Thursday, 2015-08-20

*** dave-mccowan has quit IRC00:00
*** shaohe_feng has joined #openstack-nova00:01
*** unicell has quit IRC00:02
*** unicell has joined #openstack-nova00:02
*** bnemec has joined #openstack-nova00:03
*** rushil has quit IRC00:03
openstackgerritSylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests()  https://review.openstack.org/19920500:07
openstackgerritSylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object  https://review.openstack.org/20307800:07
*** bnemec has quit IRC00:10
*** artom has quit IRC00:10
*** achanda has quit IRC00:11
*** artom has joined #openstack-nova00:12
*** dave-mccowan has joined #openstack-nova00:14
*** andreykurilin__ has quit IRC00:15
*** mnestratov has quit IRC00:16
*** shaohe_feng has quit IRC00:17
*** shoutm has quit IRC00:18
*** armax has quit IRC00:20
*** shoutm has joined #openstack-nova00:21
*** unicell has left #openstack-nova00:21
*** armax has joined #openstack-nova00:22
*** unicell has joined #openstack-nova00:22
*** eglynn has quit IRC00:22
*** armax has quit IRC00:24
*** bnemec has joined #openstack-nova00:25
*** asettle has quit IRC00:28
*** dannywil_ has quit IRC00:28
*** salv-orl_ has quit IRC00:29
*** garyk1 has quit IRC00:30
*** dims has joined #openstack-nova00:30
*** bnemec has quit IRC00:30
*** shuquan has joined #openstack-nova00:31
*** rdopiera has quit IRC00:31
*** redbeard has joined #openstack-nova00:32
*** dave-mccowan has quit IRC00:37
*** thorst has joined #openstack-nova00:38
*** bnemec has joined #openstack-nova00:44
*** mriedem has joined #openstack-nova00:45
*** tpeoples has joined #openstack-nova00:46
*** achanda has joined #openstack-nova00:48
*** mdrabe has joined #openstack-nova00:48
*** tsekiyam_ has joined #openstack-nova00:49
openstackgerritMichael Still proposed openstack/nova: libvirt: take account of disks in migration data size  https://review.openstack.org/20663200:50
openstackgerritMichael Still proposed openstack/nova: libvirt: add helper methods for getting guest devices/disks  https://review.openstack.org/20663100:50
*** VikasC has joined #openstack-nova00:50
*** xyang1 has quit IRC00:51
*** VW_ has quit IRC00:51
*** VW_ has joined #openstack-nova00:52
openstackgerritGhanshyam Mann proposed openstack/nova: Merge used-limits functional tests of v2 and v2.1  https://review.openstack.org/17421500:52
*** eglynn has joined #openstack-nova00:53
*** tsekiyama has quit IRC00:53
*** shuquan has quit IRC00:53
*** mtanino has quit IRC00:53
*** tsekiyam_ has quit IRC00:53
*** eglynn has quit IRC00:54
mriedemnice, the lxc stuff in libvirt.driver._create_domain_and_network is not even unit tested00:56
mriedemno wonder that stuff breaks00:56
*** promulo__ has joined #openstack-nova00:57
*** VW_ has quit IRC00:57
*** VikasC has quit IRC00:57
mriedemwhy don't we require CI for the lxc stuff if we're going to have it in tree?  we require third party CI for the parallels stuff00:57
*** tjones has left #openstack-nova00:57
*** vishwanathj has quit IRC00:58
*** dave-mccowan has joined #openstack-nova00:59
mriedemapmelton: i'm gonna push a fix for https://bugs.launchpad.net/nova/+bug/1370590 quick00:59
openstackLaunchpad bug 1370590 in OpenStack Compute (nova) "Libvirt _create_domain_and_network calls missing disk_info" [Medium,In progress] - Assigned to Andrew Melton (andrew-melton)00:59
*** eliqiao_pc has joined #openstack-nova00:59
mriedemmore type conversions breaking stuff00:59
*** Piet has quit IRC01:00
*** shyu has joined #openstack-nova01:00
*** promulo has quit IRC01:00
clarkb mriedem it should be easyish to do that since lxc should perform much better than qemu on those VMs01:00
clarkbbiggest problem is probably figuring out what tests to use as tempest likely will explode against lxc01:01
mriedemclarkb: yeah, and finding someone to actually get it setup and maintain it01:01
openstackgerritZhuChunzhan proposed openstack/nova: Improve the performance of get evacuated instances from a host which should be destroyed  https://review.openstack.org/21455501:01
mriedemat some point 3 midcycles ago rax was talking about doing that01:01
mriedemalaski: johnthetubaguy: do you remember or know of anything rax is planning about getting a CI job running with libvirt + lxc?01:02
*** shaohe_feng has joined #openstack-nova01:02
mriedemmeanwhile, unit tests might help...01:02
*** yanxb_ has joined #openstack-nova01:02
*** yanxb__ has quit IRC01:04
*** asettle has joined #openstack-nova01:05
*** eliqiao_pc has quit IRC01:06
*** browne has quit IRC01:06
*** baoli has joined #openstack-nova01:07
*** wanghua has joined #openstack-nova01:10
*** thorst has quit IRC01:10
*** shane-wang has joined #openstack-nova01:10
*** RuiChen has quit IRC01:11
*** diogogmt has joined #openstack-nova01:11
*** baoli has quit IRC01:12
*** haigang has joined #openstack-nova01:16
*** Vek has joined #openstack-nova01:20
openstackgerritGhanshyam Mann proposed openstack/nova: Merge limits functional tests of v2 and v2.1  https://review.openstack.org/21491701:21
openstackgerritwanghao proposed openstack/nova: Add nova-manage command for cleaning attachment  https://review.openstack.org/18453701:21
gmannalex_xu: johnthetubaguy : sdague : did you see https://review.openstack.org/#/c/174215/01:21
*** agireud has joined #openstack-nova01:22
gmannalex_xu: johnthetubaguy : sdague :  there is no RateLimitingMiddleware for v2.1 which make difference in v2 and v2.1 usedlimits and limits API01:22
gmannalex_xu: johnthetubaguy : sdague : rate limits are not populated on v2.1 side01:23
*** RuiChen has joined #openstack-nova01:23
gmannfor limits API- https://review.openstack.org/#/c/214917/01:24
*** dims has quit IRC01:25
*** artom has quit IRC01:27
*** liusheng has joined #openstack-nova01:30
*** Apoorva has quit IRC01:30
*** vilobhmm1 has quit IRC01:33
openstackgerritwanghao proposed openstack/nova: Fix mistake in UT:test_detach_unattached_volume  https://review.openstack.org/21492001:35
*** Apoorva has joined #openstack-nova01:39
*** shane-wang has quit IRC01:39
*** trpsbill has joined #openstack-nova01:40
mriedemgmann: that's a known thing for awhile, see the api_rate_limit option01:43
mriedemhttp://docs.openstack.org/kilo/config-reference/content/list-of-compute-config-options.html01:43
mriedem"(BoolOpt) Whether to use per-user rate limiting for the api. This option is only used by v2 api. Rate limiting is removed from v3 api."01:44
mriedemi don't really know the history behind that though01:44
*** liusheng has quit IRC01:45
*** lyanchih has joined #openstack-nova01:46
*** ctrath has joined #openstack-nova01:46
*** meghal has quit IRC01:48
gmannmriedem: yea01:48
gmannmriedem: but not sure if we should introduced it in mivroversion or not as it is not really used (commit msg of https://review.openstack.org/#/c/115893/)01:49
gmannmriedem: as we cannot enable it in v2.1 which is already released01:50
*** boris-42 has quit IRC01:50
mriedemgmann: well,01:50
mriedemit's obviously not totally backwards compatible with v201:50
mriedembut it doesn't seem worth re-adding in a new microversion01:51
gmannmriedem: yea, same feeling here01:51
mriedemthat help text on the api_rate_limit option has been around for a long time, probably since havana01:51
gmannmriedem: ok, then i think i should be fine as it is01:52
*** ctrath has quit IRC01:52
gmannmriedem: please check my reply on https://review.openstack.org/#/c/198944/01:53
*** haomaiwang has joined #openstack-nova01:53
gmannmriedem: actually there was file structure change in master and kilo so did manually backport01:53
*** agireud has quit IRC01:58
*** SourabhP has quit IRC01:59
*** agireud has joined #openstack-nova02:00
*** neiljerram_bb has quit IRC02:01
*** ctrath1 has joined #openstack-nova02:02
*** tpeoples has quit IRC02:02
mriedemi'll have to check it tomorrow, about burned out for the night02:02
*** shane-wang has joined #openstack-nova02:03
*** mdrabe has quit IRC02:07
*** baoli has joined #openstack-nova02:08
openstackgerritMatt Riedemann proposed openstack/nova: lxc: fix disk_info passed to _create_domain_and_network in _hard_reboot  https://review.openstack.org/21492502:09
*** haomaiwang has quit IRC02:09
*** haomaiwang has joined #openstack-nova02:10
mriedemapmelton: s1rp: ^02:10
mriedemthat should fix that lxc reboot bug02:10
*** baoli has quit IRC02:12
mriedemmikal: ^ for the bug there, we should have probably created a new bug rather than re-opening the old one that had a fix go into kilo02:15
mriedemsince the regression was different and in liberty only02:15
openstackgerritRui Chen proposed openstack/nova: Set task_state=None when booting instance failed  https://review.openstack.org/18714502:16
alex_xugmann: I think it is fine, it is optional in the v2. So the contract is user won't expect ratelimit also being here02:18
gmannmriedem: sure. Thanks02:18
gmannalex_xu: hummm. yea02:19
openstackgerritRui Chen proposed openstack/nova: Fix permission issue of server group API  https://review.openstack.org/19311702:20
*** dims has joined #openstack-nova02:21
*** lyanchih has quit IRC02:21
openstackgerritEli Qiao proposed openstack/nova: Correct expected error code for os-resetState action  https://review.openstack.org/20393002:21
*** ctrath1 has quit IRC02:22
*** kumar has joined #openstack-nova02:23
*** unicell has quit IRC02:25
*** trpsbill has quit IRC02:26
*** trpsbill has joined #openstack-nova02:26
openstackgerritMatt Riedemann proposed openstack/nova: libvirt: _create_domain_setup_lxc needs to default disk mapping as a dict  https://review.openstack.org/21492702:26
*** trpsbill has quit IRC02:27
*** kumar has quit IRC02:28
*** trpsbill has joined #openstack-nova02:28
*** trpsbill has quit IRC02:28
*** dims has quit IRC02:29
*** dims has joined #openstack-nova02:30
*** shoutm_ has joined #openstack-nova02:30
*** otter768 has quit IRC02:30
*** shoutm has quit IRC02:31
*** shane-wang has quit IRC02:34
*** dims has quit IRC02:34
*** agireud has quit IRC02:35
*** shane-wang has joined #openstack-nova02:38
*** agireud has joined #openstack-nova02:43
*** bkopilov has quit IRC02:43
*** shane-wang1 has joined #openstack-nova02:43
*** mriedem has quit IRC02:44
*** salv-orlando has joined #openstack-nova02:44
*** markvoelker has quit IRC02:46
*** shane-wang has quit IRC02:47
*** shane-wang1 has quit IRC02:47
*** salv-orlando has quit IRC02:51
*** hakimo has joined #openstack-nova02:52
*** yamahata has quit IRC02:53
*** agireud has quit IRC02:53
*** hakimo_ has quit IRC02:54
*** salv-orlando has joined #openstack-nova02:55
*** Apoorva has quit IRC02:56
openstackgerritFang Ji proposed openstack/nova: change qemu guest agent socket path  https://review.openstack.org/21493502:59
*** diogogmt has quit IRC03:01
*** lyanchih has joined #openstack-nova03:01
*** salv-orlando has quit IRC03:02
*** otter768 has joined #openstack-nova03:02
openstackgerritGhanshyam Mann proposed openstack/nova: Remove merged sample tests and file for v2 tests  https://review.openstack.org/21493703:05
*** shoutm has joined #openstack-nova03:06
*** y_sawai has joined #openstack-nova03:06
*** shoutm_ has quit IRC03:07
*** haomaiwang has quit IRC03:09
openstackgerritGhanshyam Mann proposed openstack/nova: Remove merged sample tests and file for v2 tests  https://review.openstack.org/21493703:10
*** browne has joined #openstack-nova03:10
*** haomaiwang has joined #openstack-nova03:10
*** otter768 has quit IRC03:13
openstackgerritZhenyu Zheng proposed openstack/nova: Make query to quota usage table order preserved.  https://review.openstack.org/21358403:15
*** armax has joined #openstack-nova03:15
openstackgerritwanghao proposed openstack/nova: Add nova-manage command for cleaning attachment  https://review.openstack.org/18453703:17
*** y_sawai has quit IRC03:18
*** y_sawai has joined #openstack-nova03:19
*** agireud has joined #openstack-nova03:20
*** venkat_p has joined #openstack-nova03:23
*** baoli has joined #openstack-nova03:23
*** gibi has quit IRC03:24
*** gibi has joined #openstack-nova03:24
*** baoli has quit IRC03:28
*** y_sawai has quit IRC03:35
*** rushil has joined #openstack-nova03:36
*** Kennan2 is now known as Kennan03:36
openstackgerritwangxiyuan proposed openstack/nova: Resize: Return to initial state instead of ACTIVE when resize fails  https://review.openstack.org/20237903:37
openstackgerritEli Qiao proposed openstack/nova: Raise exception.Migration earlier in REST API layer  https://review.openstack.org/21443403:37
*** VW_ has joined #openstack-nova03:38
*** tpeoples has joined #openstack-nova03:38
*** VW_ has quit IRC03:39
*** VW_ has joined #openstack-nova03:40
openstackgerritRui Chen proposed openstack/nova: Fix incorrect result of host-describe  https://review.openstack.org/21108303:44
*** rushil_ has joined #openstack-nova03:44
*** VW_ has quit IRC03:45
*** VW_ has joined #openstack-nova03:45
*** shane-wang has joined #openstack-nova03:45
*** rushil_ has quit IRC03:46
*** markvoelker has joined #openstack-nova03:46
*** rushil has quit IRC03:48
*** y_sawai has joined #openstack-nova03:48
*** VW_ has quit IRC03:50
*** lpetrut has joined #openstack-nova03:51
*** markvoelker has quit IRC03:51
*** RuiChen has quit IRC03:55
*** RuiChen has joined #openstack-nova03:55
*** y_sawai has quit IRC03:56
*** shoutm_ has joined #openstack-nova04:00
*** gibi has quit IRC04:00
*** gibi has joined #openstack-nova04:00
*** shoutm has quit IRC04:01
*** RuiChen has quit IRC04:04
*** achanda has quit IRC04:05
*** y_sawai has joined #openstack-nova04:06
*** irenab has quit IRC04:09
*** haomaiwang has quit IRC04:09
*** haomaiwang has joined #openstack-nova04:10
*** irenab has joined #openstack-nova04:11
*** Nikolay_St has quit IRC04:21
*** dave-mccowan has quit IRC04:22
*** lyanchih has quit IRC04:22
*** chhavi_ has joined #openstack-nova04:23
*** baoli has joined #openstack-nova04:24
*** baoli has quit IRC04:29
*** unicell has joined #openstack-nova04:30
*** bkopilov has joined #openstack-nova04:30
*** lpetrut has quit IRC04:31
*** fawadkhaliq has joined #openstack-nova04:32
*** salv-orlando has joined #openstack-nova04:32
*** agireud has quit IRC04:33
*** agireud has joined #openstack-nova04:35
*** salv-orlando has quit IRC04:35
*** shaohe_feng has quit IRC04:37
*** deray_ has joined #openstack-nova04:40
*** deray_ has left #openstack-nova04:40
*** tpeoples has quit IRC04:42
*** lyanchih has joined #openstack-nova04:42
*** Sukhdev has joined #openstack-nova04:43
openstackgerritMan Li Qi proposed openstack/nova: (WIP)Filter instance query result for DB who doesn't support regex  https://review.openstack.org/13017404:43
*** salv-orlando has joined #openstack-nova04:47
*** Sukhdev has quit IRC04:47
*** salv-orlando has quit IRC04:51
*** huats has quit IRC04:55
*** Sree has joined #openstack-nova04:58
*** Sukhdev has joined #openstack-nova04:58
*** zul has quit IRC05:04
*** Jeffrey4l|afk has quit IRC05:05
*** Sree has quit IRC05:06
*** achanda has joined #openstack-nova05:07
*** afazekas has joined #openstack-nova05:08
*** haomaiwang has quit IRC05:09
*** danieru has quit IRC05:10
*** y_sawai has quit IRC05:10
*** haomaiwang has joined #openstack-nova05:10
*** Jeffrey4l|afk has joined #openstack-nova05:12
*** otter768 has joined #openstack-nova05:14
*** haigang has quit IRC05:14
*** afazekas has quit IRC05:15
*** jasonpan has left #openstack-nova05:16
*** chhavi__ has joined #openstack-nova05:16
*** jasonpan has joined #openstack-nova05:16
*** shane-wang has quit IRC05:17
jasonpantest05:18
*** chhavi_ has quit IRC05:18
*** emagana has joined #openstack-nova05:18
*** kiran-r has joined #openstack-nova05:18
*** haomaiwang has quit IRC05:19
*** otter768 has quit IRC05:19
jasonpankiran-r: Hi05:19
kiran-rjasonpan: Hi05:19
kiran-rjasonpan: Good morning!05:19
jasonpankiran-r: Good afternoon!05:20
jasonpankiran-r: did your problem solve?05:21
*** irenab has quit IRC05:21
kiran-rjasonpan: Yes I did. Thanks! I added an NFS share and solved it. :)05:21
kiran-rjasonpan: I am curious about other alternatives rather than NFS.05:25
kiran-rjasonpan: I am curious about other alternatives rather than having a shared storage, How about block live migration?05:25
*** Sukhdev has quit IRC05:26
jasonpankiran-r: hmmm... I haven't tested using block live migration before.05:27
*** RuiChen has joined #openstack-nova05:28
jasonpankiran-r: In manuals, block live mirgration is "No share storage required"05:30
*** irenab has joined #openstack-nova05:32
*** Marga_ has joined #openstack-nova05:38
kiran-rjasonpan: Thats I want to try that out :)05:39
*** tpeoples has joined #openstack-nova05:41
*** agireud has quit IRC05:47
openstackgerritSylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests()  https://review.openstack.org/19920505:50
openstackgerritSylvain Bauza proposed openstack/nova: Consider that all scheduler calls are IO Ops  https://review.openstack.org/18848605:50
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler  https://review.openstack.org/18848705:50
openstackgerritSylvain Bauza proposed openstack/nova: Prepare filters for using RequestSpec object  https://review.openstack.org/19791205:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify left filters for RequestSpec  https://review.openstack.org/20267805:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object  https://review.openstack.org/20307805:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify metric-related filters for RequestSpec  https://review.openstack.org/20267605:50
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler  https://review.openstack.org/19905105:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify NUMA, PCI and num_instances filters for RequestSpec  https://review.openstack.org/20267705:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify Aggregate filters for RequestSpec  https://review.openstack.org/20267405:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify default filters for RequestSpec  https://review.openstack.org/20267505:50
openstackgerritSylvain Bauza proposed openstack/nova: Add RequestSpec methods for primitiving into dicts  https://review.openstack.org/18719705:50
openstackgerritSylvain Bauza proposed openstack/nova: Modify Affinity filter for RequestSpec  https://review.openstack.org/20267305:50
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec object in HostManager  https://review.openstack.org/19125105:50
openstackgerritSylvain Bauza proposed openstack/nova: Use the same pci_requests field for all filters and HostManager  https://review.openstack.org/19125005:50
*** danieru has joined #openstack-nova05:52
*** nkrinner has joined #openstack-nova05:52
*** shoutm_ has quit IRC05:52
*** y_sawai has joined #openstack-nova05:54
*** chhavi__ has quit IRC05:54
*** kiran-r has quit IRC05:59
*** exploreshaifali has joined #openstack-nova05:59
*** haomaiwa_ has joined #openstack-nova06:00
*** armax has quit IRC06:00
*** y_sawai has quit IRC06:00
*** chhavi__ has joined #openstack-nova06:04
*** unicell1 has joined #openstack-nova06:04
*** salv-orlando has joined #openstack-nova06:05
*** y_sawai has joined #openstack-nova06:05
*** unicell has quit IRC06:05
*** shoutm has joined #openstack-nova06:07
*** VW_ has joined #openstack-nova06:09
*** asettle has quit IRC06:09
*** haomaiwa_ has quit IRC06:09
*** haomaiwang has joined #openstack-nova06:13
*** ajayaa has joined #openstack-nova06:14
*** yanxb has joined #openstack-nova06:15
*** salv-orlando has quit IRC06:16
*** haomaiwang has quit IRC06:17
openstackgerritZhenyu Zheng proposed openstack/nova: Resize: Return to initial state instead of ACTIVE when resize fails  https://review.openstack.org/20237906:18
*** VW_ has quit IRC06:21
*** afazekas has joined #openstack-nova06:21
*** VW_ has joined #openstack-nova06:21
*** jichen has joined #openstack-nova06:23
*** ankit_ag has joined #openstack-nova06:24
*** shuquan has joined #openstack-nova06:25
*** Murali has joined #openstack-nova06:26
*** Sree has joined #openstack-nova06:27
jasonpanbauzas: Hi bauzas, I have some question of the bug #145256806:28
openstackbug 1452568 in OpenStack Compute (nova) "nova allows to live-migrate instance from one availability zone to another" [Low,Confirmed] https://launchpad.net/bugs/1452568 - Assigned to Sylvain Bauza (sylvain-bauza)06:28
*** haomaiwang has joined #openstack-nova06:29
*** shuquan has quit IRC06:30
*** salv-orlando has joined #openstack-nova06:32
*** shaohe_feng has joined #openstack-nova06:32
*** Sree has quit IRC06:33
jasonpanbauzas: modify the database availability zone I think it is needed, but not allows instance from one availability zone to another I don't think it is needed. What do you think about it?06:36
*** haigang has joined #openstack-nova06:37
*** jichen has quit IRC06:37
*** exploreshaifali has quit IRC06:37
*** afazekas has quit IRC06:37
*** ajayaa has quit IRC06:37
*** tpeoples has quit IRC06:44
*** fawadkhaliq has quit IRC06:45
*** y_sawai has quit IRC06:45
*** y_sawai has joined #openstack-nova06:47
*** haomaiwang has quit IRC06:47
*** moshele has joined #openstack-nova06:48
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex  https://review.openstack.org/21236706:49
*** ajayaa has joined #openstack-nova06:50
*** haomaiwang has joined #openstack-nova06:52
*** scheuran has joined #openstack-nova06:52
openstackgerritZhenyu Zheng proposed openstack/nova: Resize: Return to initial state instead of ACTIVE when resize fails  https://review.openstack.org/20237906:53
*** y_sawai has quit IRC06:57
*** browne has quit IRC06:59
*** Nikolay_St has joined #openstack-nova06:59
*** VikasC has joined #openstack-nova07:00
*** achanda has quit IRC07:00
*** ishant has joined #openstack-nova07:06
*** y_sawai has joined #openstack-nova07:07
*** ssurana has quit IRC07:07
openstackgerritGhanshyam Mann proposed openstack/nova: Cleanup for merging v2 and v2.1 functional tests.  https://review.openstack.org/21498507:08
*** haomaiwang has quit IRC07:09
*** fawadkhaliq has joined #openstack-nova07:09
*** haomaiwang has joined #openstack-nova07:10
*** shane-wang has joined #openstack-nova07:13
*** otter768 has joined #openstack-nova07:15
*** Nikolay_St has quit IRC07:16
*** nkrinner has quit IRC07:17
openstackgerritwanghao proposed openstack/nova: Image&block device mapping combination isn't valid  https://review.openstack.org/17198407:17
*** exploreshaifali has joined #openstack-nova07:17
*** claudiub has joined #openstack-nova07:19
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/21499107:19
*** otter768 has quit IRC07:19
*** cfriesen has quit IRC07:20
openstackgerritHans Lindgren proposed openstack/nova: Remove/deprecate conductor instance_update()  https://review.openstack.org/18915507:20
openstackgerritHans Lindgren proposed openstack/nova: Make compute manager instance updates use objects  https://review.openstack.org/18915407:20
*** shane-wang has quit IRC07:21
*** nkrinner has joined #openstack-nova07:21
*** Murali_ has joined #openstack-nova07:23
openstackgerritHans Lindgren proposed openstack/nova: Add a VolumeUsage object  https://review.openstack.org/18044207:23
openstackgerritHans Lindgren proposed openstack/nova: Replace conductor proxying calls with the new VolumeUsage object  https://review.openstack.org/18044307:23
openstackgerritHans Lindgren proposed openstack/nova: Remove and deprecate unused conductor method vol_usage_update()  https://review.openstack.org/18044407:23
*** Nikolay_St has joined #openstack-nova07:24
*** jasonpan has quit IRC07:24
*** atuvenie has joined #openstack-nova07:24
*** Murali has quit IRC07:25
*** Murali_ is now known as Murali07:25
*** romainh has joined #openstack-nova07:26
*** exploreshaifali has quit IRC07:26
*** exploreshaifali has joined #openstack-nova07:26
*** SourabhP has joined #openstack-nova07:26
*** jasonpan has joined #openstack-nova07:28
*** SourabhP_ has joined #openstack-nova07:29
*** Murali has quit IRC07:30
*** Murali_ has joined #openstack-nova07:30
*** Murali_ is now known as Murali07:30
*** SourabhP has quit IRC07:32
*** SourabhP_ is now known as SourabhP07:32
openstackgerritGhanshyam Mann proposed openstack/nova: Remove merged sample tests and file for v2 tests  https://review.openstack.org/21493707:37
*** salv-orlando has quit IRC07:40
*** jcoufal has joined #openstack-nova07:41
*** alexschm has joined #openstack-nova07:42
*** SourabhP has quit IRC07:44
*** markvoelker has joined #openstack-nova07:48
openstackgerritGhanshyam Mann proposed openstack/nova: Merge server list multi status tests between v2 and v2.1  https://review.openstack.org/20138707:49
*** IanGovett has joined #openstack-nova07:51
openstackgerritGhanshyam Mann proposed openstack/nova: Merge server list multi status tests between v2 and v2.1  https://review.openstack.org/20138707:52
openstackgerritGhanshyam Mann proposed openstack/nova: Merge block_device_mapping boot tests between v2 and v2.1  https://review.openstack.org/20140107:53
*** markvoelker has quit IRC07:53
openstackgerritGhanshyam Mann proposed openstack/nova: Merge virtual interface tests between v2 and v2.1  https://review.openstack.org/19781107:53
openstackgerritGhanshyam Mann proposed openstack/nova: Merge flavor all_extensions tests between v2 and v2.1  https://review.openstack.org/20143007:53
*** markus_z has joined #openstack-nova07:53
openstackgerritGhanshyam Mann proposed openstack/nova: Fix link's href to consider osapi_compute_link_prefix  https://review.openstack.org/20770207:53
openstackgerritGhanshyam Mann proposed openstack/nova: Move "versions" functional tests in v2.1 tests  https://review.openstack.org/20143907:54
openstackgerritGhanshyam Mann proposed openstack/nova: Merge used-limits functional tests of v2 and v2.1  https://review.openstack.org/17421507:54
*** shyama has quit IRC07:54
openstackgerritGhanshyam Mann proposed openstack/nova: Merge limits functional tests of v2 and v2.1  https://review.openstack.org/21491707:54
openstackgerritGhanshyam Mann proposed openstack/nova: Remove merged sample tests and file for v2 tests  https://review.openstack.org/21493707:54
*** shyama has joined #openstack-nova07:55
*** alexschm has quit IRC07:55
*** salv-orlando has joined #openstack-nova07:55
*** y_sawai has quit IRC07:59
*** yamahata has joined #openstack-nova08:00
*** derekh has joined #openstack-nova08:08
*** yassine has joined #openstack-nova08:09
*** haomaiwang has quit IRC08:09
*** alexschm has joined #openstack-nova08:10
*** lucasagomes has joined #openstack-nova08:10
*** haomaiwa_ has joined #openstack-nova08:10
*** shuquan has joined #openstack-nova08:12
*** afazekas has joined #openstack-nova08:18
openstackgerritZhenyu Zheng proposed openstack/nova: Make query to quota usage table order preserved.  https://review.openstack.org/21358408:19
*** asettle has joined #openstack-nova08:19
*** RuiChen has quit IRC08:21
*** lyanchih has quit IRC08:22
*** shuquan has quit IRC08:22
*** lyanchih has joined #openstack-nova08:23
*** agireud has joined #openstack-nova08:23
*** fawadkhaliq has quit IRC08:23
*** asettle has quit IRC08:24
*** fawadkhaliq has joined #openstack-nova08:26
*** bharath_ has joined #openstack-nova08:28
*** jistr has joined #openstack-nova08:28
*** shoutm has quit IRC08:29
*** exploreshaifali has quit IRC08:30
*** chhavi__ has quit IRC08:30
*** agireud has quit IRC08:33
*** sayali has quit IRC08:35
*** SourabhP has joined #openstack-nova08:35
*** lalit has quit IRC08:36
*** VW_ has quit IRC08:36
*** huats has joined #openstack-nova08:36
*** VW_ has joined #openstack-nova08:36
*** sayali has joined #openstack-nova08:37
openstackgerritGhanshyam Mann proposed openstack/nova: Fix remote_consoles miveroversion 2.8 not to run on /v3  https://review.openstack.org/21501508:38
*** VW_ has quit IRC08:40
*** huats has quit IRC08:41
*** aix has joined #openstack-nova08:41
*** scheuran has quit IRC08:44
*** shoutm has joined #openstack-nova08:44
johnthetubaguyalex_xu: gmann: it would be good to catch up quick if you folks are still around?08:44
gmannjohnthetubaguy: hi good morning08:45
openstackgerritJohn Garbutt proposed openstack/nova: Remove the useless require_admin_context decorator  https://review.openstack.org/20038208:45
*** apuimedo has joined #openstack-nova08:46
johnthetubaguygmann: hi, I was really just wondering how its all going08:46
*** huats has joined #openstack-nova08:46
johnthetubaguygmann: I uploaded some stuff around the paste.ini and deprecating the extensions08:47
gmannjohnthetubaguy: yea, I am almost done for merging the tests stuff, i will put your paste.ini patch on top of those08:47
gmannjohnthetubaguy: but i did not get chance to look on metadata validation stuff. ll look tomorrow08:48
johnthetubaguygmann: OK, I think I have it working apart from the tempest tests, well almost08:49
gmannjohnthetubaguy: ok. for functional tests we can confirm by putting that on https://review.openstack.org/#/c/214985/08:51
gmannwhere all functional tests moved to run on /v2 and /v2.1 instead of /v308:52
openstackgerritGhanshyam Mann proposed openstack/nova: Cleanup for merging v2 and v2.1 functional tests.  https://review.openstack.org/21498508:52
*** aagrawal has joined #openstack-nova08:52
*** jichen has joined #openstack-nova08:53
johnthetubaguygmann: I really don't need all those changes to make this stuff pass08:53
*** shoutm has quit IRC08:54
*** shoutm has joined #openstack-nova08:55
*** ankit_ag has quit IRC08:55
gmannjohnthetubaguy: but functional tests failure are there which needs those changes08:57
openstackgerritSudipta Biswas proposed openstack/nova: Fix scheduler code to use monitor metric objects  https://review.openstack.org/21331708:57
johnthetubaguygmann: only needs some of them it seems, I only have a handful of failures, stuff like this: https://review.openstack.org/#/c/215015/108:58
johnthetubaguyalthough I can't make that one work actually, which is confusing08:59
*** VikasC has quit IRC08:59
*** alexschm has quit IRC08:59
*** markus_z has quit IRC08:59
gmannjohnthetubaguy: yea one is https://review.openstack.org/213317 needed09:00
johnthetubaguygmann: these are the current failures: http://logs.openstack.org/85/214085/6/check/gate-nova-tox-functional/ff79a6a/console.html09:00
*** jamie_h has joined #openstack-nova09:02
tobascowhen i boot a instance i also get assigned a ipv6 local address like fe80::xxxxxxxxxxxx/64, would setting use_ipv6=False cause these to not be assigned or how can i solve that?09:04
gmannjohnthetubaguy: x-server-management-url are due to tests using /v309:05
johnthetubaguygmann: agreed09:05
* alex_xu try to catch the context09:05
*** SourabhP has quit IRC09:05
tobascowe dont have ipv6 yet and when looking up domains that has a AAAA record it tries to use ipv609:05
*** abhishekk has joined #openstack-nova09:06
gmannjohnthetubaguy:  i am running those locally on test merge changes. lets see the result09:06
johnthetubaguytobasco: please ask in #openstack for operating advice, the folks over there might have seen this issue before, sounds a bit like a networking configuration thing09:06
johnthetubaguygmann: OK, cool09:07
*** ociuhandu has quit IRC09:07
tobascojohnthetubaguy: ok, was just curious if the use_ipv6 in nova.conf would be the key, but sure i will09:07
johnthetubaguytobasco: it depends on your networking stack in use, I suspect, the other folks should be able to help with that09:08
alex_xugmann: are you working strip the extra params out?09:08
*** fawadkhaliq has quit IRC09:08
alex_xus/working/working on/09:08
openstackgerritJohn Garbutt proposed openstack/nova: api: tidy up paste.ini  https://review.openstack.org/21408509:08
gmannalex_xu: not yet. was doing test merge stuff09:09
*** PrabAsh has joined #openstack-nova09:09
*** haomaiwa_ has quit IRC09:09
alex_xugmann: ok, let me know if you need help09:09
*** haomaiwang has joined #openstack-nova09:10
*** apuimedo is now known as apuimedo|away09:10
PrabAshCan openstack migrate VM instances automatically to another node if one of the compute node goes down ?09:10
*** shaohe_feng has quit IRC09:10
gmannalex_xu: I will start tomorrow and we can discuss that. or if you have some idea you can give a try because i have to look into those yet09:10
*** alexschm has joined #openstack-nova09:11
alex_xugmann: I didn't take look at into detail yet, but I can take a look at, then let discussion tomorrow09:12
gmannalex_xu: +1 perfect.09:12
openstackgerritSergey Nikitin proposed openstack/nova: Switch to oslo.cache lib  https://review.openstack.org/20304909:13
openstackgerritJohn Garbutt proposed openstack/nova: api: tidy up paste.ini  https://review.openstack.org/21408509:15
*** shoutm has quit IRC09:15
*** otter768 has joined #openstack-nova09:16
gmannjohnthetubaguy: it passing except extension tests09:17
gmannremote console and usedlimit works fine09:17
johnthetubaguygmann: OK, interesting, I just have a fix for that I think09:17
gmannjohnthetubaguy: ohk09:18
*** markus_z has joined #openstack-nova09:18
johnthetubaguygmann: https://review.openstack.org/#/c/214085/8/nova/tests/functional/test_extensions.py,cm09:19
*** jistr has quit IRC09:19
johnthetubaguygmann: but on my change, I can't get this one to pass: https://review.openstack.org/#/c/214085/8/nova/tests/functional/api_sample_tests/test_remote_consoles.py,cm I keep getting a 400 error?09:20
*** unicell1 has quit IRC09:20
johnthetubaguyits like its pointed at the legacy mode or something09:20
*** otter768 has quit IRC09:20
*** jistr has joined #openstack-nova09:20
*** bharath_ has left #openstack-nova09:22
gmannthose are passed but test_extension_info still fail with your fix09:22
*** tbh has joined #openstack-nova09:23
*** unicell has joined #openstack-nova09:23
tbhHi johnthetubaguy, is this bp still valid case https://blueprints.launchpad.net/nova/+spec/os-hypervisors-ipaddress ?09:23
*** y_sawai has joined #openstack-nova09:24
johnthetubaguytbh: I think its fine, but it needs a spec submitting for the mitaka release, with all the API change details listed09:25
johnthetubaguygmann: whats the error you see?09:25
gmannjohnthetubaguy: its 404 for v2 get_extension09:25
johnthetubaguygmann: yeah, thats what the change fixed on my branch, odd09:26
johnthetubaguygmann: do you have all these fixes your branch? https://review.openstack.org/#/c/214085/8/nova/tests/functional/api_paste_fixture.py,cm09:27
tbhjohnthetubaguy: okay, I will try to figure out that09:27
gmannjohnthetubaguy: not those.09:27
gmannjohnthetubaguy: let me see again09:28
johnthetubaguytbh: let me know if that doesn't make sense, there are some more details on the process here: https://wiki.openstack.org/wiki/Nova/Liberty_Release_Schedule#How_do_I_get_my_code_merged.3F09:29
johnthetubaguygmann: https://review.openstack.org/#/c/214085/8/nova/tests/functional/api_sample_tests/legacy_v2/api_samples_test_base.py,cm and https://review.openstack.org/#/c/214085/8/nova/tests/functional/api_sample_tests/api_sample_base.py,cm might also be importnat09:29
gmannjohnthetubaguy: i had /v2 on paste.ini just checking for /v3 removal from it09:29
gmannjohnthetubaguy: yea09:30
johnthetubaguygmann: I am still not sure we have all the things tested quite correctly after my change, its a bit confusing09:31
tbhjohnthetubaguy:  I am thinking like, I will send host_ip from nova-api, and then I will test all the other api calls09:31
tbhis that the correct way?09:32
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: Implements detach_interface method  https://review.openstack.org/21449309:32
*** yamahata has quit IRC09:32
gmannjohnthetubaguy: for me too. may be it will clear if we make your changes on top of test merge one and see whats left?09:33
johnthetubaguytbh: not sure I understand, sounds like you want to add host_ip into one of the existing API calls, you need to submit a patch to this repo (via gerrit) to add the spec: https://github.com/openstack/nova-specs09:33
johnthetubaguygmann: there are more details here: http://docs.openstack.org/infra/manual/developers.html#working-on-specifications-and-blueprints09:34
johnthetubaguygmann: possibly09:34
johnthetubaguygmann: feels like we need some kind of coverage check, ideally09:34
gmannjohnthetubaguy: hummm09:35
tbhjohnthetubaguy: yeah09:35
johnthetubaguygmann: I don't get why the tests are posting to /v2 with microversions attached, that seems very odd09:38
gmannjohnthetubaguy: that actually goes for v21 through api fixture09:38
gmannjohnthetubaguy: we can make it on  /v2 and /v2.1 separately for v2 tests and v2.1 and microversion one resp09:40
johnthetubaguygmann: I think that would be better, run the base version on /v2 and /v2.1 then microversions only on /v2.109:40
*** gszasz has joined #openstack-nova09:40
johnthetubaguygmann: but anyways, I am still confused with my 400 error09:41
openstackgerritClaudiu Belu proposed openstack/nova: Hyper-V: Implements detach_interface method  https://review.openstack.org/21449309:41
gmannjohnthetubaguy: ok09:41
gmannjohnthetubaguy: microversion runs only for v21 Router not for v209:41
*** y_sawai has quit IRC09:42
gmannbut used /v2 which is being fixtured to "/v2: openstack_compute_api_v21"09:42
johnthetubaguygmann: yeah thats the really confusing bit09:42
johnthetubaguygmann: its not a configuration we would ever want anyone to run in production09:43
gmannjohnthetubaguy: yea, i will give try to make those clear, microversion on .v2.1 without any fixture thing09:43
johnthetubaguygmann: will need the fixture thing to test legacy_v2 code, after my paste.ini changes remove that09:43
gmannjohnthetubaguy: ahh i remember now. actually issue is with version in url which is going to be in sample files link09:44
johnthetubaguygmann: given we need to test: v2_legacy, v2.1 in legacy compat mode, v2.1 base09:44
johnthetubaguygmann: oh dear09:44
gmannjohnthetubaguy: thats why we run all on /v2 and make fixture for v2.1 and microversion to avoid different sample files09:44
alex_xunot sure the context, just note the current sample tests run on legacy compat mode also09:45
gmannjohnthetubaguy: alex_xu:  yea09:46
gmannwe do run all tests for all those 3  - v2_legacy, v2.1 in legacy compat mode, v2.1 base09:46
johnthetubaguyyeah, just being sure not to break that09:46
gmannjohnthetubaguy: not sure why you get 400 for remote console tests :).09:48
johnthetubaguygmann: so in my enviroment ConsolesV28SampleJsonTests needs the extra line: extra_extensions_to_load = ["os-access-ips"]09:48
johnthetubaguyfoudn it09:48
openstackgerritJohn Garbutt proposed openstack/nova: api: tidy up paste.ini  https://review.openstack.org/21408509:49
*** markvoelker has joined #openstack-nova09:49
gmannjohnthetubaguy: ahhh , for v21 os-access-ips is separate extension09:49
*** sudipto has joined #openstack-nova09:50
johnthetubaguyso with that, am I going for my first coffee of the day, that explains why my brain is going slowly09:50
gmannjohnthetubaguy: i put that in base class in https://review.openstack.org/#/c/214985/09:51
*** aix has quit IRC09:51
johnthetubaguygmann: makes sense09:51
johnthetubaguygmann: this is just really a hack so I can find out how far away we are really, the tempest tests are the big issue I feel09:52
gmannjohnthetubaguy: :) yea09:52
johnthetubaguygmann: alex_xu: I put up these hacked up changes, in case its useful: https://review.openstack.org/#/q/topic:bp/nova-api-deprecate-extensions,n,z09:52
gmannjohnthetubaguy: most of changes are there on tests merge thing. do you want to put your changes dependent on those or those are lot more as dependent?09:53
*** markvoelker has quit IRC09:54
*** romainh has quit IRC09:54
johnthetubaguygmann: I think its best not to merge those for now, thinking about it09:55
johnthetubaguygmann: mine can just say on its own while we work out what to do with those tempest failures09:55
gmannjohnthetubaguy: ok. sounds good.09:56
*** dims has joined #openstack-nova09:56
johnthetubaguyalex_xu: that v3 enable setting is still used in the v2.1 API router :(09:57
alex_xujohnthetubaguy: :(09:58
*** fawadkhaliq has joined #openstack-nova09:58
alex_xujohnthetubaguy: but it is ok for deprecate09:58
*** jasonpan has quit IRC10:00
johnthetubaguyalex_xu: yeah, we have to deprecate configs before we can remove them (as we want people to be able to upgrade without changing their config from the last release)10:02
alex_xujohnthetubaguy: just one more comment for the log msg10:02
*** Kennan2 has joined #openstack-nova10:03
* alex_xu can stop the baby cry~...10:03
johnthetubaguyalex_xu: yeah, the log messages could be a lot better!10:03
*** Kennan has quit IRC10:04
gmannjohnthetubaguy: need fixture for v2_legacy tests - https://review.openstack.org/#/c/214085/9/nova/tests/functional/api_sample_tests/api_sample_base.py10:04
johnthetubaguygmann: my api-paste.ini defaults to /v2 having v2.1 in legacy mode, so its not needed10:06
johnthetubaguygmann: I had to add a new fixture to move /v2 to run the legacy v2 code base10:06
*** ociuhandu has joined #openstack-nova10:06
johnthetubaguygmann: oh, I think I see what you mean now10:06
gmannjohnthetubaguy: yea for later one i think you did10:07
gmannok10:07
gmannjohnthetubaguy: ApiPasteV2CompatibleFixture may be you want to rename it10:07
gmannfor v2_legacy now10:07
johnthetubaguygmann: I have already done that: https://review.openstack.org/#/c/214085/9/nova/tests/functional/api_paste_fixture.py,cm10:08
johnthetubaguygmann: so I agree :)10:08
gmannyea10:08
gmannjohnthetubaguy: ahh i see :)10:09
*** haomaiwang has quit IRC10:09
johnthetubaguygmann: I kinda want the diff to show that stuff deleted, but yeah, it doesn't10:09
*** haomaiwang has joined #openstack-nova10:10
*** fawadkhaliq has quit IRC10:13
johnthetubaguygmann: cools, thanks for those, good spot there, I will wait to see what jenkins says, then update that from your comments10:13
gmannjohnthetubaguy: ok10:13
openstackgerritSimona Iuliana Toader proposed openstack/nova: Hyper-V: Adds Hyper-V UEFI Secure Boot  https://review.openstack.org/20958110:13
openstackgerritSimona Iuliana Toader proposed openstack/nova: Hyper-V: refactors utilsfactory  https://review.openstack.org/21452710:13
*** markus_z has quit IRC10:15
*** lyanchih has quit IRC10:16
openstackgerritJohn Garbutt proposed openstack/nova: api: deprecate the concept of extensions in v2.1  https://review.openstack.org/21459210:18
*** PrabAsh has quit IRC10:20
*** ftersin_ has joined #openstack-nova10:20
*** mnestratov has joined #openstack-nova10:23
*** atuvenie has quit IRC10:24
openstackgerritJohn Garbutt proposed openstack/nova: api: deprecate the api v2 extension configuration  https://review.openstack.org/21460110:25
johnthetubaguyalex_xu: gmann: I updated those two patches, added a slightly better log message for the first one10:25
openstackgerritSudipta Biswas proposed openstack/nova: Fix scheduler code to use monitor metric objects  https://review.openstack.org/21331710:27
alex_xujohnthetubaguy: cool, will check again after i get baby sleep success...10:32
*** xgsa has quit IRC10:33
openstackgerritSimona Iuliana Toader proposed openstack/nova: Hyper-V: Adds Hyper-V UEFI Secure Boot  https://review.openstack.org/20958110:36
openstackgerritSimona Iuliana Toader proposed openstack/nova: Hyper-V: refactors utilsfactory  https://review.openstack.org/21452710:36
*** fawadkhaliq has joined #openstack-nova10:38
gmannjohnthetubaguy: looks good, just 1 suggestion if that make sense10:38
*** lpetrut has joined #openstack-nova10:38
*** xgsa has joined #openstack-nova10:39
johnthetubaguygmann: thats cool, will take a look in a bit.10:47
*** haomaiwang has quit IRC10:48
openstackgerritSergey Nikitin proposed openstack/nova: Switch to oslo.cache lib  https://review.openstack.org/20304910:51
*** shaohe_feng has joined #openstack-nova10:53
*** shuquan has joined #openstack-nova10:53
*** sudipto has quit IRC10:55
*** sudipto has joined #openstack-nova10:56
abhishekksnikitin: hi you around?10:56
openstackgerritwanghao proposed openstack/nova: Add nova-manage command for cleaning attachment  https://review.openstack.org/18453710:56
*** unicell1 has joined #openstack-nova10:57
*** unicell has quit IRC10:58
*** shuquan has quit IRC10:58
*** mwagner_lap has quit IRC11:01
*** unicell1 has quit IRC11:04
*** alex_klimov has joined #openstack-nova11:04
*** unicell has joined #openstack-nova11:04
*** apuimedo|away is now known as apuimedo11:04
*** ftersin_ has quit IRC11:06
alex_xujohnthetubaguy: sdague we add new status but without microversions in this patch https://review.openstack.org/#/c/212065/6 is it ok?11:07
openstackgerritGhanshyam Mann proposed openstack/nova: Cleanup for merging v2 and v2.1 functional tests.  https://review.openstack.org/21498511:10
openstackgerritSudipta Biswas proposed openstack/nova: cleanup NovaObjectDictCompat from quota  https://review.openstack.org/21507811:13
*** unicell has quit IRC11:13
*** unicell has joined #openstack-nova11:13
*** jamie_h has quit IRC11:14
*** dims has quit IRC11:15
*** dims has joined #openstack-nova11:16
*** aix has joined #openstack-nova11:16
*** shaohe_feng has quit IRC11:17
*** otter768 has joined #openstack-nova11:17
*** takedakn has joined #openstack-nova11:18
johnthetubaguyalex_xu: I thought we said that is not OK11:18
johnthetubaguyalex_xu: not sure I see how that returns a new status though11:20
*** dims has quit IRC11:20
alex_xujohnthetubaguy: the 'queued' status is new11:21
johnthetubaguyalex_xu: ah, thats in the migration, I guess we do have an API to list those now11:22
*** smatzek has joined #openstack-nova11:22
*** otter768 has quit IRC11:22
alex_xujohnthetubaguy: yea. https://github.com/openstack/nova/blob/master/doc/api_samples/os-migrations/migrations-get.json11:22
johnthetubaguyalex_xu: that migration stuff needs sorting out with constants, etc, so its all more obvious11:22
alex_xujohnthetubaguy: yea11:23
alex_xujohnthetubaguy: so what we can do, revert it?11:23
openstackgerritOpenStack Proposal Bot proposed openstack/nova: Updated from global requirements  https://review.openstack.org/21499111:24
*** baoli has joined #openstack-nova11:25
johnthetubaguyalex_xu: we can, I would send an email to the approvers11:27
openstackgerritjichenjc proposed openstack/nova: Prevent build more than one default ephemeral disk  https://review.openstack.org/21376211:27
johnthetubaguy(and author)11:27
*** exploreshaifali has joined #openstack-nova11:29
alex_xujohnthetubaguy: ok11:29
*** baoli has quit IRC11:30
*** markus_z has joined #openstack-nova11:30
openstackgerritSudipta Biswas proposed openstack/nova: cleanup NovaObjectDictCompat from hv_spec  https://review.openstack.org/21509411:31
*** ftersin_ has joined #openstack-nova11:31
*** scheuran has joined #openstack-nova11:32
gmannjohnthetubaguy: just commentd on paste.ini patch failure11:32
*** dims has joined #openstack-nova11:33
*** dprince has joined #openstack-nova11:35
*** mhorban has quit IRC11:35
*** markvoelker has joined #openstack-nova11:35
*** Murali has quit IRC11:35
*** mhorban has joined #openstack-nova11:36
*** lucasagomes is now known as lucas-hungry11:37
tbhjohnthetubaguy: This is the spec I have written now https://paste.kde.org/p9fammubr, I tried to issue git review, but I got network unreachable error11:39
*** takedakn has quit IRC11:39
*** markvoelker has quit IRC11:40
tbhjohnthetubaguy: I just added ssh key, may be it takes time to reflect11:40
*** Marga_ has quit IRC11:42
*** alexschm has quit IRC11:42
*** Marga_ has joined #openstack-nova11:42
*** bkopilov has quit IRC11:42
*** rajesht has joined #openstack-nova11:42
johnthetubaguytbh: it sounds like you are behind a firewall, but I am not 100% sure11:43
*** thorst has joined #openstack-nova11:43
*** alexschm has joined #openstack-nova11:43
openstackgerritjichenjc proposed openstack/nova: conductor: update comments for rpc and use object  https://review.openstack.org/18506811:45
*** snikitin has joined #openstack-nova11:45
*** salv-orlando has quit IRC11:45
snikitindims: Dims, could you look at https://review.openstack.org/#/c/203049/ ? I added using of memcached in service groups11:46
*** afazekas has quit IRC11:46
openstackgerritAlex Xu proposed openstack/nova: Revert "Limit parallel live migrations in progress"  https://review.openstack.org/21510011:46
dimssnikitin: do we add new configuration options? would need a DocImpact tag11:47
snikitindims: ok. I'll add11:47
*** dansmith has quit IRC11:47
alex_xujohnthetubaguy: ^11:47
dimssnikitin: thanks11:48
*** dansmith has joined #openstack-nova11:48
openstackgerritSergey Nikitin proposed openstack/nova: Switch to oslo.cache lib  https://review.openstack.org/20304911:48
*** dansmith is now known as Guest9532811:48
*** romainh has joined #openstack-nova11:49
snikitindims: Done11:49
dimssnikitin: i am not happy about having to do the set_default in https://review.openstack.org/#/c/203049/10/nova/utils.py,cm11:49
dimssnikitin: can you please talk to stevemar and bknudson to see if there's a better wat11:49
dimsway11:49
dimssnikitin: you can start by logging a bug and ping them when they come online in a few hours please11:50
snikitindims: Ok, thank you11:50
dimssnikitin: thanks for the hard work, it's coming along nicely11:52
openstackgerritAlexander Schmidt proposed openstack/nova: libvirt: enable live migration with serial console  https://review.openstack.org/19103511:52
openstackgerritAlexander Schmidt proposed openstack/nova: libvirt: reallocate TCP ports for serial console during live migration  https://review.openstack.org/21510211:52
openstackgerritMarian Horban proposed openstack/nova: Added processing /compute URL  https://review.openstack.org/19530311:53
*** kolesovdv has joined #openstack-nova11:53
snikitindims: Hope we merge it soon :)11:53
dimssnikitin: please add a better commit message with details on the changes made11:53
snikitindims: Ok, I'll add11:54
*** VikasC has joined #openstack-nova11:54
dimsthanks snikitin!11:54
snikitindims: thanks for review!11:55
openstackgerritAlexander Schmidt proposed openstack/nova: WIP libvirt: Acquire TCP ports for console during live migration  https://review.openstack.org/21510211:55
*** atuvenie has joined #openstack-nova11:55
*** emagana has quit IRC11:56
openstackgerritGhanshyam Mann proposed openstack/nova: Share server POST sample file for microversion too  https://review.openstack.org/21510411:58
*** lpetrut has quit IRC12:02
jaypipesmikal, alaski, Vek,  other cores: rajesht's patch here: https://review.openstack.org/#/c/212374/ looks good to go now. reviews welcome.12:03
*** derekh is now known as derekh_afk12:03
rajeshtjaypipes: t12:04
rajeshtjaypipes: Thanks :)12:04
jaypipesrajesht: of course, no worries :)12:06
rajeshtjaypipes: I would request you to review dependent patch as well, if you have time  :)12:07
*** b3nt_pin is now known as beagles12:07
jaypipesrajesht: yup, it's on the queue :)12:08
rajeshtjaypipes: thanks a lot :)12:08
*** venkat_p has quit IRC12:09
*** salv-orlando has joined #openstack-nova12:11
*** shyu has quit IRC12:16
*** lpetrut has joined #openstack-nova12:17
*** haigang has quit IRC12:18
*** erlon has joined #openstack-nova12:19
*** takedakn has joined #openstack-nova12:20
*** burgerk has joined #openstack-nova12:21
*** asettle has joined #openstack-nova12:22
*** edmondsw has joined #openstack-nova12:23
*** asettle has quit IRC12:23
*** fawadkhaliq has quit IRC12:24
*** aysyd has joined #openstack-nova12:24
*** VikasC has quit IRC12:28
*** ociuhandu_ has joined #openstack-nova12:32
*** emagana has joined #openstack-nova12:32
*** ociuhandu has quit IRC12:33
*** ociuhandu_ is now known as ociuhandu12:33
*** unicell1 has joined #openstack-nova12:34
tbhjohnthetubaguy: still I have connectivity issue, I think you are correct12:34
tbhtbh: once the spec got approved, I will assign bp to myself12:34
*** unicell has quit IRC12:34
*** signed8bit has joined #openstack-nova12:36
*** exploreshaifali has quit IRC12:37
*** mwagner_lap has joined #openstack-nova12:37
*** lpetrut has quit IRC12:39
*** markvoelker has joined #openstack-nova12:39
openstackgerritClaudiu Belu proposed openstack/nova: WIP: Hyper-V: determine vm_com_pipe error during spawn  https://review.openstack.org/21372012:41
alex_xuanyone know Anne Genttle in which timezone?12:42
*** mriedem has joined #openstack-nova12:42
johnthetubaguyalex_xu: central, she is in austin texas I think12:43
alex_xujohnthetubaguy: thanks12:44
alex_xuemm... I guess need waiting for more few hours12:46
*** lucas-hungry is now known as lucasagomes12:48
*** garyk has joined #openstack-nova12:49
*** mriedem has quit IRC12:49
*** VW_ has joined #openstack-nova12:49
*** annegentle has joined #openstack-nova12:50
*** mriedem has joined #openstack-nova12:51
*** dave-mccowan has joined #openstack-nova12:52
*** jistr is now known as jistr|mtg12:53
*** aix has quit IRC12:54
*** signed8bit is now known as signed8bit_ZZZzz12:55
*** emagana has quit IRC12:57
openstackgerritwanghao proposed openstack/nova: Add nova-manage command for cleaning attachment  https://review.openstack.org/18453712:57
mriedemhas the xenproject ci gone off the rails?12:57
mriedemoh awesome we have a new 500 in the gate12:59
mriedem+ neutron12:59
*** lpetrut has joined #openstack-nova12:59
*** aix has joined #openstack-nova13:00
*** signed8bit_ZZZzz is now known as signed8bit13:00
*** wanghua has quit IRC13:00
*** rlrossit has joined #openstack-nova13:00
*** burgerk has quit IRC13:02
alaskimriedem: I'm not aware of any rax plans for libvirt + lxc CI.  but the containers folks are hidden away in a room somewhere and communicate with the outside world much afaict.  I can try to send a telegram or something to see if they've discussed it13:04
alaski*don't communicate13:04
mriedemsmoke signals13:04
mriedemcarrier pigeon13:04
*** VW_ has quit IRC13:05
mriedemthere are still remote uncontacted tribes living in the amazon rainforest you konw13:05
alaskiI saw a sign for someone selling homing pigeons about a week ago.  I seriously considered stopping in13:06
*** tbh has quit IRC13:06
mriedemha13:06
mriedemmakes a great gift13:06
*** dustins has joined #openstack-nova13:07
alaskiheh13:07
*** hhuang has quit IRC13:07
*** VW_ has joined #openstack-nova13:08
*** jecarey has joined #openstack-nova13:08
*** raildo-afk is now known as raildo13:10
*** VW_ has quit IRC13:14
*** ijuwang has joined #openstack-nova13:14
*** VW_ has joined #openstack-nova13:15
*** bkopilov has joined #openstack-nova13:15
*** rfolco has joined #openstack-nova13:15
*** otter768 has joined #openstack-nova13:18
*** ajayaa has quit IRC13:18
*** krtaylor has quit IRC13:18
*** VW_ has quit IRC13:19
*** ishant has quit IRC13:21
*** aagrawal has quit IRC13:22
*** wanghua has joined #openstack-nova13:22
*** otter768 has quit IRC13:23
alex_xuannegentle: hi, are you around?13:24
*** fawadkhaliq has joined #openstack-nova13:24
*** xyang1 has joined #openstack-nova13:25
*** dustins has quit IRC13:26
*** dustins has joined #openstack-nova13:26
*** fawadkhaliq has quit IRC13:28
*** jamielennox is now known as jamielennox|away13:30
*** zul has joined #openstack-nova13:30
*** krtaylor has joined #openstack-nova13:31
*** abhishekk has quit IRC13:32
mriedemrlrossit: https://bugs.launchpad.net/nova/+bug/148703813:35
openstackLaunchpad bug 1487038 in OpenStack Compute (nova) "nova.exception._cleanse_dict should use oslo_utils.strutils._SANITIZE_KEYS" [Low,Confirmed]13:35
mriedemdo it13:35
mriedem:)13:35
*** dims_ has joined #openstack-nova13:35
rlrossitmriedem: I'll take it, but I'm also building up a rant against this host_manager stuff13:35
*** zzzeek has joined #openstack-nova13:36
*** eharney has joined #openstack-nova13:37
*** bfic has joined #openstack-nova13:38
rlrossitmriedem: does it bother you as much as it bothers me that https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L195-L214 is setting self.metrics within a private function, instead of returning the metrics and setting them within update_from_compute_node() (where a bunch of other self.* are set)?13:38
rlrossitbauzas: ^13:39
*** diogogmt has joined #openstack-nova13:39
*** dims has quit IRC13:39
mriedemrlrossit: well it's called from update_from_compute_node which is called from a couple of places13:40
mriedemand update_from_compute_node is setting a bunch of attributes on the HostState object13:40
mriedemi guess you could argue that _update_metrics_from_compute_node could just return the metrics dict and the caller could set it13:40
rlrossitmriedem: dingdingding13:40
rlrossitmakes unit tests a lot easier13:40
mriedemyeah probably13:41
rlrossitwhich is what I was planning on doing in the first place13:41
mriedemhaving said that, it's not the worst thing that bothers me right now13:41
openstackgerritjichenjc proposed openstack/nova: Disallow query deleted flavor  https://review.openstack.org/21513313:43
*** dims_ has quit IRC13:44
*** mdrabe has joined #openstack-nova13:44
*** ccarmack has joined #openstack-nova13:45
*** tpeoples has joined #openstack-nova13:45
*** dims has joined #openstack-nova13:46
*** signed8bit is now known as signed8bit_ZZZzz13:46
*** ahilan has joined #openstack-nova13:47
*** fawadkhaliq has joined #openstack-nova13:48
*** signed8bit_ZZZzz is now known as signed8bit13:49
*** signed8bit is now known as signed8bit_ZZZzz13:50
*** thangp has joined #openstack-nova13:51
*** xyang1 has quit IRC13:52
*** ivasev has joined #openstack-nova13:52
bauzasrlrossit: I tend to agree with mriedem, that's not exactly something really mandatory to do13:53
*** haomaiwang has joined #openstack-nova13:53
openstackgerritjichenjc proposed openstack/nova: Disallow query deleted flavor in nova flavor-show  https://review.openstack.org/21513313:53
bauzasrlrossit: here, the main problem was due to the fact that the unittest was changed to fit the convience of the proposer13:54
bauzasat the convenience even13:54
rlrossitbauzas: true, but I also think there should be a little more unit testing around that area13:54
bauzasrlrossit: it can certainly do13:55
mriedemnova meeting in 5 minutes13:55
*** obondarev_ has joined #openstack-nova13:56
*** tdurakov has quit IRC13:58
*** tdurakov has joined #openstack-nova13:59
*** obondarev has quit IRC13:59
*** xyang1 has joined #openstack-nova13:59
*** browne has joined #openstack-nova13:59
johnthetubaguymikal tjones cburgess jgrimm adrian_otto funzo mjturek jcookekhugen irina_pov krtaylor danpb alexpilotti flip214 jaypipes garyk edleafe dims moshele anteaya Nisha sileht claudiub lxsli neiljerram markus_z swamireddy alevine tonyb andreykurilin ndipanov sc68cal akuriata artom jlvillal mnestratov kashyap aloga rgeragnov bauzas xyang tpatil med_ nic scottda nagyz dannywilson belliott dguitarbite nova-meeting time13:59
*** pkholkin has quit IRC13:59
*** pkholkin has joined #openstack-nova14:00
jrolland here I thought we had a spammer until I saw the end14:00
*** Guest95328 is now known as dansmith14:00
*** wanghua has quit IRC14:02
bauzasrlrossit: well, if you want to get MHO, I just think all of that is just coming from the fact that we're poorly interfacing HostState14:03
bauzasrlrossit: there is no reason to have a python object while we already have a NovaObject (ComputeNode)14:03
*** diogogmt has quit IRC14:04
*** aix has quit IRC14:04
rlrossitbauzas: you're saying that HostState is kind of being a redundant wrapper of ComputeNode?14:04
bauzasso that error couldn't happen because the field would be coerced - that's where I would love to see some improvements instead of adding more coverage14:04
bauzasrlrossit: totally14:04
openstackgerritZhenyu Zheng proposed openstack/nova: Make query to quota usage table order preserved.  https://review.openstack.org/21358414:05
*** signed8bit_ZZZzz has quit IRC14:06
*** ctrath has joined #openstack-nova14:06
*** afazekas has joined #openstack-nova14:07
*** jistr|mtg is now known as jistr14:08
*** aix has joined #openstack-nova14:09
*** haomaiwang has quit IRC14:09
jrolledleafe: hey, looking at https://review.openstack.org/#/c/111837/ and the patch that supercedes it... we should be able to close https://bugs.launchpad.net/nova/+bug/1352510 I think?14:10
openstackLaunchpad bug 1352510 in OpenStack Compute (nova) "Delete and re-add of same node to compute_nodes table is broken" [Medium,In progress] - Assigned to Ed Leafe (ed-leafe)14:10
*** haomaiwang has joined #openstack-nova14:10
*** ajayaa has joined #openstack-nova14:11
*** lyanchih has joined #openstack-nova14:13
edleafejroll: I'll look into it after the nova meeting14:14
markus_zjaypipes: You +W'ed it a few days ago but it ran into a merge conflict: https://review.openstack.org/#/c/206631/14:14
*** signed8bit has joined #openstack-nova14:14
jrolledleafe: yeah no rush14:14
*** kiran-r has joined #openstack-nova14:14
jrolljust looking at bug lists for fun14:14
*** ahilan has quit IRC14:15
*** kiran-r has quit IRC14:15
*** ftersin_ has quit IRC14:15
*** garyk has quit IRC14:15
*** burgerk has joined #openstack-nova14:15
*** diogogmt has joined #openstack-nova14:17
jaypipesmarkus_z: done.14:17
*** zul has quit IRC14:17
markus_zjaypipes: Thanks!14:17
*** mtanino has joined #openstack-nova14:17
openstackgerritj_king proposed openstack/nova: libvirt: Add wait for vif unplug events on destroy  https://review.openstack.org/21012614:20
*** lucasagomes is now known as lucas-afk14:21
*** drankis has joined #openstack-nova14:25
*** vishwanathj has joined #openstack-nova14:27
*** sayali has quit IRC14:27
*** vivekd has joined #openstack-nova14:29
*** albertom-afk is now known as albertom14:31
*** bfic has quit IRC14:31
*** fawadkhaliq has quit IRC14:33
*** ftersin_ has joined #openstack-nova14:33
*** diogogmt has quit IRC14:34
sudiptodims, I think we can close this one out: https://review.openstack.org/#/c/209499/314:36
*** tonytan4ever has joined #openstack-nova14:36
*** lpetrut has quit IRC14:37
*** lpetrut has joined #openstack-nova14:38
* jaypipes throws on some deadmau514:39
dimssudipto: it's not a bug :)14:39
dimssudipto: but you are on the right track14:39
dimssudipto: this one is a feature and it does not have DocImpact either14:39
sudiptodims, damn yeah - i thought Jay had them under a bug...sorry about that.14:40
* dims pokes jaypipes :)14:40
apmeltonmriedem: https://gist.github.com/ramielrowe/081deaf0c6b79aec689014:41
jichenmriedem: please let me know when you submit the update for the devref, so I can change my patch accordingly ...14:41
jaypipesdims: got it. should I create a bug and --amend the commit with a DocImpact?14:41
sudiptodims, https://review.openstack.org/#/c/156126/14:41
*** cfriesen has joined #openstack-nova14:41
jichenmriedem: or I can do it if you are ok14:41
*** tsekiyama has joined #openstack-nova14:41
sudiptothis one seems very trivial and has 3 +2s already14:41
apmeltonmriedem: there's a couple issues I've noticed, the first is with nbd devices getting leaked14:41
dimsjaypipes: for it to be added to the trivial list14:42
jaypipesdims: k, gimme a few minutes.14:42
dimsat least DocImpact otherwise i think14:42
sudiptoWonder why it's not merged yet...14:42
mriedemjichen: yeah working on it now14:42
apmeltonmriedem: the other issue is that network on the public interface just seems to drop, and I haven't been able to track down what actually causes that14:42
jichenmriedem: ok ~14:42
dimssudipto: this has a workflow +1, so just needed a "recheck"14:42
openstackgerritJohn Garbutt proposed openstack/nova: Xenapi: Handle missing aggregate metadata on startup  https://review.openstack.org/15612614:42
sudiptodims, yeah - ok let me put it there then?14:43
dimssudipto: no need to add to etherpad14:43
mriedemapmelton: ok, any lxc job would definitely start out on the experimetnal queue14:43
mriedemor non-voting check14:43
*** signed8bit has quit IRC14:43
mriedemLIBVIRT_TYPE=lxc - heh, about what i expected but wasn't sure14:43
apmeltonyea, it's surprisingly simple at this point14:44
*** signed8bit has joined #openstack-nova14:44
mriedemapmelton: i'm assuming that's all nova-network right?14:44
apmeltonyea14:44
mriedemok, i'll see if i can hack up a project-config change this afternoon14:45
mriedemwe can keep the tempest blacklist regex in nova's tree like we do for cells14:45
*** dustins has quit IRC14:45
apmeltonoh mriedem, there's another step I always forget about14:46
*** dustins has joined #openstack-nova14:46
*** takedakn1 has joined #openstack-nova14:46
apmeltonthe default devstack doesn't enable the nbd module14:46
mriedemb/c it uses libguestfs14:46
*** kolesovdv has quit IRC14:46
mriedembecause nbd + neutron kills the ubuntu kernel14:46
mriedemat least in 12.0414:46
apmeltoninteresting14:47
apmeltonlibguestfs is only used for injecting into the guest filesystem right?14:47
mriedemthat was a major gate issue for neutron jobs back in havana/icehouse14:47
mriedemyeah14:47
apmeltongah14:47
mriedembut there is an nbd option there too14:47
mriedemhttp://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/mount/nbd.py14:48
apmeltonyup14:48
apmeltonthe issue for lxc is it doesn't mount, modify the filesystem, then unmount14:48
apmeltonit actually needs to leave it connected since the container is just running on the filesystem14:48
johnthetubaguybauzas: alaski: how are with flavor or no flavor in here? https://review.openstack.org/#/c/145528/4514:49
mriedemapmelton: hmm, can we use the loop module? http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/mount/api.py#n5414:49
mriedemif the image format is raw14:50
*** takedakn has quit IRC14:50
mriedemelse maybe this todo from danpb is a longer term fix http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/mount/api.py#n6614:50
*** derekh_afk is now known as derekh14:50
*** xgsa has quit IRC14:51
apmeltoninteresting14:51
alaskijohnthetubaguy: I would still prefer not to have flavor there, but what's there is a compromise that I'm good with14:51
apmeltonthat's actually what we've ended up doing14:51
*** raildo is now known as raildo-afk14:51
dimssudipto: got another one?14:51
*** exploreshaifali has joined #openstack-nova14:52
mriedemapmelton: so do you have out of tree patches to fix this stuff?14:52
sudiptodims, yeah reading through the entire change - https://review.openstack.org/#/c/190047/14:52
apmeltonmriedem: it doesn't really "fix" nbd14:52
apmeltonit just skips it14:52
mriedemapmelton: i guess my point is, can we get around not using nbd?14:52
johnthetubaguyalaski: is your preference to make the interface independent of the flavor concept? I prefer either to having both, it has to be said14:52
mriedemapmelton: since i think we actually want nbd to die14:52
apmeltonmriedem: we all do :)14:53
sudiptodims, it does look like a candidate to me ...14:54
mriedemapmelton: and guestfs doesn't work with lxc?14:54
apmeltonI haven't tried14:54
openstackgerritJay Pipes proposed openstack/nova: Allow compute monitors in different namespaces  https://review.openstack.org/20949914:54
jaypipesdims, sudipto: done ^14:54
mriedemapmelton: ok, i guess i can circle back on that14:54
apmeltonmriedem: if it's not meant to keep the image mounted for long periods of time, it probably won't be a great solution14:54
alaskijohnthetubaguy: yes.  I think having a relationship with flavor adds unneeded complexity as well.  But there's one filter that needs something that looks like flavor, so I think we need both for now14:54
*** signed8bit has quit IRC14:55
dimsthanks jaypipes. sudipto please add that to the etherpad14:55
alaskijohnthetubaguy: but longer term I think we can deprecate and remove flavor from this object14:55
*** signed8bit has joined #openstack-nova14:55
apmeltonmriedem: let me see if I can get the work on pulling that patch upstream prioritized14:55
jaypipesalex_xu, dansmith: https://review.openstack.org/#/c/209499/ ready for your re-review. thx in advance.14:55
*** raildo-afk has quit IRC14:55
sudiptodims, sure.14:55
mriedemapmelton: ok, cool, and i'll work on the job14:55
mriedemapmelton: do you also have devstack patches for getting nbd installed?14:56
apmeltonmriedem: the only caveat for our use case is that we're using LVM to back the images14:56
*** raildo-afk has joined #openstack-nova14:56
apmeltonmriedem: no14:56
johnthetubaguyalaski: hmm, the current one seemed to have removed everything from the object thats in the flavor, to make the flavor the default14:56
*** garyk has joined #openstack-nova14:56
dimssudipto: good one, only caveat in https://review.openstack.org/#/c/190047/ is that we are adding yet another flag, but let's add it to etherpad and nova cores can choose to move it to the invalid bucket14:56
sudiptodims, you mean a new flag in the nova.conf?14:57
openstackgerritMatt Riedemann proposed openstack/nova: Handle NotFound exceptions while processing network-changed events  https://review.openstack.org/21517014:57
alaskijohnthetubaguy: right.  I would have gone the other direction.  but what's important I think is the object interface, and we can modify the internals down the road14:57
dimssudipto: y14:57
sudiptodims, ok14:57
*** raildo-afk has quit IRC14:57
*** xgsa has joined #openstack-nova14:57
johnthetubaguyalaski: true14:57
*** raildo-afk has joined #openstack-nova14:57
*** annegentle has quit IRC14:58
*** raildo-afk is now known as raildo14:58
*** annegentle has joined #openstack-nova14:58
alex_xujaypipes: thanks15:00
sudiptodims, https://review.openstack.org/#/c/192043/15:00
*** alex_klimov has quit IRC15:01
*** zul has joined #openstack-nova15:02
*** exploreshaifali has quit IRC15:02
*** alexschm has quit IRC15:03
*** vivek_nandavanam has joined #openstack-nova15:03
alex_xujaypipes: just need s/obj.ext/ext.obj/ in https://review.openstack.org/#/c/209499/15:03
*** xgsa has quit IRC15:04
*** xgsa has joined #openstack-nova15:04
jaypipesalex_xu: gah, thxnk!15:04
alex_xujaypipes: np :)15:04
sudiptodims, https://review.openstack.org/#/c/188025/ this actually has a +2 from johnthetubaguy - and seems a trivial one - but guess the commit message and a bit of code comments is needed...should we still consider it?15:04
openstackgerritJay Pipes proposed openstack/nova: Allow compute monitors in different namespaces  https://review.openstack.org/20949915:05
jaypipesalex_xu: done :)15:05
*** shoutm has joined #openstack-nova15:05
alex_xujaypipes: cool, go to +115:05
*** fawadkhaliq has joined #openstack-nova15:06
dimshttps://review.openstack.org/#/c/192043/ has a todo from Alex, but yes, sure please add it to etherpad15:06
jaypipesalex_xu: ty sir! :)15:06
openstackgerritZhenyu Zheng proposed openstack/nova: CONF.allow_resize_on_same_host should check only once in controller  https://review.openstack.org/18623615:07
alex_xujaypipes: you are welcom, sir :)15:07
*** takedakn has joined #openstack-nova15:07
*** signed8bit has quit IRC15:08
*** alejandrito has joined #openstack-nova15:08
openstackgerritZhenyu Zheng proposed openstack/nova: Add UniqueConstraint for quota usages table  https://review.openstack.org/21290815:08
dimssudipto: use the "edit message" button and edit the message yourself. once the CI passes then you can add it to etherpad (https://review.openstack.org/#/c/188025/). leave a note to the original committer says, "hope its ok i fixed the commit message"15:08
*** hemnafk is now known as hemna15:08
dimssudipto: lot of times, the patch may be stale and ci jobs may fail, so need to wait15:09
*** takedakn1 has quit IRC15:09
*** haomaiwang has quit IRC15:09
*** haomaiwang has joined #openstack-nova15:10
*** dprince has quit IRC15:12
*** nkrinner has quit IRC15:14
edleafejroll: ok, got to look at that bug, and yeah, it should have closed it, but I forgot to put Closes-bug: in the commit message15:15
openstackgerritZhenyu Zheng proposed openstack/nova: Add UniqueConstraint for quota usages table  https://review.openstack.org/21290815:16
*** diogogmt has joined #openstack-nova15:16
*** burt has joined #openstack-nova15:17
jrolledleafe: no worries15:17
sudiptodims, sure15:18
jrolljohnthetubaguy: mind marking this bug fix committed/released/whatever as needed? https://bugs.launchpad.net/nova/+bug/135251015:18
openstackLaunchpad bug 1352510 in OpenStack Compute (nova) "Delete and re-add of same node to compute_nodes table is broken" [Medium,In progress] - Assigned to Ed Leafe (ed-leafe)15:18
jroll(or point me at the right person)15:18
*** otter768 has joined #openstack-nova15:19
johnthetubaguyjroll: sure, I can do, does it not give you the permissions for that?15:19
*** kfox1111 has quit IRC15:19
*** diogogmt has quit IRC15:19
*** eharney has quit IRC15:19
jrolljohnthetubaguy: maybe it does...15:20
*** shoutm has quit IRC15:20
openstackgerritSudipta Biswas proposed openstack/nova: libvirt: Fix KeyError during LXC instance boot  https://review.openstack.org/18802515:20
jrollI'm having trouble posting a comment though, timeouts :/15:20
jrolljohnthetubaguy: april 23 merge would be in kilo release, yeah?15:21
*** dannywilson has joined #openstack-nova15:21
johnthetubaguyjroll: it should be, I usually just check in github and look at the tags to see if it made it15:21
*** shaohe_feng has joined #openstack-nova15:21
*** browne has quit IRC15:21
jrollgood call15:22
*** jistr is now known as jistr|mtg15:22
*** wanghua has joined #openstack-nova15:22
*** diogogmt has joined #openstack-nova15:22
jrolljohnthetubaguy: weird, shows      12.0.0a0 12.0.0.0b2 12.0.0.0b115:22
jrollwhich is liberty afaik?15:22
johnthetubaguyjroll: yeah, odd15:23
jrollL1 that is15:23
johnthetubaguyjroll: we do branch quite early sometimes15:23
jroll2015.1.0 was tagged apr 3015:23
jrollar RCs branched?15:23
jrollare*15:23
johnthetubaguyyep15:23
jrollaha15:23
johnthetubaguybranch at RC115:23
jrollyeah, that missed15:23
*** otter768 has quit IRC15:23
openstackgerritClaudiu Belu proposed openstack/nova: WIP: Hyper-V: determine vm_com_pipe error during spawn  https://review.openstack.org/21372015:24
johnthetubaguyit was dam close15:24
jrollrc1 was april 1015:24
* jroll marks released anyway because l115:24
johnthetubaguysounds good15:24
jrolldone, sorry for the bother, didn't realize I could much with your bugs :)15:24
bauzasjohnthetubaguy: alaski: right, I totally agree with alaski but we needed some mitigation because of a stupid filter15:26
johnthetubaguybauzas: OK, there seem to be unit test fails further up that chain at the moment15:27
bauzasjohnthetubaguy: strange, Jenkins was happy yesterday15:28
bauzasjohnthetubaguy: but I'll dig into that15:28
*** moshele has quit IRC15:28
*** scheuran has quit IRC15:28
johnthetubaguybauzas: I think it was the update this morning that upset it, no worries15:28
bauzasI'm UTC-7 so I'll look at that15:29
*** VikasC has joined #openstack-nova15:31
*** wanghua has quit IRC15:32
*** zul has quit IRC15:33
*** andreykurilin__ has joined #openstack-nova15:34
*** xgsa has quit IRC15:34
*** haomaiwang has quit IRC15:35
openstackgerritMatt Riedemann proposed openstack/nova: Clarify API microversion docs around handling 500 errors  https://review.openstack.org/21519515:36
mriedemjaypipes: alex_xu: sdague: johnthetubaguy: alaski: dansmith: jichen: ^15:36
jaypipesdanke15:36
jichendanke schon15:36
mriedemjichen: you're practically german already :)15:36
*** r-daneel has joined #openstack-nova15:36
mriedemjust in time for october15:36
jichenmriedem: yeah :)15:36
*** takedakn has quit IRC15:36
rlrossitmriedem: I left a comment on https://bugs.launchpad.net/nova/+bug/1487038, but basically do we want to remove keys in _cleanse_dict() or do we want to censor the values to stay closer to strutils?15:38
openstackLaunchpad bug 1487038 in OpenStack Compute (nova) "nova.exception._cleanse_dict should use oslo_utils.strutils._SANITIZE_KEYS" [Low,Confirmed]15:38
*** VW_ has joined #openstack-nova15:39
*** rdopiera has joined #openstack-nova15:40
jaypipesmriedem: did you want to include in there something about never backporting any patch to a stable branch that contains a microversion bump?15:42
mriedemjaypipes: i wanted to leave backport stuff to a separate change15:43
mriedemsince i think that's it's own issue15:43
jaypipesmriedem: got it. ok, +W'ing this.15:43
mriedem+W already?15:43
mriedemw/o sdague?15:43
mriedemdamn15:43
mriedemor alex15:44
mriedemi figure we should have some more concensus15:44
jaypipesmriedem: k, removed.15:44
*** armax has joined #openstack-nova15:44
openstackgerritMatt Riedemann proposed openstack/nova: Clarify API microversion docs around handling 500 errors  https://review.openstack.org/21519515:44
mriedemgotta pull it from the queue15:44
openstackgerritMatt Riedemann proposed openstack/nova: Clarify API microversion docs around handling 500 errors  https://review.openstack.org/21519515:45
openstackgerritZhenyu Zheng proposed openstack/nova: Make query to quota usage table order preserved.  https://review.openstack.org/21358415:45
mriedemjaypipes: added kenichi15:45
mriedemand -2ed so there is time to review15:45
mriedemi know sean is in seattle with family this week15:45
*** mriedem is now known as mriedem_away15:46
mriedem_awayjaypipes: alaski: btw, feel free to reapply your +2s :)15:46
alaski403 over quota15:46
*** belmoreira has joined #openstack-nova15:49
*** mwagner_lap has quit IRC15:50
*** angdraug has joined #openstack-nova15:51
*** salv-orlando has quit IRC15:52
*** hogepodge has quit IRC15:53
openstackgerritVipin Balachandran proposed openstack/nova: VMware: Delete vmdk UUID during volume detach  https://review.openstack.org/20462215:55
*** armax has quit IRC15:56
*** redbeard has quit IRC15:56
*** AlexeyElagin has joined #openstack-nova15:57
*** belmoreira has quit IRC15:58
*** eharney has joined #openstack-nova15:59
*** salv-orlando has joined #openstack-nova15:59
dansmithalaski: I know you're 403, but is the request spec thing in the pipe already?15:59
dansmithalaski: I'd be +2, but think you should vote on it before we land it if at all possible15:59
alaskithe 403 was only for +2s to mriedem_away15:59
dansmithokay16:00
alaskiI'm finishing my rebase on top and then I'll look at it16:00
alaskithinking on it now I probably should have reviewed first...16:00
openstackgerritbharaththiruveedula proposed openstack/nova-specs: Returning Compute Node IP address in hypervisor list  https://review.openstack.org/21520716:00
bauzasalaski: dansmith: I need first to rebase and update the test_relationships stuff because there was an object bump16:00
dansmithalaski: well, bauzas is about to push again for a rebase test fix16:00
dansmithyeah, that16:00
alaskiokay, I'll review then16:01
*** rajesht has quit IRC16:01
*** VW_ has quit IRC16:02
*** SourabhP has joined #openstack-nova16:02
*** ftersin_ has quit IRC16:03
*** VW_ has joined #openstack-nova16:03
*** unicell has joined #openstack-nova16:03
*** unicell1 has quit IRC16:05
*** tbh has joined #openstack-nova16:05
*** Nikolay_St has quit IRC16:05
*** SourabhP_ has joined #openstack-nova16:05
alaskidansmith: bauzas lgtm.  I'll wait for the new push and +216:06
*** mdrabe has quit IRC16:06
tbhjohnthetubaguy: finally push the patch for nova-spec for review16:06
*** VW_ has quit IRC16:06
*** VW_ has joined #openstack-nova16:07
openstackgerritAndrea Rosa proposed openstack/nova: Correct a wrong docstring  https://review.openstack.org/21521516:07
bauzasalaski: uploading now16:07
*** SourabhP has quit IRC16:08
*** SourabhP_ is now known as SourabhP16:08
*** ftersin_ has joined #openstack-nova16:08
*** yamahata has joined #openstack-nova16:09
*** jistr|mtg is now known as jistr16:10
*** VW_ has quit IRC16:11
johnthetubaguytbh: cool, glad that worked out, sorry that was such a pain16:11
*** lsmola has quit IRC16:12
bauzasokay, rebase issue, alaski don't wait for me now16:12
tbhjohnthetubaguy: I came to home and then pushed for review :)16:13
*** shaohe_feng has quit IRC16:13
openstackgerritAndrew Laski proposed openstack/nova: WIP: Add BuildRequest table and object  https://review.openstack.org/21479216:13
openstackgerritAndrew Laski proposed openstack/nova: Add persistence to the RequestSpec object  https://review.openstack.org/21175316:13
openstackgerritAndrew Laski proposed openstack/nova: Move objects registration in tests directory  https://review.openstack.org/21215916:13
johnthetubaguytbh: ah, I remember when I had to do that while my company at the time updated their firewall!16:13
*** Apoorva has joined #openstack-nova16:14
tbhjohnthetubaguy: oh :), by the way who have the rights to assign the blueprint?16:14
*** SourabhP has quit IRC16:15
johnthetubaguytbh: I should be able to move the blueprint, you mean you want to become the assignee on that?16:15
*** tjones has joined #openstack-nova16:16
tbhjohnthetubaguy: yeah I want to work on this bp, as this is my first bp16:16
johnthetubaguytbh: OK, lets start with the spec review I guess, we can update that all later on if you like?16:17
tbhjohnthetubaguy: sure16:17
*** tjones has left #openstack-nova16:18
*** krtaylor has quit IRC16:18
*** mdrabe has joined #openstack-nova16:19
*** exploreshaifali has joined #openstack-nova16:20
*** moshele has joined #openstack-nova16:20
*** kiran-r has joined #openstack-nova16:20
*** Marga_ has quit IRC16:22
*** hogepodge has joined #openstack-nova16:23
*** vivekd has quit IRC16:23
*** zul has joined #openstack-nova16:24
*** lpetrut has quit IRC16:26
*** mriedem_away is now known as mriedem16:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests()  https://review.openstack.org/19920516:27
openstackgerritSylvain Bauza proposed openstack/nova: Create RequestSpec object  https://review.openstack.org/14552816:27
openstackgerritSylvain Bauza proposed openstack/nova: Consider that all scheduler calls are IO Ops  https://review.openstack.org/18848616:27
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler  https://review.openstack.org/18848716:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify left filters for RequestSpec  https://review.openstack.org/20267816:27
openstackgerritSylvain Bauza proposed openstack/nova: Prepare filters for using RequestSpec object  https://review.openstack.org/19791216:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object  https://review.openstack.org/20307816:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify metric-related filters for RequestSpec  https://review.openstack.org/20267616:27
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler  https://review.openstack.org/19905116:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify NUMA, PCI and num_instances filters for RequestSpec  https://review.openstack.org/20267716:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify Aggregate filters for RequestSpec  https://review.openstack.org/20267416:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify default filters for RequestSpec  https://review.openstack.org/20267516:27
openstackgerritSylvain Bauza proposed openstack/nova: Add RequestSpec methods for primitiving into dicts  https://review.openstack.org/18719716:27
openstackgerritSylvain Bauza proposed openstack/nova: Modify Affinity filter for RequestSpec  https://review.openstack.org/20267316:27
openstackgerritSylvain Bauza proposed openstack/nova: Use RequestSpec object in HostManager  https://review.openstack.org/19125116:27
openstackgerritSylvain Bauza proposed openstack/nova: Use the same pci_requests field for all filters and HostManager  https://review.openstack.org/19125016:27
*** oomichi has joined #openstack-nova16:28
bauzasanyone can put their dirty hands on ^16:28
*** mwagner_lap has joined #openstack-nova16:28
dansmithalaski: you first then I'll go16:28
*** meghal has joined #openstack-nova16:29
*** Marga_ has joined #openstack-nova16:29
openstackgerritAndrew Laski proposed openstack/nova: WIP: Add BuildRequest table and object  https://review.openstack.org/21479216:29
openstackgerritAndrew Laski proposed openstack/nova: Add persistence to the RequestSpec object  https://review.openstack.org/21175316:29
openstackgerritAndrew Laski proposed openstack/nova: Move objects registration in tests directory  https://review.openstack.org/21215916:29
*** salv-orlando has quit IRC16:30
*** _kiran_ has joined #openstack-nova16:30
*** kiran-r has quit IRC16:31
* alaski passes the baton to dansmith16:31
*** matrohon has quit IRC16:32
bauzasdims: I'm just working on deprecating opts, you probably remember a discussion we had like 3 weeks ago16:33
*** browne has joined #openstack-nova16:33
dimsbauzas: vaguely :)16:33
bauzasdims: just one q before I move on, what's about http://docs.openstack.org/developer/oslo.config/opts.html#oslo_config.cfg.DeprecatedOpt and how I can have 2 opts in 2 different modules, one being deprecated and one new ?16:33
mriedemedleafe: can we leave your 500 -> anything else for a separate change16:34
mriedem?16:34
mriedemi tend to agree with that,16:34
bauzasdims: shall I move the existing opt to the new module, then mention it as deprecated, provided it's kept in the right group .16:34
bauzas?16:34
mriedembut it doesn't necessarily need to be in my change for clarifying 500 to 400+16:34
*** kashyap has quit IRC16:34
*** whenry_ has joined #openstack-nova16:34
*** lucas-afk is now known as lucasagomes16:34
dimsbauzas: move existing opt to new module and use DeprecatedOpt to mention the name and group of the old being deprecated16:34
mriedemdansmith: but then you get sloppy seconds16:35
bauzasdims: by looking at Nova, it seems we don't heavily use that kind of DeprecatedOpt, but rather just modifies the help comment16:35
mriedemgross16:35
*** jwcroppe_ has quit IRC16:35
dansmithmriedem: ....16:35
mriedembauzas: DeprecatedOpt is for moving options between groups or renames16:35
bauzasdims: okay, that was my thoughts, but since I'm moving an opt from a scheduler namespace to a compute namespace, that's a bit harsh16:35
*** sahid has joined #openstack-nova16:35
bauzasmriedem: oh ok16:36
mriedemdeprecated_for_removal kwarg is for deprecating, for removal16:36
edleafemriedem: why separate? Why not just correct it in one?16:36
mriedemedleafe: b/c this specific change is for handling a 500, not a 50116:37
mriedem501 is a separate case and i think warrants a separate change and review16:37
bauzasmriedem: okay, so I won't use the DeprecatedOpt16:37
*** fawadk has joined #openstack-nova16:37
bauzasmriedem: but rather the kwarg you mentioned16:37
*** kashyap has joined #openstack-nova16:37
*** fawadkhaliq has quit IRC16:37
bauzasmriedem: dims: thanks16:37
ctrathis there documentation anywhere that explains where logs are dumped on tox -e py27 runs?16:38
*** dprince has joined #openstack-nova16:39
*** jichen has quit IRC16:39
rdopierato the console?16:40
mriedemno16:40
*** whenry_ has quit IRC16:40
ctrathrdopiera: anywhere.. file, console, device...16:40
mriedemthe subunit stream16:40
mriedemctrath: e.g. http://logs.openstack.org/06/183706/7/check/gate-nova-python27/743f296/testrepository.subunit.gz16:40
edleafemriedem: ok - done16:41
rdopieraah, I thought when you run it manually16:41
ctrathyes, when I run it in devstack, specifically16:41
edleafemriedem: my feeling is that any 5xx is not an expected response16:43
*** armax has joined #openstack-nova16:43
mriedemi disagree16:43
mriedemspecifically for 501 and 50316:43
*** drankis has quit IRC16:43
mriedem501 - these aren't the droids you're looking for16:43
mriedem503 - shit happens16:43
*** ftersin_ has quit IRC16:45
*** jistr has quit IRC16:46
johnthetubaguymriedem: I so want to see your full list of codes16:47
johnthetubaguy5xx - my bad, 4xx - your bad16:47
*** vivekd has joined #openstack-nova16:48
*** unicell has quit IRC16:48
mriedem3xx - are these even used?16:48
mriedemwe should use 402 more16:49
mriedempayment required16:49
openstackgerritClaudiu Belu proposed openstack/nova: WIP: Hyper-V: adds oslo.windows library  https://review.openstack.org/21210216:49
*** lpetrut has joined #openstack-nova16:49
bauzasdims: eh, I got again one ticket for you, what's https://github.com/openstack/nova/blob/master/nova/scheduler/opts.py ?16:50
bauzasdims: I looked at the comment, but I just discover that one16:50
*** lpetrut has quit IRC16:50
ctrathmriedem: don't forget about 418 - I'm a teapot16:50
*** lpetrut has joined #openstack-nova16:50
*** Kennan2 has quit IRC16:50
bauzasdims: it means that anytime that we change an opt, we need to change that as well ?16:51
*** lpetrut has quit IRC16:51
dimsbauzas: yes, for the oslo-config-generator to pick things up16:51
*** lpetrut has joined #openstack-nova16:51
*** VW_ has joined #openstack-nova16:51
*** lpetrut has quit IRC16:52
johnthetubaguybauzas: hence to push to move opts all into a more central location/directory16:52
bauzasokay16:52
*** chaochin has quit IRC16:52
bauzasgood to know, I missed that one16:52
*** park_ has quit IRC16:52
bauzasso any change to an opt should impact that file, gotcha16:53
bauzasat least additions16:53
*** mingy has joined #openstack-nova16:53
*** atuvenie has quit IRC16:53
*** ftersin_ has joined #openstack-nova16:55
*** SourabhP has joined #openstack-nova16:55
*** sahid has quit IRC16:56
*** armax has quit IRC16:57
jrollmriedem: just swap out 500 for 402, payment required in form of bug fixes16:58
*** romainh has left #openstack-nova16:58
mriedemjroll: that's what i tried to pull on cfriesen yesterday16:59
jrollhah16:59
mriedemapmelton: as you can see in https://review.openstack.org/#/c/214927/ - that method is pretty wonky, not even tested16:59
mriedembut has been around a long time16:59
*** derekh has quit IRC17:00
bauzasmmm, I'm really wondering on how to properly move one opt from one component to another and provide a nice compat'...17:01
*** tpeoples has quit IRC17:01
jrollbauzas: deprecated_group?17:01
jrollI feel like that's a thing17:01
*** VW_ has quit IRC17:01
*** armax has joined #openstack-nova17:02
*** VW_ has joined #openstack-nova17:02
*** afazekas has quit IRC17:02
bauzasjroll: well, the thing is - we need to send the batsignal to operators that they have to provide that conf flag not in the same nova.conf file17:03
jrolloh, different service17:03
bauzasjroll: I feel there is no alternative than just creating a new opt and explicitely duplicate that17:03
*** vilobhmm1 has joined #openstack-nova17:03
bauzasyeah, s/component/service my bad17:03
*** tsekiyama has quit IRC17:04
jrollyeah, I'm not sure :/17:04
bauzasokay, nvm, I'm gonna shot something and we'll see17:05
*** achanda has joined #openstack-nova17:05
*** Kennan has joined #openstack-nova17:05
*** jcoufal has quit IRC17:06
*** park_ has joined #openstack-nova17:06
*** VW_ has quit IRC17:07
apmeltonmriedem: it's tested here https://github.com/openstack/nova/blob/master/nova/tests/unit/virt/libvirt/test_driver.py#L10584-L1070517:07
*** SourabhP has quit IRC17:08
*** markus_z has quit IRC17:08
*** chaochin has joined #openstack-nova17:09
*** tpeoples has joined #openstack-nova17:10
*** nic has joined #openstack-nova17:10
*** Nikolay_St has joined #openstack-nova17:12
*** Kennan has quit IRC17:14
*** park_ has quit IRC17:14
*** chaochin has quit IRC17:14
*** lyanchih has quit IRC17:16
*** unicell has joined #openstack-nova17:16
*** oomichi has quit IRC17:17
openstackgerritEli Qiao proposed openstack/nova: Catch Exception when detach encryptied volume  https://review.openstack.org/21525217:17
*** chaochin has joined #openstack-nova17:19
*** otter768 has joined #openstack-nova17:20
*** Kennan has joined #openstack-nova17:20
*** park_ has joined #openstack-nova17:20
mingyedleafe: ping17:21
*** aix has quit IRC17:22
edleafemingy: https://blogs.gnome.org/markmc/2014/02/20/naked-pings/17:22
edleafemingy: but anyway... pong17:22
edleafemingy: did you have a chance to review?17:23
*** ssurana has joined #openstack-nova17:23
mingyedleafe: yeah so I checked this last night and you were right and I was wrong, it returns the empty generator object.17:23
mingyedleafe: turns out I misunderstood how the yield keyword works in python17:23
edleafemingy: yeah, it doesn't have to hit the yield to return a generator17:24
edleafemingy: so are we cool?17:24
mingywe're cool :)17:24
edleafe:)17:24
*** Kennan has quit IRC17:24
*** otter768 has quit IRC17:24
*** chaochin has quit IRC17:24
*** park_ has quit IRC17:25
*** signed8bit has joined #openstack-nova17:26
*** achanda has quit IRC17:27
openstackgerritRoman Bogorodskiy proposed openstack/nova: Sync L3Driver, NullL3 interface with LinuxNetL3  https://review.openstack.org/18900117:28
*** oomichi has joined #openstack-nova17:28
*** VikasC has quit IRC17:28
mingyas for the ping thing... I just kind of saw people doing it on irc so I followed suite. I always thought it was a mannerism, I never knew it did things to the client.17:29
mingybut yeah ok will take note in the future17:29
dansmithmingy: for what it's worth, I do *not* subscribe to the naked pings thing :)17:31
*** meghal has quit IRC17:32
*** meghal has joined #openstack-nova17:32
dansmithmingy: meaning, I think so-called "naked pings" are fine :)17:32
cfriesenjaypipes: bauzas: is https://review.openstack.org/#/c/203044/ of any interest to you?  We hit this issue when working with SRIOV devices capable of 128+ VFs.17:33
*** yamahata has quit IRC17:34
mriedemapmelton: ok, missed that since it's not unit tested directly17:34
mriedemhowever,17:34
mriedemapmelton: you'll notice [] is passed in for disk_info rather than {}17:34
mriedemhttps://github.com/openstack/nova/blob/master/nova/tests/unit/virt/libvirt/test_driver.py#L1063117:34
mriedemand _is_booted_from_volume is mocked out17:35
mriedemso that's why it doesn't explode17:35
apmeltonD:17:35
mriedemit at least makes my job easier in updating that test rather than writing a new one17:35
melwittdansmith: "Jenkins: Please make this go away" :)17:36
mriedemdansmith: ping17:36
*** oomichi has quit IRC17:36
*** ahilan has joined #openstack-nova17:37
mingydansmith: thanks for letting me know. although now that also makes me more confused. verily, I need more training data to make a judgment on proper IRC etiquette.17:37
*** park_ has joined #openstack-nova17:37
dansmithmelwitt: heh17:37
*** nic has quit IRC17:37
*** chaochin has joined #openstack-nova17:38
mriedemrlrossit: per your question before i left for lunch,17:39
mriedemrlrossit: i wasn't thinking of replacing the '_pass' in nova with the list in oslo.utils,17:39
bauzascfriesen: sure, reloading the PCI stuff in my head is always good to me since I think we have to fix that17:39
mriedembut somehow supplement17:39
*** whenry_ has joined #openstack-nova17:39
mriedemlike check in the oslo.utils list and check '_pass' in key17:40
bauzascfriesen: that tho means it will take time for me to review that one17:40
mriedemrlrossit: unless you can come up with a fancy regex way that handles both17:40
*** Kennan has joined #openstack-nova17:40
mriedemrlrossit: i guess think about it for awhile. i just know that if you have a 'password' key in the dict, that method won't filter it out right now17:40
mriedemand we have things in strutils that do just this today, so it seems we should leverage those somehow17:41
*** salv-orlando has joined #openstack-nova17:42
melwittmriedem: I didn't know we're supposed to use six.text_type() for logging exception objects, this is the first time I see it https://review.openstack.org/#/c/215170/1/nova/compute/manager.py,cm17:43
*** emagana has joined #openstack-nova17:43
*** oomichi has joined #openstack-nova17:43
mriedemmelwitt: yar17:43
mriedembecause of unicode17:43
melwittnoted. I think I missed it in a recent review17:44
*** vilobhmm1 has quit IRC17:44
*** whenry_ has quit IRC17:44
*** vilobhmm1 has joined #openstack-nova17:45
mriedemsomeone in asia is cursing your name17:46
rlrossitmriedem: my question was aimed more towards how we want the function to operate. Like do we want to remove the keys from the dict, or do we want to substitute the value with "***"?17:46
mriedem"UnicodeDecodeError?! curse you, melwitt!!!"17:46
*** salv-orlando has quit IRC17:46
mriedemrlrossit: probably remove as it does today17:46
melwittlol!17:46
* melwitt hides17:46
rlrossitmriedem: alrighty17:46
mriedemrlrossit: that's just a guess, honestly i didn't know that method existed until today17:47
cfriesenbauzas: no worries, it's a trivial fix.  the only reason we let it slip was that someone had mentioned that the config options might get reorganized.17:47
*** krtaylor has joined #openstack-nova17:47
mriedemrlrossit: as noted in the bug, it seems like oslo.messaging's log notifier should actually be doing the mask_password bit17:47
mriedemto catch anything not filtered at the top17:47
mriedemdims: ^17:47
bauzascfriesen: yeah was about to leave a comment on that purpose17:47
mriedemrlrossit: here http://git.openstack.org/cgit/openstack/oslo.messaging/tree/oslo_messaging/notify/_impl_log.py#n4117:48
*** lpetrut has joined #openstack-nova17:48
dimswhat bug are we looking at mriedem?17:48
rlrossitdims: https://bugs.launchpad.net/nova/+bug/148703817:48
openstackLaunchpad bug 1487038 in OpenStack Compute (nova) "nova.exception._cleanse_dict should use oslo_utils.strutils._SANITIZE_KEYS" [Low,Confirmed]17:48
mriedemdims: it seems http://git.openstack.org/cgit/openstack/oslo.messaging/tree/oslo_messaging/notify/_impl_log.py#n41 should call strutils.mask_password on the jsonified message17:49
cfriesenbauzas: I can get the change rebased to current...if you've got any suggestions for changes, please add them to the review.17:49
*** _kiran_ has quit IRC17:50
apmeltonndipanov: around? I've got a question about some boot from volume/blockdevice code in the libvirt driver17:51
jaypipescfriesen: /me looks17:53
*** vilobhmm11 has joined #openstack-nova17:54
*** vilobhmm1 has quit IRC17:54
*** achanda has joined #openstack-nova17:55
openstackgerritCale Rath proposed openstack/nova: Purge soft-deleted instances cmd  https://review.openstack.org/20375117:55
*** vilobhmm1 has joined #openstack-nova17:55
*** dustins has quit IRC17:56
*** tsekiyama has joined #openstack-nova17:56
*** signed8bit is now known as signed8bit_ZZZzz17:56
*** oomichi has quit IRC17:58
*** vilobhmm11 has quit IRC17:58
*** penick has joined #openstack-nova17:58
openstackgerritMatt Riedemann proposed openstack/nova: libvirt: _create_domain_setup_lxc needs to default disk mapping as a dict  https://review.openstack.org/21492717:59
dimsmriedem: we can for M17:59
mriedemdims: sure18:00
*** gyee has joined #openstack-nova18:00
*** gyee has quit IRC18:00
*** VW_ has joined #openstack-nova18:01
*** gyee has joined #openstack-nova18:03
*** meghal has quit IRC18:04
*** signed8bit_ZZZzz is now known as signed8bit18:06
*** lpetrut has quit IRC18:06
*** worship has joined #openstack-nova18:10
*** worship has left #openstack-nova18:10
*** SourabhP has joined #openstack-nova18:10
*** bapalm has quit IRC18:11
rlrossitmriedem: so should I move forward with a fix for nova for now?18:11
*** bapalm has joined #openstack-nova18:14
*** angdraug has quit IRC18:14
*** nic has joined #openstack-nova18:14
*** burgerk has quit IRC18:16
*** whenry_ has joined #openstack-nova18:17
*** armax has quit IRC18:18
*** rdopiera has quit IRC18:18
*** vivek_nandavanam has quit IRC18:18
*** moshele has quit IRC18:21
*** vivekd has quit IRC18:21
jaypipescfriesen: reviewed.18:22
melwittbauzas: I just saw your tweet. you don't think AZs are for providing application availability by spreading across zones?18:23
cfriesenjaypipes: thanks.  trying to sort out your comment.  We did run into problems with this constant, and we did have to make it bigger.  (I18:24
*** achanda has quit IRC18:24
*** oomichi has joined #openstack-nova18:25
cfriesenjaypipes: we currently have it hardcoded to 0xff18:25
cfriesenjaypipes: let me dig up the original bug report18:26
jaypipescfriesen: just set your whitelist values to *. That MAX_FUNC value is only ever used when reading the CONF.pci_passthrough_whitelist values.18:26
*** fawadk has quit IRC18:27
*** signed8bit is now known as signed8bit_ZZZzz18:27
*** tbh has quit IRC18:27
*** ftersin_ has quit IRC18:27
*** signed8bit_ZZZzz is now known as signed8bit18:28
*** meghal has joined #openstack-nova18:28
*** tonytan4ever has quit IRC18:29
bauzasmelwitt: yes I do, but there is a big difference between providing segregation and failure isolation18:30
bauzasmelwitt: because AZs are on the same controlplane, right?18:31
cfriesenjaypipes: you mean set the func portion of the PCI whitelist to *?   (sorry, not my area of expertise)18:31
mriedemrlrossit: sure18:31
bauzasmelwitt: so, saying that AZs will prevent you from a failure is just a pony story18:31
mriedemwe can always -2 it :)18:31
*** mingy has quit IRC18:32
bauzasmelwitt: we even don't isolate the network, right ? 2 AZs can be on the same subnet nope ?18:32
jaypipescfriesen: yeah (or just set address to *:*:*.*18:32
melwittbauzas: yes, it won't save you from everything18:32
* bauzas wonders how AZs are isolated from a Neutron POV18:32
*** kiran-r has joined #openstack-nova18:33
bauzasmelwitt: so, that's an unicorn to say that OpenStack AZs are about failure domains - that's not true, hence my very angried ping18:33
bauzasmelwitt: that said, I know that jaypipes has a dream18:33
cfriesenjaypipes: but what if I don't want to whitelist all the functions?  (like some are in use by the host)?18:33
bauzasand I second him on that18:33
bauzasbut I just wonder if cells can't be our unicorn for having separate failure domains18:34
melwittbauzas: yeah, I don't know how/if one can designate different subnets to different AZs. I hope that's possible18:34
jaypipescfriesen: there very well *could* be bugs in this stuff, but I'm pretty sure the patch poster is confusing the FUNC in the PCI configuration address with the SR-IOV VF identifier (which is quite different and is like the composite of the slot, the func identifier, and the bus.18:34
cfriesenjaypipes: yeah, the commit message could easily be messed up.18:34
*** Sukhdev has joined #openstack-nova18:34
*** jwcroppe has joined #openstack-nova18:35
cfriesenjaypipes: will get someone to take a look18:35
jaypipescfriesen: what does `sudo virsh nodedev-list | grep pci` show you?18:35
*** dave-mcc_ has joined #openstack-nova18:36
alaskibauzas: cells provides more separation, but it isn't complete.  there's still a shared global db at the top18:36
melwittbauzas: I think I understand your view now. AZs can provide some resiliency to failure but it's not a catch all18:37
*** dustins has joined #openstack-nova18:37
bauzasalaski: sure, that's just a better place for talking about failure domains than AZs18:37
*** davideagnello has quit IRC18:38
*** VW_ has quit IRC18:38
bauzasalaski: to be clear, I'm making a distinction between a failure domain and resiliency18:38
openstackgerritAhilan Rajadeva proposed openstack/nova: Update to the reworked monitor plugin interface to pass in the MonitorMetric Object.  https://review.openstack.org/21477218:38
bauzasalaski: having failure domains doesn't mean your cloud is resilient, you can still have SPOFs18:38
*** davideagnello has joined #openstack-nova18:39
*** dave-mccowan has quit IRC18:39
bauzasalaski: that just means that if a crucial component is failing, the whole set will drop down18:39
*** angdraug has joined #openstack-nova18:39
*** signed8bit is now known as signed8bit_ZZZzz18:39
alaskibauzas: sure.  but there's still a point that can take out all the cells18:39
alaskibut it's still a better situation than now18:39
bauzasalaski: sure, but at least you have a consistent view of a possible failure for your whole cell that wouldn't impact the other cell18:40
cfriesenjaypipes: trying to track down a system configured appropriately18:40
jaypipescfriesen: k. pastebin it when you've found one :)18:40
bauzasalaski: that's why I'm saying that's a failure domain, not a resiliency stuff18:41
jaypipescfriesen: FYI, the term "FUNC" in the PCI configuration address was an unfortunate pick. :)18:41
*** signed8bit_ZZZzz is now known as signed8bit18:41
*** lucasagomes is now known as lucas-dinner18:43
openstackgerritSylvain Bauza proposed openstack/nova: Move CPU and RAM allocation ratios to ResourceTracker  https://review.openstack.org/21527818:43
*** dannywilson has quit IRC18:44
*** dannywilson has joined #openstack-nova18:46
openstackgerritSudipta Biswas proposed openstack/nova: cleanup NovaObjectDictCompat from quota  https://review.openstack.org/21507818:48
alaskibauzas: I see.  a cell is a failure domain, but it's within a larger failure domain of a region18:48
*** whenry_ has quit IRC18:48
openstackgerritSudipta Biswas proposed openstack/nova: cleanup NovaObjectDictCompat from hv_spec  https://review.openstack.org/21509418:49
openstackgerritSudipta Biswas proposed openstack/nova: cleanup NovaObjectDictCompat from hv_spec  https://review.openstack.org/21509418:49
rlrossitoh mriedem, also remember that https://review.openstack.org/#/c/197135/ exists :)18:49
*** davidlenwell has quit IRC18:50
*** davideag_ has joined #openstack-nova18:50
cfriesenjaypipes: I tracked down our original bug report, it contained the following:  http://paste.openstack.org/show/422746/       note the addresses as high as 0000:04:10.14     Now I suppose it's possible we're doing something wrong with setting up the whitelist...18:51
*** davidlenwell has joined #openstack-nova18:51
*** davideagnello has quit IRC18:52
*** kiran-r has quit IRC18:53
*** oomichi has quit IRC18:53
*** meghal has quit IRC18:54
*** shyama has quit IRC18:55
*** afazekas has joined #openstack-nova18:55
openstackgerritSylvain Bauza proposed openstack/nova: Move CPU and RAM allocation ratios to ResourceTracker  https://review.openstack.org/21527818:56
*** ociuhandu has quit IRC18:58
*** davidlenwell has quit IRC18:58
jaypipescfriesen: if you run `lspci | cut -d' ' -f1` on your machine, pretty sure you won't see a value above 7 for the final field :)18:58
*** davidlenwell has joined #openstack-nova18:59
mriedemrlrossit: yeah yeah19:01
mriedemsheehs19:01
rlrossitmriedem: I know I can bug you about it because it's not possible for you to get any more annoyed at me :)19:01
mriedemrlrossit: try peeing on my living room floor19:02
*** meghal has joined #openstack-nova19:02
*** afazekas has quit IRC19:03
jaypipescfriesen: note that the domain part of the PCI address space doesn't show if your system only has domain 0. To show it always, do `lspci -D | cut -d' ' -f1`19:04
*** dannywilson has quit IRC19:04
*** dannywilson has joined #openstack-nova19:05
openstackgerritBrian Elliott proposed openstack/nova: Don't "lock" the DB on expand dry run  https://review.openstack.org/21528819:07
openstackgerritJay Pipes proposed openstack/nova: Allow compute monitors in different namespaces  https://review.openstack.org/20949919:07
*** achanda has joined #openstack-nova19:08
jaypipesahilan: yay! just saw your patch :) will review shortly.19:10
*** emagana has quit IRC19:10
*** dannywilson has quit IRC19:10
*** dannywilson has joined #openstack-nova19:10
ahilanjaypipes: Thanks :-)19:10
bauzasjaypipes: I think you should be interested in https://review.openstack.org/#/c/215278/19:10
jaypipesbauzas: you wrote that, so ... nope.19:11
jaypipes:P19:11
bauzasjaypipes: DB upgrade and object bump are following19:11
bauzasjaypipes: arrrgh19:11
cfriesenjaypipes: thanks for the clarification.  it's looking like we may have had a bug generating the whitelist (since fixed) which led to the issue with exceeding the MAX_FUNC value.19:11
jaypipesbauzas: coolio, thanks for tackling this!19:11
bauzasjaypipes: I'm in the US, I can ask the crew to make a stop by Florida :p19:11
jaypipescfriesen: no worries, mate, lemme know if you need assistance on anything further. happy to help.19:11
jaypipesbauzas: :)19:12
*** oomichi has joined #openstack-nova19:21
*** zul has quit IRC19:23
*** dims_ has joined #openstack-nova19:26
*** vishwana_ has joined #openstack-nova19:26
*** meghal has quit IRC19:27
*** ijw_ has joined #openstack-nova19:27
*** signed8bit is now known as signed8bit_ZZZzz19:27
*** ttx has quit IRC19:27
*** rex_lee has quit IRC19:27
*** jamielennox|away has quit IRC19:27
*** signed8bit_ZZZzz is now known as signed8bit19:28
*** dims has quit IRC19:28
*** DuncanT has quit IRC19:28
*** ijw has quit IRC19:28
*** adam_g has quit IRC19:28
*** oomichi has quit IRC19:28
*** davidlenwell has quit IRC19:28
*** r-daneel has quit IRC19:28
*** vishwanathj has quit IRC19:28
*** dguitarbite has quit IRC19:28
*** serverascode has quit IRC19:28
*** toan has quit IRC19:28
*** toan has joined #openstack-nova19:29
*** irenab has quit IRC19:29
*** dave-mccowan has joined #openstack-nova19:30
*** davidlenwell has joined #openstack-nova19:31
*** DuncanT has joined #openstack-nova19:31
*** meghal has joined #openstack-nova19:31
sudiptoahilan, Just looking at your patch...trying to understand, what advantage does it give us to pass the monitormetric object as a method argument? :)19:32
*** serverascode has joined #openstack-nova19:32
*** signed8bit is now known as signed8bit_ZZZzz19:32
*** dave-mcc_ has quit IRC19:33
*** VW_ has joined #openstack-nova19:33
*** ttx has joined #openstack-nova19:33
*** meghal has quit IRC19:33
*** achanda has quit IRC19:34
*** rex_lee has joined #openstack-nova19:34
*** yassine has quit IRC19:36
*** adam_g has joined #openstack-nova19:36
*** adam_g has quit IRC19:36
*** adam_g has joined #openstack-nova19:36
*** signed8bit_ZZZzz is now known as signed8bit19:36
*** Marga_ has quit IRC19:40
ahilansudipto: this allows metric plugin to set the MonitorMetric fields and not the base19:41
*** meghal has joined #openstack-nova19:41
*** Marga_ has joined #openstack-nova19:41
*** dguitarbite has joined #openstack-nova19:41
*** salv-orlando has joined #openstack-nova19:41
sudiptoahilan, yeah my question is - if we instantiate the object inside the populate_metrics_object - will it not work?19:42
sudiptoi guess that way the populate_metrics_object name could change to get_metrics_object or something...just a thought :)19:42
*** VW_ has quit IRC19:43
ahilanSudipto: So if we add new fields to MonitorMetric Object, then each metric plugin can update their field in their metric plugin code and no need to modify the base19:43
*** VW_ has joined #openstack-nova19:43
*** r-daneel has joined #openstack-nova19:44
sudiptoahilan, hmm i got the intent...i didn't mean that - i just meant - if your metric_object instantiation could have been just done inside each plugin (then like you said - the plugin anyway knows what to populate it with)...nevermind!19:44
*** sudipto has quit IRC19:46
mriedemapmelton: good news, i was able to bring up a devstack with lxc19:46
mriedemthat exceeded my expectations19:46
*** Vek has quit IRC19:46
apmeltonwoo!19:46
apmeltonhaha19:46
*** Vek has joined #openstack-nova19:47
*** armax has joined #openstack-nova19:47
*** alejandrito has quit IRC19:47
mriedembad news is, nova boot fails with NoValidHost :)19:47
*** jamielennox|away has joined #openstack-nova19:47
apmelton:/19:47
*** jamielennox|away is now known as jamielennox19:47
*** VW_ has quit IRC19:48
apmeltonmriedem: did you run "sudo modprobe nbd"?19:48
mriedemnope19:49
*** alejandrito has joined #openstack-nova19:49
openstackgerritRyan Rossiter proposed openstack/nova: Search for more keys when clearing passwords  https://review.openstack.org/21530819:51
openstackgerritj_king proposed openstack/nova: libvirt: Add wait for vif unplug events on destroy  https://review.openstack.org/21012619:51
*** setuid has quit IRC19:52
openstackgerritBrian Haley proposed openstack/nova: Disable IPv6 on bridge devices  https://review.openstack.org/19805419:53
openstackgerritAndrew Melton proposed openstack/nova: libvirt: refactor _create_domain_setup_lxc to use Image.get_model  https://review.openstack.org/21531019:53
openstackgerritAndrew Melton proposed openstack/nova: virt: allow for direct mounting of LocalBlockImages  https://review.openstack.org/21531119:53
apmeltonmriedem: ^^19:53
apmeltonI'm not sure if devstack supports the LVM image backend for Nova, but with those patches Nova will skip nbd when using the LVM backend19:54
*** andreykurilin__ has quit IRC19:55
*** jichen has joined #openstack-nova19:56
*** setuid has joined #openstack-nova19:59
*** ssurana has quit IRC20:02
*** tjones1 has joined #openstack-nova20:03
*** ssurana has joined #openstack-nova20:03
*** annegentle has quit IRC20:04
*** annegentle has joined #openstack-nova20:04
*** IanGovett has quit IRC20:06
*** tjones1 has left #openstack-nova20:08
*** alejandrito has quit IRC20:09
*** browne has quit IRC20:11
*** artom has joined #openstack-nova20:12
*** browne has joined #openstack-nova20:12
*** achanda has joined #openstack-nova20:13
*** alejandrito has joined #openstack-nova20:13
*** irenab has joined #openstack-nova20:15
*** irenab has quit IRC20:15
*** ajayaa has quit IRC20:19
*** achanda has quit IRC20:19
*** emagana has joined #openstack-nova20:20
mriedemapmelton: sudo modprobe nbd fixed the boot20:20
mriedemi need to get that into devstack20:20
*** ndipanov has joined #openstack-nova20:20
apmeltonmriedem: good to hear20:21
*** VW_ has joined #openstack-nova20:22
*** jichen has quit IRC20:24
*** exploreshaifali has quit IRC20:30
alaskimriedem: I just hit an instance of https://bugs.launchpad.net/neutron/+bug/1485969.  should I just recheck, or is a e-r change needed as well?20:30
openstackLaunchpad bug 1485969 in neutron "test_dualnet_multi_prefix_dhcpv6_stateless failed due to "RuntimeError: Set changed size during iteration" in quota code" [Undecided,In progress] - Assigned to Ihar Hrachyshka (ihar-hrachyshka)20:30
mriedemrecheck20:30
mriedemi need to get the e-r query up20:30
*** VW_ has quit IRC20:30
*** VW_ has joined #openstack-nova20:31
alaskiok.  I saw the failure on the test volume boot pattern test in Nova20:31
mriedemyeah, it's a recent failure20:31
*** achanda has joined #openstack-nova20:32
*** meghal has quit IRC20:33
*** Sukhdev has quit IRC20:34
*** meghal has joined #openstack-nova20:35
*** ndipanov has quit IRC20:35
*** VW_ has quit IRC20:35
*** signed8bit is now known as signed8bit_ZZZzz20:37
bauzasmelwitt: alaski: meh, hitting http://logs.openstack.org/78/215278/2/check/gate-tempest-dsvm-cells/a55bcab/logs/screen-n-cell-region.txt.gz?level=TRACE20:39
*** alex_klimov has joined #openstack-nova20:39
bauzasmelwitt: alaski: this is not something I remember20:39
melwitthas anyone seen before mock not working as expected running unit tests with the py34 env? I'm seeing a thing where I set side_effect=[..., ..., ..., ...] and the first call of the mocked function isn't returning the thing I put in side_effect20:39
*** zul has joined #openstack-nova20:39
mriedemapmelton: nice, nova stop lxc-test - killed the connection to my devstack vm20:39
bauzasmelwitt: alaski: but I was pretty off the cells bus for the last 3 months so I probably missed that20:39
apmeltonmriedem: D: yea, I haven't quite figured that one out yet20:40
bauzasmelwitt: mmm, none I've seen and tho I had some problems :)20:40
bauzas(now speaking of the side_effect)20:40
mriedemapmelton: oh double nice, now i can't get back in via ssh20:41
mriedemsuweeet20:41
alaskibauzas: that would be due to https://review.openstack.org/#/c/202593/.  it's not a problem unless it happens multiple times for the same db update.  right now it's exposing races that were hidden before20:41
mriedemthanks nbd20:41
*** signed8bit_ZZZzz is now known as signed8bit20:41
apmeltonmriedem: should have warned ya to make sure you had a second way in20:41
*** boris-42 has joined #openstack-nova20:41
bauzasalaski: oh20:43
*** thangp has quit IRC20:43
bauzasalaski: any referenced bug that I could recheck against ?20:43
alaskibauzas: did it cause a failure?20:44
bauzasalaski: yep20:44
bauzasalaski: https://review.openstack.org/#/c/215278/20:44
alaskibauzas: I'm not aware of an open bug for it.  I haven't seen it cause a failure yet20:44
alaskihadn't20:44
bauzaslemme recheck and see20:45
bauzasI seriously don't think my change can't be responsible for that :)20:45
melwittalaski: I'm glad you know what it is because I was looking and coming up with nothing20:45
*** gszasz has quit IRC20:47
alaskimelwitt: I helped revert that patch originally so the exception was familiar :)20:47
*** emagana has quit IRC20:47
melwittnice :)20:48
bauzasokay, so I'm probably unlucky then :)20:48
*** exploreshaifali has joined #openstack-nova20:48
apmeltonmriedem: I was poking around devstack and apparently we can set NOVA_BACKEND=LVM, and that'll get everything set up20:49
apmeltonI've gotta run for the evening, but I'll start testing out my patchsets with that asap20:49
*** armax has quit IRC20:49
mriedemapmelton: sounds good, thanks20:50
mriedemdims_: do you remember the way to run a specific nova change from devstack? i thought that was by setting the NOVA_BRANCH var in stackrc20:50
mriedemmtreinish: ^ if you're around20:50
*** meghal has quit IRC20:50
alaskibauzas: I don't think it's related to those TRACES.  the failure is http://logs.openstack.org/78/215278/2/check/gate-tempest-dsvm-cells/a55bcab/logs/screen-n-api.txt.gz#_2015-08-20_19_39_28_30320:51
alaskiand that instance isn't in any of the trace logs20:51
dims_mriedem: one sec20:52
*** ivasev has quit IRC20:53
dims_        NOVA_REPO=https://review.openstack.org/p/openstack/nova20:54
dims_        NOVA_BRANCH=refs/changes/50/5050/120:54
dims_mriedem: is that what you are looking for?20:54
mriedemdims_: yup, perfect, thanks20:55
bauzasalaski: that's not related to my change either :)20:55
bauzasanyway, I won't take too much time on that one, let's see what a anonym recheck does (booh)20:55
melwittlifeless: are you aware of any issues with setting a mock side_effect = [Mock(), Mock(), Mock(), Mock()] in py34? I have a test where it works fine in py27 but in py34 the first call of the mocked function calls through and doesn't return the Mock() I set in side_effect. am I doing something wrong?20:57
*** achanda has quit IRC20:57
*** achanda_ has joined #openstack-nova20:57
bauzasmelwitt: haypo is on your TZ today20:59
bauzasmelwitt: he could probably help you with that20:59
bauzasmelwitt: did you reproduced locally ?21:00
*** dannywil_ has joined #openstack-nova21:00
melwittbauzas: yeah I'm only seeing it locally, that is, I haven't uploaded what I've got yet because it's failing locally21:00
*** ndipanov has joined #openstack-nova21:01
bauzasmelwitt: okay, any paste I could look at ?21:02
*** dannywilson has quit IRC21:03
bauzasmelwitt: what's strange with your comment is that [Mock(), Mock(), Mock(), Mock()] is not callable, right?21:04
bauzasside_effect is waiting a callable AFAIK21:04
*** meghal has joined #openstack-nova21:05
alaskiI thought it wanted an iterable?21:05
*** aysyd has quit IRC21:06
melwittyeah, I'm doing this style https://docs.python.org/3/library/unittest.mock.html#quick-guide21:06
melwitt>>> mock.side_effect = [5, 4, 3, 2, 1]21:06
* melwitt assembling a paste21:06
bauzasmelwitt: my bad, it accepts an iterable https://docs.python.org/3/library/unittest.mock.html#unittest.mock.Mock.side_effect21:07
bauzasmelwitt: but if an iterable, then you have to yield21:08
bauzasmelwitt: I guess you want to mock 3 calls of the same method for example ?21:09
melwittbauzas: right.21:09
*** yamahata has joined #openstack-nova21:10
bauzasmelwitt: so with 3 different results?21:10
bauzasmelwitt: because here, you actually don't care about the result, all of them are mocks21:11
bauzasmelwitt: if so, just provide a function which returns a mock, that will work everytime21:11
haypomelwitt: i recall vaguely a change on side_effect with list. don't use side_effect with a list :)21:13
melwittbauzas: thanks for the suggestion, I'll try that21:13
haypomelwitt: or use .return_value, sorry i didn't read the whole discussion21:14
melwitthaypo: the current docs shows it as a valid way to use https://docs.python.org/3/library/unittest.mock.html#quick-guide so I'd be interested in what's different now21:14
haypomelwitt: unittest.mock is in python 3.3+, but nova uses the third party mock module on python 2. there are subtle differences between the two modules21:15
haypoas the one you hitted21:15
bauzasmelwitt: yeah, return_value is even better for an identical result21:15
*** dprince has quit IRC21:16
melwitthaypo: I understand that, but isn't the doc I just linked talking about the python 3 builtin? that's where I'm hitting the problem. py27 works for me21:16
haypom.side_effect=[1,2]; [m(), m()]) must return [1, 2]21:16
*** thorst has quit IRC21:16
*** meghal has quit IRC21:17
haypomelwitt: can i see your code? do you have a paste?21:17
*** derekh has joined #openstack-nova21:17
dims_melwitt: works for me - http://paste.openstack.org/show/422763/21:17
dims_melwitt: can you try the same as haypo pasted?21:18
*** zul has quit IRC21:18
*** erlon has quit IRC21:18
haypoi have the same behaviour on python 3.4 (unittest.mock) and python 2.7 (mock 1.0 and mock 1.3)21:20
melwitthaypo dims_: this is what I have http://pastebin.com/DCYQRck721:20
*** annegentle has quit IRC21:20
*** annegentle has joined #openstack-nova21:20
haypomelwitt: i don't understand why you use side_effect with a list of mocks. if you don't reuse mocks, use side_effect=lambda: mock.Mock(), or simply side_effect=mock.Mock21:21
*** eharney has quit IRC21:21
melwitthaypo: yeah, I realized from bauzas suggestion that I can do it another way. I just thought what I had before is supposed to also work21:22
ericksonsantosmelwitt, I think return_value should be good in your case, unless you want different results21:22
*** rfolco has quit IRC21:23
bauzasmelwitt: I think you were probably not calling your mocked method within an iterable21:23
bauzasmelwitt: because it's necessary21:24
openstackgerritMatt Riedemann proposed openstack/nova: libvirt: only get bdm in _create_domain_setup_lxc if booted from volume  https://review.openstack.org/21534221:24
bauzasor call the .__next__21:24
bauzassomething like mocked.side_effect = [1,2@21:24
bauzasoops21:24
bauzas[1, 2]21:24
bauzasand thezn21:24
bauzasmocked.__next__()21:25
bauzasmocked.__next__()21:25
*** krtaylor has quit IRC21:26
*** armax has joined #openstack-nova21:26
haypobauzas: hum? mock.__next__() returns a new mock21:27
melwittbauzas: yeah, I don't know. the examples show you don't have to21:27
melwitt>>> mock.side_effect = [5, 4, 3, 2, 1]21:27
melwitt>>> mock(), mock(), mock()21:27
melwitt(5, 4, 3)21:27
bauzasanyway, keep it simple I guess21:31
mtreinishmriedem: yeah, what dims_ said21:32
mriedemstill stacking21:32
*** dustins has quit IRC21:32
*** vilobhmm1 has quit IRC21:34
alaskimelwitt: I've used it the way you tried to, there are other examples of that in the Nova codebase.  so the concept in theory shoudl be working21:34
*** lucas-dinner has quit IRC21:34
*** ahilan has quit IRC21:34
*** ndipanov has quit IRC21:35
*** rlrossit has left #openstack-nova21:35
*** Piet has joined #openstack-nova21:35
melwitthm, fails the same way with side_effect=lambda *a, **k: mock.Mock() in py34 only. I did not expect that21:35
melwittfirst call goes through to the real code even though I mocked the function being called21:36
*** vilobhmm1 has joined #openstack-nova21:36
*** vilobhmm11 has joined #openstack-nova21:37
*** smatzek has quit IRC21:38
*** dannywil_ has quit IRC21:40
*** vilobhmm1 has quit IRC21:40
*** ccarmack has left #openstack-nova21:40
*** annegentle has quit IRC21:42
*** VW_ has joined #openstack-nova21:43
*** edmondsw has quit IRC21:44
bauzasmelwitt: mmm21:48
bauzasmelwitt: weirod21:48
bauzasweirdo even21:48
bauzasmelwitt: failback to return_value21:48
bauzasmelwitt: I'm tho pretty sure that side_effect is already called by the py34 tests21:49
melwittbauzas: yeah :( I thought about it but I think that can't work if I need a new instance for every call. else I have to rethink my test. I'm trying to test whether I get a new thing back vs a cached thing21:49
*** alejandrito has quit IRC21:49
*** bapalm has quit IRC21:49
bauzasso you wanna mock the instance creation ?21:50
bauzasand give a fake Instance?21:50
melwittbauzas: I want _create_facade to always return a new/different facade. and that should only happen if a cached facade isn't found21:51
bauzasI see, so yeah a side_effect is needed there IMHO21:52
*** bapalm has joined #openstack-nova21:52
*** alejandrito has joined #openstack-nova21:52
*** dannywilson has joined #openstack-nova21:54
*** signed8bit has quit IRC21:55
*** meghal has joined #openstack-nova21:55
*** bapalm has quit IRC21:57
lifelessmelwitt: what implementation of mock are you using ?21:58
*** diogogmt_ has joined #openstack-nova22:00
openstackgerritCale Rath proposed openstack/nova: Purge soft-deleted instances cmd  https://review.openstack.org/20375122:00
*** diogogmt has quit IRC22:00
*** diogogmt_ is now known as diogogmt22:00
*** burt has quit IRC22:02
*** bapalm has joined #openstack-nova22:03
openstackgerritCale Rath proposed openstack/nova: Purge soft-deleted instances cmd  https://review.openstack.org/20375122:04
openstackgerritAndrew Laski proposed openstack/nova: Update Flavor obj_reset_changes to match signature  https://review.openstack.org/21535822:04
melwittlifeless, bauzas: I think I have finally found through more digging that it's not related to mock. the calls are actually different between py27 and py34 because of the different state of a global variable in sqlalchemy/db/api.py between the two envs. sorry about that :(22:04
*** claudiub has quit IRC22:05
openstackgerritAndrew Laski proposed openstack/nova: WIP: Add BuildRequest table and object  https://review.openstack.org/21479222:07
openstackgerritAndrew Laski proposed openstack/nova: Add persistence to the RequestSpec object  https://review.openstack.org/21175322:07
*** bapalm has quit IRC22:08
lifelessmelwitt: np22:11
bauzasmelwitt: soooooooo specific :/22:12
bauzasglad you found it22:12
*** bapalm has joined #openstack-nova22:14
melwittbauzas: at least I have the lambda returning the Mock() now! better than the list :)22:16
*** oomichi has joined #openstack-nova22:17
*** krtaylor has joined #openstack-nova22:18
mriedemalaski: is https://review.openstack.org/#/c/215358/ fixing a bug or you're going to build on that?22:18
*** bapalm has quit IRC22:19
*** ctrath has quit IRC22:20
*** annegentle has joined #openstack-nova22:20
*** zul has joined #openstack-nova22:26
*** oomichi has quit IRC22:26
*** alex_klimov has quit IRC22:28
*** jecarey has quit IRC22:28
*** exploreshaifali_ has joined #openstack-nova22:29
*** bapalm has joined #openstack-nova22:29
*** exploreshaifali_ has quit IRC22:30
*** exploreshaifali has quit IRC22:31
*** thorst has joined #openstack-nova22:32
*** annegentle has quit IRC22:33
*** annegentle has joined #openstack-nova22:33
*** tpeoples has quit IRC22:34
*** bapalm has quit IRC22:34
*** mriedem is now known as mriedem_away22:35
*** bapalm has joined #openstack-nova22:35
*** annegentle has quit IRC22:38
*** xyang1 has quit IRC22:38
*** flwang1 has quit IRC22:39
*** rfolco has joined #openstack-nova22:39
*** emagana has joined #openstack-nova22:40
*** bapalm has quit IRC22:43
*** bapalm has joined #openstack-nova22:43
bauzasmelwitt: alaski: grrr, just listening a Trove presentation, hearing "so, if using AZs, if your master DB is failing from AZ1, then the slave one will still be good"22:44
bauzasand that said, the presenter is the Trove PTL22:44
bauzasso I think we have a huuuuge problem with leaving people understand what AZs mean in Nova22:44
bauzassounds like the contract is not fulfilled, and I don't know how to fix that22:45
bauzasI should probably write a blogpost22:45
bauzasjaypipes: ^22:46
*** nic has quit IRC22:46
melwittbauzas: that would be interesting. the topic has come up a lot22:47
bauzasnow writing a -dev thread22:48
*** bapalm has quit IRC22:48
*** dannywilson has quit IRC22:48
*** bapalm has joined #openstack-nova22:51
*** mdrabe has quit IRC22:51
*** VikasC has joined #openstack-nova22:53
*** thorst has quit IRC22:55
*** dannywilson has joined #openstack-nova22:57
*** tsbot_ has quit IRC23:01
*** asettle has joined #openstack-nova23:03
*** ijuwang has quit IRC23:05
*** achanda_ has quit IRC23:05
*** alejandrito has quit IRC23:06
*** diogogmt has quit IRC23:12
*** diogogmt has joined #openstack-nova23:14
*** emagana has quit IRC23:14
*** hemna is now known as hemnafk23:16
*** achanda has joined #openstack-nova23:16
*** dannywilson has quit IRC23:19
*** VW_ has quit IRC23:25
*** zzzeek has quit IRC23:25
*** VW_ has joined #openstack-nova23:25
*** r-daneel has quit IRC23:27
*** flwang1 has joined #openstack-nova23:29
*** dannywilson has joined #openstack-nova23:30
*** VW_ has quit IRC23:30
*** otter768 has joined #openstack-nova23:30
*** VW_ has joined #openstack-nova23:34
*** cfriesen has quit IRC23:35
*** SourabhP has quit IRC23:43
*** angdraug has quit IRC23:45
*** Marga_ has quit IRC23:46
*** salv-orlando has quit IRC23:46
*** Marga_ has joined #openstack-nova23:47
*** dannywilson has quit IRC23:56
*** VW_ has quit IRC23:56
*** VW_ has joined #openstack-nova23:57

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!