Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-07-09
14:59:22 TheJulia mgagne: circling back to the discussion from last week, it has been a really long time since I've dug into that level of nova, but I suspect the reason for the additional call is to attempt to ensure that we have the correct latest data as opposed to an initial or possibly empty copy of network_info...
15:01:58 mriedem melwitt: you'll want to check the scheduler / placement meeting logs from this morning specifically about the state of the 'reshaper' series and the need for some oversight on that, especially given the short timeline we have to land stuff for that before feature freeze
15:01:59 openstackgerrit Mike Lowe proposed openstack/nova master: Option needed for ceph rbd erasure coding support https://review.openstack.org/581055
15:13:05 openstackgerrit Chris Dent proposed openstack/nova master: Use nova.db.api directly https://review.openstack.org/543262
15:14:18 mgagne TheJulia: it doesn't make sense to me because a new instance of instance_metadata.InstanceMetadata is created (so no stalled data here) and network_metadata is forcefully passed to the constructor. You could take the object, access network_metadata attribute and munge it instead just like before without calling netutils.get_network_metadata directly in _get_network_metadata.
15:16:08 mgagne but network_metadata isn't an object which isn't cool neither =)
15:16:25 mgagne (just a side note)
15:18:16 mgagne but I think it's fine anyway, I managed to patch network_metadata in my change
15:21:52 openstackgerrit Mathieu Gagné proposed openstack/nova master: Add support for multiple fixed-ips in metadata https://review.openstack.org/580742
15:37:51 dansmith mriedem: I know this is not very helpful, but look at my comments here on yikun's thing: https://review.openstack.org/#/c/573628
15:38:06 sahid stephenfin, jaypipes, i have some interesting fix https://review.openstack.org/#/c/532168/ if you can have a look
15:38:29 sahid they are related to how we compute hugepages, we have some issue that could make th VM running OOMkilled
15:38:49 dansmith mriedem: I don't want to derail him further for no reason, so if you think it's worth the bloat to just move forward I guess I can go with that
15:40:50 dansmith in fact, just dropping that patch and tweaking a few things in the one that modifies group would probably be the only change
15:40:56 dansmith so I volunteer to do that for him if you want
16:02:18 mriedem dansmith: replied
16:03:25 mriedem dansmith: would you leave this? https://review.openstack.org/#/c/579113/
16:03:25 TheJulia mgagne: That seems reasonable to me, I think nova in the past did async information generation, but I'm not sure that is the case anymore
16:05:34 dansmith mriedem: yeah I'm cool with keeping the policies->policy deprecation stuff
16:11:01 openstackgerrit Merged openstack/nova stable/pike: Block deleting compute services which are hosting instances https://review.openstack.org/580496
16:19:22 mriedem ok, i'm fine with dropping the separate object if you want, it's kind of pointless when he couldn't re-use InstanceGroupPolicy.create() and it's just a data bag
16:19:58 mriedem dansmith: ^ - sorry, had to jump into a call
16:27:26 openstackgerrit Margarita Shakhova proposed openstack/nova master: Do not skip migrations in _destroy_evacuated_instances() https://review.openstack.org/563623
16:31:49 openstackgerrit Merged openstack/nova master: move lookup of provider from _new_allocations() https://review.openstack.org/579920
16:43:55 openstackgerrit Stephen Finucane proposed openstack/nova master: objects: Add RequestSpec.network_metadata https://review.openstack.org/564442
16:43:56 openstackgerrit Stephen Finucane proposed openstack/nova master: scheduler: Start utilizing RequestSpec.network_metadata https://review.openstack.org/564452
16:43:57 openstackgerrit Stephen Finucane proposed openstack/nova master: conf: Add '[neutron] physnets' and related options https://review.openstack.org/564440
16:43:58 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Start populating NUMACell.network_metadata field https://review.openstack.org/564441
16:44:36 openstackgerrit Jay Pipes proposed openstack/nova master: delete consumers which no longer have allocations https://review.openstack.org/581086
16:51:01 openstackgerrit Jay Pipes proposed openstack/nova master: delete consumers which no longer have allocations https://review.openstack.org/581086
16:59:30 openstackgerrit Jay Pipes proposed openstack/nova master: placement: delete auto-created consumers on fail https://review.openstack.org/579921
17:23:42 openstackgerrit Matt Riedemann proposed openstack/nova master: Heal allocations with incomplete consumer information https://review.openstack.org/574488
17:23:43 openstackgerrit Matt Riedemann proposed openstack/nova master: Refactor _heal_instances_in_cell https://review.openstack.org/577896
17:23:44 openstackgerrit Matt Riedemann proposed openstack/nova master: Use consumer generation in _heal_allocations_for_instance https://review.openstack.org/577905
17:26:30 openstackgerrit Merged openstack/python-novaclient master: Rename --endpoint-override to --os-endpoint-override https://review.openstack.org/580991
18:08:38 openstackgerrit Jay Pipes proposed openstack/nova master: delete consumers which no longer have allocations https://review.openstack.org/581086
18:11:13 mriedem dansmith: melwitt: Kevin_Zheng: gibi: i'm +2 on the 'handling a down cell' spec now https://review.openstack.org/#/c/557369/
18:15:09 dansmith mriedem: I'm closing in on finishing surgery on yikun's patches
18:15:14 dansmith I think this'll be a lot nicer
18:15:28 dansmith drops a whole patch and removes a lot of confusing stuff like group.policy.policy
18:16:20 openstackgerrit Chris Dent proposed openstack/nova master: make incomplete_consumer_project_id a valid UUID https://review.openstack.org/580358
18:19:26 openstackgerrit Jay Pipes proposed openstack/nova master: delete consumers which no longer have allocations https://review.openstack.org/581086
18:19:53 mriedem1 dansmith: ok, just don't leave a glove in the torso when you sow it up
18:22:43 dansmith sew?
18:22:52 jaypipes :)
18:22:56 mriedem female pig
18:22:56 dansmith not planning to "sow" anything in this patch set
18:23:02 jaypipes always be pedantic about surgery.
18:23:38 mriedem don't crush any piglets during feeding on this series
18:23:43 mriedem is what i meant
18:42:20 melwitt mriedem: acks
18:43:27 efried I thought it was like "reap what you sow" kind of sow.
18:45:41 mriedem don't fear the reaper
19:09:51 openstackgerrit Dan Smith proposed openstack/nova master: Add rules column to instance_group_policy table. https://review.openstack.org/560832
19:09:52 openstackgerrit Dan Smith proposed openstack/nova master: Refactor policies to policy in InstanceGroup DB model https://review.openstack.org/579113
19:09:53 openstackgerrit Dan Smith proposed openstack/nova master: Add policy to InstanceGroup object https://review.openstack.org/563375
19:31:27 openstackgerrit Merged openstack/nova stable/pike: Delete allocations from API if nova-compute is down https://review.openstack.org/580498
19:46:04 tssurya mriedem: thanks so much for the spec review, I will address the nits in some time (after dinner) and ping you back.
19:51:47 openstackgerrit Merged openstack/nova stable/pike: Cleanup RP and HM records while deleting a compute service. https://review.openstack.org/580499
20:10:39 openstackgerrit Merged openstack/nova-specs master: Handling a down cell https://review.openstack.org/557369
20:35:03 openstackgerrit Jay Pipes proposed openstack/nova master: delete consumers which no longer have allocations https://review.openstack.org/581086
20:42:33 openstackgerrit Dan Smith proposed openstack/nova master: Add policy to InstanceGroup object https://review.openstack.org/563375
20:57:43 openstackgerrit Matt Riedemann proposed openstack/nova master: Add nova-manage placement sync_aggregates https://review.openstack.org/575912
21:08:06 openstackgerrit Eric Fried proposed openstack/nova master: Test for unsanitized consumer UUID https://review.openstack.org/581137
21:16:32 mriedem dansmith: should i be rebasing the rest of that policy series then?
21:16:55 dansmith mriedem: I was going to get to it, but if you want to yeah
21:17:14 mriedem go ahead if you've got it local
21:17:22 mriedem i've got a hard stop in about 45 minutes
21:17:48 dansmith I haven't done any of it yet
21:18:03 dansmith I gotta run an errand and then will be back for a few
21:18:08 dansmith so I can try to do it then,
21:18:28 dansmith although the top of the series was also needing a rebase so I figured it wasn't too bad to just break it a little further down
21:18:57 mriedem i'll do what i can quick
21:19:04 mriedem he's gonna do the api bit tomorrow
21:19:31 dansmith okay
21:28:31 openstackgerrit Jay Pipes proposed openstack/nova master: update project/user for consumer in allocation https://review.openstack.org/581139
21:29:47 mriedem dansmith: your change looks good, just one question in https://review.openstack.org/#/c/563375/34/nova/objects/instance_group.py@283
22:10:42 openstackgerrit Matt Riedemann proposed openstack/nova master: Add policy field to ServerGroup notification object https://review.openstack.org/563401
22:10:43 openstackgerrit Matt Riedemann proposed openstack/nova master: Change the ServerGroupAntiAffinityFilter to adapt to new policy https://review.openstack.org/571166
22:10:44 openstackgerrit Matt Riedemann proposed openstack/nova master: Adapt _validate_instance_group_policy to new policy model https://review.openstack.org/571465
22:22:54 openstackgerrit Matt Riedemann proposed openstack/nova master: Heal allocations with incomplete consumer information https://review.openstack.org/574488
22:22:55 openstackgerrit Matt Riedemann proposed openstack/nova master: Refactor _heal_instances_in_cell https://review.openstack.org/577896
22:22:56 openstackgerrit Matt Riedemann proposed openstack/nova master: Use consumer generation in _heal_allocations_for_instance https://review.openstack.org/577905
22:30:48 openstackgerrit Chris Dent proposed openstack/nova master: Update some placement docs to reflect modern times https://review.openstack.org/581151
23:02:50 openstackgerrit Jay Pipes proposed openstack/nova master: delete consumers which no longer have allocations https://review.openstack.org/581086
23:02:51 openstackgerrit Jay Pipes proposed openstack/nova master: update project/user for consumer in allocation https://review.openstack.org/581139
23:04:15 openstackgerrit Eric Fried proposed openstack/nova master: Test for unsanitized consumer UUID https://review.openstack.org/581137
23:06:00 openstackgerrit Jay Pipes proposed openstack/nova master: placement: delete auto-created consumers on fail https://review.openstack.org/579921
23:31:44 openstackgerrit Eric Fried proposed openstack/nova master: return 404 when no consumer found in allocs https://review.openstack.org/579163
23:44:01 openstackgerrit Eric Fried proposed openstack/nova master: Update some placement docs to reflect modern times https://review.openstack.org/581151
#openstack-nova - 2018-07-10
01:02:58 takashin oomichi: Would you review https://review.openstack.org/#/c/529520/ again? It got one +2.
01:04:11 openstackgerrit Mike Lowe proposed openstack/nova master: Option needed for ceph rbd erasure coding support https://review.openstack.org/581055
01:06:15 oomichi takashin: thanks for doing this, +A
01:06:42 takashin oomichi: Thanks!
01:34:01 yikun mriedem, dansmith, Thanks for your help on group policy patch~and I will address left comments today.
02:08:29 openstackgerrit huanhongda proposed openstack/nova stable/queens: [Stable Only] Remove soft-deleted instances from quota_usages https://review.openstack.org/579093
04:17:44 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Add policy to InstanceGroup object https://review.openstack.org/563375
04:17:45 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Add policy field to ServerGroup notification object https://review.openstack.org/563401
04:17:46 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Change the ServerGroupAntiAffinityFilter to adapt to new policy https://review.openstack.org/571166
04:17:47 openstackgerrit Yikun Jiang (Kero) proposed openstack/nova master: Adapt _validate_instance_group_policy to new policy model https://review.openstack.org/571465

Earlier   Later