Thursday, 2016-03-31

*** ljxiash has joined #openstack-nova00:00
*** haomaiwa_ has quit IRC00:01
*** akshai has joined #openstack-nova00:01
*** haomaiwang has joined #openstack-nova00:01
*** ljxiash_ has joined #openstack-nova00:01
*** mylu has quit IRC00:02
*** stevemar has quit IRC00:02
*** mvk_ has quit IRC00:03
*** ljxiash has quit IRC00:05
*** Sukhdev has quit IRC00:06
*** fawadkhaliq has quit IRC00:06
*** ljxiash_ has quit IRC00:06
*** mylu has joined #openstack-nova00:07
*** fawadkhaliq has joined #openstack-nova00:07
*** p01nt3r75 has quit IRC00:08
*** dimtruck is now known as zz_dimtruck00:08
*** salv-orl_ has quit IRC00:09
*** akshai_ has joined #openstack-nova00:09
*** zehua has joined #openstack-nova00:10
*** zehua has left #openstack-nova00:10
*** akshai has quit IRC00:13
*** fawadkhaliq has quit IRC00:13
*** ZZelle has quit IRC00:14
*** rk4n has quit IRC00:14
*** markvoelker has joined #openstack-nova00:15
*** fawadkhaliq has joined #openstack-nova00:16
*** fawadkhaliq has quit IRC00:21
*** markvoelker_ has joined #openstack-nova00:22
*** diogogmt has joined #openstack-nova00:22
*** mriedem has joined #openstack-nova00:23
*** davideag_ has quit IRC00:23
*** fawadkhaliq has joined #openstack-nova00:23
*** markvoelker has quit IRC00:25
*** mtanino has quit IRC00:29
*** Swami has quit IRC00:33
*** Sukhdev has joined #openstack-nova00:33
*** Sukhdev has quit IRC00:33
*** sdake has quit IRC00:34
*** sdake has joined #openstack-nova00:36
*** mvk_ has joined #openstack-nova00:37
*** Jeffrey4l_ has quit IRC00:38
*** ccarmack has quit IRC00:39
*** ducttape_ has joined #openstack-nova00:39
*** mylu has quit IRC00:39
*** mkrcmari__ has quit IRC00:40
*** achanda has quit IRC00:41
*** efoley_ has quit IRC00:42
*** mylu has joined #openstack-nova00:45
*** Marga_ has quit IRC00:48
*** ducttape_ has quit IRC00:50
*** mnestratov|2 has quit IRC00:50
*** vladikr has quit IRC00:52
*** dave-mccowan has quit IRC00:53
*** stevemar has joined #openstack-nova00:58
*** dave-mccowan has joined #openstack-nova00:58
*** haomaiwang has quit IRC01:01
*** haomaiwang has joined #openstack-nova01:01
*** jamielennox is now known as jamielennox|away01:01
*** nic has quit IRC01:02
*** stevemar has quit IRC01:03
*** ducttape_ has joined #openstack-nova01:04
*** mylu has quit IRC01:07
*** mylu has joined #openstack-nova01:08
*** aysyd has quit IRC01:08
*** ivase has quit IRC01:09
*** vilobhmm11 has quit IRC01:09
*** ivase has joined #openstack-nova01:09
*** ducttape_ has quit IRC01:09
*** vilobhmm11 has joined #openstack-nova01:10
*** yuywz has joined #openstack-nova01:12
*** Goneri has quit IRC01:12
*** apoorvad has quit IRC01:13
*** ivase_ has joined #openstack-nova01:13
*** fawadkhaliq has quit IRC01:14
*** fawadkhaliq has joined #openstack-nova01:14
*** ivase has quit IRC01:17
*** yuywz has quit IRC01:20
*** fawadkhaliq has quit IRC01:20
openstackgerritZhenyu Zheng proposed openstack/nova: remove glance deprecated config  https://review.openstack.org/29953501:20
*** houming has joined #openstack-nova01:22
*** fawadkhaliq has joined #openstack-nova01:23
*** vilobhmm11 has quit IRC01:23
*** fawadkhaliq has quit IRC01:24
*** fawadkhaliq has joined #openstack-nova01:25
*** fawadkhaliq has quit IRC01:27
*** vishwanathj has joined #openstack-nova01:30
openstackgerritZhenyu Zheng proposed openstack/nova: config options: centralize section: "glance"  https://review.openstack.org/25715301:32
*** browne has quit IRC01:34
*** baoli has joined #openstack-nova01:37
*** baoli_ has joined #openstack-nova01:39
*** stevemar has joined #openstack-nova01:39
*** achanda has joined #openstack-nova01:41
*** baoli_ has quit IRC01:42
*** baoli has quit IRC01:42
*** Daisy has joined #openstack-nova01:43
mriedemdansmith: can we please just go back to the good old days where you can add new APIs to cinder and nova can assume they are just available and the policy settings are the same as the defaults forever https://review.openstack.org/#/c/41298/01:45
*** efried has quit IRC01:45
*** achanda has quit IRC01:46
yonglihemriedem:  ping01:47
openstackgerritMohammed Naser proposed openstack/nova: Convert MonitorMetric to use FloatField for values  https://review.openstack.org/29970801:48
openstackgerritMohammed Naser proposed openstack/nova: Convert MonitorMetric to use FloatField for values  https://review.openstack.org/29970801:49
yongliheshaohe is on vacation, and the Intel SRIOV CI comments to Neutron,  vnic_type is macvtap.01:49
*** sdake_ has joined #openstack-nova01:49
*** ducttape_ has joined #openstack-nova01:49
*** Marga_ has joined #openstack-nova01:49
*** sdake has quit IRC01:50
*** Daisy has quit IRC01:51
*** Daisy has joined #openstack-nova01:51
*** ljxiash has joined #openstack-nova01:52
*** gjayavelu has joined #openstack-nova01:52
*** Marga_ has quit IRC01:54
*** Daisy has quit IRC01:55
*** Daisy has joined #openstack-nova01:55
*** achanda has joined #openstack-nova01:56
*** pushkaru has joined #openstack-nova01:57
mriedemyonglihe: hi01:58
mriedemyonglihe: ok, can you update the third party CI wiki please with that information?01:58
mriedemyonglihe: are there any plans to run that against nova?01:58
mriedemyonglihe: because the wiki says it runs against nova today01:59
mriedemOpenStack Programs: Neutron,Nova01:59
*** haomaiwang has quit IRC01:59
yonglihemriedem: ok, i will update the wiki.01:59
*** Daisy has quit IRC02:00
*** mtanino has joined #openstack-nova02:00
*** Daisy has joined #openstack-nova02:00
*** unicell1 has quit IRC02:02
*** ducttape_ has quit IRC02:02
yonglihefor running against nova:  SRIOV need neutron been starting up, it's almost a new CI for Intel PCI/SRIOV CI. do you think we better to do that?  we have a SRIOV test run against the nova.02:03
*** Daisy has quit IRC02:03
*** Daisy has joined #openstack-nova02:03
*** ducttape_ has joined #openstack-nova02:03
mriedemyonglihe: i'm confused,02:04
*** zz_dimtruck is now known as dimtruck02:04
mriedemyonglihe: there is intel nfv ci, intel sriov ci, and intel networking ci02:04
mriedemyou said intel sriov ci uses macvtap and only runs on neutron, correct?02:04
mriedemwhat do the other 2 test?02:04
yongliheyeah,  the other 2 test is not  operated by me,  we have escalate your questing internally. so someone else will update you about this.02:05
yonglihequesting->question02:06
*** liunian_ has joined #openstack-nova02:06
openstackgerritPushkar Umaranikar proposed openstack/nova: Config options: Centralize netconf options  https://review.openstack.org/29431202:07
openstackgerritPushkar Umaranikar proposed openstack/nova: Config options: Centralize resource tracker options  https://review.openstack.org/29421102:07
*** ducttape_ has quit IRC02:08
*** gjayavelu has quit IRC02:09
*** Marga_ has joined #openstack-nova02:10
yonglihemriedem: SRIOV CI wiki updated. sriov ci  runs on neutron.  according to wiki: Intel-Networking-CI runs on  Neutron,  Intel-NFV-CI runs on Nova.02:11
*** mylu has quit IRC02:11
*** jianghuaw has joined #openstack-nova02:12
*** thorst has quit IRC02:13
*** thorst has joined #openstack-nova02:13
mriedemyonglihe: ok, thanks, i look forward to the details.02:13
openstackgerritMatt Riedemann proposed openstack/nova: Fix swap-volume for non-admin users  https://review.openstack.org/29971502:14
*** Goneri has joined #openstack-nova02:14
*** Marga_ has quit IRC02:14
gmannmriedem: just replied in ML, so in  https://review.openstack.org/29971502:15
gmannmriedem: if nova allow swap volume for non-admin then with your patch it should do attach/detach explicitly right?02:15
*** Jeffrey4l has joined #openstack-nova02:17
mriedemgmann: well, if non-admin it just shouldn't call cinder's migrate_volume_completion API02:17
mriedemmy fix isn't checking admin vs non-admin02:17
mriedemits' checking to see if the volume migration_status field is set02:18
mriedemif it is, we're an admin and it is a call initiated by cinder02:18
gmannmriedem: yea02:18
gmannand in case non-admin user call initiated on nova02:18
mriedemi might work on a tempest test for the non-admin swap-volume case tomorrow02:18
mriedemunless you want to :)02:19
jianghuawhi, all. If I want to avoid running python34 testing on all unit tests which located under a specific directory, can I just add this directory into nova/tests-py3.txt? or I have to add each test case in it? Thanks.02:19
gmannin that case volume migration_status filed would not be present right02:19
mriedemgmann: right02:19
mriedemjianghuaw: nova/tests-py3.txt is a blacklist file02:20
gmannmriedem: so in that case, new volume attach and old detach is not happened which was done in  migrate_volume_completion on cinder side02:20
mriedemso add tests that you don't want to run to that file for py3402:20
mriedemgmann: i'm not following the question02:20
gmannmriedem: actually i was thinking the case where swap volume call is initiated by non-admin on nova side02:21
gmannin that case  migrate_volume_completion will not be called as  volume migration_status will be unset02:21
*** salv-orlando has joined #openstack-nova02:21
gmannso attach new volume and detach old volume on server would not happen (which is done in  cinder->migrate_volume_completion)02:22
*** thorst has quit IRC02:22
gmannI think long back we moved that part from nova to cinder side02:22
jianghuawmriedem: thanks. Can I just add a single line for the directory? e.g. I have several test cases under nova/tests/unit/virt/xenapi/plugins, can I just add a line as: nova.tests.unit.virt.xenapi.plugins to exclude all of those tests under that path?02:23
mriedemgmann: are you sure that doesn't happen in the virt driver's swap_volume method?02:23
*** Goneri has quit IRC02:23
gmannmriedem: humm, let me check that. from manager it was removed02:24
mriedemjianghuaw: nova.tests.unit.virt.xenapi.plugins is already in there02:24
gmannbut i will check and see if that happens02:24
mriedemgmann: ok, i want to write a tempest test before this change goes in anyway and depend on that to see it passing,02:24
mriedembut it's getting late for me so i'm going to sign off02:24
gmannmriedem: yea that is nice idea. I will add tempest one and once you are in tomorrow can see the result02:25
mriedemgmann: cool, thanks02:25
gmannmriedem: Thanks, have a GN02:25
*** mriedem has quit IRC02:25
jianghuawmriedem: indeed. But py34 seems not exclude the new test when I added a new one in the same path today.02:26
*** xyang1 has quit IRC02:26
*** ljxiash has quit IRC02:26
jianghuawthat's why I'm thinking if I have to add a line for each test.02:26
*** ljxiash has joined #openstack-nova02:27
*** browne has joined #openstack-nova02:28
*** ljjjustin_ has joined #openstack-nova02:30
*** markvoelker_ has quit IRC02:31
*** zehua_ has joined #openstack-nova02:31
*** zehua_ has quit IRC02:31
*** hoangcx has joined #openstack-nova02:31
*** ljxiash has quit IRC02:32
*** salv-orlando has quit IRC02:32
*** markvoelker_ has joined #openstack-nova02:32
jianghuawmriedem: oh. I misunderstood the failure. the py34 testing failed due to "urllib2 is not available with py3x". So actually it fails at syntax parsing. I think I've fixed that. So indeed the path way works at here.02:34
jianghuawThanks mriedem.02:34
*** vilobhmm11 has joined #openstack-nova02:34
*** markvoel_ has joined #openstack-nova02:36
*** liunian_ has quit IRC02:36
*** mylu has joined #openstack-nova02:36
*** salv-orlando has joined #openstack-nova02:36
*** yamahata has quit IRC02:38
*** haomaiwang has joined #openstack-nova02:38
*** markvoelker_ has quit IRC02:40
*** liunian_ has joined #openstack-nova02:41
*** liunian_ has quit IRC02:43
*** ljjjustin_ is now known as zehua02:44
*** mylu has quit IRC02:44
*** mylu has joined #openstack-nova02:44
*** liunian_ has joined #openstack-nova02:46
*** mylu_ has joined #openstack-nova02:48
*** Marga_ has joined #openstack-nova02:48
*** mylu has quit IRC02:49
*** salv-orlando has quit IRC02:52
*** Marga_ has quit IRC02:53
*** liunian_ has quit IRC02:54
*** liunian_ has joined #openstack-nova02:58
*** haomaiwang has quit IRC03:01
*** markvoelker has joined #openstack-nova03:01
*** haomaiwang has joined #openstack-nova03:01
*** hoangcx has quit IRC03:02
*** jianghuaw has quit IRC03:02
*** achanda has quit IRC03:03
*** markvoelker_ has joined #openstack-nova03:04
*** markvoel_ has quit IRC03:04
openstackgerritJiajun Liu proposed openstack/nova: config options: centralize cinder options  https://review.openstack.org/29819103:04
*** sdake has joined #openstack-nova03:07
*** Marga_ has joined #openstack-nova03:07
*** Marga_ has quit IRC03:07
*** fawadkhaliq has joined #openstack-nova03:08
*** Marga_ has joined #openstack-nova03:08
*** markvoelker has quit IRC03:08
*** Marga_ has quit IRC03:08
*** hoangcx has joined #openstack-nova03:08
*** Marga_ has joined #openstack-nova03:09
*** sdake_ has quit IRC03:10
*** pumarani__ has joined #openstack-nova03:10
*** apoorvad has joined #openstack-nova03:11
*** pushkaru has quit IRC03:14
*** apoorvad has quit IRC03:19
*** thorst has joined #openstack-nova03:19
*** tangchen has joined #openstack-nova03:21
*** pkoniszewski has quit IRC03:24
*** coreywright has quit IRC03:24
*** garyk1 has joined #openstack-nova03:24
*** jplopezgu_ has joined #openstack-nova03:25
*** fawadkhaliq has quit IRC03:26
*** fawadkhaliq has joined #openstack-nova03:26
*** ajo_ has joined #openstack-nova03:27
*** jeblair_ has joined #openstack-nova03:27
*** thorst has quit IRC03:27
*** haomaiw__ has joined #openstack-nova03:28
*** rhagarty__ has joined #openstack-nova03:28
*** keekz_ has joined #openstack-nova03:30
*** krotscheck has joined #openstack-nova03:30
*** hamzy_ has joined #openstack-nova03:31
*** dmellado_ has joined #openstack-nova03:31
*** scottda_ has joined #openstack-nova03:31
*** pkoniszewski has joined #openstack-nova03:31
*** fawadkhaliq has quit IRC03:32
*** rpodolyaka_ has joined #openstack-nova03:32
*** fawadkhaliq has joined #openstack-nova03:32
*** jlopezgu has quit IRC03:32
*** jplopezgu_ is now known as jlopezgu03:32
*** keekz has quit IRC03:32
*** dougwig has quit IRC03:32
*** jeblair has quit IRC03:32
*** ajo has quit IRC03:32
*** inara` has quit IRC03:32
*** haomaiwang has quit IRC03:32
*** rhagarty_ has quit IRC03:32
*** hamzy has quit IRC03:32
*** krotscheck_dcm has quit IRC03:32
*** scottda has quit IRC03:32
*** dmellado has quit IRC03:32
*** inara has joined #openstack-nova03:32
*** aorourke_ has joined #openstack-nova03:33
*** fmccrthy has quit IRC03:33
*** beagles has quit IRC03:33
*** qeelee has quit IRC03:33
*** rpodolyaka has quit IRC03:33
*** andrewbogott has quit IRC03:33
*** serverascode has quit IRC03:33
*** tpatzig has quit IRC03:33
*** BlackDex has quit IRC03:33
*** aorourke has quit IRC03:33
*** adreznec has quit IRC03:33
*** tpatzig has joined #openstack-nova03:33
*** rpodolyaka_ is now known as rpodolyaka03:34
*** b3nt_pin has joined #openstack-nova03:34
*** adreznec has joined #openstack-nova03:34
*** fmccrthy has joined #openstack-nova03:34
*** b3nt_pin is now known as Guest4039603:34
*** Jeffrey4l_ has joined #openstack-nova03:34
*** scottda_ is now known as scottda03:34
*** Jeffrey4l has quit IRC03:37
*** BlackDex has joined #openstack-nova03:38
*** andrewbogott has joined #openstack-nova03:38
*** dougwig has joined #openstack-nova03:38
*** serverascode has joined #openstack-nova03:39
*** pumarani__ has quit IRC03:41
*** liunian_ has quit IRC03:42
*** coreywright has joined #openstack-nova03:42
*** links has joined #openstack-nova03:44
*** liunian_ has joined #openstack-nova03:47
openstackgerritJianghua Wang proposed openstack/nova: XenAPI: add unit test for the plugin - glance  https://review.openstack.org/29953303:47
*** yarkot_ has joined #openstack-nova03:51
*** liunian_ has quit IRC03:51
*** salv-orlando has joined #openstack-nova03:52
*** yarkot_ has quit IRC03:53
*** salv-orlando has quit IRC03:55
*** liunian_ has joined #openstack-nova03:56
*** ljxiash has joined #openstack-nova03:57
*** pushkaru has joined #openstack-nova03:57
*** mylu_ has quit IRC03:58
*** fawadkhaliq has quit IRC03:58
openstackgerritVladik Romanovsky proposed openstack/nova: Add tag column to vifs and bdm  https://review.openstack.org/26401403:59
openstackgerritVladik Romanovsky proposed openstack/nova: Add metadata objects for device tagging  https://review.openstack.org/26401503:59
*** fawadkhaliq has joined #openstack-nova03:59
openstackgerritVladik Romanovsky proposed openstack/nova: Device tagging user visible changes  https://review.openstack.org/26401803:59
openstackgerritVladik Romanovsky proposed openstack/nova: Add compute manager device tagging support  https://review.openstack.org/26401703:59
openstackgerritVladik Romanovsky proposed openstack/nova: Libvirt driver implementation of device tagging  https://review.openstack.org/26401603:59
*** fawadkhaliq has quit IRC04:00
*** fawadkhaliq has joined #openstack-nova04:01
*** haomaiw__ has quit IRC04:01
*** ivase has joined #openstack-nova04:01
*** liunian_ has quit IRC04:01
*** haomaiwang has joined #openstack-nova04:01
*** ljxiash has quit IRC04:01
*** liunian_ has joined #openstack-nova04:01
*** irenab_ has joined #openstack-nova04:02
*** fawadkhaliq has quit IRC04:02
*** fawadkhaliq has joined #openstack-nova04:02
*** ivase_ has quit IRC04:03
*** achanda has joined #openstack-nova04:04
*** tangchen has quit IRC04:04
*** irenab has quit IRC04:04
*** irenab_ is now known as irenab04:04
*** anush_ has joined #openstack-nova04:04
*** liunian_ has quit IRC04:05
*** liunian_ has joined #openstack-nova04:08
*** achanda has quit IRC04:09
*** achanda has joined #openstack-nova04:09
*** baoli has joined #openstack-nova04:10
*** liunian_ has quit IRC04:13
*** baoli has quit IRC04:14
*** salv-orlando has joined #openstack-nova04:16
*** mylu has joined #openstack-nova04:17
*** liunian_ has joined #openstack-nova04:17
*** pushkaru has quit IRC04:21
*** markvoelker_ has quit IRC04:21
*** salv-orlando has quit IRC04:23
*** liunian_ has quit IRC04:27
*** sdake_ has joined #openstack-nova04:28
*** fawadkhaliq has quit IRC04:29
*** liunian_ has joined #openstack-nova04:29
*** fawadkhaliq has joined #openstack-nova04:30
*** sdake has quit IRC04:30
*** baoli has joined #openstack-nova04:31
*** dimtruck is now known as zz_dimtruck04:32
*** EinstCrazy has joined #openstack-nova04:33
*** efoley has joined #openstack-nova04:34
*** mylu has quit IRC04:34
*** Jeffrey4l_ has quit IRC04:34
*** rk4n has joined #openstack-nova04:35
*** rk4n has quit IRC04:35
*** baoli has quit IRC04:36
*** vilobhmm11 has quit IRC04:41
*** mylu has joined #openstack-nova04:42
*** moshele has joined #openstack-nova04:45
*** vilobhmm11 has joined #openstack-nova04:47
*** yamahata has joined #openstack-nova04:47
*** liunian_ has quit IRC04:49
*** efoley has quit IRC04:50
*** liunian_ has joined #openstack-nova04:51
*** liunian_ has quit IRC04:52
*** vishwanathj has quit IRC04:53
*** EinstCra_ has joined #openstack-nova04:57
*** haukebruno has joined #openstack-nova04:58
*** rcernin has joined #openstack-nova04:59
*** gjayavelu has joined #openstack-nova05:00
*** liunian_ has joined #openstack-nova05:00
*** haomaiwang has quit IRC05:01
*** EinstCrazy has quit IRC05:01
openstackgerritJianghua Wang proposed openstack/nova: XenAPI: add unit test for the plugin - glance  https://review.openstack.org/29953305:01
*** haomaiwang has joined #openstack-nova05:01
*** liunian_ has quit IRC05:05
*** liunian_ has joined #openstack-nova05:11
*** liunian_ has quit IRC05:11
*** nikhil has quit IRC05:13
*** dave-mccowan has quit IRC05:13
*** nikhil has joined #openstack-nova05:14
*** sdake_ is now known as sdake05:14
*** jaosorior has joined #openstack-nova05:14
*** fawadkhaliq has quit IRC05:15
*** fawadkhaliq has joined #openstack-nova05:16
*** liunian_ has joined #openstack-nova05:21
*** unicell has joined #openstack-nova05:23
*** salv-orlando has joined #openstack-nova05:23
*** liunian_ has quit IRC05:25
*** Sukhdev has joined #openstack-nova05:26
*** salv-orlando has quit IRC05:28
*** mtanino has quit IRC05:31
*** baoli has joined #openstack-nova05:32
*** thorst has joined #openstack-nova05:32
*** gjayavelu has quit IRC05:36
*** lpetrut has joined #openstack-nova05:36
*** baoli has quit IRC05:37
*** thorst has quit IRC05:37
*** fawadkhaliq has quit IRC05:37
*** sridharg has joined #openstack-nova05:38
*** Jeffrey4l_ has joined #openstack-nova05:39
*** mylu has quit IRC05:39
*** fawadkhaliq has joined #openstack-nova05:39
*** browne has quit IRC05:40
*** liunian_ has joined #openstack-nova05:42
*** pcaruana has quit IRC05:49
*** mtanino has joined #openstack-nova05:50
*** Jeffrey4l_ has quit IRC05:54
*** e0ne has joined #openstack-nova05:56
*** gjayavelu has joined #openstack-nova05:57
*** ihrachys has joined #openstack-nova05:57
*** ivase_ has joined #openstack-nova05:57
*** Sukhdev has quit IRC05:57
*** Swami has joined #openstack-nova05:59
*** fawadkhaliq has quit IRC05:59
*** fawadkhaliq has joined #openstack-nova06:00
*** ivase has quit IRC06:00
*** haomaiwang has quit IRC06:01
*** gjayavelu has quit IRC06:01
*** haomaiwang has joined #openstack-nova06:01
*** mtanino has quit IRC06:03
*** fawadkhaliq has quit IRC06:04
*** fawadkhaliq has joined #openstack-nova06:08
*** mnestratov|2 has joined #openstack-nova06:11
*** ihrachys has quit IRC06:11
*** nkrinner has joined #openstack-nova06:12
*** rk4n has joined #openstack-nova06:12
*** mpavone has joined #openstack-nova06:15
*** mjura has joined #openstack-nova06:16
*** unicell has quit IRC06:16
*** unicell has joined #openstack-nova06:17
*** lpetrut has quit IRC06:17
*** woodster_ has joined #openstack-nova06:18
*** fawadkhaliq has quit IRC06:19
*** fawadkhaliq has joined #openstack-nova06:19
*** esracelik has joined #openstack-nova06:20
*** daemontool has joined #openstack-nova06:22
*** anush_ has quit IRC06:22
*** yamahata has quit IRC06:23
*** e0ne has quit IRC06:24
*** Swami has quit IRC06:24
*** liunian__ has joined #openstack-nova06:29
*** liunian_ has quit IRC06:29
*** thorst has joined #openstack-nova06:34
*** sdake_ has joined #openstack-nova06:37
*** sdake has quit IRC06:40
*** garyk1 has quit IRC06:40
*** thorst has quit IRC06:42
*** sdake has joined #openstack-nova06:45
*** rk4n has quit IRC06:46
*** hoangcx has quit IRC06:47
*** hoangcx has joined #openstack-nova06:48
*** salv-orlando has joined #openstack-nova06:48
*** sdake_ has quit IRC06:48
*** fawadkhaliq has quit IRC06:48
*** fawadkhaliq has joined #openstack-nova06:49
*** tesseract has joined #openstack-nova06:49
*** tesseract is now known as Guest1103706:49
*** baoli has joined #openstack-nova06:50
*** salv-orlando has quit IRC06:51
*** rk4n has joined #openstack-nova06:52
*** baoli has quit IRC06:55
*** rk4n has quit IRC06:56
*** ihrachys has joined #openstack-nova06:58
*** moshele has quit IRC07:00
*** haomaiwang has quit IRC07:01
*** haomaiwang has joined #openstack-nova07:01
*** moshele has joined #openstack-nova07:02
*** daemontool has quit IRC07:04
*** rk4n has joined #openstack-nova07:05
*** gongysh has quit IRC07:06
*** liverpooler has joined #openstack-nova07:06
*** Daisy_ has joined #openstack-nova07:07
*** ivase has joined #openstack-nova07:07
*** Daisy has quit IRC07:10
*** ivase_ has quit IRC07:10
*** Daisy has joined #openstack-nova07:11
*** Daisy__ has joined #openstack-nova07:12
*** Daisy___ has joined #openstack-nova07:12
*** rk4n has quit IRC07:14
*** Daisy_ has quit IRC07:14
*** e0ne has joined #openstack-nova07:15
*** Daisy has quit IRC07:16
*** rdopiera has joined #openstack-nova07:16
*** Daisy___ has quit IRC07:16
*** Daisy__ has quit IRC07:16
*** Daisy has joined #openstack-nova07:16
*** pushkaru has joined #openstack-nova07:17
*** Daisy_ has joined #openstack-nova07:18
*** e0ne has quit IRC07:18
*** Daisy has quit IRC07:18
*** Daisy has joined #openstack-nova07:20
*** atuvenie has joined #openstack-nova07:21
*** Daisy_ has quit IRC07:23
*** gongysh has joined #openstack-nova07:23
*** Daisy has quit IRC07:24
*** takashin has joined #openstack-nova07:26
*** jlanoux has joined #openstack-nova07:26
*** jlanoux_ has joined #openstack-nova07:28
*** jlanoux has quit IRC07:28
*** p01nt3r75 has joined #openstack-nova07:29
*** zenoway has joined #openstack-nova07:30
*** stevemar has quit IRC07:33
*** omnipresent has joined #openstack-nova07:33
*** zenoway has quit IRC07:35
*** pushkaru has quit IRC07:38
*** wuhg has joined #openstack-nova07:38
*** zenoway has joined #openstack-nova07:39
*** omnipresent has quit IRC07:40
*** liverpooler has quit IRC07:40
*** liverpooler has joined #openstack-nova07:41
*** thorst has joined #openstack-nova07:41
*** pcaruana has joined #openstack-nova07:41
*** coreywright has quit IRC07:47
*** thorst has quit IRC07:47
*** permalac has joined #openstack-nova07:48
*** moshele has quit IRC07:48
*** armax_ has joined #openstack-nova07:49
*** moshele has joined #openstack-nova07:49
*** rk4n has joined #openstack-nova07:49
*** mgoddard has joined #openstack-nova07:50
*** armax has quit IRC07:52
*** armax_ is now known as armax07:52
*** ihrachys has quit IRC07:53
*** jistr has joined #openstack-nova07:53
*** matrohon has joined #openstack-nova07:54
openstackgerritZhenyu Zheng proposed openstack/nova: config options: centralize section: "crypto"  https://review.openstack.org/27391607:56
*** fawadkhaliq has quit IRC07:59
*** cristicalin has joined #openstack-nova07:59
*** haomaiwang has quit IRC08:01
*** haomaiwang has joined #openstack-nova08:01
*** fawadkhaliq has joined #openstack-nova08:01
Kevin_Zhengmikal: Hi, can you remove your -2 on https://review.openstack.org/#/c/272949/ , since Newon is open and I have a dependent patch on this one.08:03
*** fawadkhaliq has quit IRC08:03
*** coreywright has joined #openstack-nova08:03
*** fawadkhaliq has joined #openstack-nova08:04
*** permalac has quit IRC08:06
*** salv-orlando has joined #openstack-nova08:07
*** salv-orlando has quit IRC08:09
*** Daisy has joined #openstack-nova08:14
*** PaulMurray has joined #openstack-nova08:15
*** Daisy has quit IRC08:15
PaulMurraygood morning everyone08:15
*** Daisy has joined #openstack-nova08:15
*** sdake_ has joined #openstack-nova08:17
*** salv-orlando has joined #openstack-nova08:18
*** sdake has quit IRC08:19
*** ndipanov has joined #openstack-nova08:20
*** sdake has joined #openstack-nova08:22
*** salv-orlando has quit IRC08:22
*** sdake_ has quit IRC08:24
*** permalac has joined #openstack-nova08:25
*** woodster_ has quit IRC08:27
*** Daisy_ has joined #openstack-nova08:27
*** e0ne has joined #openstack-nova08:28
*** e0ne has quit IRC08:29
*** ihrachys has joined #openstack-nova08:30
*** ndipanov has quit IRC08:30
*** Daisy has quit IRC08:30
*** achanda has quit IRC08:30
*** Daisy_ has quit IRC08:31
*** sfinucan has joined #openstack-nova08:33
*** stevemar has joined #openstack-nova08:33
*** achanda has joined #openstack-nova08:34
*** e0ne has joined #openstack-nova08:34
*** e0ne has quit IRC08:35
*** akshai_ has quit IRC08:37
*** fawadkhaliq has quit IRC08:37
*** yassine has joined #openstack-nova08:37
openstackgerritLi Yingjun proposed openstack/nova-specs: Hypervisor notifications spec  https://review.openstack.org/29980708:37
*** yassine is now known as Guest4145908:38
*** Guest41459 is now known as yassou08:38
*** stevemar has quit IRC08:39
*** fawadkhaliq has joined #openstack-nova08:40
*** daemontool has joined #openstack-nova08:41
*** ivase_ has joined #openstack-nova08:41
*** ivase has quit IRC08:44
*** thorst has joined #openstack-nova08:45
*** ljxiash has joined #openstack-nova08:46
*** tbachman has quit IRC08:46
*** cristicalin has quit IRC08:47
*** hoangcx has quit IRC08:47
*** tbachman has joined #openstack-nova08:47
*** daemontool has quit IRC08:48
*** achanda has quit IRC08:48
*** achanda has joined #openstack-nova08:49
*** ljxiash has quit IRC08:50
*** baoli has joined #openstack-nova08:51
*** josecastroleon has joined #openstack-nova08:51
*** garyk2 has joined #openstack-nova08:52
*** thorst has quit IRC08:52
*** achanda has quit IRC08:52
*** zioproto has joined #openstack-nova08:53
garyk2markus_z: please see https://review.openstack.org/218639 - that should address the bug that we spoke about yesterday08:53
zioprotohello there. I tried to submit a change into Openstack https://review.openstack.org/#/c/295865/ . So far I had positive review. Should I worry about the -1 from Xen Project CI ?08:55
*** garyk2 has quit IRC08:55
*** garyk1 has joined #openstack-nova08:55
*** baoli has quit IRC08:55
*** hoangcx has joined #openstack-nova08:56
*** pece has joined #openstack-nova08:56
*** daemontool has joined #openstack-nova08:58
*** achanda has joined #openstack-nova08:59
*** rdopiera has quit IRC09:00
*** Daisy has joined #openstack-nova09:00
*** haomaiwang has quit IRC09:01
*** haomaiwang has joined #openstack-nova09:01
*** Daisy_ has joined #openstack-nova09:01
*** rdopiera has joined #openstack-nova09:01
* sheel looking for review of https://review.openstack.org/#/c/280371/09:02
gibizioproto: as far as I see the failure is not related to your change set so you should retrigger the Xen run to see if it succeeds. You can do that by adding a comment 'xen: recheck' to your review09:02
*** vilobhmm11 has quit IRC09:03
*** Daisy_ has quit IRC09:03
*** fawadkhaliq has quit IRC09:05
*** Daisy has quit IRC09:06
markus_zgaryk1: thanks! I'll take a look. I'm not yet sure if we released RC3, if yes, I'll propose it for a backport.09:09
*** fawadkhaliq has joined #openstack-nova09:09
*** Daisy has joined #openstack-nova09:10
*** ihrachys has quit IRC09:11
openstackgerritStephen Finucane proposed openstack/nova: config options: centralize section "workarounds"  https://review.openstack.org/29808309:11
openstackgerritStephen Finucane proposed openstack/nova: config options: Centralise 'utils' options  https://review.openstack.org/29877509:11
*** Daisy has quit IRC09:11
zioprotogibi: so just click on 'Reply ...' on top with text 'xen: recheck' and leave Code-Review and Workflow with 0 0 ?09:11
zioprotogibi: I hope I got it right https://review.openstack.org/#/c/295865/509:15
gibizioproto: you did it right. :) Xen will redo the verification of your patchset09:17
*** fhorvath has joined #openstack-nova09:18
*** derekh has joined #openstack-nova09:19
*** eliqiao has quit IRC09:19
*** danpb has joined #openstack-nova09:19
openstackgerritStephen Finucane proposed openstack/nova: config options: Centralise 'monkeypatch' options  https://review.openstack.org/29877609:20
openstackgerritStephen Finucane proposed openstack/nova: Improve 'monkey_patch' conf options documentation  https://review.openstack.org/29879609:20
markus_zgaryk1: Could you add the "closes-bug 1563684" to that patch please? Or is Vipin double-checking it?09:23
openstackbug 1563684 in OpenStack Compute (nova) "VMware: rebuild an instance failed" [Medium,New] https://launchpad.net/bugs/1563684 - Assigned to Vipin Balachandran (vbala)09:23
*** salv-orlando has joined #openstack-nova09:23
openstackgerritVasyl Saienko proposed openstack/nova: Allow to update resource per single node.  https://review.openstack.org/28748509:24
*** fawadkhaliq has quit IRC09:25
*** lpetrut has joined #openstack-nova09:25
*** achanda has quit IRC09:26
*** Jeffrey4l_ has joined #openstack-nova09:26
*** fawadkhaliq has joined #openstack-nova09:26
*** Daisy has joined #openstack-nova09:29
*** e0ne has joined #openstack-nova09:30
*** lpetrut has quit IRC09:31
*** lpetrut has joined #openstack-nova09:31
*** sdake_ has joined #openstack-nova09:32
*** sdake has quit IRC09:32
*** sdake has joined #openstack-nova09:33
*** gszasz has joined #openstack-nova09:35
*** bogdano has joined #openstack-nova09:35
*** sdake_ has quit IRC09:36
*** e0ne has quit IRC09:37
*** takashin has left #openstack-nova09:39
*** salv-orlando has quit IRC09:39
*** Daisy has quit IRC09:40
*** Daisy has joined #openstack-nova09:40
*** fawadkhaliq has quit IRC09:41
*** atuvenie has quit IRC09:42
bauzasgood morning nova09:42
*** e0ne has joined #openstack-nova09:42
*** fawadkhaliq has joined #openstack-nova09:42
*** daemontool has quit IRC09:43
*** mkrcmari__ has joined #openstack-nova09:44
*** daemontool has joined #openstack-nova09:44
*** fawadkhaliq has quit IRC09:44
*** Daisy has quit IRC09:44
*** ndipanov has joined #openstack-nova09:45
*** liunian__ has quit IRC09:46
*** mvk_ has quit IRC09:47
gmannmriedem:  Added tempest tests https://review.openstack.org/#/c/299830/09:48
gmannI will add Depends-On(or you add) once we have results with current code09:49
*** liunian_ has joined #openstack-nova09:49
*** thorst has joined #openstack-nova09:50
*** fawadkhaliq has joined #openstack-nova09:50
*** mvk_ has joined #openstack-nova09:52
*** mkrcmari__ has quit IRC09:55
*** yassou has quit IRC09:55
*** thorst has quit IRC09:56
*** rmart04 has joined #openstack-nova09:59
*** haomaiwang has quit IRC10:01
*** haomaiwang has joined #openstack-nova10:01
*** cdent has joined #openstack-nova10:01
*** eliqiao has joined #openstack-nova10:01
*** sdague has joined #openstack-nova10:03
*** sdake has quit IRC10:03
*** rook-lappy has quit IRC10:05
*** fawadkhaliq has quit IRC10:09
*** ihrachys has joined #openstack-nova10:10
*** atuvenie has joined #openstack-nova10:10
*** fawadkhaliq has joined #openstack-nova10:10
*** mkrcmari__ has joined #openstack-nova10:13
openstackgerritSean Dague proposed openstack/nova: Handle pre-migration flavor creation failures in the crusty old API  https://review.openstack.org/29962510:13
*** e0ne has quit IRC10:15
*** atuvenie has quit IRC10:17
*** mvk_ has quit IRC10:17
*** e0ne has joined #openstack-nova10:17
*** hoangcx has quit IRC10:18
*** fawadkhaliq has quit IRC10:18
*** fawadkhaliq has joined #openstack-nova10:19
*** e0ne has quit IRC10:20
*** EinstCra_ has quit IRC10:21
*** e0ne has joined #openstack-nova10:21
*** e0ne has quit IRC10:22
*** e0ne has joined #openstack-nova10:23
*** atuvenie has joined #openstack-nova10:23
*** yassine has joined #openstack-nova10:23
*** e0ne has quit IRC10:24
*** yassine is now known as Guest2368710:24
*** Guest23687 is now known as yassou10:25
*** fawadkhaliq has quit IRC10:25
*** mvk_ has joined #openstack-nova10:28
*** ociuhandu has quit IRC10:30
*** smatzek has joined #openstack-nova10:32
*** mkrcmari__ has quit IRC10:32
*** atuvenie has quit IRC10:34
*** stevemar has joined #openstack-nova10:35
*** fawadkhaliq has joined #openstack-nova10:39
*** stevemar has quit IRC10:39
*** ivase_ has quit IRC10:41
*** permalac has quit IRC10:44
liunian_I'm really puzzled ,why not nova provide snapshot function?10:46
*** fawadkhaliq has quit IRC10:49
ndipanovliunian_, it does provide several ways to do it10:49
ndipanovliunian_, see nova image-create cli command10:50
liunian_yes, but i think ,it is too complicated。10:51
*** salv-orlando has joined #openstack-nova10:51
liunian_https://blueprints.launchpad.net/nova/+spec/nova-snapshot         it is my bp。10:51
*** doug-fish has joined #openstack-nova10:51
*** baoli has joined #openstack-nova10:52
*** dmellado_ is now known as dmellado10:52
*** thorst has joined #openstack-nova10:54
*** baoli has quit IRC10:57
markus_ztonyb: regarding libvirt test job: I pushed https://review.openstack.org/#/c/299840/ Could you have a look please?10:59
openstackgerritNikola Dipanov proposed openstack/nova: pci: Add utility method for getting the MAC addr  https://review.openstack.org/26258510:59
openstackgerritNikola Dipanov proposed openstack/nova: neutron: Update the port with a MAC address for PFs  https://review.openstack.org/26258610:59
markus_ztonyb: The test job is merged to the "project.config" but a test run fails: http://logs.openstack.org/20/299320/2/experimental/gate-tempest-dsvm-nova-libvirt-kvm-apr/ee9c4ef/logs/devstacklog.txt.gz11:00
*** haomaiwang has quit IRC11:01
*** ociuhandu has joined #openstack-nova11:01
*** thorst has quit IRC11:01
*** haomaiwang has joined #openstack-nova11:01
*** moshele has quit IRC11:03
*** moshele has joined #openstack-nova11:03
ndipanovmoshele, when you get a sec: https://review.openstack.org/#/c/262585/11:05
*** liunian_ has quit IRC11:09
*** cdent has quit IRC11:11
*** Kevin_Zheng has quit IRC11:12
openstackgerritsahid proposed openstack/nova: libvirt: add "get_job_info" to Guest's object  https://review.openstack.org/25074011:12
openstackgerritsahid proposed openstack/nova: libvirt: add method to configure max downtime when migrating  https://review.openstack.org/25073911:12
openstackgerritsahid proposed openstack/nova: libvirt: add "abort_job" to Guest's object  https://review.openstack.org/25073811:12
openstackgerritsahid proposed openstack/nova: libvirt: add method "migrate" to Guest's object  https://review.openstack.org/25036511:12
*** fhorvath has quit IRC11:12
openstackgerritYaoZheng proposed openstack/nova: After migrate in-use volume the BDM information lost  https://review.openstack.org/28843311:13
*** cdent has joined #openstack-nova11:14
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: [extensions] cache discovered extensions  https://review.openstack.org/28713111:16
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Add support of caching  https://review.openstack.org/29986011:16
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Discover extensions from contrib dir with pkgutil  https://review.openstack.org/29986111:16
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Cache discover_version method  https://review.openstack.org/29986211:16
*** e0ne has joined #openstack-nova11:17
*** e0ne has quit IRC11:18
*** unicell has quit IRC11:22
*** jlvillal_ has joined #openstack-nova11:22
*** adreznec has quit IRC11:22
*** tbachman has quit IRC11:22
*** allen_gao has quit IRC11:22
*** jlvillal has quit IRC11:22
*** dhellmann has quit IRC11:22
*** allen_ga- has joined #openstack-nova11:22
*** dhellmann_ has joined #openstack-nova11:22
*** unicell has joined #openstack-nova11:23
*** adreznec has joined #openstack-nova11:23
*** e0ne has joined #openstack-nova11:23
*** jlvillal_ is now known as jlvillal11:23
*** jlvillal is now known as Guest8757811:24
*** claudiub has joined #openstack-nova11:25
*** Guest40396 is now known as b3nt_pin11:26
*** e0ne has quit IRC11:26
*** dhellmann_ is now known as dhellmann11:26
*** b3nt_pin is now known as beagles11:27
*** tbachman has joined #openstack-nova11:30
*** vladikr has joined #openstack-nova11:32
*** rtheis has joined #openstack-nova11:33
*** e0ne has joined #openstack-nova11:35
*** vladikr has quit IRC11:39
*** thorst has joined #openstack-nova11:41
*** e0ne has quit IRC11:41
*** haukebruno has quit IRC11:42
cdentsdague: if you've not already started I'll go ahead and see about intergrating microversion-parse into nova11:43
*** zenoway has quit IRC11:44
*** zenoway has joined #openstack-nova11:45
*** zenoway has quit IRC11:49
*** vladikr has joined #openstack-nova11:52
*** rfolco has joined #openstack-nova11:53
openstackgerritsahid proposed openstack/nova: virt: reserved number of mempages on compute host  https://review.openstack.org/29249911:53
*** sahid has joined #openstack-nova11:57
*** e0ne has joined #openstack-nova11:59
*** zenoway has joined #openstack-nova11:59
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: [extensions] cache discovered extensions  https://review.openstack.org/28713111:59
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Cache discover_version method  https://review.openstack.org/29986211:59
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Add support of caching  https://review.openstack.org/29986011:59
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Discover extensions from contrib dir with pkgutil  https://review.openstack.org/29986112:00
*** e0ne has quit IRC12:00
*** haomaiwang has quit IRC12:01
*** moshele has quit IRC12:01
*** haomaiwang has joined #openstack-nova12:01
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: [extensions] cache discovered extensions  https://review.openstack.org/28713112:03
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Cache discover_version method  https://review.openstack.org/29986212:03
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Add support of caching  https://review.openstack.org/29986012:03
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Discover extensions from contrib dir with pkgutil  https://review.openstack.org/29986112:03
*** vladikr has quit IRC12:05
openstackgerritZhenyu Zheng proposed openstack/nova: config options: centralize section "cloudpipe"  https://review.openstack.org/27294912:09
*** aysyd has joined #openstack-nova12:09
*** ayoung has quit IRC12:10
*** sdague_ has joined #openstack-nova12:11
*** e0ne has joined #openstack-nova12:11
*** aysyd has quit IRC12:11
*** sfinucan has quit IRC12:11
*** efried has joined #openstack-nova12:12
*** e0ne has quit IRC12:13
openstackgerritZhenyu Zheng proposed openstack/nova: config options: centralize section: "crypto"  https://review.openstack.org/27391612:14
*** aysyd has joined #openstack-nova12:16
*** ducttape_ has joined #openstack-nova12:16
*** pradk has joined #openstack-nova12:16
openstackgerritSean Dague proposed openstack/nova: remove glance deprecated config  https://review.openstack.org/29953512:17
openstackgerritsahid proposed openstack/os-vif: test: use real UUID in all UUID fields  https://review.openstack.org/29940412:17
*** sfinucan has joined #openstack-nova12:18
*** vladikr has joined #openstack-nova12:18
*** EinstCrazy has joined #openstack-nova12:19
*** tbachman has quit IRC12:19
*** moshele has joined #openstack-nova12:19
*** dave-mccowan has joined #openstack-nova12:20
*** rodrigods has quit IRC12:21
*** rodrigods has joined #openstack-nova12:22
openstackgerritSean Dague proposed openstack/nova: remove the ability to disable v2.1  https://review.openstack.org/29989312:26
*** baoli has joined #openstack-nova12:27
*** baoli has quit IRC12:27
*** baoli has joined #openstack-nova12:27
moshelendipanov: done12:28
openstackgerritSean Dague proposed openstack/nova: WIP: warn if using the legacy v2 api stack  https://review.openstack.org/29989512:31
*** ivase has joined #openstack-nova12:31
*** haomaiwang has quit IRC12:32
*** ducttape_ has quit IRC12:33
*** baoli_ has joined #openstack-nova12:34
*** stevemar has joined #openstack-nova12:36
ndipanovmoshele, thanks12:36
*** ivase_ has joined #openstack-nova12:36
*** ivase has quit IRC12:36
*** raildo-afk is now known as raildo12:36
*** p01nt3r75 has quit IRC12:37
*** baoli has quit IRC12:37
*** ivase has joined #openstack-nova12:38
*** daemontool_ has joined #openstack-nova12:39
*** tbachman has joined #openstack-nova12:39
*** burgerk has joined #openstack-nova12:39
*** ivase_ has quit IRC12:40
*** delatte has quit IRC12:40
*** stevemar has quit IRC12:40
johnthetubaguyalaski: are you around to talk about policy things?12:41
alaskijohnthetubaguy: yeah12:41
johnthetubaguyalaski: I was thinking more generally about the fine grained policy stuff, like: os_compute_api:servers:create:attach_volume12:42
*** daemontool has quit IRC12:42
johnthetubaguyalaski: I wonder if we want to transition to be more opinionated, so there is something like os_compute_api:supports_volumes that turns on or off a whole load of things12:42
johnthetubaguyalaski: I am really thinking about interoperability12:43
johnthetubaguyits just that corse grained thing, might we want we want to expose to users, a bit like you are proposing12:43
sdague_johnthetubaguy: honestly, I think we should get it loaded to code first12:44
sdague_then build a fascility for aliases and groups in policy, so that we can start renaming things in sensible ways12:44
alaskiagreed12:44
sdague_but until that's code annotated, it's going to be a mess to manage12:44
johnthetubaguyyeah, that makes sense12:44
openstackgerritGleb Stepanov proposed openstack/nova: [WIP] Add console log offset  https://review.openstack.org/29247312:45
johnthetubaguymake upgrade work, so we can then actually go fix things12:45
alaskijohnthetubaguy: I like the idea, but it breaks backwards compat because currently deployers can configure things that wouldn't work at a coarse grained level12:45
sdague_right, exactly12:45
*** ZZelle has joined #openstack-nova12:45
johnthetubaguyalaski: yeah, it would totally need a deprecation cycle and alias stuff, etc12:45
sdague_it gets made a ton easier if people are only specifying overrides in policy.json12:46
sdague_and a lot of people are running with empty policy files12:46
alaskijohnthetubaguy: yeah. I originally has some of that in my proposal and had to back it out when claudiub pointed out that current policies weren't necessarily expressable like that12:46
alaskisdague_: yep12:46
sdague_I also think scope wise, policies into code with no change on names, might actually fit in a cycle12:47
johnthetubaguysdague_: +1 getting most folks onto empty policy files12:47
sdague_the rest definitely won't12:47
*** houming has quit IRC12:47
johnthetubaguysdague: yeah, thats sounds about right12:47
sdague_johnthetubaguy: so, you have a chance to look at the api-ref patches yet?12:47
johnthetubaguyso I totally by just getting the defaults into code, as a step12:47
johnthetubaguysdague: crud, I did just after the API meeting, then got distracted... arg12:48
sdague_I think we all have a similar vision of where to get to eventually, we just need to keep the scope narrow enough to make real progress12:48
alaski+112:48
johnthetubaguyOK, I wasn't so sure we all have a similar end vision12:49
johnthetubaguybut sounds like I am wrong, sounds like we are shooting in very similar directions12:49
*** daemontool__ has joined #openstack-nova12:49
johnthetubaguytotally agreed getting the defaults into code makes sense as a first step, so we can do the alias thing12:49
*** e0ne has joined #openstack-nova12:50
*** dmellado is now known as dmellado|lunch12:50
johnthetubaguyalaski: is it worth reducing the scope of that spec a little, so it just looks at moving the defaults into code, running gate on an empty policy file?12:50
alaskiI think we all agree that we need a better way to express capabilities/endpoints/whatever. we may not yet agree on exactly the form, but first we need a way to even transition12:50
sdague_alaski: ++12:51
*** e0ne has quit IRC12:51
sdague_and nothing about this precludes anything I've heard from folks as expressed end states12:51
sdague_ok, got to drop for a bit. johnthetubaguy if you could take a look at - https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:wip_api_docs2 would be great12:52
johnthetubaguyI mean it stops the canonical defaults living in a keystone database, but thats no bad thing12:52
*** daemontool_ has quit IRC12:52
edleafejohnthetubaguy: can you remove your -2 from https://review.openstack.org/#/c/282419/ ?12:52
sdague_johnthetubaguy: right, we always thought that was a bad idea though12:52
johnthetubaguysdague_: within the Nova discussions certainly12:53
*** liusheng has quit IRC12:53
johnthetubaguysdague: is this commented out code here on purpose? https://review.openstack.org/#/c/292420/10/api-ref/ext/rest_parameters.py@7812:53
alaskijohnthetubaguy: I can reduce the scope, sure. we lose the ability to build capability exposure on top so I think we want a second spec to add something for that12:53
johnthetubaguyalaski: yeah, I think we want to talk about that separately really12:54
*** liusheng has joined #openstack-nova12:54
johnthetubaguyalaski: once we expose that, its a micro versioned API that lives for ever, I guess, hence me thinking about very course grained things that are consistent across all deployments12:54
*** daemontool__ has quit IRC12:55
tbachmangaryk1: you around?12:55
alaskijohnthetubaguy: sure. I've very intentionally avoided discussing the API. My spec only deals with what we can glean internally from policy rules12:55
johnthetubaguysdague_: maybe we ignore https://review.openstack.org/#/c/292420/10/api-ref/ext/rest_parameters.py@78 and fix it up in a follow on patch, given we have a +2 on there12:55
tbachmangaryk1: was wondering if you could give your blessings to the latest iteration of: https://review.openstack.org/#/c/28807612:56
*** pushkaru has joined #openstack-nova12:56
alaskijohnthetubaguy: I think to your concern, https://review.openstack.org/#/c/289405/ is already a separate spec for the API side of this12:57
*** sdague_ has quit IRC12:57
*** penick has quit IRC12:57
*** edmondsw has joined #openstack-nova12:57
johnthetubaguyalaski: OK, so I should re-read to see how that other bit helps us12:57
*** cdelatte has joined #openstack-nova12:57
*** rlrossit has joined #openstack-nova12:58
*** links has quit IRC12:58
alaskijohnthetubaguy: what I'm proposing is the mechanism that allows us to link policies into a coarse grained definition, but it doesn't get into what definitions we use12:58
*** daemontool__ has joined #openstack-nova12:59
*** daemontool__ is now known as daemontool13:00
*** achanda has joined #openstack-nova13:00
*** Goneri has joined #openstack-nova13:01
*** Daisy has joined #openstack-nova13:01
johnthetubaguyalaski: my gut says get the defaults in code, and work out some alias system for those as a first step, on its own, I am just not sure about doing the coarse grained thing until we have a clear user of it13:01
johnthetubaguysdague: there is commented out code in the plugin, maybe tidy that up in a follow on patch?13:01
*** mdrabe has joined #openstack-nova13:04
*** jaosorior has quit IRC13:05
*** achanda has quit IRC13:06
*** mvk_ has quit IRC13:06
*** jaosorior has joined #openstack-nova13:06
*** Daisy has quit IRC13:06
*** zul has quit IRC13:08
openstackgerritClaudiu Belu proposed openstack/nova: Fixes python 3 unit tests  https://review.openstack.org/27881013:08
openstackgerritGleb Stepanov proposed openstack/nova: Fix wrong watermark state during live migration  https://review.openstack.org/27961613:08
openstackgerritGleb Stepanov proposed openstack/nova: Fix wrong watermark state during live migration  https://review.openstack.org/27961613:10
*** akshai has joined #openstack-nova13:10
*** rfolco has quit IRC13:11
* alex_xu try to review api-ref patch also, but looks like he need more knowledge about sphinx13:12
*** penick has joined #openstack-nova13:13
*** ctrath has joined #openstack-nova13:13
*** yamahata has joined #openstack-nova13:13
*** liverpooler has quit IRC13:15
*** baoli_ has quit IRC13:16
tbachmangaryk1: thanks!13:17
*** salv-orlando has quit IRC13:18
*** rfolco has joined #openstack-nova13:19
*** salv-orlando has joined #openstack-nova13:19
*** mtanino has joined #openstack-nova13:20
*** pwnall has joined #openstack-nova13:21
*** Goneri has quit IRC13:22
*** anush_ has joined #openstack-nova13:22
*** cleong has joined #openstack-nova13:22
*** garyk1 has quit IRC13:23
*** salv-orl_ has joined #openstack-nova13:24
sdaguejohnthetubaguy: yeh, at lot of it is logging13:25
*** salv-orlando has quit IRC13:26
sdaguejohnthetubaguy: there isn't really a great way to conditionally enable that stuff, so right now leaving it there for people to know where the important parts are is probably useful13:26
johnthetubaguysdague: oh, I see what you mean, I figured there would be a purpose to that13:26
*** ccarmack has joined #openstack-nova13:27
*** zul has joined #openstack-nova13:27
sdaguefor instance, you can't print debug sphinx, that's all swallowed13:29
sdagueyou have to use it's logger13:29
*** akshai has quit IRC13:29
sdagueand where that logger is in the object model changes depending on context you are in13:29
*** tritro has joined #openstack-nova13:30
*** diogogmt has quit IRC13:30
*** akshai has joined #openstack-nova13:30
*** knikolla has joined #openstack-nova13:32
johnthetubaguysdague: I wish I knew about that thing when I was looking at the feature classification stuff, seems worth leaving in13:32
*** e0ne has joined #openstack-nova13:32
sdaguejohnthetubaguy: yeh... there are some hard fought discoveries in this code13:33
sdagueI read a lot of sphinx / docutils code, plus asked dhellmann many questions13:33
mnaserwhos the best person to bother regarding the objects code in nova13:33
mnaseri've resolved a bug but tests are not failing because the object change so it wants a version update, however, the old version was broken to start with from the first place13:34
mnaserso i'm not sure how to go about doing this13:34
rlrossitmnaser: either dansmith or myself. If I can't answer the question, I can just toss it over to him13:34
mnaserthis is regarding patch: https://review.openstack.org/#/c/299708/13:34
rlrossitmnaser: I'll take a look13:34
mnaserthank you :)13:34
*** diogogmt has joined #openstack-nova13:35
*** stevemar has joined #openstack-nova13:36
alex_xujohnthetubaguy: sdague are we safe to remove extension options https://github.com/openstack/nova/blob/master/nova/api/openstack/__init__.py#L49 ? if yes, I will cook a patch for those.13:36
*** Goneri has joined #openstack-nova13:37
dansmithmnaser: yeah, you can't do that.. also, jaypipes is the person to look at that13:37
sdaguealex_xu: I don't know on that one honestly13:37
dansmithmnaser: this floatness came up at some point and I thought he had a solution for it (which might be to just integerify it)13:37
sdagueI feel like that probably needs to be an austin conversation13:37
johnthetubaguyalex_xu: hmm, the v2.1 ones I would say yes, not so sure about v2.0 ones, I would rather they died when the v2.0 code died13:37
*** mriedem has joined #openstack-nova13:37
johnthetubaguysdague: you are right though, its a summit conversation I think13:37
sdagueI still think the extensions thing is going to catch a lot of folks by surprise13:38
mnaserdansmith: I see... I will try to get in touch with him to see what would be the ideal solution in that case13:38
alex_xujohnthetubaguy: they are separaed option between v2 and v2.1 I think13:38
johnthetubaguysdague: so if we have policy being only overrides, it would be an easier call, for sure13:38
mnaserBecause all values were returning 0, and now that we have it working with that patch internally, CPU usage is *much* better distributed13:38
dansmithmnaser: I just added him to the review13:38
mnaserCool, thank you so much :)13:38
*** salv-orl_ has quit IRC13:38
sdaguejohnthetubaguy: yeh13:38
johnthetubaguyalex_xu: yep, its two different set of variables13:38
dansmithmnaser: which hypervisor?13:38
mnaserdansmith: kvm13:38
dansmithmnaser: okay, well, anyway, punt to jay13:39
mnaseryep, i'll follow up once he replies and hopefully we can get this resolved (and justified to be backported to liberty)13:39
mnaserbut that's another discussion :d13:39
alex_xujohnthetubaguy: ok, got it, let's wait after summit13:39
*** ametts has joined #openstack-nova13:39
*** mylu has joined #openstack-nova13:40
sdaguedansmith: I updated - https://review.openstack.org/#/c/299625/ with a test13:40
dansmithsdague: I saw, thanks13:40
*** stevemar has quit IRC13:41
*** lixiaoy1 has quit IRC13:41
*** salv-orlando has joined #openstack-nova13:43
rlrossitmnaser: dansmith: I went hunting down the changes from when we had this discussion last time. The commits of interest are https://github.com/openstack/nova/commit/ddb3c4e26f3a908de7836a8fac48360afa6aa1d7 and https://github.com/openstack/nova/commit/87e1d8fab2b5888e711f07001025600fe709aaa713:43
*** burgerk has quit IRC13:44
*** pushkaru has quit IRC13:44
*** haomaiwang has joined #openstack-nova13:44
dansmithrlrossit: ah, yeah, thanks.. that's what I thought we did.. integer and *= 10013:44
mnaserrlrossit you know what, i think the https://github.com/openstack/nova/commit/ddb3c4e26f3a908de7836a8fac48360afa6aa1d7 patch might be missing from liberty13:45
dansmithmnaser: maybe this fixes it for you on an older verson but isn't related?13:45
mnaser(or at least the ubuntu release)13:45
*** permalac has joined #openstack-nova13:45
mnaserbecause when i was troubleshooting, that code was not there13:45
rlrossitoh yeah, I suppose that wouldn't be in liberty13:45
*** cdent has quit IRC13:45
mnaserdansmith: yep, not a thing in liberty - https://github.com/openstack/nova/blob/stable/liberty/nova/objects/monitor_metric.py13:46
rlrossitmnaser: you want to backport that?13:46
*** cdent has joined #openstack-nova13:46
*** klkumar has joined #openstack-nova13:46
mnaseri will do that later today, i will drop this review and do that13:46
openstackgerritCédric LECOMTE proposed openstack/nova: Archive instance_actions and instance_actions_event  https://review.openstack.org/29947413:46
dansmithmnaser: you'll need both if they're not there I think right?13:46
mnaserthe first one is there (from august)13:47
rlrossitI think the second was much earlier13:47
mnaserthe second one from november was not13:47
dansmithokay13:47
mnaserso just this one -- https://github.com/openstack/nova/commit/ddb3c4e26f3a908de7836a8fac48360afa6aa1d713:47
rlrossitmnaser: I'll also rearrange the bug you opened to point to the old bug to make the bugs people happier13:48
mnaserplease :)13:48
rlrossitmnaser: should be. you can add me as a reviewer too when you get that backported13:48
mnaserwill do, i will abandon this patch13:48
rlrossit(If you didn't know, gerrit can really help you out by just clicking the "Cherry Pick" button on the commit you want to backport)13:48
mnaseris there a wiki page or something that can help with that?13:49
mnaserrlrossit : nevermind, got it - http://docs.openstack.org/project-team-guide/stable-branches.html#proposing-fixes13:50
*** ametts has quit IRC13:50
rlrossitmnaser: I just got there myself :)13:50
*** ayoung has joined #openstack-nova13:50
rlrossityou beat me to it13:50
*** baoli has joined #openstack-nova13:50
mnaseralright, ill check this out later and take care of it13:51
*** xyang1 has joined #openstack-nova13:52
*** esracelik has quit IRC13:52
*** e0ne has quit IRC13:52
*** salv-orlando has quit IRC13:52
*** mtanino has quit IRC13:52
*** Daisy has joined #openstack-nova13:53
*** permalac has quit IRC13:53
*** pushkaru has joined #openstack-nova13:53
*** diogogmt has quit IRC13:54
*** e0ne has joined #openstack-nova13:54
*** pushkaru has quit IRC13:54
*** dmellado|lunch is now known as dmellado13:54
*** ametts has joined #openstack-nova13:54
rlrossitmriedem: how do you mark a bug as duplicate in LP?13:56
*** e0ne has quit IRC13:56
*** mriedem has left #openstack-nova13:56
*** mriedem has joined #openstack-nova13:57
*** ducttape_ has joined #openstack-nova13:57
mriedemrlrossit: there is a link in the top right13:57
rlrossitah found it. thanks13:57
rlrossitmriedem: and then if we will be backporting a bug that is in fix released, do we need to say it's still a bug in a different branch, or do we not have to care about that?13:59
mriedemrlrossit: you can nominate it for another series13:59
*** anush_ has quit IRC13:59
mriedemand then someone on the bug team can accept/reject that14:00
mriedemyou could also just tag with *-backport-potential14:00
rlrossitI'll do the latter14:00
*** mgoddard_ has joined #openstack-nova14:00
*** haomaiwang has quit IRC14:01
mriedemlooks like we need to either blacklist more xenapi unit tests from py34 or find some patches that replace mox usage with stub_out14:01
*** haomaiwang has joined #openstack-nova14:01
*** eharney has joined #openstack-nova14:02
openstackgerritsahid proposed openstack/nova: libvirt: remove live migrate workaround for an unsupported ver  https://review.openstack.org/29948914:02
openstackgerritsahid proposed openstack/nova: libvirt: introduces object to handle domain xml migration  https://review.openstack.org/29949014:02
openstackgerritsahid proposed openstack/nova: libvirt: move graphic/serial consoles check to pre_live_migration  https://review.openstack.org/25470914:02
openstackgerritsahid proposed openstack/nova: libvirt: add "get_job_info" to Guest's object  https://review.openstack.org/25074014:02
openstackgerritsahid proposed openstack/nova: libvirt: add method to configure max downtime when migrating  https://review.openstack.org/25073914:02
sdaguehonestly, blacklisting xenapi seem like the better option14:02
openstackgerritsahid proposed openstack/nova: libvirt: add "abort_job" to Guest's object  https://review.openstack.org/25073814:02
*** salv-orlando has joined #openstack-nova14:02
openstackgerritsahid proposed openstack/nova: libvirt: add method "migrate" to Guest's object  https://review.openstack.org/25036514:02
*** anush_ has joined #openstack-nova14:02
*** achanda has joined #openstack-nova14:02
*** cdent_ has joined #openstack-nova14:03
mriedemsdague: because it's easier or because we know stub_out has it's own issues with staticmethods?14:03
*** mgoddard has quit IRC14:03
*** cdent has quit IRC14:03
*** cdent_ has quit IRC14:04
dansmithmriedem: are you seeing the problem in xen tests, or you know that xen tests are the problem?14:04
sdaguemriedem: because it's easier, and the xenapi tests use mox in non trivial ways14:04
dansmithbecause the stub out staticmethod thing manifests somewhere other than the test that does the problematic behavior14:04
mriedemhttp://logs.openstack.org/25/299625/2/check/gate-nova-python34-db/75276a4/console.html#_2016-03-31_10_32_46_71014:04
mriedemnova.tests.unit.virt.xenapi.test_vmops.VMOpsTestCase.test_get_vm_opaque_ref_raises_instance_not_found14:04
*** cdent has joined #openstack-nova14:04
mriedemdansmith: failed on your top of series flavor migration chnage14:04
mriedem*change14:04
* dansmith waits an eternity for it to load14:05
*** rook-lappy has joined #openstack-nova14:05
dansmithah, lovely14:06
*** eharney has quit IRC14:06
dansmiththat one is not hard to replace with mock, but yeah14:06
*** keekz_ is now known as keekz14:07
*** penick has quit IRC14:07
dansmithI feel like just increasing the size of the blacklist isn't really helping the problem, because later if we need to really start working on py3 compat, we'll just have a higher mountain to climb14:07
dansmithbut I also don't want to work on it muself14:07
mriedemi have a separate question14:08
*** sdague has quit IRC14:08
*** achanda has quit IRC14:08
mriedemdid anyone know that swap-volume's api has been broken since 2013?14:08
dansmithlike, I wonder what the py3 test is getting us if we're just adding to the blacklist any time something fails14:08
*** sdague has joined #openstack-nova14:08
mriedemswap-volume is the API that's been broken at least twice :(14:09
*** cdent has quit IRC14:09
*** akshai has quit IRC14:10
* dansmith doesn't know anything about it14:10
mriedemhttps://review.openstack.org/#/c/299715/14:10
mriedembasically, it's been hacked up so badly to fit cinder's volume migration scenario, that it's really an admin-only API for nova now14:10
mriedembecause it totally sharts itself for non-admins14:10
*** akshai has joined #openstack-nova14:11
dansmithwow, those are some huge except handlers you have there14:12
*** rfolco has quit IRC14:12
*** coreywright has quit IRC14:12
mriedemlatent, let those mellow14:12
dansmithheh14:13
bauzasp!spy14:14
openstackgerritSean Dague proposed openstack/nova: remove the ability to disable v2.1  https://review.openstack.org/29989314:16
openstackgerritSean Dague proposed openstack/nova: WIP: warn if using the legacy v2 api stack  https://review.openstack.org/29989514:16
tbachmanp!spy14:17
*** burgerk has joined #openstack-nova14:18
edleafeok, can someone point me to a link that explains the 'p!spy' usage?14:19
mriedemso vish added swap-volume way back in the day14:19
danpbmriedem: dansmith the length of the py34 blacklist has reduced from 453 to 216  tests, since it was first enabled, so we seem to be making decent progress despite occassionally adding more tests to the blacklist14:19
*** cdent has joined #openstack-nova14:19
mriedemthen cinder started calling it as part of their volume migration scenario,14:19
mriedemand added a callback to cinder from swap-volume to tell cinder that the operation was done14:19
mriedemwhich is admin-only14:19
*** cdent has quit IRC14:20
mriedemso that broke swap-volume for non-admin useres14:20
dansmithdanpb: okay14:20
mriedemthen later, they also removed swap-volume doing explicit detach/attach calls to cinder, so it only works via the admin-only api call to cinder14:20
mriedemthat's api break the second14:20
mriedemeffectively making swap-volume only useful for the admin-initiated volume migration from the cinder side14:20
openstackgerritChuck Carmack proposed openstack/nova: DB API changes for the nova-manage quota_usage_refresh command  https://review.openstack.org/29944114:21
mriedemand is not useful at all when initiated from nova, as admin or user14:21
*** hogepodge has joined #openstack-nova14:21
*** eharney has joined #openstack-nova14:21
mriedemDuncanT: so nova should probably not only make swap-volume an admin-only API, it should probably fail in the compute API if the old volume's migration_status is not 'migrating', right?14:21
*** rook-lappy has quit IRC14:22
mriedemDuncanT: because that's what cinder is validating when nova calls migrate_volume_completion14:22
johnthetubaguymriedem: dansmith: I have a feeling sfinucan might fancy helping create a extra specs object, a bit like image_meta and props, do we want to hold that kind of thing till after the summit?14:22
johnthetubaguys/extra/flavor extra/14:23
mriedemjohnthetubaguy: is it a new spec?14:23
sahidmarkus_z: can you remind me why you want to reuse same ports instead of just release ports when we destroy an instance here: https://review.openstack.org/#/c/275800/ ?14:23
mriedemor specless bp with just lots of conversion work14:23
johnthetubaguymriedem: well image_meta snook in as part of mitaka-objects, but it feels like a separate specless bp to me14:23
johnthetubaguyyeah, the later14:24
dansmithmriedem: definitely something needing discussion, IMHO14:24
dansmithI do not think it's just a specless bit of mindless conversion14:24
*** cdent has joined #openstack-nova14:24
mriedemif it needs discussion then it needs a spec14:24
mriedemand it waits til after the summit14:24
dansmithimage properties is a lot more straightforward I tthink14:24
johnthetubaguydansmith: mriedem: OK, that fair enough, I don't really know the true horrors living in there14:25
johnthetubaguyI was assuming its a me too thing like image_meta14:25
*** gongysh has quit IRC14:25
*** coreywright has joined #openstack-nova14:25
dansmithjohnthetubaguy: it's also a thing that would have to be serialized in the db, inside another object which we already serialize, etc14:26
sahiddansmith: why do you think it's lot more complicated ?14:26
*** rfolco has joined #openstack-nova14:26
dansmithjohnthetubaguy: much more work than just the imagemeta object which is used for parsing metadata from glance and a few things from sysmeta14:26
dansmithsahid: because ^14:27
*** haukebruno has joined #openstack-nova14:27
sahidhum that is because we will have to serialize that object inside the instance object ?14:28
dansmithno, it's serialized inside of flavor14:28
mriedemflavor extra specs would be serialized in flavor which is serialized in instance14:28
*** links has joined #openstack-nova14:28
johnthetubaguydansmith: hmm, and the API sets stuff, etc, there are some details we should straighten out there14:28
sahidi see14:28
dansmithmriedem: flavor is not serialized inside flavor in the db14:29
*** e0ne has joined #openstack-nova14:29
dansmithmriedem: it is over the wire though14:29
mriedemdansmith: flavor extra specs you mean?14:29
*** e0ne has quit IRC14:29
dansmithmriedem: sorry, flavor is not serialized inside instance in the database14:29
mriedemright, it's in instance_extra14:30
mriedembut instance.flavor14:30
johnthetubaguyyeah, old and new, etc14:30
openstackgerritSean Dague proposed openstack/nova: remove glance deprecated config  https://review.openstack.org/29953514:30
johnthetubaguysfinucan: did you get all that ^14:31
openstackgerritDan Smith proposed openstack/nova: Migrate compute node resource information to Inventory objects  https://review.openstack.org/27931314:31
openstackgerritDan Smith proposed openstack/nova: Add Service.get_minimum_version_multi() for multiple binaries  https://review.openstack.org/27931214:31
openstackgerritDan Smith proposed openstack/nova: Make compute_node_statistics() use new schema  https://review.openstack.org/28693114:31
openstackgerritDan Smith proposed openstack/nova: Move resource provider staticmethods to proxies  https://review.openstack.org/29952614:31
rlrossitdansmith: just a thought as I'm working through this frustration on the FixedIP dict compat stuff... do we want to make a plan to (at some point) migrate over the network api to use the existing versioned objects?14:32
sfinucanjohnthetubaguy: Reading through it now14:32
rlrossitthese divided dicts/objects are very frustrating to work with14:32
dansmithrlrossit: I'm not sure what you mean14:32
mriedemrlrossit: the nova.network.model stuff you mean?14:33
rlrossitmriedem: yeah14:33
mriedembeagles had a thing up for that awhile back14:33
rlrossitand converting over the dicting in https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py14:33
*** anush_ has quit IRC14:34
markus_zsahid: Let me refresh the context in my head14:34
*** Jeffrey4l_ has quit IRC14:35
*** Jeffrey4l_ has joined #openstack-nova14:35
*** nkrinner has quit IRC14:35
mriedemrlrossit: dansmith was doing some of that a couple of releases ago, it gets weird though14:35
dansmithyeah, it's really ugly14:36
mriedemrlrossit: because the nova.objects.* for networks are modeled on nova-network,14:36
dansmithI did the networkrequest work which makes it cleaner over the wire14:36
mriedemand don't fit well for neutron14:36
rlrossitwell... so is dictcompat stuff14:36
mriedemrlrossit: i don't tihnk you understand14:36
mriedemthe level of fugly14:36
rlrossitso I'm guessing these are two conflicting formats that will never be able to be settled until n-net dies?14:37
markus_zsahid: IIRC it's because a multiple rebuild of the same instance would otherwise cause a port exhaustion14:37
mriedemthe versioned network objects we have today are more for nova-network based on our nova db tables for nova-network14:37
mriedemanything we use them for for the neutron api is just shoe-horning that in14:37
dansmiththings like FloatingIP and FixedIP are yeah14:37
rlrossitI actually don't think we use them in the neutron api stuff at all14:38
rlrossitunless I'm forgetting someting14:38
markus_zsahid: The "_create_serial_console_devices()" method also gets called multiple times in a live-migration which would result in a wrong xml.14:39
mriedemneutronv2.api really just uses network and network_request14:39
*** READ10 has joined #openstack-nova14:39
markus_zsahid: That's the things I remember. Would another approach make more sense?14:39
rlrossitmriedem: so this appears to be a can of worms? I probably don't want to look at it then until I finish the dict compat stuff14:40
rlrossitone level of frustration at a time14:40
*** jpomeroy has joined #openstack-nova14:41
sahidmarkus_z: yes, about the rebuild thing, so for libvirt we use something like destroy/spawn which is basically same process for hard_reboot14:41
mriedemrlrossit: yeah i'd probably avoid that can of worms for now14:41
sahidi think we should just to release ports when we destroy14:41
dansmithrlrossit: it's like a ticking can of worms14:42
sahidin _create_serial_console_devices what is exactly the problem ? can't we just look t the xml and see if serial consoles are already defined ?14:42
*** mvk_ has joined #openstack-nova14:42
*** gongysh has joined #openstack-nova14:42
rlrossitdansmith: ticking? does that mean if I pick it up it will blow up and cover me in worm shrapnel?14:43
openstackgerritGleb Stepanov proposed openstack/nova-specs: Add offset to console log  https://review.openstack.org/30000614:43
sahidi'm not sure it's the good solution to add all of this code just to keep using same ports whereas we can just release them and acquire them when we need14:44
dansmithrlrossit: it might yes, and worm shrapnel is gross14:44
markus_zsahid: I think we can. We have to, to be honest. The xml must be updated to avoid a port conflict on the target host.14:44
sahidthen your patch can't be backported, and if you just change where we release ports we can probably backport it14:44
*** e0ne has joined #openstack-nova14:45
markus_zsahid: yeah, I inherited it and the original author is not available anymore.14:45
rlrossitdansmith: so now that (I think) I've found my problems with these actually dicts vs. objects things, it looks like the changes in https://review.openstack.org/#/c/267768/ won't be necessary14:45
*** jpomeroy has left #openstack-nova14:45
sahidmarkus_z: i see, so do you want to continue to work on it or do you want me to jump on ?14:45
bauzasdansmith: have you seen my comment on the havana migration creating old-style flavors ?14:45
rlrossitdansmith: but... do we still want me to at least keep that patch around for the unit tests it adds? they're good for coverage at least I think14:46
*** jlanoux_ has quit IRC14:46
dansmithrlrossit: yeah14:46
dansmithbauzas: yeah, I replied right?14:46
markus_zsahid: I wouldn't say no if you want to take over :)14:46
bauzasheh14:46
bauzasdansmith: I'm lagging today14:46
dansmithbauzas: I'm not sure what you're really saying there, other than that you like FIXME better than NOTE and that you agree the havana migration is ugly14:47
sahidmarkus_z: ok no problem i take it14:47
*** andrewbogott has quit IRC14:47
*** andrewbogott has joined #openstack-nova14:47
markus_zsahid: That's really nice, thank you! I set a workflow -1 and when you patches are ready I abandon mine.14:48
markus_z*your14:48
sdagueon the realm of things to delete - https://review.openstack.org/#/c/29953514:48
sdaguecleans up glance a bit before we start making new changes14:48
bauzasdansmith: sorry, my point was that maybe we should see how to remove the havana migration script rather than doing that kind of online migration in test.py14:48
sahidmarkus_z: cool14:48
dansmithbauzas: right, we definitely have to do that, but it's a bigger effort14:48
*** links has quit IRC14:48
bauzasdansmith: that's a big thing ? I maybe misunderstood that14:49
*** wuhg has quit IRC14:49
*** links has joined #openstack-nova14:49
dansmithbauzas: yeah, the implications of removing that from our first migration have wide-reaching impacts14:49
*** lyarwood has quit IRC14:49
dansmithbauzas: all of our docs assume those default flavors are there, for example, as does devstack, etc14:49
bauzasdansmith: I see, the main problem is that devstack/tempest users wait for a flavor #42 for m1.tiny as an example ?14:51
*** jaosorior_ has joined #openstack-nova14:52
sdaguethe gate will be fine14:52
sdaguebut the docs are a huge issue14:52
*** jaosorior has quit IRC14:53
bauzasI see14:53
sdaguebecause nova has been this way for litterally ever14:53
dansmithbauzas: right, tempest actually creates its own flavors14:53
dansmithsdague: although, grenade will fail right? it uses the default flavors no?14:53
sdaguedansmith: yeh, it does today I think14:53
dansmithfor the test instances14:53
dansmithyeah14:53
bauzasokay, this is the problem right ? https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/migrate_repo/versions/216_havana.py#L8814:54
*** jwcroppe has quit IRC14:54
sdaguethough, the reality is we're going to create a few base flavors in devstack anyway14:54
*** jwcroppe has joined #openstack-nova14:54
dansmithsdague: yep, and if we do that then grenade should be okay14:55
sdagueif they aren't in nova. The only reason we don't also download an image, is the time is too long14:55
dansmithI gotta run out for a bit, back in about an hour14:55
*** gongysh has quit IRC14:56
*** gjayavelu has joined #openstack-nova14:57
*** sdake has joined #openstack-nova14:57
*** mgoddard_ has quit IRC14:57
*** mgoddard has joined #openstack-nova14:58
*** ljxiash has joined #openstack-nova14:58
*** jwcroppe has quit IRC14:59
*** mtanino has joined #openstack-nova15:00
*** haomaiwang has quit IRC15:01
*** pece has quit IRC15:01
*** haomaiwang has joined #openstack-nova15:01
*** e0ne has quit IRC15:01
mriedemsdague: on https://review.openstack.org/#/c/299535/ why not just set glance.api_servers in the ConfFixture so we don't have to copy that all over the place?15:02
*** ametts has quit IRC15:02
*** rdopiera has quit IRC15:04
*** stevemar has joined #openstack-nova15:06
*** jistr has quit IRC15:07
*** ametts has joined #openstack-nova15:07
openstackgerritRyan Rossiter proposed openstack/nova: Remove DictCompat from FixedIP  https://review.openstack.org/26595115:07
*** mdrabe has quit IRC15:07
*** mtanino has quit IRC15:09
*** jistr has joined #openstack-nova15:09
*** mtanino has joined #openstack-nova15:11
*** salv-orlando has quit IRC15:11
*** andymaier has joined #openstack-nova15:12
openstackgerritJohn Garbutt proposed openstack/nova: Config options: Centralize consoleauth options  https://review.openstack.org/29311815:13
ndipanovmriedem, https://review.openstack.org/#/c/262585/ <-- this one should be ready now15:14
mriedemlooking15:14
sdaguemriedem: because it would be good to know when we need it15:15
sdagueinstead of just massively papering over it15:15
mriedemsdague: i figured you'd say something like that15:15
doffmSo the configdrive contains instance metadata. If we modify instance metadata, and then reboot or perform a cold migration... do we expect the config drive to get rebuilt?15:16
mriedemndipanov: that series is getting hammered with job timeouts on vexxhost15:16
doffmOr is config drive a one time... first time you launch... thing.15:16
mriedemndipanov: you must be jinxed15:16
*** zz_dimtruck is now known as dimtruck15:16
*** jlanoux has joined #openstack-nova15:17
mriedemsdague: if you remove the api_insecure=False from self.flags i'll +2 it15:17
*** klkumar has quit IRC15:17
*** cdent has left #openstack-nova15:18
*** cdent has joined #openstack-nova15:18
*** links has quit IRC15:19
*** rk4n has quit IRC15:19
openstackgerritSean Dague proposed openstack/nova: remove glance deprecated config  https://review.openstack.org/29953515:20
sdaguemriedem: done15:21
*** rk4n has joined #openstack-nova15:21
*** mdrabe has joined #openstack-nova15:21
*** armax_ has joined #openstack-nova15:22
alaskidoffm: I would only expect it to get rebuilt for 'rebuild'. But I don't know what the intent was15:23
*** jaosorior_ has quit IRC15:24
*** vks has joined #openstack-nova15:24
*** ayoung has quit IRC15:24
doffmalaski: I can't find anywhere where it would be re-built on reboot. Currently we attempt to rebuild it when doing a migration or resize. (libvirt)15:25
doffmI'm guessing that wasn't intended though.15:25
*** armax has quit IRC15:25
*** armax_ is now known as armax15:25
*** ayoung has joined #openstack-nova15:25
alaskiyeah. there may be a reason it needs to be rebuilt, but I doubt the intention was to change what's in there at all15:26
*** mjura has quit IRC15:27
alaskiI don't know who would have knowledge around that15:27
*** jistr has quit IRC15:28
*** jistr has joined #openstack-nova15:28
*** jdurgin1 has joined #openstack-nova15:31
mriedemndipanov: for sriov-v pf, is it required that the port is created in neutron first and then provided to nova when booting the server?15:31
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: [extensions] cache discovered extensions  https://review.openstack.org/28713115:32
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Cache discover_version method  https://review.openstack.org/29986215:32
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Add support of caching  https://review.openstack.org/29986015:32
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Discover extensions from contrib dir with pkgutil  https://review.openstack.org/29986115:32
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: [tests] initialize client object inside setUp  https://review.openstack.org/30003215:32
*** dimtruck is now known as zz_dimtruck15:34
ndipanovmriedem, I think that's true for any sriov port15:35
ndipanovmriedem, you can't pass vnic-type via --nic15:35
openstackgerritMatt Riedemann proposed openstack/nova: remove glance deprecated config  https://review.openstack.org/29953515:35
mriedemndipanov: ok, that's what i thought15:35
mriedemndipanov: left comments on the top patch, i think we should only call that populate mac addr method in the request.port_id case15:36
vkswhy openstack agents can't access DB directly , instead agents do a RPC to plugin to get any additional data ?15:36
mriedemjust to tighten it up a bit15:36
openstackgerritRoman Podoliaka proposed openstack/nova: Config options: centralize common console options  https://review.openstack.org/22211315:36
openstackgerritRoman Podoliaka proposed openstack/nova: Config options: make console a group  https://review.openstack.org/30003615:36
*** markvoelker has joined #openstack-nova15:37
*** e0ne has joined #openstack-nova15:37
openstackgerritGleb Stepanov proposed openstack/nova: [WIP] Add console log offset  https://review.openstack.org/29247315:38
*** tritro has quit IRC15:39
*** e0ne has quit IRC15:40
*** markvoelker has quit IRC15:40
*** markvoelker_ has joined #openstack-nova15:41
*** tobe has joined #openstack-nova15:43
*** efoley has joined #openstack-nova15:44
*** e0ne has joined #openstack-nova15:44
*** Swami has joined #openstack-nova15:44
*** e0ne has quit IRC15:46
*** EinstCrazy has quit IRC15:47
*** e0ne has joined #openstack-nova15:47
doffmgibi: So i'm currently looking at: https://review.openstack.org/#/c/299045/ and hanrong has probably correctly pointed out that, when migrating, instances shouldn't have affinity with themselves.15:48
doffmgibi: I can't find anywhere in the soft-affinity code to take this in to account. Do you know if its done somewhere?15:48
*** tobe has quit IRC15:48
*** markvoelker_ has quit IRC15:50
openstackgerritEd Leafe proposed openstack/nova: Make compute_node_statistics() use new schema  https://review.openstack.org/28693115:51
*** markvoelker has joined #openstack-nova15:51
gibidoffm: I don't know top of my head, I have to look into a bit. However if the problem is reproducible then we can be pretty sure that we don't have such a code in the code patch15:52
gibis/patch/path/15:52
migisdague: hi15:52
*** cdelatte has quit IRC15:52
migisdague: I have question around your change: https://review.openstack.org/#/c/283261/15:52
*** klkumar has joined #openstack-nova15:53
*** gszasz has quit IRC15:53
*** eharney has quit IRC15:53
*** zz_dimtruck is now known as dimtruck15:54
*** Guest87578 is now known as jlvillal15:55
doffmgibi: I'll investigate some more, and try writing a test to check the affinity weigher.15:55
gibidoffm: hm, as a second thought. if we talk about soft-affinity then we have less problems as even if a VM is soft-affin to itself it stil can be migrated as soft-affinity allows breaking it15:55
doffmIts true Its not such a big problem as hard affinity.15:55
*** markvoelker has quit IRC15:55
doffmBut the weights will be incorrect, especially for small server groups.15:55
*** jaypipes has joined #openstack-nova15:55
gibidoffm: probably you are right and the soft-affinity is broken this way as well15:56
*** markvoelker has joined #openstack-nova15:56
doffmOk, well I'll check. And file a bug if neccessary.15:56
*** baoli_ has joined #openstack-nova15:56
*** rmart04 has quit IRC15:56
*** markvoelker_ has joined #openstack-nova15:57
*** baoli has quit IRC15:57
gibidoffm: thanks15:57
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Add support of caching  https://review.openstack.org/29986015:57
*** e0ne has quit IRC15:58
mriedemalaski: i suspect that compute:swap_volume in the policy.json is not actually used anywhere, is there an easier way to tell than just checking the authorizer calls in the volume attachment update REST API code?15:59
*** mgoddard_ has joined #openstack-nova15:59
mriedemor just removing it i guess15:59
gibidoffm: basically here https://github.com/openstack/nova/blob/master/nova/scheduler/weights/affinity.py#L49 you have to take into account if the instance being scheduled is on the host and remove that from the calculation15:59
*** shangxdy has joined #openstack-nova15:59
*** shangxdy_ has joined #openstack-nova15:59
*** jlanoux_ has joined #openstack-nova16:00
*** eharney has joined #openstack-nova16:00
*** angdraug has joined #openstack-nova16:00
andreykurilinmriedem sdague: hi! Can you look at my update for spec(first revision was published long time ago in augest) about microversion support in novaclient? please... it is very simple change16:00
*** jlanoux has quit IRC16:00
*** haomaiwang has quit IRC16:01
*** markvoelker has quit IRC16:01
mriedemandreykurilin: link?16:01
andreykurilinmriedem: https://review.openstack.org/#/c/21120616:01
*** sdake has quit IRC16:01
*** haomaiwang has joined #openstack-nova16:01
*** mgoddard has quit IRC16:02
mriedemandreykurilin: seems we should also move that to specs/liberty/implemented/ right?16:02
mriedemthe bp was marked as implemented in liberty https://blueprints.launchpad.net/python-novaclient/+spec/api-microversion-support16:02
andreykurilinmriedem: yes. there is a separate patch16:03
andreykurilinwith +2 from you:)16:03
mriedemoh on top of it16:03
mriedemi see16:03
mriedemtouche16:03
*** Daisy has quit IRC16:04
*** Daisy has joined #openstack-nova16:04
*** Jeffrey4l_ has quit IRC16:04
andreykurilinmridem: also, you use this patch as an example of moving specs to implemented in your latest notification mail:) http://lists.openstack.org/pipermail/openstack-dev/2016-March/089570.html16:04
*** irenab_ has joined #openstack-nova16:05
*** irenab has quit IRC16:05
*** irenab_ is now known as irenab16:05
*** Jeffrey4l_ has joined #openstack-nova16:05
mriedemthat was a doppleganger16:05
andreykurilinlol16:05
dansmithsahid: how does this break before the next patch? https://review.openstack.org/#/c/292499/1216:05
*** achanda has joined #openstack-nova16:05
dansmithsahid: does it just mean reserved calculation is wrong, or instances fail to boot or what?16:05
dansmithoh, you already squashed16:06
*** rlrossit has quit IRC16:07
sahiddansmith: yes yes, i changed my mind, instead of overiding the __init__'s method which you said is not good i prefered to make libvirt driver to set that new field16:08
*** salv-orlando has joined #openstack-nova16:08
*** thorst is now known as thorst_afk16:08
*** pcaruana has quit IRC16:08
*** Daisy has quit IRC16:09
dansmithsahid: how did it fail though?16:10
*** e0ne has joined #openstack-nova16:11
*** achanda has quit IRC16:11
dansmithsahid: I ask because I would have thought that was okay, so I want to make sure we're not missing something16:11
sahiddansmith: from my point of view...16:12
sahidlibvirt constructs the new version of PageTopology without to set that 'reserved' field, this object is stored in database16:12
*** Jeffrey4l_ has quit IRC16:12
*** rfolco has quit IRC16:12
sahidthen it get deserialized and explode16:12
sahidmakes sense?16:12
dansmithno, because deserializing an older object without the field should be fine16:13
dansmithsahid: do you have some logs or something you can point me to?16:13
sahiddansmith: you can see some in PS1016:14
dansmithoh, his comment, okay16:14
*** unicell1 has joined #openstack-nova16:15
mriedemandreykurilin: i commented on ps616:15
sahidyes i was not able to reproduce the case (i did not try too much actually)16:15
*** unicell has quit IRC16:15
mriedemandreykurilin: seems like most of this should just go in the novaclient devref16:15
dansmithsahid: I think that you're covering up this issue by squashing16:15
dansmithsahid: when really, you need to tolerate that field being missing16:16
mriedemsince it's talking about things that are not even close to happening today, like nova raising the minimum microversion past v2.116:16
alaskimriedem: yeah, there's no good way to track down policy usage16:16
dansmithlet me go look at the notification code16:16
*** jschlueter has quit IRC16:17
mriedemalaski: oh sneaky16:17
mriedemtest_volume_ops_invalid_task_state16:17
mriedemoops16:17
mriedem_action = '%s:%s' % (scope, action)16:17
mriedemnova.compute.api.API.swap_volume has a decorator that checks it16:18
alaskimriedem: I started working on a hacking check for it briefly, but it has the same problem as notifications. many ways to call it and no assurance that all branches get hit16:18
mriedemso not only do we have the policy checks in the REST API using different values, we also have the compute API policy check16:18
alaskiyeah16:18
*** fawadkhaliq has joined #openstack-nova16:19
sahiddansmith: yes, so if my thinking was not the root cause i just have covered the issue16:19
dansmithsahid: yeah, I think you should split these back up because it's going to be easier to reproduce the issue.. what did you try to reproduce?16:19
dansmithsahid: just apply the first patch and boot an instance with the proper config in place?16:20
*** ljxiash has quit IRC16:20
dansmithsahid: the exception on the first line of his log output is getting hidden so we can't tell the true origin I think16:20
sahiddansmith: yes, the i have restarted all processes16:20
dansmithand that's what we need to figure out16:20
*** manjeets has joined #openstack-nova16:21
openstackgerritSujitha proposed openstack/nova: Config options: Centralize compute options  https://review.openstack.org/29593816:21
sahiddansmith: i see, so i will split these back up and do futher investigations16:22
*** cdent has quit IRC16:22
openstackgerritCale Rath proposed openstack/nova: Make generate_password more secure  https://review.openstack.org/21009216:22
*** cdent has joined #openstack-nova16:23
openstackgerritChris Dent proposed openstack/nova: WIP initial use of microversion_parse  https://review.openstack.org/30007616:23
openstackgerritChris Dent proposed openstack/nova: WIP Initial support for both microversion headers  https://review.openstack.org/30007716:23
jaypipesdansmith, edleafe, cdent: would be great to get https://review.openstack.org/#/c/286931/ merged today.16:23
dansmithsahid: yeah, and I think that if you can reproduce it I have a path forward for you, but I need to look at some o.vo code and see the actual failure you get first16:23
andreykurilin__mriedem: agreed, as soon as I have final version of this update, I'll copy a part of it to novaclient's doc16:23
cdentsdague: those ^ are just wips for now but might be of interest and I could do with some guidance on where to take things16:24
dansmithjaypipes: edleafe: looks like edleafe pushed up a change on top of my push this morning.. what else changed?16:24
*** gagehugo has joined #openstack-nova16:24
sahiddansmith: ack16:24
dansmithoh, the coalesce thing16:24
*** sdake has joined #openstack-nova16:24
bauzasI don't want to be nitpicking, but is the spec merged ?16:25
*** burt has joined #openstack-nova16:26
*** cdelatte has joined #openstack-nova16:26
edleafedansmith: yeah, just the coalesce change in my patch16:26
cdentjaypipes: yar +1 now, with edleafe's adjustments16:26
mriedembauzas: it is https://blueprints.launchpad.net/nova/+spec/compute-node-inventory-newton16:27
mriedemwell, the bp is approved16:27
dansmithI dont' really feel like I can approve that code, both because of my close-ness and lack of understanding any of it16:27
bauzasokay, will review it tonight16:27
dansmithso I propose other cores who do, or who have less integrity :P16:27
*** browne has joined #openstack-nova16:28
* edleafe bites his lip 16:28
*** sdake has quit IRC16:28
dansmithedleafe: :(16:29
*** sdake has joined #openstack-nova16:29
*** apoorvad has joined #openstack-nova16:29
manjeetsis there any limit on nova list ? in my case I created more than 1200 instances but see only 1000 when i do nova list16:29
openstackgerritCale Rath proposed openstack/nova: Make generate_password more secure  https://review.openstack.org/21009216:29
*** yamahata has quit IRC16:30
*** sdake has quit IRC16:30
*** davideagnello has joined #openstack-nova16:30
jaypipesmriedem: the blueprint is approved, but the spec still isn't. https://review.openstack.org/#/c/260048/16:31
*** e0ne has quit IRC16:32
cdentjaypipes: how/where did we leave the placement of the db queries in https://review.openstack.org/#/c/284895/  (Add a get_by_uuid for aggregates)16:32
jaypipescdent: I said "cool if you do the nova.db.sqlalchemy.api approach to keep it consistent with what's there now" and let doffm refactor it all later.16:33
jaypipescdent: lemme do a re-review of that...16:33
cdentthanks16:34
*** penick has joined #openstack-nova16:34
*** whenry has joined #openstack-nova16:34
* jaypipes opens gerrty16:34
jaypipesor even gertty.16:34
*** mylu has quit IRC16:34
* edleafe would love to one day get gertty to actually work16:35
cdentedleafe: Yeah, I keep trying it and it keeps failing in weird ways16:36
cdentIt's also too hard to read commentary on earlier patchsets16:36
*** fawadkhaliq has quit IRC16:36
*** fawadkhaliq has joined #openstack-nova16:37
dansmithyeah, the visual interface to gertty is just not usable IMHO16:38
dansmithreally hard to tell where the cursor thing is, really strange rules for moving between fields, etc16:38
dansmithI customized the colors to make it a little better, but.. it's just not for me I think16:38
*** shangxdy_ has quit IRC16:38
*** mattymo has left #openstack-nova16:39
edleafeIt must be like emacs - once you figure out how to use it, everything else sucks16:39
*** shangxdy_ has joined #openstack-nova16:39
*** sfinucan has quit IRC16:39
*** jdurgin1 has quit IRC16:40
*** cdent has quit IRC16:41
*** fawadkhaliq has quit IRC16:41
*** mpavone has quit IRC16:41
*** fawadkhaliq has joined #openstack-nova16:41
*** matrohon has quit IRC16:42
*** cdent has joined #openstack-nova16:43
*** cdelatte has quit IRC16:45
*** sahid has quit IRC16:47
*** e0ne has joined #openstack-nova16:47
*** sdake has joined #openstack-nova16:48
*** e0ne has quit IRC16:50
manjeetsis there any limit on nova list ? in my case I created more than 1200 instances but see only 1000 when i do nova list16:52
openstackgerritMark Doffman proposed openstack/nova: Make Aggregate metadata functions work with API db  https://review.openstack.org/29735016:52
openstackgerritMark Doffman proposed openstack/nova: Add aggregates tables to the API db.  https://review.openstack.org/29480616:52
openstackgerritMark Doffman proposed openstack/nova: Make Aggregate host operations work against API db  https://review.openstack.org/29659716:52
openstackgerritMark Doffman proposed openstack/nova: Make AggregateList.get_ return API & cell db items  https://review.openstack.org/29606216:52
openstackgerritMark Doffman proposed openstack/nova: Make Aggregate.get operation favor the API db  https://review.openstack.org/29601416:52
mriedemmanjeets: see the osapi_max_limit config option16:53
mriedemit defaults to 100016:53
*** moshele has quit IRC16:53
mriedemdanpb: ndipanov: have you ever heard of a case where non-admins find swap-volume useful?16:54
manjeetsthanks mriedem which file ?16:54
mriedemmanjeets: http://docs.openstack.org/developer/nova/sample_config.html16:54
Prophidysmriedem, if you can recheck this one : https://review.openstack.org/#/c/299474/4 I followed the ML decision16:55
manjeetsmriedem thanks :)16:55
*** zenoway has quit IRC16:55
danpbmriedem: i'm not particularly familiar with it but i bet it would be  people running pets who care16:55
dansmithyeah, I know people that care about it,16:55
dansmithnot sure if they care about non-admins though16:55
mriedemi was going to say, people, like admins?16:56
dansmithcan non-admins migrate their storage on the cinder side?16:56
mriedemb/c it's been busted for non-admins for a looong time16:56
mriedemno16:56
dansmithit's only really useful in nova for that case, right/16:56
dansmithso what's the point of having it be available to non-admins?16:56
mriedemi guess? vish added it in havana with little detail on the use case16:56
mriedemthen later cinder started using it for their volume migration scenario16:56
dansmithI think it's only really for volume migration16:57
dansmithand I think it was intended to be not only admins, but really just for cinder16:57
*** thorst_afk is now known as thorst16:57
dansmithwhich, as you know, makes for the best APIs around16:57
*** rpodolyaka is now known as rpodolyaka_afk16:57
mriedemok, i'm just trying to figure out if nova should restrict to admin or cinder should loosen up their migrate_volume_completion API to admin-or-owner,16:57
*** gjayavelu has quit IRC16:57
mriedemsince it was originally not admin-only,16:58
mriedembut has been changed twice to make it now require admin16:58
mriedemwe never had tempest testing for this so no one noticed the api changes16:58
dansmithheh16:58
dansmithalso good makins' for an API16:58
mriedemyeah,16:58
mriedemso given that,16:59
*** shangxdy_ has quit IRC16:59
mriedemi'm inclined to just say eff it and make swap-volume admin-only in nova16:59
*** shangxdy has quit IRC16:59
*** pwnall has quit IRC16:59
mriedem'because it's been this way for years, if you didn't know, you probably already forked it'16:59
dansmithwell, I definitely agree with the "eff it" part16:59
*** mylu has joined #openstack-nova16:59
*** cdelatte has joined #openstack-nova16:59
mriedembtw, we should probably change the channel topic since i think someone is auto-pinging me every time i join17:00
dansmithI'm fine with that17:00
mriedemwe do'nt have a newton schedule yet, so i'm thinking a link to the review dashboard17:00
*** jistr has quit IRC17:00
*** ChanServ sets mode: +o dansmith17:00
*** haomaiwang has quit IRC17:01
mriedemjust add "Reviews: https://goo.gl/1vTS0Z"17:01
Prophidysmriedem, thank you17:01
*** mgoddard has joined #openstack-nova17:01
*** diogogmt has joined #openstack-nova17:01
*** dansmith changes topic to "This channel is for Nova development. For support of Nova deployments, please use #openstack - Reviews: https://goo.gl/1vTS0Z"- PTL says: if it's latent let it mellow, if it's 500 flush it down"17:01
*** haomaiwa_ has joined #openstack-nova17:01
alaskihah17:01
mriedemheh, i see what you did there17:01
*** dansmith changes topic to "This channel is for Nova development. For support of Nova deployments, please use #openstack - Reviews: https://goo.gl/1vTS0Z - PTL says: if it's latent let it mellow, if it's 500 flush it down"17:01
*** mgoddard_ has quit IRC17:02
*** ChanServ sets mode: -o dansmith17:02
*** dedery has joined #openstack-nova17:03
*** pwnall has joined #openstack-nova17:04
mriedembauzas: for a release note that fixes a bug by changing a default policy value, for a thing that didn't work without it, do you think that goes in upgrades, fixes or other?17:04
*** jlanoux_ has quit IRC17:04
*** Guest11037 has quit IRC17:05
*** yamahata has joined #openstack-nova17:05
mriedemi'm thinking other17:05
*** ZZelle has quit IRC17:05
*** ptm_away has joined #openstack-nova17:05
*** ZZelle has joined #openstack-nova17:06
*** rcernin has quit IRC17:06
*** derekh has quit IRC17:06
*** sridharg has quit IRC17:07
*** haukebruno has quit IRC17:08
*** ihrachys has quit IRC17:08
*** dedery has quit IRC17:09
*** PaulMurray has quit IRC17:09
*** angdraug has quit IRC17:10
*** haukebruno has joined #openstack-nova17:13
*** ptm_away has quit IRC17:13
*** woodster_ has joined #openstack-nova17:14
*** achanda has joined #openstack-nova17:15
mriedemone thing i think we can all agree on is chocolate covered marshmallows in egg shapes are good17:19
*** penick has quit IRC17:21
cdentmriedem: wrong17:22
*** haukebruno has quit IRC17:23
*** zenoway has joined #openstack-nova17:24
mriedemcdent: don't let the easter bunny hear you say that17:25
*** haukebruno has joined #openstack-nova17:25
*** daemontool has quit IRC17:25
cdentThe easter bunny and I disagree a fair bit.17:25
jaypipescdelatte: done.17:26
jaypipesgrrr.... cdent, done.17:26
cdentneed me to change my name?17:26
jaypipescdent: though it doesn't really matter since the dependent change has the black mark of 02 death.17:26
jaypipes-2 death.17:27
jaypipesffs my typing sucks today.17:27
*** penick has joined #openstack-nova17:27
cdentjaypipes: Yeah, I'm totally confused on what is or is important and is or is not going to get some attention and whether I can reorder things to make them legit.17:27
*** fawadkhaliq has quit IRC17:28
*** fawadkhaliq has joined #openstack-nova17:30
jaypipesmriedem, johnthetubaguy: I am not able to set the URL for this specification: https://blueprints.launchpad.net/nova/+spec/compute-node-inventory-newton. Can you set it to https://review.openstack.org/#/c/260048/ please?17:31
jaypipesmriedem, johnthetubaguy: and would it be possible to approve https://review.openstack.org/#/c/260048/? :)17:31
mriedemjaypipes: you can't17:31
mriedembecause the mitaka bp has that link for it's spec already17:31
mriedemjaypipes: i think i already tried that once17:31
*** yonglihe has quit IRC17:32
jaypipesmriedem: it says I don't have permission to set the link...17:32
mriedemthis guy https://blueprints.launchpad.net/nova/+spec/compute-node-inventory17:32
mriedemi'm not really sure why https://blueprints.launchpad.net/nova/+spec/compute-node-inventory was marked complete honestly17:32
jaypipesmriedem: I don't know either. I certainly didn't do that.17:33
mriedemi suspect john did17:33
jaypipesmriedem: what do I need to do? Abandon 279313 and resubmit a whole new spec?17:33
mriedemi do'nt think so...17:34
mriedemsec17:34
jaypipesmriedem: I can certainly do so for all remaining specs in the resource-providers series. and in doing that I can make them depend on each other appropriately in git.17:34
*** anush_ has joined #openstack-nova17:35
mriedemjaypipes: do you mean drop compute-node-inventory-newton in lp and we just re-open compute-node-inventory ?17:35
jaypipesmriedem: no, I mean the specs in Gerrit...17:36
jaypipesmriedem: I can leave you and john to do what you want with the blueprints in LP.17:36
*** sdague has quit IRC17:36
mriedemi was just going to remove the 260048 from the compute-node-inventory bp in lp17:37
jaypipesmriedem: tell you what, don't do that...17:37
*** ctrath has quit IRC17:37
*** jwcroppe has joined #openstack-nova17:38
*** dimtruck is now known as zz_dimtruck17:38
*** sdague has joined #openstack-nova17:38
*** rlrossit has joined #openstack-nova17:38
jaypipesmriedem: lemme push a whole new set of specs, with proper dependencies, with -newton at the end of each blueprint link. we can then obselete the original blueprints in LP and get a new set of blueprints associated with the correct Gerrit patches, all with proper dependencies in git.17:39
*** anush_ has quit IRC17:39
*** ducttape_ has quit IRC17:39
*** dims_ has quit IRC17:39
mriedemsounds fine to me...gonna be quite a bit of churn17:40
*** gjayavelu has joined #openstack-nova17:40
*** dims has joined #openstack-nova17:40
jaypipesmriedem: yeah, but you and dansmith have been asking me to fix up the dependency chain for a while now and I've been putting that off... might as well just get it done :)17:40
*** ctrath has joined #openstack-nova17:40
*** zz_dimtruck is now known as dimtruck17:41
mriedemsounds like cdent wants that also17:41
mriedemi can't speak for that dansmith guy17:41
dansmiththere is really no hope of pleasing me, so I'd go with what mriedem  wants17:41
cdentmriedem: that's not quite what I wanted, I was more sort of flailing in a kind of "what's going on? where's my hat? how come this is over there and I'm upside down?"17:42
*** mylu has quit IRC17:42
cdentA newly refined stack may help that.17:43
mriedemyou know what i want?17:44
mriedemhttps://review.openstack.org/#/c/295118/17:44
mriedemlooks like i have to fix it17:44
*** READ10 has quit IRC17:45
openstackgerritMatt Riedemann proposed openstack/nova: Set 'libvirt.sysinfo_serial' to 'none' in RealTimeServersTest  https://review.openstack.org/29511817:45
*** ccarmack has quit IRC17:45
mriedemsomeone want to +W that ^?17:45
*** fawadkhaliq has quit IRC17:46
*** fawadkhaliq has joined #openstack-nova17:46
*** fawadkhaliq has quit IRC17:47
sdaguemriedem: done17:47
*** fawadkhaliq has joined #openstack-nova17:47
*** fawadkhaliq has quit IRC17:47
*** fawadkhaliq has joined #openstack-nova17:48
*** fawadkhaliq has quit IRC17:48
dansmithmriedem: so, we need one more ack on the top flavor patch, and then a full set on the very bottom one17:48
-openstackstatus- NOTICE: Job failures for missing traceroute packages are in the process of being fixed now, ETA 30 minutes to effectiveness for new jobs17:48
dansmithmriedem: we're pretty close and I think we can/should merge it this week17:49
jaypipesdansmith: link? I'll do it.17:49
*** fawadkhaliq has joined #openstack-nova17:49
dansmiththis is the top one: https://review.openstack.org/#/c/295310/2817:49
*** jaosorior has joined #openstack-nova17:49
dansmithand the bottom one in that stack17:49
openstackgerritChris Dent proposed openstack/nova: WIP Initial support for both microversion headers  https://review.openstack.org/30007717:49
openstackgerritChris Dent proposed openstack/nova: Initial use of microversion_parse  https://review.openstack.org/30007617:50
*** fawadkhaliq has quit IRC17:50
dansmithI also think we need to get jaypipes' stats thing merged because three of us have been stepping on each other with it17:50
*** mriedem has quit IRC17:50
dansmithbut I'm also kinda chicken17:50
*** ducttape_ has joined #openstack-nova17:50
dansmithI guess I should just do it because I wrote the code on top of it that wouldn't work if it was broken17:50
*** fawadkhaliq has joined #openstack-nova17:50
*** mriedem has joined #openstack-nova17:51
dansmitheff it17:53
*** gyee has joined #openstack-nova17:53
*** unicell has joined #openstack-nova17:54
*** klkumar has quit IRC17:54
*** vks has quit IRC17:55
openstackgerritChris Dent proposed openstack/nova: WIP Initial support for both microversion headers  https://review.openstack.org/30007717:55
openstackgerritChris Dent proposed openstack/nova: Initial use of microversion_parse  https://review.openstack.org/30007617:55
* cdent is touched17:55
*** boden has joined #openstack-nova17:55
*** unicell1 has quit IRC17:56
dansmithis docimpact still a thing?17:57
*** boden has left #openstack-nova17:58
*** ccarmack has joined #openstack-nova17:58
jaypipesdansmith: 295310 wallaby'd17:58
dansmiththanks17:58
dansmithI will have everyone note that I just ran this command:17:58
dansmithgit checkout master -b sins-of-the-father17:59
sdaguedansmith: DocImpact creates a bug in the nova tracker now instead of the openstack-manuals tracker18:00
openstackgerritDan Smith proposed openstack/nova: Remove flavor seeding from the base migration  https://review.openstack.org/30012718:00
dansmithsdague: okay that was the thing I was hearing and wondering about I guess18:00
*** haomaiwa_ has quit IRC18:01
dansmithsdague: so for the flavor base migration thing (^) should I do a docimpact?18:01
*** haomaiwang has joined #openstack-nova18:01
*** jaosorior has quit IRC18:01
sdaguedansmith: or just push the bug directly to openstack-manuals explaining what needs to change in the manuals18:01
sdagueI expect you'll need to follow up on the bug with the details regardless18:02
*** vilobhmm11 has joined #openstack-nova18:02
openstackgerritMatt Riedemann proposed openstack/nova: Make swap-volume an admin-only API by default  https://review.openstack.org/29971518:02
mriedem""18:03
mriedem"In a time long ago and a land far far away"18:03
mriedems/land/galaxy/?18:03
mriedemor don't want to get sued18:03
dansmithsdague: meaning let it auto-create and then add openstack-manuals to the bug?18:04
dansmithmriedem: I'm just not that familiar with star wars.. figure this was close enough18:04
*** danpb has quit IRC18:04
*** haukebruno has quit IRC18:05
mriedemdansmith: you could do that18:05
mriedemrlrossit: policy file changes, your favorite https://review.openstack.org/29971518:05
rlrossitoh right, I was going to look at that spec18:05
*** anush_ has joined #openstack-nova18:06
*** fawadk has joined #openstack-nova18:06
*** fawadkhaliq has quit IRC18:07
*** rtheis has quit IRC18:08
*** rtheis has joined #openstack-nova18:09
*** fawadk has quit IRC18:09
*** gyee has quit IRC18:10
mriedemdansmith: can you post something to the operators ML about this https://review.openstack.org/#/c/300127/ ?18:10
*** jwcroppe_ has joined #openstack-nova18:11
bauzasmriedem: dansmith: oh, the consensus is now to remove that?18:11
dansmithmriedem: I either want to wait until I have a look at all the failures, or not include that link and just meta-discuss it, so which do you want?18:11
*** zenoway has quit IRC18:11
bauzas<318:11
dansmithbauzas: was there any question/18:11
mriedemdansmith: meta is fine18:11
mriedemor both18:11
dansmithmriedem: damn.18:11
bauzasdansmith: about https://review.openstack.org/#/c/300127/18:11
*** Sukhdev has joined #openstack-nova18:12
*** anush_ has quit IRC18:12
* bauzas was a bit off today because of the French general strike needing me to help my daughters :(18:12
*** jwcroppe has quit IRC18:12
mriedembauzas: if we don't remove them from the initial migration, you have to run the online data migrations thing right away with new deployments18:12
dansmithbauzas: I said here: https://review.openstack.org/#/c/295310/28/nova/test.py18:12
dansmithbauzas: that we clearly needed to do that, I just don't want to wrap it all up in this work18:12
sdaguedansmith: that's the normal process anyway, but you have to explain what should change.18:13
*** lpetrut has quit IRC18:13
bauzasdansmith: okay, it seems I maybe misunderstood your thoughts18:13
dansmithsdague: what is normal process, the bug thing?18:13
sdaguedansmith: normal process ==18:13
sdagueDocImpact -> triggers creation of bug in the nova tracker18:13
*** devananda has quit IRC18:14
sdaguesomeone from nova team updates bug with details that are sufficient for the openstack-manuals team to make the fixes we are suggesting18:14
dansmithsdague: right, okay got it18:14
sdagueadd openstack-manuals to the bug18:14
*** gyee has joined #openstack-nova18:14
sdaguemark nova part as invalid18:14
dansmithokay18:15
sdaguedansmith: shouldn't https://review.openstack.org/#/c/300127/ have the reno explaining that you must now manually create flavors before nova works ?18:15
*** Sukhdev has quit IRC18:15
*** tbachman has quit IRC18:15
dansmithsdague: that patch is going to probably need lots more, I just wanted a run to see if anything else is broken :)18:16
dansmithsdague: I figure that's blocked on probable devstack changes, the manual stuff being ready, etc18:16
dansmithI put it in the commit message18:16
dansmithI'll do the docimpact in a sec, re-push, and throw a -W on it18:16
*** diogogmt_ has joined #openstack-nova18:16
*** diogogmt has quit IRC18:17
*** diogogmt_ is now known as diogogmt18:17
*** baoli_ has quit IRC18:17
mriedemrlrossit: https://review.openstack.org/#/c/115749/18:17
mriedemrlrossit: resurrect that whydontcha18:17
*** zengine has joined #openstack-nova18:18
rlrossitI bet dansmith is gonna love me for this one18:18
ildikovjaypipes: hmm, I think thanks for abandoning that patch :)18:19
*** zul has quit IRC18:20
ildikovjaypipes: I thought to reuse it, but in the current stage I'm not sure how much of will be reusable anyway18:20
*** zul has joined #openstack-nova18:22
*** cdelatte has quit IRC18:22
*** zenoway has joined #openstack-nova18:24
*** dims_ has joined #openstack-nova18:24
*** dims has quit IRC18:24
rlrossit"and have 1 and 8215 different commits each, respectively."18:25
rlrossitboy this is going to be fun18:25
*** rfolco has joined #openstack-nova18:25
*** salv-orlando has quit IRC18:26
*** devananda has joined #openstack-nova18:27
*** penick has quit IRC18:27
*** zenoway has quit IRC18:28
*** fawadkhaliq has joined #openstack-nova18:29
jaypipesildikov: yeah, that was super old :)18:31
*** Marga_ has quit IRC18:32
*** zul has quit IRC18:32
*** zul has joined #openstack-nova18:32
*** cdelatte has joined #openstack-nova18:33
doffmgibi: https://bugs.launchpad.net/nova/+bug/156455918:33
openstackLaunchpad bug 1564559 in OpenStack Compute (nova) "Instances should not have soft-affinity with themselves." [Low,New] - Assigned to Mark Doffman (mjdoffma)18:33
openstackgerritMatt Riedemann proposed openstack/nova: Make git clean actually remove covhtml  https://review.openstack.org/30014018:33
ildikovjaypipes: yeah, I did not really maintained that last patch as it wasn't decided at that point how the change should be reflected on the Nova API18:33
openstackgerritJeremy Stanley proposed openstack/nova: List system dependencies for running common tests  https://review.openstack.org/20067218:33
doffmjdurgin, dansmith: Could you take another look at https://review.openstack.org/#/c/241777/ Its a bug fix that has been sitting for a while.18:35
*** delattec has joined #openstack-nova18:36
*** mvk_ has quit IRC18:36
*** cdent has quit IRC18:38
*** gjayavelu has quit IRC18:38
*** cdelatte has quit IRC18:39
*** dave-mccowan has quit IRC18:41
*** delatte has joined #openstack-nova18:41
*** delatte has quit IRC18:42
*** delattec has quit IRC18:44
*** efoley has quit IRC18:46
dansmithsdague: so I see 69 instances of catching exception to avoid state corruption in compute manager alone18:46
dansmithsdague: and several places use a common decorator to do it, which means we end up with that case many more than that in just one file18:46
dansmithsdague: given the ramifications of us corrupting our own state so that we disallow any new operations after a failure, what else would you suggest?18:47
dansmithshort of "do all the task things" of course :)18:47
openstackgerritAndrey Kurilin proposed openstack/python-novaclient: Add support of caching  https://review.openstack.org/29986018:47
lifelesssdague: did you get that constraints question answered? I did follow up  but I think you'd EOD'd...18:51
sdaguedansmith: in reference to - https://review.openstack.org/#/c/298997/ ?18:51
dansmithsdague: yeah18:51
sdaguelifeless: the question was mostly if there was a clever way to hack requirements.txt so that you can install a library from a specific git commit for spot testing purposes only18:52
dansmithholy necrotic thread batman!18:53
sdaguedansmith: so this moves from error on shutdown to random leaks of things, right?18:53
dansmithsdague: it actually closes leaks..18:53
dansmithif we don't continue here, we leave things unterminated across us and cinder18:53
dansmithwhich normal people can't resolve18:53
dansmithand we don't emit the notification for instance shutdown18:54
sdaguedansmith: ok, well it still seems like we should be logging the bdm in question as well so it's clear what didn't get cleaned up on the otherside, because if cinder is down, then that just gets orphaned in an odd state, right?18:55
openstackgerritJeremy Stanley proposed openstack/nova: List system dependencies for running common tests  https://review.openstack.org/20067218:55
dansmithsdague: what gets orphaned?18:55
dansmithsdague: the volume or the bdm/18:55
sdaguethe volume on the cinder side will be in the wrong state, right?18:56
*** zenoway has joined #openstack-nova18:56
openstackgerritClenimar Filemon proposed openstack/nova: Use Identity v3 params when connecting to Ironic  https://review.openstack.org/30015418:56
dansmithsdague: depends on the failure, but any of them after this one fails are guaranteed to be bad right?18:56
dansmithsdague: so if we're talking to one backend that is broken but the rest aren't we don't clean up the others18:57
lifelesssdague: yes, grep for edit-constraints in devstack18:57
dansmithwhich is the reason we already exclude a bunch of them by name18:57
dansmithsdague: I believe the reason cinder allows things other than nova to call force-detach is also to allow this sort of cleanup18:57
dansmithsdague: but if we die here, then it requires nova intervention as well18:58
sdagueok, so I now get that, though it still sucks that the best we have is Exception18:58
dansmithsdague: well, we exclude a bunch by name, but given how bad it is if we die here, I think we should be careful18:59
dansmithlike we try to be lots of other places18:59
dansmiththe other bad case here,18:59
dansmithis if you're trying to delete an instance to create another and attach your volumes elsewhere,18:59
sdaguehowever it still seems like we should log something more specific about the bdm in question. This is going to be a WARN with a giant stack trace right?18:59
dansmithall your volumes are completely hosed and stuck to this until you get support intervention18:59
dansmithinstead of just the bad one18:59
lifelessif you need to emulate try:finally for some reason, BaseException is your friend19:00
lifelessbut really think hard before doing that19:00
dansmithsdague: no stack trace just a warning19:00
*** zenoway has quit IRC19:00
sdaguedansmith: ok so shutdown instance19:00
sdagueWARNING Ignoring unknown exception19:00
sdaguewhat does that even mean19:00
dansmithsdague: it includes the exception name19:00
dansmithor are you saying you want a trace?19:00
*** haomaiwang has quit IRC19:01
sdaguesure, but it doesn't include the bdm19:01
openstackgerritJeremy Stanley proposed openstack/nova: List system dependencies for running common tests  https://review.openstack.org/20067219:01
*** haomaiwang has joined #openstack-nova19:01
sdagueI want enough left in the logs so that the devices that never got talked to on the cinder side are there19:01
sdaguea warn should have enough info to be actionable19:01
sdagueso something that an admin could realize ... oh there is a think potentially screwing on these vols now19:02
dansmithsdague: okay, so volume_id then?19:02
sdaguedansmith: sure, I don't know everything in that structure that might be useful19:02
sdaguebut it seems like breadcrumbs would help19:02
mriedemvolume_id might not be set if boot from volume failed...19:02
dansmithsdague: we log the instance already, so I'm not sure..19:03
dansmithyeah19:03
*** lpetrut has joined #openstack-nova19:03
dansmithinstance should be enough19:03
dansmithbecause you can just go see what is still attached to that instance in cinder19:03
sdaguewe log the instance, but this is a fail inside a bdm loop19:03
mriedembdm.device_path should be set though19:03
sdaguewe should actually say which bdm is the cause19:03
dansmithmriedem: device_path isn't really useful is it?19:03
*** openstackgerrit has quit IRC19:03
dansmithsdague: we don't have uuids on BDMs so there's really not much identifying information, other than volume id if it's set19:04
mriedemdansmith: well device_path is part of the bdm unique constraint (if that existed)19:04
*** dave-mccowan has joined #openstack-nova19:04
*** openstackgerrit has joined #openstack-nova19:04
mriedemthe uniqueness constraints for bdm today are volume_id and device_path19:04
mriedem*device_name19:04
sdaguewell all those calls are bdm.volume_id19:04
sdagueso I kind of assume that's a thing19:04
dansmithyeah, I guess so given they don't check first19:05
sdagueunless we're using Except to catch attr error19:05
mriedemyar19:05
mriedemvol_bdms = [bdm for bdm in bdms if bdm.is_volume]19:05
mriedemthey do check i think19:05
dansmithah19:05
mriedemi remember looking at this before19:05
dansmithright, but this is volumes only,19:05
mriedemb/c of the gate race failure here19:05
dansmithbut are you saying BFV might not have a volume_id here even though it is_volume?19:05
dansmithotherwise I don't know how it could be making this call to cinder19:05
mriedemdansmith: i'd have to look but i think that's possible19:06
mriedemif boot from volume where source_type is image/snapshot/blank fails19:06
mriedemb/c that's what gives us the volume_id from cinder and we set it on the bdm19:06
mriedemdown in nova.virt.block_device19:06
sdaguein which case the Exception catch is actually masking that fact, because we'll end up with a typeerror or something19:06
dansmithmriedem: so we might be calling terminate here, but without expecting it to work for that case or something?19:06
*** penick has joined #openstack-nova19:06
mriedemwhich should be in the prep_block_devices lock that is shared with the build_resources clock19:06
mriedem*lock19:06
sdaguewhich is actually part of the concern in my mind19:06
dansmithsdague: not if the connector is how we find it19:07
dansmithsdague: even if we did, we need to terminate the rest of the volumes19:07
sdaguedansmith: which is fine19:07
dansmithokay, so regardless, bdm.volume_id will not crash19:07
*** penick has quit IRC19:07
sdaguehowever if we're going to steamroll grind to success, we should at least say which bodies we left behidn19:07
dansmithit might be none or something,19:07
dansmithbut it's not going to kill us here,19:07
dansmith_diana_: so I think the outcome of that is, please add bdm.volume_id to that log message19:08
mriedemdansmith: sdague: hold the phone19:09
*** Marga_ has joined #openstack-nova19:09
openstackgerritAugustina Ragwitz proposed openstack/nova-specs: Add project id validation with keystone  https://review.openstack.org/29433719:09
dansmithsdague: I would say if we're going to do that we should add it to all the ignores above right?19:09
sdaguedansmith: yeh, I think that would be worthwhile as well19:09
_diana_dansmith: will do, thanks19:09
*** cdent has joined #openstack-nova19:10
*** Sukhdev has joined #openstack-nova19:10
sdagueit will be helpful to track down when things went wrong in people's environments, and what resources might be in inconsistent states19:10
dansmithmriedem: volume_id is nullable, which I hope means it's always set, but might be none19:10
mriedemdansmith: commented, look at the comment from me in terminate_instance19:10
dansmithmriedem: it'll be set or None right?19:11
dansmithyep, it will19:12
*** cdelatte has joined #openstack-nova19:12
dansmith_diana_: so I would, at the top, try to find the best identification you can for the bdm19:12
mriedemblock_device_obj[key] = db_block_device[key]19:12
mriedemdansmith: yeah19:12
dansmith_diana_: volume_id if it's non-None, and then the device_name, and if nothing else set it to sdague's email address19:13
_diana_lol19:13
sdaguedansmith: :P19:13
dansmithI'm not joking, who is joking?19:14
dansmithokay, I'm joking19:14
dansmithbut not really19:14
dansmithno, I am19:14
* dansmith channels his inner izzard19:14
*** penick has joined #openstack-nova19:14
*** ociuhandu has quit IRC19:15
mriedemto see how it works with the bfv case, you'd have to add a sleep or something in the flow that gives you time to take down the cinder-api when nova tries to create the volume19:15
*** penick has quit IRC19:16
mriedemif nova creates the volume and we fail (or timeout) while waiting for it to be available, then we clean that up ourselves19:16
openstackgerritAugustina Ragwitz proposed openstack/nova-specs: Add project id validation with keystone  https://review.openstack.org/29433719:16
mriedemi remember in one of my changes that i tried changing "vol_bdms = [bdm for bdm in bdms if bdm.is_volume]" to check bdm.volume_id but ndipanov didn't like that19:17
*** zenoway has joined #openstack-nova19:18
mriedem_diana_: fyi https://review.openstack.org/#/c/258118/19:19
_diana_mriedem: Thanks! I'll review the logs here and that review, and check back in a bit with any questions.19:21
mriedemif bdm.volume_id is None, we don't get to Exception19:26
mriedemwe get VolumeNotFound19:26
mriedemhttps://github.com/openstack/cinder/blob/master/cinder/api/contrib/volume_actions.py#L233-L23619:26
mriedemdansmith: _diana_: sdague: ^19:26
mriedemso it should be safe to use bdm.volume_id in the generic Exception block19:26
dansmithmriedem: well, it sounds like we'll grab an id first and be able to log it for all the cases19:26
sdaguemriedem: cool19:26
dansmithbut yeah, cool19:26
*** jamielennox|away is now known as jamielennox19:26
mriedembtw, this brings back some bad memories of that gate bug19:27
mriedemsdague: ever seen failures like this? http://logs.openstack.org/40/300140/1/check/gate-tempest-dsvm-full/0a6f567/console.html19:28
openstackgerritJay Pipes proposed openstack/nova-specs: resource-providers: Migrate compute node inventory  https://review.openstack.org/30017519:28
openstackgerritJay Pipes proposed openstack/nova-specs: resource-providers: generic resource pools  https://review.openstack.org/30017619:28
mriedemjenkins hated my little .gitignore patch https://review.openstack.org/#/c/300140/19:28
openstackgerritJay Pipes proposed openstack/nova-specs: resource-providers: Move allocation fields  https://review.openstack.org/30017719:28
openstackgerritJay Pipes proposed openstack/nova-specs: resource-providers: Scheduler filters in DB  https://review.openstack.org/30017819:28
mriedemsdague: nvm, i see you asked the same in infra19:28
jaypipesmriedem, dansmith, johnthetubaguy: k ^^ is four specs in the resource-providers blueprint series, properly ordered and dependency linked.19:29
*** sileht has quit IRC19:30
*** apoorvad has quit IRC19:30
jaypipesmriedem: if you wouldn't mind setting the spec URL of https://blueprints.launchpad.net/nova/+spec/compute-node-inventory-newton to https://review.openstack.org/#/c/300175/ that would be appreciated.19:31
mriedemjaypipes: done19:32
doffmalaski: Do you think it would be worthwhile to add the insertion of cell0 to the mappings to one of the test fixtures?19:32
alaskidoffm: I think so. for the most part we really just want to assume it's there19:34
doffmalaski: Thanks. I guess I'll add to Database fixture.19:36
*** ociuhandu has joined #openstack-nova19:37
dansmithdoffm: alaski: you think it should be integrated at that level and not a fixture of its own or something?19:37
dansmithperhaps the flavors-exist-since-dinosaurs thing has me a little overly sensitive19:37
doffmdansmith: I really don't know. I will basically accept whatever you suggest here.19:38
alaskiI don't have a strong stance on existing vs new fixture. just that a fixture would be good19:38
dansmithhow many of our tests will have to have that mapping in place?19:39
dansmithonly scheduling tests19:39
doffmAnything that touches the scheduling.19:39
dansmithor tests that hit the create api19:39
doffmOr building of instances really.19:39
mriedemso 90% of the functional tests19:39
alaskior list/show instances19:39
dansmithI dunno, I think it'd be nice to know what code requires that to be in place by being rather explicit19:39
dansmithmriedem: sure the functional test base should have it19:40
*** vilobhmm11 has quit IRC19:40
*** krtaylor has quit IRC19:40
dansmithalaski: list show through the api19:40
dansmithright?19:40
*** vilobhmm11 has joined #openstack-nova19:40
alaskiright19:40
dansmithso of the non-mocked cases of that, I think it's almost all in fuctional tests, an then tests specifically testing this stuff19:40
dansmither, unit tests specifically covering this19:41
*** mgoddard has quit IRC19:41
dansmithgive how easy it is to just use a fixture if you need it19:41
dansmithI kinda think, separate fixture until it's clear that's too painful19:41
dansmithjust MHO19:41
rlrossitdansmith: I hope your blast from the past works19:42
openstackgerritRyan Rossiter proposed openstack/nova: Make neutronapi get_floating*() methods return objects  https://review.openstack.org/11574919:42
openstackgerritRyan Rossiter proposed openstack/nova: Give the FloatingIP object an external_id field, because neutron  https://review.openstack.org/11580119:42
alaskidansmith: sure, seems reasonable to me19:42
doffmOk.19:42
*** ociuhandu has quit IRC19:43
*** sdake_ has joined #openstack-nova19:43
dansmithrlrossit: I wish you luck19:43
sdaguemriedem: https://review.openstack.org/#/c/299893/ what you asked for yesterday19:43
sdagueand alex_xu pointed out that we actually do have a warning for the legacy v2.0 stack already19:44
*** sdake has quit IRC19:44
sdaguehttps://github.com/openstack/nova/blob/master/nova/api/openstack/compute/__init__.py#L5819:44
*** ociuhandu has joined #openstack-nova19:44
mriedemsdague: ahh, i was looking for something like that in https://github.com/openstack/nova/blob/master/nova/api/openstack/__init__.py yesterday19:45
sdaguemriedem: yeh, me too, he caught in on the review that I pushed that added it another place19:46
mriedemoh sdague https://review.openstack.org/#/c/299893/2/nova/api/openstack/compute/versions.py19:47
mriedemyou know i have to -2 b/c of that whitespace change right19:47
mriedemsdague: touch up the reno and i'll +W quick19:49
*** jwcroppe has joined #openstack-nova19:49
openstackgerritSean Dague proposed openstack/nova: remove the ability to disable v2.1  https://review.openstack.org/29989319:50
sdaguedone, and cleaned up the whitespace19:51
*** jwcroppe_ has quit IRC19:51
sdaguethat's what happens when I delete a method, then have to add part of it back in19:51
mriedem+W19:51
mriedemthanks19:51
mriedemsdague: i guess you can just abandon the top patch19:54
sdaguemriedem: yeh, I parked it for now mostly as a reminder19:54
*** ihrachys has joined #openstack-nova19:54
-openstackstatus- NOTICE: Any jobs which erroneously failed on missing traceroute packages should be safe to recheck now19:55
*** sdake_ has quit IRC19:56
*** jaugustine has joined #openstack-nova19:57
*** mylu has joined #openstack-nova19:58
openstackgerritSaverio Proto proposed openstack/nova: Allow snapshot_volume_backed in pause/suspended  https://review.openstack.org/29586519:58
sdaguemriedem: I'm looking at the summit planning etherpad19:59
sdagueand I'm a bit concerned that "API: Policy/Capability Discoverbility and Microversion Docs" is 17 lbs of dirt in a 5 lb bag. There are like 20 bullets in that item19:59
*** haomaiwang has quit IRC20:01
*** haomaiwa_ has joined #openstack-nova20:01
mriedemsdague: that reminds me, i was going to look that over again before the meeting in an hour20:01
mriedemsdague: are we going to be talking about the api docs in the nova time-boxed sessions, or just friday meetup? are you going to be talking about any of that in the xp tuesday?20:02
sdaguemriedem: I have a xp session proposed for that20:03
sdagueI'd be fine with that and a small slice on friday20:03
sdagueand encourage nova folks to come to the xp session20:03
mriedemok, and the first session on thursday i penciled in ironic, which might also be a tuesday xp talk20:03
sdagueI think we should dedicate a whole block to "discoverable policy in code"20:04
sdaguethat's going to be important to make sure we actually give it time20:04
*** rlrossit has quit IRC20:05
sdagueI also proposed another bit on what's the actual path to nova network deprecation20:06
mriedemmaybe we move johnthetubaguy's feature classification thing to thursday20:07
mriedemthat would free up the last 2 slots on wed for api talks20:07
sdagueyeh, that might be best. We definitely want to earmark one block entirely about policy in code20:07
sdagueprobably the first one20:07
*** salv-orlando has joined #openstack-nova20:08
sdagueis os-brick mostly about privsep?20:08
sdaguebecause privsep and libraries seems like a thing we should propose into xp day20:09
mriedemyeah, which i'm not sure there is actually anything to go over there20:09
sdagueas we seem to continue to be confused here20:09
mriedemthe cinder session is really going to be all multi-attach i think20:09
sdagueabout how anything lands20:09
sdaguemriedem: yeh, I'd just call it multi-attach then20:09
*** Sukhdev has quit IRC20:09
sdaguemriedem: ok, who has the todo to get privsep / os-brick onto the xp discussion list?20:10
openstackgerritSaverio Proto proposed openstack/nova: Allow snapshot_volume_backed in pause/suspended  https://review.openstack.org/29586520:10
mriedemsdague: you want it?20:13
mriedemelse i can add it20:14
sdaguenot really, I was actually just poking dims_20:14
*** jwcroppe has quit IRC20:14
*** baoli has joined #openstack-nova20:15
*** _diana_ is now known as diana_clarke20:15
*** efoley has joined #openstack-nova20:15
*** sileht has joined #openstack-nova20:15
*** smatzek has quit IRC20:15
*** baoli_ has joined #openstack-nova20:15
*** browne has quit IRC20:19
*** baoli has quit IRC20:19
*** ctrath has quit IRC20:20
dansmithalaski: doffm: so, on this cell0 command20:21
*** browne has joined #openstack-nova20:21
dansmithalaski: doffm: I feel like that's mechanical enough that we can/should tie it into something else20:21
dansmithpeople will only ever run that once and on an empty database20:22
dansmithcan't we, like, just do it at the end of api_db sync if we notice it's missing or something?20:22
sdaguedansmith: ++ yes please :)20:22
alaskidansmith: how would you specify which db to use for cell0?20:22
dansmithalaski: well, we'd need something, but I just feel like we're talking about an explosion of new commands here20:23
dansmithand everyone will have to do this,20:23
dansmithnot just multi-cell people20:23
sdaguealaski: the connect strings have to be in the nova.conf already, right?20:23
dansmithwell,20:23
alaskisdague: this can be a new database if desired20:23
dansmithyeah20:23
alaskidansmith: right, there is a little more to bootstrap nova with cells20:23
dansmithalaski: so we talked about a single command to move both dbs forward if you're a single-cell deploy20:24
alaskidansmith: if we can solve how to specify the db rolling it into something else seems great20:24
dansmithhow about we make that command and do it there?20:24
dansmithI don't really mind if new godaddy type deploys have to run a command to make sure they have theirs split,20:24
dansmithbut most people are going to want to run "get me a database"20:25
dansmithand do both sets of migrations and any other setup that they need to be able to go create their flavors (see what I did there?) and images20:25
alaskisure. in my head these have always been low level commands that we could layer a high level command on top of20:25
*** penick has joined #openstack-nova20:25
sdagueyeh, I think it's important that we get that high level command before cell0 shows up20:26
dansmithalaski: we could also have a conf variable called "cell-o-one" that short circuits a lot of setup steps like this20:26
*** fawadkhaliq has quit IRC20:27
dansmithI dunno, I just ... feel like we're adding a ton of steps and we need to have a straightforward path for the 90% case20:27
dansmithand20:27
dansmithsince this is't going to be done in newton,20:27
dansmithif we have a bunch of new shit people have to do just to upgrade to newton regardless20:27
dansmiththat just seems uncool20:27
*** jlopezgu has left #openstack-nova20:27
sdagueI have no concern having the ability to do more complicated things as long as there still is an easy path for the small scale case20:27
*** fawadkhaliq has joined #openstack-nova20:27
sdaguedansmith: ++20:27
alaskiwell, cell0 should be used in newton20:27
dansmithalaski: right, I know20:28
dansmiththat means they'll have to do it20:28
dansmitheven though no more than one cell in newton20:28
*** krtaylor has joined #openstack-nova20:28
dansmithand before someone else brings it up, the difference I see between this and default flavors is that this is not a thing that varies based on your size20:28
dansmitheveryone needs this record20:28
*** ihrachys has quit IRC20:28
*** apoorvad has joined #openstack-nova20:28
mriedemdansmith: you can drop the -2 on this now https://review.openstack.org/#/c/248780/20:29
dansmithmriedem: don20:29
dansmithe20:29
alaskidansmith: sdague so the options we're looking at are have the proposed nova-manage command with support for explicit db info, add a config for that which people can set and roll the setup into something else, all of the above and provide deployer choice. right?20:30
sdaguemriedem: I'm going to have to drop before the meeting, anything you need out of me in advance of it?20:30
alaskidansmith: the reason I haven't been overly concerned about this is setting up cells is something that will require some form of interaction from deployers20:31
alaskiwe can ease the initial upgrade path, but there will always be commands for this20:31
dansmithalaski: right, but I don't want newton non-cell people with 10-node deploys to have to run a bunch of commands for cells when (a) they can't have two and (b) they don't need it20:31
mriedemsdague: don't think so, i was just going through the agenda,20:31
mriedemsdague: dansmith: alaski: jaypipes: johnthetubaguy: i'm hoping to meet (maybe a hangout) next week to go over the design summit schedule20:32
sdagueright, the promiss of cells v2 was it was a seamless upgrade for non cells people20:32
mriedemsdague: when does the tuesday xp get voted on?20:32
sdaguemriedem: next week20:32
mriedemtuesday's TC meeting?20:32
alaskimriedem: sounds good20:32
sdagueprobably later than that20:32
dansmithalaski: there is no way to re-create  or change the record with this command, which means to me that its only ever run once, and I just feel like we could absorb multiple things into a "setup my cell" command or something20:32
dansmithalaski: which does the host mapping thing, and this, and whatever else20:33
dansmithI dunno20:33
mriedemsdague: hmm, kind of have to wait on the nova sessions being scheduled until after the tuesday stuff is in place20:33
*** stevemar has quit IRC20:33
sdaguedansmith / alaski could we etherpad a set of pseudo commands that we'd expect people would have to run in a few different scenarios, and work backwards from that?20:33
*** stevemar has joined #openstack-nova20:33
alaskisdague: bauzas is working on that20:34
*** cleong has quit IRC20:34
mriedemi think the project sessions have to be in the schedule ~4/14, so i guess we could sort it out the week after next20:34
alaskidansmith: that's reasonable. I think where we differ is that I'm expecting a third command that wraps the cell0/host_mapping thing. which still allows for doing it all manually.20:34
sdaguemriedem: right, I was intending to get this nailed next week, we should have in/out by wed.20:35
sdaguealaski: ok, it seems like this conversation about how many commands, and of what type gets easier if we actually see them on paper somewhere20:35
dansmithalaski: that would be the fourth.. that's the concern I have, this is now three complicated commands that take db urls that everyone would have to run20:35
*** cdelatte has quit IRC20:35
dansmithalaski: for most people with only an api and single cell db, we should be able to automate all of that in a single "nova-manage cells setup --one-cell" or somethng20:36
dansmithand in reality,20:36
dansmithsince cellsv1 people can't convert yet,20:36
dansmith*everyone* will run the same command I think20:36
dansmithso until they can be separate, it just seems like work for the sake of it20:36
*** ccarmack has quit IRC20:36
*** vilobhmm11 has quit IRC20:37
dansmithbut, I agree we should lay out the steps as they are now and figure out what we can consolidate20:37
*** vilobhmm11 has joined #openstack-nova20:37
*** vilobhmm11 has quit IRC20:37
alaskiyeah20:37
alaskiI'm not opposing a single command to do it all. I just want to allow for power users to customize20:37
sdaguealaski: right, I don't think anyone's said that wouldn't be an option20:38
alaskiyeah, just seems like top-down vs bottom-up right now20:38
dansmithalaski: it's mostly about not wanting to risk releasing newton with no unified approach20:39
dansmithso I'd rather build with that from the start20:39
dansmithand anything we add right now, has to be done with an understanding of how the simple case goes20:39
sdagueright, because retrofiting easy user experience late in the game never works, we need that use case in the mix as all the commands get designed20:40
alaskiokay. my experience is very much the heavily customized deployment, so it's good to get this pushback20:41
dansmithyeah, I know, RAX _wants_ a bucket of parts :)20:41
*** tbachman has joined #openstack-nova20:41
dansmithnone of my customers want that :)20:41
*** unicell has quit IRC20:42
alaskiit seems easy enough to me, sync two dbs, setup a cell0, setup a first cell, then add cells as necessary. pretty much all scriptable20:42
*** fawadkhaliq has quit IRC20:42
alaskibut I see your points20:42
dansmithheh20:42
dansmithalaski: you left out two steps you know20:43
*** ayoung has quit IRC20:43
sdagueright, if only every web shop with a 10 node deploy had rax experienced ops team :)20:43
alaskisdague: only one way to get that experience :)20:43
alaskidoffm: melwitt are y'all following along?20:44
dansmithuse hard-to-use software?20:44
alaskiI'm kidding, but sometimes you just do it and learn20:45
*** fawadkhaliq has joined #openstack-nova20:45
*** artom has quit IRC20:45
mriedemdoffm is at buffalo wild wings20:46
melwittalaski: sort of :)20:46
alaskithe gist of it for me is that we should roll the logic from your nova-manage command and doffms cell0 command into one easy to use button20:47
*** Bjoern_ has joined #openstack-nova20:47
*** Bjoern_ is now known as BjoernT20:48
melwittah, okay20:48
alaskibasically roll with all defaults at the start, and add customization controls later20:48
alaskithough transport_url is still tricky here20:48
dansmithalaski: you can add the custom command alongside the thing that makes the unified command do the same thing with defaults20:49
dansmithI just want to see them built along the way20:49
dansmithnot in n3 us sit back and say "how do we automate the mess we made?"20:49
alaskiokay20:50
*** ccarmack has joined #openstack-nova20:50
*** dave-mccowan has quit IRC20:51
*** browne has quit IRC20:51
*** tbachman has quit IRC20:53
*** browne has joined #openstack-nova20:54
*** rfolco has quit IRC20:56
*** stevemar_ has joined #openstack-nova20:56
*** knikolla has quit IRC20:56
*** stevemar has quit IRC20:56
mriedemadrian_otto akuriata alevine alexpilotti aloga andreykurilin anteaya artom auggy 20:58
mriedembauzas belliott belmoreira bobball cburgess claudiub danpb dguitarbite _diana_ 20:58
mriedemdiana_clarke dims duncant edleafe efried flip214 funzo garyk gcb gjayavelu 20:58
mriedemirina_pov jaypipes jcookekhugen jgrimm jichen jlvillal jroll kashyap klindgren20:58
mriedemkrtaylor lbeliveau lxsli macsz markus_z mdorman med_ mikal mjturek mnestratov 20:58
mriedemmoshele mrda nagyz ndipanov neiljerram nic Nisha PaulMurray raildo rgeragnov 20:58
mriedemsc68cal scottda sdague sileht sorrison swamireddy thomasem thorst tjones tonyb 20:58
mriedemtpatil tpatzig xyang rdopiera sarafraj woodster20:58
mriedemnova meeting in 2 minutes20:58
*** tbachman has joined #openstack-nova20:59
*** edmondsw has quit IRC21:00
*** mylu has quit IRC21:00
*** haomaiwa_ has quit IRC21:01
openstackgerritAndrew Laski proposed openstack/nova: Change BuildRequest to contain a serialized instance  https://review.openstack.org/29845521:01
alaskidansmith: I need to see what breaks, but ^21:01
dansmithokay21:01
*** haomaiwang has joined #openstack-nova21:01
*** aysyd has quit IRC21:02
alaskiit cleans things up quite a bit, but a lot had to change to get there21:03
*** Goneri has quit IRC21:04
*** unicell has joined #openstack-nova21:05
*** zengine has quit IRC21:06
*** thorst has quit IRC21:07
*** thorst has joined #openstack-nova21:08
*** cdent has quit IRC21:08
openstackgerritAnusha Unnam proposed openstack/nova: config options: Centralize xenapi driver options  https://review.openstack.org/30021721:09
tonybdansmith: So because I didn't know who Warren Haynes was I listened to a couple of albums the other day21:11
*** mylu has joined #openstack-nova21:12
dansmithtonyb: good man21:12
tonybdansmith: I don't think going to a concert would be my cup'o'tea but I'd listen to the albums again21:12
dansmithtonyb: heh, okay21:12
dansmithtonyb: it's mostly continuous jamming for two hours21:12
dansmithso it's pretty awesome show material21:12
*** thorst has quit IRC21:12
tonybdansmith: :)21:13
dansmithbut kudos for at least exploring :)21:13
* tonyb will listen to most things at least once21:13
*** thorst has joined #openstack-nova21:14
auggytonyb: dansmith iggy pop was pretty awesome21:14
dansmithwell, I didn't see you show up at the venue with the real music, so I assumed it was at least tolerable21:15
auggyhe had so much energy, i can only imagine what he must have been like when he was younger haha21:15
dansmithcocaine is quite the wonder drug21:15
auggyha21:15
tonyb:)21:16
auggyi mean besides that, he was just an energetic guy i'm sure21:16
dansmithI figure it would really ruin the illusion if we ever saw each other on this side of the planet, much less in this town21:16
*** wuhg has joined #openstack-nova21:16
*** baoli_ has quit IRC21:17
auggyhaha well in this case i even ducked out right at the encore21:17
auggyi strained my back a bit ago and i needed to head home :(21:17
dansmithso dd we, because "being trampled" is not part of the experience for me21:17
auggyluckily the iggy pop concert was a few blocks from my apartment21:17
*** Sukhdev has joined #openstack-nova21:18
auggyyeah i was not even gonna try to get out of keller in that crowd21:18
dansmiththere are things you can yell to the crowd to help speed it up21:18
dansmithbut you have to be ready for it21:18
*** eharney has quit IRC21:18
dansmithlike "free flannel in the parking lot!"21:18
*** thorst has quit IRC21:19
auggymaybe that would work at your concert ;)21:19
auggyi think "free cocaine in the parking lot" would have worked better for mine hahaha21:19
*** rmevans has joined #openstack-nova21:19
dansmithheh, no, my concert was all old people21:19
auggywhat, old people in portland don't wear flannel??21:19
dansmithI just mean portland in general, but good call.. know your audience21:19
dansmithnot these old people21:20
auggyoh right, these were all the yuppies from hillboro21:20
auggyactually i was surprised at the age range of the audience at iggy pop21:21
auggylike a good 1/3 were probably into the stooges back in the day21:21
auggy1/3 maybe their grandparents were into the stooges21:21
*** rtheis has quit IRC21:21
*** efried has quit IRC21:21
auggyand then 1/3 genx'ers like me ;)21:21
openstackgerritGage Hugo proposed openstack/nova: Verify pci_passthrough_whitelist fields  https://review.openstack.org/30021921:22
*** pmathews has joined #openstack-nova21:22
*** achanda has quit IRC21:22
mriedemtonyb: you should have announced this in the nova meeting https://review.openstack.org/#/c/289255/21:23
*** mylu has quit IRC21:23
tonybmriedem: but we were going for a record!21:23
mriedemi know21:24
tonybmriedem: you're right I shoudl have.21:24
auggyyeah i joined the meeting late and you all had left already :(21:24
*** baoli has joined #openstack-nova21:24
* auggy missed the party :(((21:24
*** mylu has joined #openstack-nova21:27
*** vilobhmm11 has joined #openstack-nova21:28
*** knikolla has joined #openstack-nova21:28
*** haplo37 has joined #openstack-nova21:29
mriedemhas anyone figured out if there is easy public transit from the austin airport to the hotel areas?21:29
auggywhen i was in austin a couple years ago for YAPC::NA the public transit was, erm, lacking21:29
*** sdake has joined #openstack-nova21:30
alaskiI was provided this link http://www.capmetro.org/airport/, but I haven't looked into it yet21:30
mriedemmaybe i'll just pay mtreinish to pick me up in his rental ferrari21:30
mriedemha21:31
mriedem$1.75 Premium fare21:31
mriedemif premium is $1.75, what's standard?21:31
auggyhttps://goo.gl/maps/FgWEv4nnm4n21:31
sarafrajmriedem: Austin lacks in public transport especially from airport. I suggest using ride hailing apps.21:31
mtreinishI have a rental ferrari?21:31
mriedemsarafraj: even that capmetro thing isn't good?21:32
mriedemi thought austin was overrun with hippy liberals, this is like a crazy paradox21:32
auggyit looks like ther's "100-Airport Flyer-OB"21:32
auggythat's the bus that came up in the google maps i plugged in21:33
sarafrajmriedem: I am from Austin and I prefer uber or lyft over all others21:33
manjeetsmriedem this time I have osapi_max_limit = 4000 in nova.conf i checked conf after re deployment but still see  1000 returned by nova list21:33
mriedemmanjeets: did you restart the nova-api service?21:33
manjeetsi re deployed openstack21:33
manjeetsi did nova_api restart first it did not work then i just re deployed21:34
auggywow that capmetro thing looks pretty sweet21:34
auggythat's what gmaps had suggested, i just didn't know what 100 Airport Flyer was21:34
mriedemmanjeets: ok, well, not sure then. but it's end of working hours for me, so i'm signing off.21:35
*** mriedem has quit IRC21:35
manjeetsmriedem no worries21:35
*** fawadkhaliq has quit IRC21:38
*** ivase has quit IRC21:39
openstackgerritJay Pipes proposed openstack/nova: Add Service.get_minimum_version_multi() for multiple binaries  https://review.openstack.org/27931221:39
*** fawadkhaliq has joined #openstack-nova21:39
jaypipesdansmith: ^^ just rebasing...21:40
*** fawadkhaliq has quit IRC21:40
*** fawadkhaliq has joined #openstack-nova21:41
openstackgerritJay Pipes proposed openstack/nova: Move resource provider staticmethods to proxies  https://review.openstack.org/29952621:41
*** mdrabe has quit IRC21:41
openstackgerritJay Pipes proposed openstack/nova: Migrate compute node resource information to Inventory objects  https://review.openstack.org/27931321:41
*** rushiagr has quit IRC21:42
*** haplo37 has quit IRC21:42
*** cdelatte has joined #openstack-nova21:43
*** fawadkhaliq has quit IRC21:44
*** fawadkhaliq has joined #openstack-nova21:44
*** burt has quit IRC21:44
*** zenoway has quit IRC21:45
*** gyee has quit IRC21:46
*** rushiagr has joined #openstack-nova21:47
*** tojuvone_ has joined #openstack-nova21:47
*** angdraug has joined #openstack-nova21:48
*** gyee has joined #openstack-nova21:48
*** burgerk has quit IRC21:49
*** tojuvone has quit IRC21:51
*** gagehugo has quit IRC21:52
*** jaugustine has quit IRC21:55
*** mylu has quit IRC21:56
*** krtaylor has quit IRC21:57
*** zenoway has joined #openstack-nova21:57
*** artom has joined #openstack-nova21:59
artomdansmith, still around? Can we quickly discuss https://review.openstack.org/#/c/264015/20/nova/objects/metadata.py@76 ?21:59
*** jschlueter has joined #openstack-nova22:00
dansmithartom: which part?22:00
dansmithI haven't looked at the comments since my last reply22:00
*** haomaiwang has quit IRC22:01
*** xyang1 has quit IRC22:01
artomdansmith, the to_dict method generating guest-visible stuff, thus making it an API22:01
*** haomaiwang has joined #openstack-nova22:01
*** amotoki has joined #openstack-nova22:01
dansmithokay22:01
*** zenoway has quit IRC22:02
artomdansmith, so, I definitely agree that having API stuff buried in objects is dangerous.22:03
dansmithyeah22:04
*** BjoernT has quit IRC22:04
dansmithI think it probably needs to be closer to the metadata code we have now22:04
*** unicell has quit IRC22:04
dansmitha consumer of the objects and not part of them22:04
*** ametts has quit IRC22:09
artomdansmith, right, so in https://review.openstack.org/#/c/264018/19/nova/api/metadata/base.py@346 instead of just calling to_dict, get the fields one by one and stick the values in the outputted dict...22:10
*** Goneri has joined #openstack-nova22:10
artom(My initial reasoning was having the object be responsible for their own attributes, but this is clearly a case where that doesn't apply)22:10
dansmithartom: right, well, the organization is nice of course, I just think it's setting us up to fail22:10
artomAnd then add some tests in functional/api_sample_tests...22:11
dansmithyeah, something to make sure the format sticks22:11
artomdansmith, oh yeah. I read your comments and though "Damn, dude's got a point." :)22:11
dansmithI also think you probably want to be more structural with your handling of them and not just loop over the fields,22:12
dansmithbecause that means we add something and it just goes in, or we change the type or something and it's reflected there22:12
*** Bjoern_ has joined #openstack-nova22:12
dansmithbut I say move them over there and then we can argue about the method22:12
artomYeah, I wasn't planning to... Essentially, we can't "trust" the object in the metadata handler22:12
dansmithright22:12
dansmithit's model, not view22:12
artom'tis settled then.22:14
artomvladikr, if you're seeing this ^^^22:14
dansmithI wil try to figure out what vladikr's deal is with the base class too22:15
dansmithtomorrowish22:15
*** thorst has joined #openstack-nova22:16
artomThe registered stuff? He was saying the problem is that other example of non registered base classes aren't used as fields in other objects.22:16
artomSo it looks like this field-ness is what's causing problems here22:16
dansmithah22:16
dansmithwell, then we have to fix the test22:16
*** erlon has quit IRC22:17
*** vladikr has quit IRC22:18
*** cdelatte has quit IRC22:19
*** andymaier has quit IRC22:25
*** vilobhmm11 has quit IRC22:26
*** vladikr has joined #openstack-nova22:27
*** vilobhmm11 has joined #openstack-nova22:27
*** ducttape_ has quit IRC22:30
*** dimtruck is now known as zz_dimtruck22:31
edleafeauggy: (reading scrollback) I've seen Iggy many times. The last was in 1980!22:34
claudiubhah, interesting news... :D http://fossbytes.com/microsoft-buys-canonical-kills-ubuntu-linux-forever/?fb_action_ids=1102810479758847&fb_action_types=og.comments22:35
claudiub*empire music starts playing*22:36
edleafeclaudiub: A little early on this side of the world...22:36
tbachmanlol22:37
tbachmanlet the fun begin :)22:37
tbachmanI know I’ll fall prey to one of these22:37
*** thorst has quit IRC22:37
*** baoli has quit IRC22:38
*** unicell has joined #openstack-nova22:38
*** dave-mccowan has joined #openstack-nova22:40
*** vladikr_ has joined #openstack-nova22:43
*** vladikr has quit IRC22:44
*** ayoung has joined #openstack-nova22:45
*** lpetrut has quit IRC22:46
*** rmevans has quit IRC22:49
*** dave-mccowan has quit IRC22:50
*** sdague has quit IRC22:50
*** akshai has quit IRC22:51
*** browne has quit IRC22:55
*** sdake has quit IRC22:57
*** penick has quit IRC22:58
*** browne has joined #openstack-nova22:59
*** haomaiwang has quit IRC23:01
*** akshai has joined #openstack-nova23:02
*** haomaiwa_ has joined #openstack-nova23:02
*** rwmjones has quit IRC23:03
*** akshai_ has joined #openstack-nova23:03
*** akshai has quit IRC23:06
*** rwmjones has joined #openstack-nova23:09
*** sdake has joined #openstack-nova23:09
*** Bjoern_ has quit IRC23:10
*** knikolla has quit IRC23:10
*** salv-orlando has quit IRC23:11
*** sdake_ has joined #openstack-nova23:12
*** sdake has quit IRC23:13
*** amotoki has quit IRC23:13
*** pwnall has quit IRC23:15
*** tjones_zzzz has left #openstack-nova23:17
*** READ10 has joined #openstack-nova23:24
*** ayoung has quit IRC23:26
*** pradk has quit IRC23:26
*** efried has joined #openstack-nova23:34
*** mriedem has joined #openstack-nova23:36
* tbachman wonders if his april fools joke will be getting +2’s on his nova patch :-)23:38
*** thorst has joined #openstack-nova23:38
tbachmanI guess it would be funnier to get -2s23:38
*** rk4n has quit IRC23:39
tbachmanwhich is worse — a false -2, or a false +2?23:39
* tbachman goes back to his lurking corner23:39
jrollfalse -2 is worse, a false +2 means you only have to bribe one more core23:40
jroll:P23:40
tbachmanlol23:41
*** sdake_ has quit IRC23:43
*** ivase has joined #openstack-nova23:46
*** thorst has quit IRC23:46
*** ivase has quit IRC23:50
*** knikolla has joined #openstack-nova23:53
*** cdelatte has joined #openstack-nova23:53
*** unicell has quit IRC23:57
*** Bofu2MBP has joined #openstack-nova23:59

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