Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-09-21
18:05:06 sdague mtreinish: http://logs.openstack.org/74/501874/2/gate/gate-nova-python35/0dcede7/console.html#_2017-09-21_18_02_47_155101 - is that an stestr issue?
18:05:21 sdague worker 7 just hung, and eventually that was a fail
18:11:51 mriedem stvnoyes: the bdm attachment ids stuff in the migrate_data object is going to get weird when we have multiattach
18:12:04 mriedem since in your change it's a 1:1 between volume id and attachment id
18:12:06 mriedem although,
18:12:23 mriedem i guess you wouldn't have a volume attached multiple times to the same instance, unless we're migrating it
18:12:27 mriedem so maybe i'm overthinking things
18:15:54 openstackgerrit Ildiko Vancsa proposed openstack/nova-specs master: Add multiattach support to Nova https://review.openstack.org/499777
18:29:05 mriedem and just when i was about to -1 this change
18:38:14 mriedem dansmith: so for your 2nd change in the series to honor the global limit,
18:38:19 mriedem i was posting this comment, but can't
18:38:33 mriedem "So today we iterate the cells in order and just pull $limit instances and if we keep going to another cell we subtract how many we pulled out of the previous cell for a new limit on the next cell. Now we're going to be pulling $limit instances out of each cell concurrently, merge sorting them all and then enforcing the limit below. That is more traffic, right? Maybe it's negligible.",
18:38:34 mriedem "I was wondering if we wanted to divide the limit by the number of cells and then pull that many instances from each. So if we had 2 cells and the limit was 1000 (which is the default in the API), we'd just pull 500 instances from each cell and merge sort them."
18:44:03 dansmith mriedem: we can't do that partitioning of the limit until/unless we support going back to the cells for more instances
18:44:24 dansmith mriedem: otherwise we'd hit the end of one limit batch, assume there are no more in that cell that sort after anything in other cells and then get out of sync
18:44:39 dansmith mriedem: we basically have to have at least $limit results from each cell to prevent that
18:44:55 dansmith or emulate it by refilling our batch when we run out before we continue the merge
18:45:29 dansmith that distinction is the thing that made us think we couldn't do this easily initially and why we went down the path of punting on the problem
18:45:54 dansmith also, yes it may be more traffic at times, but something else I was discussing with people was:
18:46:33 dansmith if each tenant has N instances now, and then later those N instances are spread across a bunch of cells, this list operation is still pulling the same number of rows from the DB in total, but smaller pieces
18:46:51 dansmith the db driver doesn't even fetch partial results today or with this, it fetches everything in the result
18:47:27 dansmith does that make sense at all?
18:50:14 mriedem yeah, hadn't considered the case that there are like 200 instances in cell1 and 600 insteances in cell2, if we split the limit to 500 each, we'd end up with 700 instead of 800
18:51:10 dansmith well, there's that, but we'd also be lossy in that we'd ignore things from cell1 that sorted before some in cell2, but just didn't make the limit cut
18:51:34 dansmith and if those were ever after max_limit/N cells, you'd never be able to limit them ever
18:51:56 dansmith er, never be able to list them
19:15:58 openstackgerrit Eric Fried proposed openstack/nova master: Get auth from context for glance endpoint https://review.openstack.org/490057
19:16:30 efried sdague mordred ^ removed conf auth for glance
19:17:09 efried dangit, tried to re-author, but failed. Now the co-authored line in the commit message doesn't make sense. Fixing.
19:17:37 openstackgerrit Eric Fried proposed openstack/nova master: Get auth from context for glance endpoint https://review.openstack.org/490057
19:19:21 mordred efried: lgtm
19:20:00 efried cool
19:20:46 mriedem stvnoyes: ok comments inline https://review.openstack.org/#/c/463987/
19:21:34 stvnoyes mriedem: thanks i'll take a look
19:25:37 openstackgerrit Sean Dague proposed openstack/nova master: Support qemu >= 2.10 https://review.openstack.org/505673
19:37:11 openstackgerrit Merged openstack/nova master: Transform servergroup.create notification https://review.openstack.org/454023
19:41:03 openstackgerrit Merged openstack/nova master: Transform servergroup.delete notification https://review.openstack.org/483902
19:41:30 openstackgerrit Merged openstack/nova master: Move libvirts dmcrypt support to privsep. https://review.openstack.org/490737
19:42:03 openstackgerrit Merged openstack/nova master: api-ref: fix default sort key when listing servers https://review.openstack.org/506227
20:06:43 mriedem dansmith: replied in https://review.openstack.org/#/c/504984/8/nova/tests/functional/compute/test_instance_list.py - not sure what i'm missing there why the slice is needed
20:07:28 dansmith heh
20:07:41 dansmith mriedem: that should be slicing self.instances not uuids
20:08:11 dansmith I'm in the middle of fixing migration uuid stuff in my tree so I can't I update it right now
20:08:15 dansmith but I will when I'm done
20:08:29 mriedem ok yeah that makes more sense :)
20:08:55 dansmith optionally you can locally test and then I'll slap something on the end, or you can push a fix
20:09:26 mriedem i had to rebase the entire thing, so we can slap it on the end
20:09:37 dansmith ack
20:10:00 dansmith I would be happy if we merge everything at least up to the pagination one this week
20:10:12 mriedem i'll try to push it to the limit
20:10:19 dansmith +1
20:13:10 mriedem does that mean all these other tests doing "uuids = [inst['uuid'] for inst in insts]" should be using self.instances?
20:15:01 dansmith no,
20:15:32 dansmith just that one was intending to compare the list to the snipped list of known instances and sliced the wrong list
20:15:33 dansmith I mean
20:15:41 mriedem alright left a diff in the comments
20:15:47 dansmith point me to something, I could have made the same mistake I guess, but..
20:15:56 dansmith cool, in the limit one?
20:16:00 mriedem yeah
20:16:35 dansmith cool thanks
20:16:36 dansmith also
20:17:01 dansmith our migration and rt stuff is hell on earth
20:17:02 dansmith that is all
20:17:40 mriedem i thought branson, mo was hell on earth?
20:17:51 dansmith we have a new contender
20:20:23 dansmith mriedem: that one was checking that the list of uuids was the same as returned and as sorted
20:30:23 mriedem in test_get_sorted ?
20:30:52 mriedem ah ok, nvm then
20:31:37 mriedem oh yeah, for the migration uuid stuff, we still have a couple of outstanding bugs related to cleaning up allocations during a failed move
20:32:06 dansmith that you'll want to get in before this so you can backport them I guess?
20:32:10 mriedem yeah
20:32:17 mriedem been meaning to at least write the recreate tests
20:32:30 mriedem maybe i can do that at 10pm tonight after all others are asleep
20:32:34 openstackgerrit Chris Dent proposed openstack/nova master: Move project_id and user_id to Allocation object https://review.openstack.org/500410
20:32:35 openstackgerrit Chris Dent proposed openstack/nova master: [placement] Allow _set_allocations to delete allocations https://review.openstack.org/501051
20:32:35 openstackgerrit Chris Dent proposed openstack/nova master: WIP: [placement] POST /allocations to set allocations for >1 consumers https://review.openstack.org/500073
20:42:15 mriedem shit 1 week for forum topics?
20:42:56 cdent barely digested the ptg, too soon!
20:43:14 mriedem i threw up half of it
20:43:21 mriedem couldn't even digest it all
20:43:33 openstackgerrit Eric Berglund proposed openstack/nova master: Add PowerVM hypervisor configuration doc https://review.openstack.org/505665
20:58:55 mriedem ha http://forumtopics.openstack.org/
20:59:06 mriedem so apparently this has been getting communicated all month, i've just missed it
20:59:11 mriedem not sure why i'd be distracted...
20:59:21 mriedem but so has everyone else
20:59:59 cdent huh
21:00:37 cdent first I started hearing about it was last week
21:02:48 mriedem god man idk, we just had the ptg and i was already annoyed with how much stuff we talked about which has no chance of making queens
21:04:52 melwitt I feel that way too but try to remind myself that ppl just wanted to make sure placement would be able to model the stuff they want to do whenever nested RPs is ready
21:05:18 mriedem and then how to use placement to make your toast
21:05:23 melwitt I think that could have been accomplished with fewer hours than were spent, but yeah
21:05:28 mriedem with the "grape jelly" trait
21:05:50 melwitt heh
21:06:44 dansmith we had already been over a lot of that stuff before
21:06:49 cdent I think it’s a bit unfair to expect people not to want to talk when they only have a chance to really talk maybe 2-4 times a year
21:06:52 dansmith I was fairly annoyed, but I'm sure nobody noticed
21:06:55 mriedem the forum is now supposed to be the place to get the ops and users in the same room, so i kind of just want to get topic submissions from those groups
21:07:38 mriedem i have accepted that everytime we get together we're going to talk about stuff that we aren't going to do in that release
21:08:40 mriedem just hard to juggle that plus knowing we have bugs to fix in pike yet, and things to get done in queens, without thinking about all the future things, it's overwhelming
21:09:03 cdent it is indeed, but it is that overwhelmedness that makes people want to talk

Earlier   Later