/nova master: Add a microversion for consumer generation support https://review.openstack.org/565604 | 14:12 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Unchain _get_phynet_info from _get_port_vnic_info https://review.openstack.org/564443 | 14:18 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Add 'create_resource_requests' to network API https://review.openstack.org/564444 | 14:18 |
| openstackgerrit | Stephen Finucane proposed openstack/nova master: network: Retrieve tunneled status in '_get_network_info' https://review.openstack.org/564445 | 14:18 |
| leakypipes | fried_rice: +2 on reshaper spec. I have a few nits but it appropriately captures the spirit of the discussion and the agreed-upon implementation. | 14:38 |
| fried_rice | leakypipes: Roger that, thanks for reviewing. I will be respinning fyi. | 14:39 |
| fried_rice | leakypipes: I would like to discuss the performance impact that edmondsw brought up, related to our needing to gather allocations on every periodic. | 14:39 |
| leakypipes | fried_rice: I don't see why we'd need to do that on every periodic run. | 14:40 |
| fried_rice | leakypipes: The obvious answer is to cache like we do the provider_tree - which is something you suggested in passing (adding allocations to ProviderTree) | 14:40 |
| leakypipes | fried_rice: in fact, pretty sure if we do that, Surya and CERN will run screaming. | 14:41 |
| fried_rice | leakypipes: Chicken/egg. Only the virt driver can tell us whether we're doing a migration, but unless we make it a two-step process (update_provider_tree() -> indicates migration needed -> discover and send allocations -> get altered allocations back).... | 14:42 |
| fried_rice | back in 10. | 14:43 |
| leakypipes | fried_rice: ack. that two-stage process *might* be worthwhile investigating, but could be done if performance woes ensue from this original approach. | 14:44 |
| fried_rice | leakypipes: Another option is only doing this thing on compute startup. Problem there is, if it fails, how to redrive? | 14:51 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add requested_resources field to RequestSpec https://review.openstack.org/567267 | 15:03 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Add bandwidth related standard resource classes https://review.openstack.org/570847 | 15:13 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Transfer port.resource_request to the scheduler https://review.openstack.org/567268 | 15:13 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Send resource allocations in the port binding https://review.openstack.org/569459 | 15:13 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Test boot with more ports with bandwidth request https://review.openstack.org/573317 | 15:13 |
| *** ttsiouts has quit IRC | 15:28 | |
| *** ttsiouts has joined #openstack-placement | 15:30 | |
| *** cdent has joined #openstack-placement | 15:40 | |
| *** cdent has quit IRC | 15:40 | |
| *** e0ne has quit IRC | 15:41 | |
| mriedem | fried_rice: we could likely set some kind of db flag to indicate we have or haven't done the migratoin check for a given instance | 15:52 |
| mriedem | set a flag when starting, and then set again when done - even if we didn't need to migrate anything | 15:53 |
| mriedem | could be a system_metadata flag per instance. | 15:53 |
| mriedem | that's a lot of db updates if you're restarting 1000 computes at once | 15:53 |
| fried_rice | mriedem: Mm. And sounds pretty complicated. | 15:54 |
| fried_rice | mriedem: When would you set the flag? | 15:54 |
| mriedem | whenever you start checking the migrations - on compute start, but thread it out | 15:54 |
| mriedem | for the ironic flavor migration and my patch to migrate old bdm attachment records, we thread that out | 15:54 |
| mriedem | and they are idempotent, i.e. they know when they don't need to do anything | 15:54 |
| mriedem | fried_rice: https://review.openstack.org/#/c/549130/ | 15:55 |
| mriedem | might only need to set a flag once - get all instances on a given host that don't have the system_metadata flag set (nrp-migrated=True?) | 15:55 |
| mriedem | so for instances that don't have the flag on that host on startup, process them and then set the flag as you process each one | 15:56 |
| mriedem | then it's a one time thing, not during every periodic | 15:56 |
| mriedem | dansmith: ^ | 15:56 |
| fried_rice | mriedem: So on compute start, set the flag; | 15:57 |
| fried_rice | then in update, if the flag is set, collect allocations and send them to the virt driver | 15:57 |
| fried_rice | then in update_provider_tree, only make migration-y changes if allocations is not None (and otherwise, what, continue to model the old way?) | 15:57 |
| fried_rice | then in update_from_provider_tree, check whether allocations is not None and if so, do the /migrator thing | 15:57 |
| fried_rice | if that succeeds, unset the flag | 15:57 |
| fried_rice | otherwise leave it set for the next periodic | 15:57 |
| dansmith | we can't thread it out I think, | 15:57 |
| dansmith | because we have to convert them all at the same time as updating our inventory right? | 15:58 |
| dansmith | we need to do this before we let the rest of the startup procedure complete | 15:58 |
| fried_rice | yeah, it wouldn't be a per-instance flag. | 15:58 |
| mriedem | ok so then that's update_available_resources as fried_rice said above | 15:58 |
| dansmith | not sure we need a marker, because if we don't complete, we're just going to re-generate the new state of the world | 15:58 |
| dansmith | mriedem: right, but one-shot head of time, and if it doesn't complete we bail | 15:58 |
| mriedem | point is, i'd like to only do this once per startup, and not refresh the wold every periodic to know if we've already done this | 15:59 |
| fried_rice | dansmith: The issue is that we're trying to avoid needing to collect allocations every periodic. | 15:59 |
| dansmith | yeah, has to be that way I think | 15:59 |
| fried_rice | because that's expensive. | 15:59 |
| dansmith | fried_rice: agreed, which is why it should be one specific operation we call at startup | 15:59 |
| mriedem | that's why i suggested a flag | 15:59 |
| dansmith | mriedem: but per instance is not a thing | 15:59 |
| dansmith | you have to convert them all at once | 15:59 |
| fried_rice | dansmith: agree | 15:59 |
| mriedem | but how do you know if you don't need to run it on every other restart of the servioce? | 15:59 |
| fried_rice | ^ | 16:00 |
| mriedem | if this is a one time thing, we shouldn't check this on every startup if we can help it | 16:00 |
| dansmith | checking to see if the current allocations look like what we expect at service start is less expensive than pulling flags for all our instances and then doing the migration right? | 16:00 |
| mriedem | if it's not per-instance, then it's per-host i guess | 16:01 |
| mriedem | but we don't have host metadata to store random stuff | 16:01 |
| dansmith | or, we just check our own inventory.. if we have migrated our inventory, our instances had to go with it | 16:01 |
| fried_rice | The other thing leakypipes and I discussed above was having a separate ComputeDriver method that asks provider_migrate_needed(provider_tree). We only pull allocations to send to update_provider_tree if that answers True. | 16:01 |
| mriedem | if pulling all of the current allocations isn't a big deal, then meh | 16:01 |
| mriedem | i'm pretty sure we already do pull allocations during every update_available_resources periodic | 16:01 |
| dansmith | we have to update our inventory and allocations atomically, so if our inventory looks good (which is inexpensive to pull and we have to do it anyway) then we're good | 16:01 |
| dansmith | else we migrate | 16:01 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova master: Fix unbound local when saving an unchanged RequestSpec https://review.openstack.org/573768 | 16:02 |
| fried_rice | If we don't think the expense of pulling allocations is an issue, then we're good as written. Do you think this is something we can just defer until/unless it's a problem? | 16:02 |
| dansmith | defer what until when? | 16:03 |
| mriedem | this is what i was thinking of, which is different https://github.com/openstack/nova/blob/master/nova/compute/resource_tracker.py#L1253 | 16:03 |
| fried_rice | superdansmith: defer <figuring out a way to avoid pulling the allocations every time> until <someone (CERN) complains that it's killing them> | 16:04 |
| fried_rice | s/every time/every periodic/ | 16:04 |
| dansmith | defer not effectively doing the migration on every periodic? | 16:04 |
| mriedem | i'm thinking about that periodic that CERN had to disable | 16:05 |
| mriedem | b/c it killed them on startup | 16:05 |
| dansmith | fried_rice: mriedem: can we hangout this real quick to make sure we're on the same page/ | 16:05 |
| mriedem | wasn't that pulling allocatoins? | 16:05 |
| dansmith | yeah, CERN definitely needs this to be right | 16:05 |
| dansmith | mriedem: yes | 16:05 |
| dansmith | associations actually, but.. | 16:05 |
| mriedem | i have to leave here in about 5 minutes | 16:05 |
| dansmith | okay, I'm trying to avoid a conflict too | 16:06 |
| mriedem | i could after i'm back | 16:06 |
| dansmith | oh okay | 16:06 |
| mriedem | my afternoon is openish | 16:06 |
| dansmith | mine isn't, but.. | 16:06 |
| dansmith | but agree, this has to be relatively low impact from the start | 16:06 |
| dansmith | they had to actively remove the association sync to function at all, this'll definitely break them | 16:06 |
| dansmith | (if we don't do it well) | 16:07 |
| *** cdent has joined #openstack-placement | 16:08 | |
| cdent | fried_rice: I'm still planning to look at the shaping spec today, but I'm running a bit behind schedule: slept for 19 hours | 16:09 |
| fried_rice | cdent: cool mon. | 16:09 |
| cdent | combo of jet lag and the remainder of conference-cold | 16:09 |
| fried_rice | dansmith: I can (sigh) start an etherpad with some options... | 16:10 |
| *** ttsiouts has quit IRC | 16:11 | |
| * cdent adds etherpaddingn to fried_rice's cv | 16:14 | |
| fried_rice | vay | 16:14 |
| *** ttsiouts has joined #openstack-placement | 16:18 | |
| *** ttsiouts has quit IRC | 16:22 | |
| fried_rice | dansmith, mriedem, leakypipes, cdent: https://etherpad.openstack.org/p/reshape-provider-tree-performance | 16:28 |
| fried_rice | edmondsw: ^ | 16:28 |
| fried_rice | bbl | 16:29 |
| *** fried_rice is now known as fried_rolls | 16:29 | |
| leakypipes | sweet.. "old jaypipes" is gone now. | 16:30 |
| cdent | finally died | 16:31 |
| cdent | so old | 16:31 |
| leakypipes | yep | 16:31 |
| leakypipes | the ghost finally moved on. | 16:31 |
| fried_rolls | for now. In previous pads it took him a while to show up. | 16:33 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | 16:46 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Isolate placement database config https://review.openstack.org/541435 | 16:46 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Ensure that os-traits sync is attempted only at start of process https://review.openstack.org/553857 | 16:46 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Add PLACEMENT_DB_ENABLED=True to the nova-next job https://review.openstack.org/564067 | 17:02 |
| -openstackstatus- NOTICE: The Zuul scheduler was offline briefly to clean up from debugging a nodepool issue, so changes uploaded or approved between 16:50 and 17:15 UTC may need to be rechecked or reapproved (all already queued changes are in the process of being reenqueued now) | 17:24 | |
| *** edleafe has joined #openstack-placement | 17:33 | |
| *** figleaf has quit IRC | 17:35 | |
| *** edleafe is now known as figleaf | 18:13 | |
| openstackgerrit | Chris Dent proposed openstack/nova stable/queens: Ensure resource class cache when listing usages https://review.openstack.org/573811 | 18:36 |
| cdent | mriedem: ^ no conflicts after all, just gerrit not liking file moves | 18:39 |
| mriedem | ah | 18:39 |
| *** fried_rolls is now known as fried_rice | 18:43 | |
| *** bhagyashri_s has joined #openstack-placement | 20:00 | |
| *** bhagyashris_ has quit IRC | 20:03 | |
| *** bhagyashris_ has joined #openstack-placement | 20:11 | |
| *** bhagyashri_s has quit IRC | 20:15 | |
| *** e0ne has joined #openstack-placement | 20:27 | |
| fried_rice | mriedem, cdent, leakypipes, dansmith, edmondsw: Only Chris expressed a preference on https://etherpad.openstack.org/p/reshape-provider-tree-performance | 20:39 |
| fried_rice | Shall I move forward with "conditioned by virt driver"? | 20:39 |
| fried_rice | (NB: I proposed a tweak to the code flow that doesn't entail an extra method) | 20:39 |
| leakypipes | fried_rice: I haven't looked at the etherpad much yet, sorry :( | 20:43 |
| leakypipes | fried_rice: can we discuss more on the etherpad and on Monday? | 20:43 |
| fried_rice | sure. I'm leaving a note on the spec that will hopefully make sure we don't forget the issue. | 20:43 |
| leakypipes | k, thx | 20:44 |
| fried_rice | figleaf: Can you have a look at https://review.openstack.org/#/c/573113/ when you get a chance please? | 21:22 |
| fried_rice | figleaf: Basically need you to be my second +2 if you can follow the sql | 21:23 |
| *** e0ne has quit IRC | 22:00 | |
| *** nicolasbock has quit IRC | 22:06 | |
| *** fried_rice is now known as efried | 22:35 | |
| figleaf | efried: sorry, already started the weekend. Will look on Monday. | 23:15 |
| *** figleaf is now known as edleafe | 23:15 | |
| openstackgerrit | Chris Dent proposed openstack/nova master: Optional separate database for placement API https://review.openstack.org/362766 | 23:30 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Isolate placement database config https://review.openstack.org/541435 | 23:30 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Ensure that os-traits sync is attempted only at start of process https://review.openstack.org/553857 | 23:30 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Add PLACEMENT_DB_ENABLED=True to the nova-next job https://review.openstack.org/564067 | 23:31 |
| openstackgerrit | Chris Dent proposed openstack/nova master: Extract part of PlacementFixture to placement https://review.openstack.org/568359 | 23:56 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!