Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-10-18
18:55:48 jaypipes efried: just rebases or anything functional?
18:56:13 jaypipes efried: I'm gonna pull the top patch (385693) and run tests locally.
18:56:13 efried jaypipes Almost entirely rebases, fixing nits, a couple of bugs. I don't think there was anything functional - lemme skim real quick...
18:56:19 jaypipes kk
18:56:23 openstackgerrit Chris Dent proposed openstack/nova master: [placement] manage cache headers for resource classes https://review.openstack.org/513161
18:56:33 edleafe jaypipes: I know you got a ton to catch up with, but the alternate hosts series is ready for review, starting with https://review.openstack.org/#/c/486215/
18:56:49 jaypipes edleafe: rock on, will do that as soon as I get those tests running.
18:56:51 efried jaypipes Okay, should I assume you're taking the reins back at this point, or do you want to do some pair dev on that series?
18:57:14 jaypipes efried: oh, I can take it over again, no prob.
18:57:17 efried jaypipes I'm totally prepared to fix those test cases if you have more important (or jaypipes-needin) things to do.
18:57:37 jaypipes efried: nope, I'm good to take it back
18:57:52 jaypipes appreciate you, edleafe and dansmith pushing on this while I was off
18:59:33 efried jaypipes At some point you should give https://review.openstack.org/#/c/510244/ a once-over, make sure it gels with what you conceived at the PTG.
19:00:05 jaypipes efried: yep, it's near the top of my queue.
19:00:09 efried coo
19:02:38 efried jaypipes The only changes of any substance were: 1) https://review.openstack.org/#/c/377138/50..51/nova/objects/resource_provider.py -- please double-check me to make sure I referenced the exception from the right spot
19:03:13 efried jaypipes 2) https://review.openstack.org/#/c/377215/48/nova/objects/resource_provider.py@1275 moved no-op reformatting stuff to a new patch
19:04:36 efried jaypipes That's it. The rest were rebases & spelling.
19:05:18 jaypipes efried: coolio. thx.
19:05:38 mriedem dansmith: this newton backport of artom's looks good to me now https://review.openstack.org/#/c/471356/ - it's kind of a frankenpatch with squashes but does the job
19:05:48 mriedem would be the last one for newton before we release and tag for eol
19:07:00 efried jaypipes And as soon as you come up for air, would like your input on the possibility of ripping out some of the existing aggregate code until we can figure out how tf it's ever going to work with traits & nested & numbered.
19:08:03 jaypipes efried: well, on that, if it isn't currently affecting anything (other than read queries which cdent has patches up to address perf issues), I'd just as soon leave it be for now. We have lots of other things to get done by SYD
19:08:20 dansmith yeah
19:08:22 dansmith me too
19:08:28 efried jaypipes That's just it: it very well may be affecting "things".
19:10:14 jaypipes efried: well, I'm perfectly happy to see your thoughts on it. just not a huge priority unless you can show me it's causing bugs.
19:10:33 jaypipes efried: besides the general "tech debt bugs" which I certainly admit there are many.
19:10:47 jaypipes ok, lemme read this granular spec :)
19:13:28 mriedem dansmith: john put up a patch for the ironic flavor migratoin allocatoins thing https://review.openstack.org/#/c/513085/2
19:13:40 mriedem seems ok to me
19:14:24 dansmith mriedem: yeah, sorry, I'm just behind
19:14:32 mriedem np, i just got back from running errands
19:15:13 mriedem your context targeting thing in pike is also happy for ci now https://review.openstack.org/#/c/512456/
19:15:28 mriedem sdague: ^ we need to get that released soon
19:16:50 dansmith okay I can hit that one quick
19:16:54 dansmith because I know it's perfect
19:17:00 mriedem :)
19:21:09 sdague mriedem: yeh, that seems fine. I wasn't originally sure that the sc reinflates over that, but it looks like it does
19:26:21 efried cdent What was the thing where you could modify an allocations/{consumer_uuid} - is that gonna be via PUT or POST?
19:26:24 melwitt mriedem: have you seen live migration job fails like this before? raises InstanceNotFound because of VIR_ERR_NO_DOMAIN? https://github.com/openstack/nova/blob/stable/ocata/nova/virt/libvirt/host.py#L553
19:26:42 cdent efried: that’s always PUT, and it always replaces
19:27:04 cdent the post version is in progress at: https://review.openstack.org/#/c/500073/
19:27:15 efried cdent Okay. I'm looking at your schema here https://review.openstack.org/#/c/510625/7/nova/api/openstack/placement/handlers/allocation.py and I think it's a little strict. Will see if you "fixed" it via ^.
19:27:31 melwitt mriedem: the "target build notification" patch for ocata failed the live migration job on that and I was wondering if it's just something to recheck or if it's something more
19:28:08 cdent efried: strict how? (entirely possible my jsonschema fu doesn’t allow for handstands)
19:28:56 efried cdent I think you might have done it, but still pondering. Basically by saying minProperties=1 in a couple places, you would be disallowing resources: {} which was (I thought) how you would say you're deleting those pieces of the allocation.
19:29:12 efried Though I guess omitting that chunk entirely would also work, since it's replacing.
19:29:17 efried So maybe moot.
19:29:36 cdent minProperties is disallowing anything other than the resource provider keyed object
19:30:09 cdent which is required on PUT, but will not be required on POST
19:30:36 mriedem melwitt: was just looking at that one
19:30:58 efried cdent So what's the POST supposed to do? Oh, allow you to modify allocations for multiple consumers at once?
19:31:00 mriedem melwitt: it's this http://logs.openstack.org/25/511325/1/check/gate-tempest-dsvm-multinode-live-migration-ubuntu-xenial/5130700/logs/libvirt/qemu/instance-00000004.txt.gz
19:31:01 mriedem known issue
19:31:06 mriedem qemu-system-x86_64: /build/qemu-GGGtkw/qemu-2.8+dfsg/block/io.c:1514: bdrv_co_pwritev: Assertion `!(bs->open_flags & BDRV_O_INACTIVE)' failed.
19:31:20 melwitt okay. I looked at it earlier and it's not related to that patch but I hesitated to blindly recheck
19:31:24 cdent efried: yes, so the the migration uuid stuff that dansmith has is not racey
19:31:35 melwitt mriedem: cool, thanks
19:31:50 mriedem melwitt: https://bugs.launchpad.net/nova/+bug/1706377
19:31:51 openstack Launchpad bug 1706377 in OpenStack Compute (nova) "(libvirt) live migration fails on source host due to "Assertion `!(bs->open_flags & BDRV_O_INACTIVE)' failed."" [Undecided,Confirmed]
19:32:03 mriedem we can't fingerprint that in e-r because the failure is in the qemu instance logs, which we don't index
19:32:15 melwitt gotcha
19:34:46 jaypipes efried: k, just read the granular resource requests spec. lgtm.
19:35:10 efried jaypipes Cool, thanks for going through it.
19:35:44 efried cdent Is there a patch somewhere that's adding the project_id/user_id to the GET /allocations/{consumer_uuid} response?
19:36:25 cdent yeah, it’s an antecedent to the thing that adds the schema: https://review.openstack.org/#/c/500073/
19:36:44 cdent sorry wrong paste
19:36:58 cdent https://review.openstack.org/#/c/512420/
19:38:25 mriedem melwitt: now you just need to find another stable core to approve it :)
19:38:29 efried cdent Okay, and where does the api-ref get updated?
19:38:40 efried cdent This sucker: https://developer.openstack.org/api-ref/placement/#list-allocations
19:38:58 mriedem efried: https://github.com/openstack/nova/tree/master/placement-api-ref/source ?
19:39:22 mriedem or the actual infra that updates the page?
19:39:47 cdent efried: it’s not done yet. I haven’t decided where to put it. Could go in a more generic https://review.openstack.org/#/c/510626/
19:40:13 cdent efried: so please feel free to put a -1 somewhere in there that says “oi, you’ve not done the docs yet"
19:40:47 efried cdent Okay, cool, that's what I wanted to know. (mriedem specifically the doc change that adds the project/user id to that api-ref piece)
19:42:12 cdent efried: thanks for the attention to detail, the content of that stack has expanded a lot since I started, so some steps have been lost
19:42:53 efried cdent Swhat I'm here for. Eventually I'll be useful for something other than pointing out typos and missed paperwork.
19:43:22 dansmith efried: don't forget rebasing
19:43:34 efried Mm
19:43:49 dansmith you can rebase a mofo like a sonofabee
19:44:54 mriedem sdague: can you get this pike backport and the one below?
19:45:33 mlavalle mriedem: any chance we can get nova team eyes on the lastest revisions of https://review.openstack.org/#/c/375580/ and https://review.openstack.org/#/c/502306/
19:45:42 dansmith edleafe: cdent: so that limits or {} thing.. limits is nullable in reqspec.. are we sure we need do that thing?
19:46:02 dansmith also, I'm not clear on what part of this series requires this when we didn't before
19:46:37 mriedem mlavalle: honestly i think the bandwidth provider thing is not going to happen in queens
19:46:45 mriedem given the various dependencies and overall complexity
19:47:39 mlavalle mriedem: dependencies meaning functionality that is being developed in the Nova side?
19:47:45 mriedem mlavalle: yes
19:48:01 mriedem there are some major challenges to overcome before we can even get to bandwidth provider
19:48:56 mriedem mlavalle: for example, it relies on the compute pushing the bandwidth allocations and we're moving away from the computes trying to create allocations
19:53:02 openstackgerrit Chris Dent proposed openstack/nova master: [placement] manage cache headers for usages https://review.openstack.org/513174
19:54:34 edleafe dansmith: it caused multiple unit test failures without the 'or {}
19:54:58 edleafe dansmith: not sure if the change back to including limits makes that no longer the case.
19:55:04 dansmith edleafe: lol. well we should put that in the comment then, eh?
19:55:16 dansmith edleafe: I'd like to understand why we're making that change if we are
19:55:20 openstackgerrit Jay Pipes proposed openstack/nova master: placement: adds REST API for nested providers https://review.openstack.org/384807
19:55:21 openstackgerrit Jay Pipes proposed openstack/nova master: placement: update client to set parent provider https://review.openstack.org/385693
19:55:22 jaypipes efried: fixed. ^

Earlier   Later