Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-01-10
01:00:47 morgan got information from -qa channel. no problem now.
01:43:11 Kevin_Zheng ping oomichi
01:43:41 openstackgerrit Takashi NATSUME proposed openstack/nova: Avoid forcing translation on logging calls https://review.openstack.org/413876
01:44:35 Kevin_Zheng ping alex_xu
01:51:09 alex_xu Kevin_Zheng: hi, i'm here
01:51:32 Kevin_Zheng Ah, I have a question
01:51:40 Kevin_Zheng http://git.openstack.org/cgit/openstack/nova/tree/nova/api/validation/parameter_types.py#n327
01:54:11 Kevin_Zheng what does this suppose to do?
01:54:55 alex_xu Kevin_Zheng: that is schema for metadata, probably share between mulitple APIs
01:55:22 alex_xu Kevin_Zheng: I guess, you want to ask what is suppose to do with patternProperties?
01:55:27 Kevin_Zheng yeah, I understand but what does the regex mean
01:55:29 Kevin_Zheng yeah
01:55:36 Kevin_Zheng cos we tested yesterday
01:55:48 Kevin_Zheng it worked very strange
01:55:56 Kevin_Zheng you can test
01:56:10 alex_xu Kevin_Zheng: it limits the valid characters, only accept a to z, A to Z and 0 to 9, etc...
01:56:19 Kevin_Zheng looks like metadata key cannot work with chinese but value can
01:56:30 alex_xu Kevin_Zheng: what kind of strange thing you have seen?
01:56:40 Kevin_Zheng so is it desinged like this or what
01:56:44 alex_xu Kevin_Zheng: yes, that schema didn't limit the value
01:56:54 Kevin_Zheng Ah, OK
01:57:16 Kevin_Zheng I thought it should limit the value also
01:57:42 alex_xu Kevin_Zheng: I remember there still have problem with metadata
01:57:59 Kevin_Zheng so we only limit the metadata key now
02:01:54 alex_xu Kevin_Zheng: there is another propose about metadata, but i'm not sure whether it mention the value or not. and I can't find the link now, and I can't access launchpad...
02:02:07 Kevin_Zheng ok
02:17:35 mriedem sean-k-mooney: are you going to request an os-vif release for https://review.openstack.org/#/c/410135/ or should i?
02:19:40 openstackgerrit Matt Riedemann proposed openstack/python-novaclient: DNM: Test devstack revert for neutronclient functional test failures https://review.openstack.org/417645
02:28:52 openstackgerrit Takashi NATSUME proposed openstack/nova-specs: Amend cold-migration-with-target-ocata spec https://review.openstack.org/397193
02:28:58 openstackgerrit Takashi NATSUME proposed openstack/nova-specs: Abort Cold Migration https://review.openstack.org/334732
02:29:04 openstackgerrit Takashi NATSUME proposed openstack/nova-specs: List/show all server migration types https://review.openstack.org/407237
02:29:12 openstackgerrit Takashi NATSUME proposed openstack/nova-specs: Fix formatting errors and warnings https://review.openstack.org/407364
02:36:52 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571
02:40:02 openstackgerrit Huan Xie proposed openstack/nova: xenapi: support the hotplug of a neutron port https://review.openstack.org/365505
03:10:36 Kevin_Zheng mriedem: looks like the client gate has been fixed
03:13:48 mriedem Kevin_Zheng: yeah it was aacd8e0d043367d84ff9e50bad83c5c361852980 in neutron
03:13:51 mriedem but it's a bug in novalcient
03:13:53 mriedem *novaclient
03:13:59 mriedem i'm pushing a fix now, unless someone else already did
03:14:41 openstackgerrit Matt Riedemann proposed openstack/python-novaclient: Fix functional tests to deal with multiple networks https://review.openstack.org/418190
03:14:42 mriedem Kevin_Zheng: ^
03:15:29 Kevin_Zheng Thanks
03:31:29 gcb mriedem, I updated https://review.openstack.org/#/c/401827/ according to your comments, maybe it's ready now
04:00:48 openstackgerrit Takashi NATSUME proposed openstack/python-novaclient: Microversion 2.42 - Enable cold migration with target host https://review.openstack.org/406707
05:32:56 openstackgerrit Wang Qing wu proposed openstack/nova: WIP(3): Introduce base PowerVM driver (console) https://review.openstack.org/409402
06:30:38 openstackgerrit Takashi NATSUME proposed openstack/nova: Refactor a test method including 7 test cases https://review.openstack.org/395436
06:31:01 openstackgerrit Takashi NATSUME proposed openstack/nova: Remove mox in unit/api/openstack/test_wsgi.py https://review.openstack.org/414941
06:31:08 openstackgerrit Takashi NATSUME proposed openstack/nova: Refactor a test method including 3 test cases https://review.openstack.org/394738
06:31:50 openstackgerrit Takashi NATSUME proposed openstack/nova: api-ref: Fix parameters in os-hosts.inc https://review.openstack.org/416444
06:57:26 openstackgerrit jichenjc proposed openstack/nova: add zvm migration object https://review.openstack.org/414090
08:03:06 openstackgerrit Taku Izumi proposed openstack/nova: virt/ironic: Implement rescue and unrescue https://review.openstack.org/416487
08:18:26 openstackgerrit melanie witt proposed openstack/nova: WIP Count resources to check quota for cells https://review.openstack.org/416521
08:18:27 openstackgerrit melanie witt proposed openstack/nova: WIP Add online migration to move quotas to API database https://review.openstack.org/410946
08:18:27 openstackgerrit melanie witt proposed openstack/nova: Make Quotas object favor the API database https://review.openstack.org/410945
08:19:33 openstackgerrit Huan Xie proposed openstack/nova: XenAPI: Fix vif plug problem during VM rescue/unrescue https://review.openstack.org/413469
08:24:17 openstackgerrit Zhenyu Zheng proposed openstack/nova: Support tag instances when boot https://review.openstack.org/394321
08:30:16 openstackgerrit Béla Vancsics proposed openstack/nova: Transform instance.trigger_crash_dump notification https://review.openstack.org/396225
08:31:58 openstackgerrit Matt Riedemann proposed openstack/python-novaclient: Fix functional tests to deal with multiple networks https://review.openstack.org/418190
08:40:18 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add sort_key white list for server list/detail https://review.openstack.org/415142
08:40:56 openstackgerrit Zhenyu Zheng proposed openstack/nova: Add new policy for server list/detail with all_tenants https://review.openstack.org/415330
08:44:59 openstackgerrit Alex Xu proposed openstack/nova: [POC][placement] Add Trait and TraitList objects https://review.openstack.org/376199
08:47:52 openstackgerrit Zhenyu Zheng proposed openstack/nova: Check destination_type when booting with bdm provided https://review.openstack.org/402372
08:48:08 openstackgerrit Zhenyu Zheng proposed openstack/nova: Don't delete neutron port when attach failed https://review.openstack.org/403568
08:51:21 alex_xu Kevin_Zheng: ping
08:51:38 alex_xu Kevin_Zheng: I guess you miss the comment in the main page of https://review.openstack.org/#/c/408571/
08:52:01 alex_xu Kevin_Zheng: "we should make the strip extra properties method to support patternProperties https://github.com/openstack/nova/blob/master/nova/api/validation/__init__.py#L113"
08:53:46 Kevin_Zheng let me check
08:58:18 Kevin_Zheng what do you mean by this?
08:58:34 Kevin_Zheng OK, I just got it
08:58:56 Kevin_Zheng How, do you have any suggestions?
08:59:19 alex_xu Kevin_Zheng: even patternProperties enable the parameter prefix with '_', but it is still stripped out
08:59:57 alex_xu Kevin_Zheng: we shouldn't strip the parameters which match the regex in the patterProperties, so we need to support that at https://github.com/openstack/nova/blob/master/nova/api/validation/__init__.py#L113
09:17:52 andreykurilin mriedem: hi! I wonder if you found an issue yesturday. Did you?)
14:28:47 eglynn bearing in mind that the placement service landed after newton-3, IIRC
14:28:52 eglynn (and was uncertain up to the last, IIRC)
14:28:58 eglynn meaning that it would have been well-nigh impossible for deployment tooling to include support for it in newton-derived distros
14:29:00 mriedem eglynn: we were running CI on it well before then
14:29:33 bauzas mriedem: you know that we had problems with us communicating the placement API
14:29:38 bauzas mriedem: in the relnotes
14:29:41 eglynn deployment tooling can't start supporting stuff that still effectively WIP until after feature freeze
14:30:01 bauzas mriedem: mostly because our prelude section wasn't included in the 14.0.0 release but in a dot release
14:30:09 bauzas (thanks to reno)
14:30:29 bauzas so, most people deploying Newton hadn't paid much attention to that fresh new service
14:31:20 cdent So what are the realistic options from the deployment tooling standpoint eglynn ?
14:31:34 cdent (as in, how do we move forward, from your perspective)
14:32:40 mriedem we had everything in place by FF
14:32:50 eglynn cdent: 1. support a fallback to the old resource tracker data in ocata, then roll forward to based scheduling decisions on the placement API in first minor point release after ocata
14:33:21 mriedem we can't do that in a stable point release
14:33:23 eglynn cdent: 2. revert the scheduler cut-over patch downstream-only in ocata GA, then roll forward as in option #1
14:33:26 cdent how then to _force_ people to run placement, as that's the opposite side of the concern, as far as I can tell
14:34:10 mriedem cdent: i guess the argument is it's still optional in ocata, with warnings and other communication that it's going to be required in pike, for realz
14:34:25 mriedem at which point everyone will have given up on placement and gone to work on k8s
14:34:26 mriedem :)
14:34:37 cdent I've seen that quote before somewhere.
14:35:48 bauzas meh, we should just mention the issue page for k8s, period.
14:35:53 eglynn cdent: in option #1, I would envisage the placement API being rolled out, but not depended on, in the ocata GA ... the forcing function would be the post-ocata roll-forward to the new scheduler decision making in the post-ocata release
14:36:23 cdent An important thing to remember here is that the use of the placement api in ocata is just filtering the available nodes, it is not "scheduling decisions".
14:36:33 cdent That's a perhaps subtle but important distinction
14:37:10 cdent the filer sheduler will now filter 10 instead of 1000 nodes (or whatever)
14:37:10 eglynn cdent: fair enough ... put another way, the scheduler logic without which placement decisions become impossible

Earlier   Later