Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-07-19
22:14:21 mriedem my favorite weather guy, randy brock, went into real estate and i still want to find him to make a joke about his forecast for a great deal on selling me a house
22:16:42 cdent jaypipes: gibi’s func test flapping is the result of differing sort order on the sharing_providers dict
22:16:59 cdent I noted that on the bug: https://bugs.launchpad.net/nova/+bug/1705231
22:17:00 openstack Launchpad bug 1705231 in OpenStack Compute (nova) "Placement returns no allocation candidate for request that needs both compute resources and custom shared resources" [High,Triaged]
22:18:17 cdent messes up the order of the joins which for reasons unknown to me messes up the query
22:28:50 openstackgerrit melanie witt proposed openstack/nova-specs master: Add a new section: "Upgrade impact" to the template https://review.openstack.org/456756
22:30:31 openstackgerrit OpenStack Proposal Bot proposed openstack/python-novaclient master: Updated from global requirements https://review.openstack.org/485358
22:42:59 mriedem melwitt: so the counting quotas changes are all approved,
22:43:08 mriedem once that last one is done i think we can close that bp unless there is something i'm missing,
22:43:17 mriedem the last 2 changes in the series are the quotas in api db bp
22:44:18 melwitt mriedem: yeah. the rest will be adding test coverage and the like
22:44:44 mriedem speaking of
22:44:45 mriedem http://logs.openstack.org/52/484152/6/check/gate-novaclient-dsvm-functional-neutron-ubuntu-xenial/2bee19d/testr_results.html.gz
22:44:52 mriedem so what i'm seeing in this test is,
22:44:56 mriedem we resize down,
22:44:59 mriedem usage changes, as expected
22:45:12 mriedem b/c no more reservations and the instance vpus/memory_mb is changed in the compute
22:45:21 mriedem then i revert the resize
22:45:26 mriedem usage changes back to the original flavor,
22:45:35 mriedem which makes sense, but in this case, it's a revert back *up*
22:45:51 mriedem at which point i could be over quota
22:46:17 mriedem oh f yeah https://review.openstack.org/#/c/416521/63/nova/compute/api.py@3040
22:46:21 mriedem you check for that
22:46:21 melwitt right. but the thing fails?
22:46:35 mriedem i think that was one of the more recent changes in that patch when we talked about it this week
22:46:49 mriedem the only thing that failed was my assertion that usage didn't change
22:46:50 mriedem which is wrong
22:46:59 mriedem i was just worried we weren't checking quota on revert up
22:47:00 mriedem but we are
22:47:02 mriedem so awesome
22:47:05 melwitt oh, good
22:47:12 melwitt phew
22:47:15 mriedem ha
22:49:52 openstackgerrit Matt Riedemann proposed openstack/python-novaclient master: Adjust test_resize_down_revert to account for counting quotas https://review.openstack.org/484152
23:10:13 mriedem good lord i guess the global lock in the functional tests is there for a reason http://logs.openstack.org/35/485335/1/check/gate-nova-tox-functional-ubuntu-xenial/854268f/testr_results.html.gz
23:10:43 mriedem oh it's something else
23:14:07 cfriesen in the context of cells v2, where do nova-consoleauth/nova-cert/nova-novncproxy run? on the nova-api node(s)?
23:15:15 openstackgerrit Jay Pipes proposed openstack/nova master: [placement] test shared custom resource provider https://review.openstack.org/485088
23:15:16 openstackgerrit Jay Pipes proposed openstack/nova master: placement: alloc candidates only shared resources https://review.openstack.org/484900
23:15:32 mriedem cfriesen: that's a question for melwitt
23:16:00 mriedem https://review.openstack.org/#/q/topic:bp/convert-consoles-to-objects+status:open
23:16:01 openstackgerrit Jay Pipes proposed openstack/nova master: placement: proper JOIN order for shared resources https://review.openstack.org/485088
23:18:57 cfriesen mriedem: I guess nova-cert is now gone, so it's just the other two
23:21:02 mriedem riht
23:21:04 mriedem *right
23:23:25 cfriesen mriedem: you wouldn't happen to know if bug 1691545 hits when using just one cell in newton?
23:23:26 openstack bug 1691545 in OpenStack Compute (nova) ocata "Significant increase in DB connections with cells" [High,Fix committed] https://launchpad.net/bugs/1691545 - Assigned to melanie witt (melwitt)
23:34:17 melwitt cfriesen: nova-consoleauth should be run globally for the deployment. currently nova-novncproxy should run globally too but after this series, it should be run per cell https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/convert-consoles-to-objects
23:34:57 melwitt cfriesen: that bug affects one cell yes. the fix was backported to newton though
23:35:20 cfriesen melwitt: I saw that, we haven't picked up that change yet
23:35:26 melwitt ah, okay
#openstack-nova - 2017-07-20
00:00:42 openstackgerrit Michael Still proposed openstack/nova master: Only setup iptables for metadata if using nova-net https://review.openstack.org/480765
00:15:25 mikal Does anyone here understand how os-vif is tested in the gate? I'm trying to understand how it configures privsep.
00:15:45 mikal I'm hoping that will explain why my privsep change fails with something which looks at lot like a configuration error.
00:46:45 openstackgerrit Merged openstack/nova master: Remove mox from nova.tests.unit.api.openstack.compute.test_limits https://review.openstack.org/477781
00:55:17 openstackgerrit Merged openstack/nova master: api-ref: Fix an expand button in os-quota-sets https://review.openstack.org/485075
01:02:42 openstackgerrit Merged openstack/nova master: Use uuid for id in os-services API https://review.openstack.org/464280
01:03:28 openstackgerrit Merged openstack/nova master: Handle uuids in os-hypervisors API https://review.openstack.org/471899
01:20:49 openstackgerrit Lars Kellogg-Stedman proposed openstack/nova master: provide interface-scoped nameserver information https://review.openstack.org/467699
01:24:15 Kevin_Zheng mriedem: should https://review.openstack.org/#/c/459493/ depend on https://review.openstack.org/#/c/484170/ ?
01:28:30 mriedem Kevin_Zheng: either that or we have to set instance.tags = instance_tags in conductor manager schedule_and_build_instance before it sends the instance to the compute
01:28:35 mriedem where the instance.create notification happens
01:28:43 mriedem that would be a workaround until my change
01:33:37 openstackgerrit Merged openstack/nova master: Count instances to check quota https://review.openstack.org/416521
01:34:30 mriedem melwitt: boing ^
02:15:57 mriedem bauzas: https://review.openstack.org/#/c/483565/ is ready for your "Flux de travail" when you get up
02:24:58 openstackgerrit Merged openstack/python-novaclient master: Updated from global requirements https://review.openstack.org/485358
02:28:34 melwitt mriedem: yeeeeeahhhh \o/
02:37:07 openstackgerrit Matt Riedemann proposed openstack/python-novaclient master: Expect id and disabled_reason in GET /os-services response https://review.openstack.org/485409
02:37:10 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/485410
02:58:25 melwitt mriedem: have you ever seen a job failure like this before? trying to figure out if it's me or not http://logs.openstack.org/73/484973/3/check/gate-tempest-dsvm-neutron-linuxbridge-ubuntu-xenial/9ec53dc/console.html#_2017-07-20_01_23_56_831458
02:58:38 mriedem yes i think so
02:58:42 mriedem it's a timeout
02:59:11 melwitt hm, so entire job timeout
02:59:19 mriedem something like that
03:23:20 TheJulia mriedem: re https://review.openstack.org/#/c/215385 Closer, is_volume is blowing up with a stack trace in the nova compute log. I don't have a good handle on the interior of the block device objects in nova, but I partially wonder if it is the tempest test scenario. A glance/sanity check would be greatly appreciated.
03:23:35 openstackgerrit Merged openstack/nova master: remove the very old unmaintained wsgi scripts https://review.openstack.org/479027
03:24:30 mriedem TheJulia: i think it's that the DriverVolumeBlockDevice object that wraps a BlockDeviceMapping object doesn't proxy the is_volume attribute
03:25:15 TheJulia That was kind of what I was suspecting at a glance, but I should have gone to sleep like an hour ago
03:25:21 mriedem https://github.com/openstack/nova/blob/master/nova/virt/block_device.py#L231
03:26:08 TheJulia joy!
03:28:01 mriedem i'll leave comments
03:28:13 TheJulia is_block_device_mapping ?
03:28:19 TheJulia mriedem: thanks!
03:28:35 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/485410
03:28:58 mriedem done
03:29:51 mriedem it's my fault for asking to use the mappings in the block_device_info list passed into spawn() to save the db roundtrip
03:30:03 mriedem i'd say we workaround it in pike and cleanup it up with a TODO in queens
03:30:09 mriedem *clean it up
03:39:00 TheJulia mriedem: I'm always for saving a roundtrip to the database
03:45:57 openstackgerrit Merged openstack/nova master: Remove key_manager.api_class hack https://review.openstack.org/484501
03:46:45 openstackgerrit Merged openstack/nova master: Remove reverts_task_state decorator from swap_volume https://review.openstack.org/481447
04:01:06 openstackgerrit Matt Riedemann proposed openstack/python-novaclient master: WIP: Microversion 2.53 - services and hypervisors using UUIDs https://review.openstack.org/485435
04:04:57 openstackgerrit Merged openstack/nova master: Use URIOpt https://review.openstack.org/479215
04:10:35 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/485410
04:23:03 openstackgerrit Julia Kreger proposed openstack/nova master: Ironic: Support boot from Cinder volume https://review.openstack.org/215385
04:43:04 openstackgerrit John Haan proposed openstack/nova-specs master: Support volume_type with BDM paramter https://review.openstack.org/466595
04:57:25 openstackgerrit John Haan proposed openstack/nova-specs master: Support volume_type with BDM paramter https://review.openstack.org/466595
05:02:07 openstackgerrit John Haan proposed openstack/nova-specs master: Support volume_type with BDM paramter https://review.openstack.org/466595
05:16:59 openstackgerrit John Haan proposed openstack/nova-specs master: Support volume_type with BDM paramter https://review.openstack.org/466595

Earlier   Later