| *** dave-mccowan has joined #openstack-nova | 00:04 | |
| *** mikeit has quit IRC | 00:08 | |
| *** pixelbeat_ has quit IRC | 00:08 | |
| *** _diana_ has quit IRC | 00:09 | |
| *** wanghua has joined #openstack-nova | 00:19 | |
| *** humble_ has joined #openstack-nova | 00:27 | |
| *** morgabra has quit IRC | 00:28 | |
| *** morgabra has joined #openstack-nova | 00:29 | |
| *** wanghua has quit IRC | 00:29 | |
| *** lyarwood has quit IRC | 00:31 | |
| *** Kennan2 is now known as Kennan | 00:33 | |
| *** lyarwood has joined #openstack-nova | 00:33 | |
| *** amotoki has joined #openstack-nova | 00:39 | |
| *** otter768 has joined #openstack-nova | 00:41 | |
| *** amotoki has quit IRC | 00:43 | |
| *** promulo__ has joined #openstack-nova | 00:44 | |
| *** promulo has quit IRC | 00:48 | |
| *** humble__ has joined #openstack-nova | 00:49 | |
| *** humble_ has quit IRC | 00:52 | |
| *** mikeit has joined #openstack-nova | 00:54 | |
| *** jasondotstar has quit IRC | 00:54 | |
| *** dims__ has joined #openstack-nova | 00:55 | |
| *** dims_ has quit IRC | 00:56 | |
| *** mikeit has quit IRC | 00:56 | |
| *** huazhihao has joined #openstack-nova | 01:01 | |
| *** dims__ has quit IRC | 01:06 | |
| *** Piet has joined #openstack-nova | 01:06 | |
| *** dims_ has joined #openstack-nova | 01:07 | |
| *** baoli has joined #openstack-nova | 01:10 | |
| *** gmann has joined #openstack-nova | 01:11 | |
| *** dims_ has quit IRC | 01:15 | |
| *** baoli has quit IRC | 01:15 | |
| *** _diana_ has joined #openstack-nova | 01:15 | |
| *** otter768 has quit IRC | 01:16 | |
| *** liusheng has quit IRC | 01:18 | |
| *** huazhihao has quit IRC | 01:22 | |
| *** gmann has quit IRC | 01:26 | |
| *** shyu has quit IRC | 01:29 | |
| *** Alexandra_ has quit IRC | 01:29 | |
| *** dave-mccowan has quit IRC | 01:34 | |
| *** zhenguo has joined #openstack-nova | 01:34 | |
| *** gmann has joined #openstack-nova | 01:35 | |
| *** dave-mccowan has joined #openstack-nova | 01:37 | |
| *** lyanchih has joined #openstack-nova | 01:38 | |
| *** jasondotstar has joined #openstack-nova | 01:38 | |
| *** amotoki has joined #openstack-nova | 01:40 | |
| *** _diana_ has quit IRC | 01:41 | |
| *** shyu has joined #openstack-nova | 01:42 | |
| *** amotoki has quit IRC | 01:44 | |
| *** _diana_ has joined #openstack-nova | 01:45 | |
| *** Longgeek_ has joined #openstack-nova | 01:46 | |
| *** Longgeek has quit IRC | 01:49 | |
| *** _diana_ has quit IRC | 01:49 | |
| *** baoli has joined #openstack-nova | 01:50 | |
| *** _diana_ has joined #openstack-nova | 01:50 | |
| *** ssurana has quit IRC | 01:51 | |
| *** jamielennox is now known as jamielennox|away | 01:51 | |
| *** jasondotstar has quit IRC | 01:55 | |
| *** _diana_ has quit IRC | 01:56 | |
| *** baoli has quit IRC | 01:58 | |
| *** annashen has joined #openstack-nova | 02:00 | |
| *** jamielennox|away is now known as jamielennox | 02:02 | |
| *** ssurana has joined #openstack-nova | 02:03 | |
| *** _diana_ has joined #openstack-nova | 02:04 | |
| *** annashen has quit IRC | 02:05 | |
| *** yingjun has joined #openstack-nova | 02:06 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Add missing rules in polcy.json https://review.openstack.org/188235 | 02:06 |
|---|---|---|
| *** annashen has joined #openstack-nova | 02:07 | |
| *** ssurana has quit IRC | 02:07 | |
| *** annashen has quit IRC | 02:08 | |
| *** _diana_ has quit IRC | 02:08 | |
| *** annashen has joined #openstack-nova | 02:08 | |
| *** mutoulbj has joined #openstack-nova | 02:09 | |
| openstackgerrit | lyanchih proposed openstack/nova: Creating instance fail when inject ssh key in cells mode https://review.openstack.org/197514 | 02:18 |
| *** baoli has joined #openstack-nova | 02:20 | |
| *** annashen has quit IRC | 02:21 | |
| *** ssurana has joined #openstack-nova | 02:21 | |
| *** achanda has joined #openstack-nova | 02:24 | |
| *** yingjun has quit IRC | 02:24 | |
| *** yingjun has joined #openstack-nova | 02:25 | |
| *** ssurana has quit IRC | 02:26 | |
| *** annashen has joined #openstack-nova | 02:26 | |
| *** Kevin_Zheng has joined #openstack-nova | 02:29 | |
| *** asettle has joined #openstack-nova | 02:32 | |
| *** achanda has quit IRC | 02:32 | |
| *** coolsvap|away is now known as coolsvap | 02:36 | |
| *** amotoki has joined #openstack-nova | 02:41 | |
| *** bkopilov has quit IRC | 02:45 | |
| *** amotoki has quit IRC | 02:45 | |
| *** Longgeek_ has quit IRC | 02:47 | |
| *** Longgeek has joined #openstack-nova | 02:47 | |
| *** tojuvone has joined #openstack-nova | 02:48 | |
| *** ssurana has joined #openstack-nova | 02:48 | |
| *** annashen has quit IRC | 02:50 | |
| *** coolsvap is now known as coolsvap|away | 02:50 | |
| *** hakimo has joined #openstack-nova | 02:51 | |
| *** dave-mccowan has quit IRC | 02:52 | |
| *** ssurana has quit IRC | 02:53 | |
| *** armax has joined #openstack-nova | 02:53 | |
| *** hakimo_ has quit IRC | 02:55 | |
| *** piyanai has quit IRC | 02:56 | |
| *** baoli has quit IRC | 02:56 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Add missing rules in polcy.json https://review.openstack.org/188235 | 03:00 |
| openstackgerrit | Alex Xu proposed openstack/nova: Add missing rules in polcy.json https://review.openstack.org/188235 | 03:01 |
| *** dave-mccowan has joined #openstack-nova | 03:02 | |
| *** armax has quit IRC | 03:03 | |
| *** annashen has joined #openstack-nova | 03:03 | |
| *** annashen has quit IRC | 03:05 | |
| *** humble_ has joined #openstack-nova | 03:07 | |
| *** humble__ has quit IRC | 03:10 | |
| *** amotoki has joined #openstack-nova | 03:15 | |
| *** Longgeek_ has joined #openstack-nova | 03:16 | |
| *** otter768 has joined #openstack-nova | 03:17 | |
| *** Longgeek has quit IRC | 03:18 | |
| *** otter768 has quit IRC | 03:21 | |
| *** annashen has joined #openstack-nova | 03:23 | |
| *** achanda has joined #openstack-nova | 03:26 | |
| *** annashen has quit IRC | 03:27 | |
| *** nihilifer has joined #openstack-nova | 03:28 | |
| *** yamahata has quit IRC | 03:28 | |
| *** tobe has joined #openstack-nova | 03:31 | |
| *** RuiChen has quit IRC | 03:33 | |
| *** oomichi has quit IRC | 03:34 | |
| *** humble__ has joined #openstack-nova | 03:34 | |
| *** nihilifer has quit IRC | 03:34 | |
| *** achanda has quit IRC | 03:35 | |
| *** achanda has joined #openstack-nova | 03:35 | |
| *** humble_ has quit IRC | 03:37 | |
| *** wanghua has joined #openstack-nova | 03:38 | |
| *** jamielennox is now known as jamielennox|away | 03:38 | |
| *** achanda has quit IRC | 03:40 | |
| *** humble__ has quit IRC | 03:40 | |
| *** yamahata has joined #openstack-nova | 03:40 | |
| *** annashen has joined #openstack-nova | 03:41 | |
| *** RuiChen has joined #openstack-nova | 03:41 | |
| *** dave-mccowan has quit IRC | 03:43 | |
| *** jamielennox|away is now known as jamielennox | 03:49 | |
| *** ssurana has joined #openstack-nova | 03:49 | |
| *** humble_ has joined #openstack-nova | 03:50 | |
| *** bkopilov has joined #openstack-nova | 03:51 | |
| *** mutoulbj has quit IRC | 03:51 | |
| *** amotoki has quit IRC | 03:52 | |
| *** mutoulbj has joined #openstack-nova | 03:52 | |
| *** Jeffrey4l has joined #openstack-nova | 03:52 | |
| *** wanghua has quit IRC | 03:53 | |
| *** coolsvap|away is now known as coolsvap | 03:53 | |
| *** Jeffrey4l has quit IRC | 03:53 | |
| *** Jeffrey4l has joined #openstack-nova | 03:53 | |
| *** ssurana has quit IRC | 03:54 | |
| *** baoli has joined #openstack-nova | 03:57 | |
| *** baoli has quit IRC | 04:01 | |
| *** amotoki has joined #openstack-nova | 04:06 | |
| *** ildikov has quit IRC | 04:06 | |
| *** shyu has quit IRC | 04:07 | |
| *** deepthi has joined #openstack-nova | 04:08 | |
| *** dims_ has joined #openstack-nova | 04:12 | |
| *** annashen has quit IRC | 04:12 | |
| *** amotoki has quit IRC | 04:14 | |
| *** wanghua has joined #openstack-nova | 04:15 | |
| *** dims__ has joined #openstack-nova | 04:16 | |
| *** lyanchih has quit IRC | 04:17 | |
| *** dims_ has quit IRC | 04:18 | |
| *** humble_ has quit IRC | 04:18 | |
| *** amotoki has joined #openstack-nova | 04:19 | |
| *** dims__ has quit IRC | 04:20 | |
| *** pradipta has joined #openstack-nova | 04:20 | |
| *** dims__ has joined #openstack-nova | 04:21 | |
| *** shuquan has joined #openstack-nova | 04:24 | |
| *** dims__ has quit IRC | 04:26 | |
| *** Sukhdev has joined #openstack-nova | 04:28 | |
| *** ijw has quit IRC | 04:31 | |
| *** shuquan has quit IRC | 04:34 | |
| *** coolsvap is now known as coolsvap|brb | 04:35 | |
| *** shuquan has joined #openstack-nova | 04:39 | |
| *** amotoki has quit IRC | 04:40 | |
| *** liusheng has joined #openstack-nova | 04:41 | |
| *** irenab has quit IRC | 04:41 | |
| *** shuquan has quit IRC | 04:44 | |
| *** annashen has joined #openstack-nova | 04:47 | |
| *** annashen has quit IRC | 04:49 | |
| *** ssurana has joined #openstack-nova | 04:50 | |
| *** tobe has quit IRC | 04:51 | |
| *** nihilifer has joined #openstack-nova | 04:53 | |
| *** ssurana has quit IRC | 04:54 | |
| *** asettle has quit IRC | 04:55 | |
| *** yingjun has quit IRC | 04:58 | |
| *** _gryf|afk is now known as _gryf | 04:59 | |
| *** lyanchih has joined #openstack-nova | 05:01 | |
| *** ijw has joined #openstack-nova | 05:01 | |
| *** lpetrut has joined #openstack-nova | 05:03 | |
| *** ijw_ has joined #openstack-nova | 05:05 | |
| *** coolsvap|brb is now known as coolsvap | 05:06 | |
| *** Marga_ has joined #openstack-nova | 05:07 | |
| *** Marga_ has quit IRC | 05:07 | |
| *** ijw has quit IRC | 05:07 | |
| *** Marga_ has joined #openstack-nova | 05:08 | |
| *** pradipta has quit IRC | 05:08 | |
| *** ijw_ has quit IRC | 05:10 | |
| *** masayukig_ has joined #openstack-nova | 05:16 | |
| *** lpetrut has quit IRC | 05:17 | |
| *** otter768 has joined #openstack-nova | 05:18 | |
| *** rotbeard has joined #openstack-nova | 05:20 | |
| *** lyanchih has quit IRC | 05:21 | |
| *** otter768 has quit IRC | 05:22 | |
| *** nkrinner has joined #openstack-nova | 05:23 | |
| *** Sukhdev has quit IRC | 05:24 | |
| *** shyu has joined #openstack-nova | 05:26 | |
| *** arnaud____ has joined #openstack-nova | 05:27 | |
| *** ankit_ag has joined #openstack-nova | 05:28 | |
| *** coolsvap is now known as coolsvap|away | 05:29 | |
| *** yingjun has joined #openstack-nova | 05:37 | |
| *** lpetrut has joined #openstack-nova | 05:39 | |
| *** Murali has joined #openstack-nova | 05:41 | |
| *** Sukhdev has joined #openstack-nova | 05:43 | |
| *** shuquan has joined #openstack-nova | 05:44 | |
| *** irenab has joined #openstack-nova | 05:45 | |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Remove unnecessary ValueError exception https://review.openstack.org/190525 | 05:45 |
| *** abhishekk has joined #openstack-nova | 05:47 | |
| *** arnaud____ has quit IRC | 05:50 | |
| *** lyanchih has joined #openstack-nova | 05:51 | |
| *** ig0r__ has joined #openstack-nova | 05:51 | |
| *** ig0r_ has quit IRC | 05:55 | |
| *** sudipto has joined #openstack-nova | 05:56 | |
| *** jichen has joined #openstack-nova | 06:00 | |
| *** lyanchih has quit IRC | 06:01 | |
| *** lyanchih has joined #openstack-nova | 06:02 | |
| *** Nikolay_St has quit IRC | 06:03 | |
| *** radek__ has joined #openstack-nova | 06:05 | |
| *** ijw has joined #openstack-nova | 06:05 | |
| *** pradipta has joined #openstack-nova | 06:06 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for quota_class_create/update https://review.openstack.org/160206 | 06:08 |
| openstackgerrit | Alex Xu proposed openstack/nova: Cleanup quota_class unittest with appropriate request context https://review.openstack.org/150351 | 06:08 |
| *** ijw_ has joined #openstack-nova | 06:08 | |
| *** Nikolay_St has joined #openstack-nova | 06:08 | |
| *** wanghua has quit IRC | 06:09 | |
| *** wanghua has joined #openstack-nova | 06:09 | |
| *** dims_ has joined #openstack-nova | 06:10 | |
| *** ijw has quit IRC | 06:11 | |
| *** ijw_ has quit IRC | 06:13 | |
| *** zhenguo has quit IRC | 06:13 | |
| *** dims_ has quit IRC | 06:15 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/197213 | 06:20 |
| *** baoli has joined #openstack-nova | 06:22 | |
| *** lyanchih has quit IRC | 06:22 | |
| *** moshele has joined #openstack-nova | 06:24 | |
| *** arnaud____ has joined #openstack-nova | 06:24 | |
| *** shuquan has quit IRC | 06:24 | |
| *** shuquan has joined #openstack-nova | 06:25 | |
| *** moshele has quit IRC | 06:29 | |
| *** moshele has joined #openstack-nova | 06:33 | |
| *** moshele has quit IRC | 06:35 | |
| *** ianbrown has quit IRC | 06:35 | |
| *** moshele has joined #openstack-nova | 06:35 | |
| *** scheuran has joined #openstack-nova | 06:35 | |
| *** tobe has joined #openstack-nova | 06:38 | |
| *** zhenguo has joined #openstack-nova | 06:41 | |
| *** ociuhandu has joined #openstack-nova | 06:42 | |
| *** lyanchih has joined #openstack-nova | 06:43 | |
| *** jwcroppe has joined #openstack-nova | 06:43 | |
| *** Sukhdev has quit IRC | 06:48 | |
| *** garyk has joined #openstack-nova | 06:51 | |
| openstackgerrit | garyk proposed openstack/nova: Cells: fix indentation for configuration variable declaration https://review.openstack.org/198615 | 06:52 |
| *** belmoreira has joined #openstack-nova | 06:53 | |
| *** exploreshaifali has joined #openstack-nova | 06:53 | |
| *** Sukhdev has joined #openstack-nova | 06:57 | |
| openstackgerrit | Eric Brown proposed openstack/nova: Replace openssl calls with cryptography lib https://review.openstack.org/198246 | 06:57 |
| *** moshele has quit IRC | 06:59 | |
| *** moshele has joined #openstack-nova | 06:59 | |
| openstackgerrit | Eric Brown proposed openstack/nova: Replace openssl calls with cryptography lib https://review.openstack.org/198246 | 07:00 |
| openstackgerrit | Eric Brown proposed openstack/nova: Replace openssl calls with cryptography lib https://review.openstack.org/198246 | 07:02 |
| *** pbandzi has joined #openstack-nova | 07:02 | |
| *** ijw has joined #openstack-nova | 07:04 | |
| *** browne has quit IRC | 07:04 | |
| *** wanghua has quit IRC | 07:05 | |
| *** wanghua has joined #openstack-nova | 07:05 | |
| *** tobe has quit IRC | 07:08 | |
| *** ijw has quit IRC | 07:09 | |
| *** kashyap has joined #openstack-nova | 07:12 | |
| *** kashyap has quit IRC | 07:13 | |
| *** achanda has joined #openstack-nova | 07:13 | |
| *** kashyap has joined #openstack-nova | 07:14 | |
| openstackgerrit | liaonanhai proposed openstack/nova: Add user_id info: add 'user_id' to REST os-simple-tenant-usage output https://review.openstack.org/151522 | 07:15 |
| *** Sukhdev has quit IRC | 07:15 | |
| *** ociuhandu has quit IRC | 07:16 | |
| *** wanghua has quit IRC | 07:17 | |
| *** wanghua has joined #openstack-nova | 07:18 | |
| *** mpaolino has joined #openstack-nova | 07:18 | |
| *** otter768 has joined #openstack-nova | 07:18 | |
| *** humble_ has joined #openstack-nova | 07:19 | |
| *** tobe has joined #openstack-nova | 07:20 | |
| *** humble_ has quit IRC | 07:22 | |
| *** wanghua has quit IRC | 07:22 | |
| *** wanghua has joined #openstack-nova | 07:23 | |
| *** lyanchih has quit IRC | 07:23 | |
| *** otter768 has quit IRC | 07:23 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: libvirt: mlnx_direct vif type removal https://review.openstack.org/181671 | 07:24 |
| *** humble_ has joined #openstack-nova | 07:24 | |
| *** lyanchih has joined #openstack-nova | 07:24 | |
| *** achanda has quit IRC | 07:25 | |
| *** mikedillion has joined #openstack-nova | 07:26 | |
| *** exploreshaifali has quit IRC | 07:26 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova-specs: Spec to add vif 'net-id' in virtual interfaces API https://review.openstack.org/198622 | 07:26 |
| *** arnaud____ has quit IRC | 07:27 | |
| *** mpavone has joined #openstack-nova | 07:28 | |
| *** wanghua has quit IRC | 07:28 | |
| *** humble_ has quit IRC | 07:29 | |
| *** mikedillion has left #openstack-nova | 07:29 | |
| *** humble_ has joined #openstack-nova | 07:29 | |
| *** markus_z has joined #openstack-nova | 07:29 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova-specs: Spec to add vif 'net-id' in virtual interfaces API https://review.openstack.org/198622 | 07:30 |
| *** ndipanov has joined #openstack-nova | 07:30 | |
| *** ndipanov has quit IRC | 07:32 | |
| *** ndipanov has joined #openstack-nova | 07:32 | |
| *** humble_ has quit IRC | 07:34 | |
| *** rdopiera has joined #openstack-nova | 07:34 | |
| *** humble_ has joined #openstack-nova | 07:34 | |
| *** humble_ has quit IRC | 07:35 | |
| *** humble_ has joined #openstack-nova | 07:36 | |
| *** achanda has joined #openstack-nova | 07:36 | |
| *** humble__ has joined #openstack-nova | 07:37 | |
| *** kaisers has joined #openstack-nova | 07:37 | |
| *** kaisers has quit IRC | 07:37 | |
| gmann | johnthetubaguy: alex_xu: oomichi: jaypipes: please have a look - https://review.openstack.org/#/c/198622/ | 07:38 |
| gmann | johnthetubaguy: alex_xu: oomichi: jaypipes: ^^ for v2.1 difference for vif net-id | 07:39 |
| *** humble__ has quit IRC | 07:39 | |
| *** wanghua has joined #openstack-nova | 07:40 | |
| *** humble_ has quit IRC | 07:40 | |
| alex_xu | gmann: I'm thinking we need a patch remove extension entry from v2.1 extension API, and we need back-port that patch | 07:42 |
| *** ildikov has joined #openstack-nova | 07:42 | |
| *** humble_ has joined #openstack-nova | 07:43 | |
| *** wanghua has quit IRC | 07:46 | |
| gmann | alex_xu: yes. | 07:49 |
| gmann | alex_xu: in morning oomichi maid a point that if we do that then this will also be change in v2.1 base? even backward compatible | 07:50 |
| gmann | alex_xu: :) what you say? | 07:50 |
| *** annashen has joined #openstack-nova | 07:50 | |
| *** sahid has joined #openstack-nova | 07:51 | |
| alex_xu | gmann: you mean remove extension entry from v2.1 base? | 07:51 |
| openstackgerrit | Andreas Scheuring proposed openstack/nova: Add macvtap as new virtual interface (vif) type to Nova's libvirt driver. https://review.openstack.org/182283 | 07:51 |
| gmann | alex_xu: yes | 07:51 |
| *** mutoulbj has quit IRC | 07:51 | |
| alex_xu | gmann: yes, we should remove from v2.1 base | 07:52 |
| wanghao | Guys, I need your help to review this spec: "Support force detach volume to nova"(https://review.openstack.org/#/c/84048/) | 07:52 |
| *** haypo has joined #openstack-nova | 07:52 | |
| alex_xu | gmann: but I think I didn't the point | 07:52 |
| wanghao | ping johnthetubaguy | 07:53 |
| gmann | alex_xu: i mean if we remove that extension from v2.1 extension list then it it will make extension list response different than v2 one | 07:53 |
| gmann | alex_xu: and if we do now as v2.1 is released then would not it be backward incompatible change for users using extension-list API? | 07:54 |
| *** annashen has quit IRC | 07:55 | |
| *** mutoulbj has joined #openstack-nova | 07:55 | |
| alex_xu | gmann: that extension list is configurable in the v2. so clients shouldn't expect one extension is existed in all the openstack deployment. | 07:55 |
| wanghao | johnthetubaguy: Hi~ john, I have updated the design for "Support force detach volume to nova" according reviewers' comments. Would you help me to check the design is ok? | 07:55 |
| *** lpetrut has quit IRC | 08:00 | |
| *** achanda has quit IRC | 08:02 | |
| *** achanda has joined #openstack-nova | 08:02 | |
| gmann | alex_xu: but cloud deployed with that extension would expect that extension to be returned from extension-list when they move from v2 to v2.1 | 08:03 |
| *** kaisers has joined #openstack-nova | 08:03 | |
| *** baoli has quit IRC | 08:04 | |
| *** kaisers has quit IRC | 08:04 | |
| *** ijw has joined #openstack-nova | 08:04 | |
| alex_xu | gmann: There should be a document to say 'v2.1 is equal to v2 except strct validation and without vif net-id extension' | 08:04 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: api：pass filters to method service_get_all https://review.openstack.org/198512 | 08:05 |
| alex_xu | gmann: and we can't just say we have this extension in the deployment, then user can't find that properties from the api, that will break the user. | 08:05 |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: libvirt: make live_migration_uri flag dependent on virt_type https://review.openstack.org/175780 | 08:05 |
| gmann | alex_xu: yea, agree. | 08:06 |
| gmann | alex_xu: that will be really false information for users | 08:06 |
| alex_xu | gmann: yea | 08:06 |
| gmann | alex_xu: but removing extension from extension-list in v2.1 base is also changing v2.1 base after release :) | 08:07 |
| alex_xu | gmann: emm...infinite loop~ | 08:08 |
| gmann | alex_xu: but we have to do that if we are not going to fix vif net-id in v2.1 base | 08:08 |
| gmann | alex_xu: heh | 08:08 |
| *** yamahata has quit IRC | 08:09 | |
| *** achanda has quit IRC | 08:09 | |
| *** ijw has quit IRC | 08:10 | |
| *** achanda has joined #openstack-nova | 08:10 | |
| *** lucasagomes has joined #openstack-nova | 08:11 | |
| gmann | johnthetubaguy: ^^ your thought on that? | 08:11 |
| gmann | alex_xu: do you know the right doc to mention about the this difference of v2 and v2.1? | 08:12 |
| *** exploreshaifali has joined #openstack-nova | 08:12 | |
| *** Marga_ has quit IRC | 08:13 | |
| alex_xu | gmann: just v2.1 api ref doc? | 08:13 |
| *** Marga_ has joined #openstack-nova | 08:14 | |
| *** lyarwood has quit IRC | 08:14 | |
| *** lyarwood has joined #openstack-nova | 08:14 | |
| *** nkrinner has quit IRC | 08:15 | |
| gmann | alex_xu: ok, Thanks | 08:15 |
| *** karimb has joined #openstack-nova | 08:15 | |
| *** jichen_ has joined #openstack-nova | 08:15 | |
| *** subscope has joined #openstack-nova | 08:17 | |
| gmann | alex_xu: anyways i will push patches to remove that extension from v2.1 list and see if any objection/suggestion | 08:17 |
| gmann | alex_xu: but i will be able to do that tomorrow. | 08:17 |
| alex_xu | gmann: yea, thanks, ust let see what happen~ | 08:17 |
| *** nkrinner has joined #openstack-nova | 08:18 | |
| *** jichen has quit IRC | 08:18 | |
| *** jichen_ is now known as jichen | 08:18 | |
| bauwser | morning Nova | 08:20 |
| bauwser | oops | 08:20 |
| *** bauwser is now known as bauzas | 08:20 | |
| *** shuquan has quit IRC | 08:20 | |
| *** shuquan has joined #openstack-nova | 08:21 | |
| ndipanov | mornin' | 08:22 |
| kashyap | Morning. | 08:24 |
| *** lyanchih has quit IRC | 08:25 | |
| *** lyanchih has joined #openstack-nova | 08:26 | |
| *** shyu has quit IRC | 08:27 | |
| alex_xu | afternoon~ | 08:29 |
| *** achanda has quit IRC | 08:29 | |
| *** masayukig_ has quit IRC | 08:33 | |
| *** masayukig_ has joined #openstack-nova | 08:33 | |
| *** moshele has quit IRC | 08:34 | |
| *** moshele has joined #openstack-nova | 08:37 | |
| *** yassine_ has joined #openstack-nova | 08:37 | |
| lxsli | Morning | 08:41 |
| johnthetubaguy | gmann: alex_xu: the extension list is not really part of v2.1, its most useful for the v2.0 compatibility, so I have less worries about that bit | 08:45 |
| johnthetubaguy | gmann: alex_xu: although I would feel better if we just removed the extension list for v2.1 in a new micro version, but that might be getting ahead of ourselves | 08:46 |
| johnthetubaguy | BobBall: did you see this: http://jogo.github.io/lastcomment/ | 08:49 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Fix checks of resize down of instance created with volume https://review.openstack.org/186643 | 08:50 |
| *** mgoddard has joined #openstack-nova | 08:55 | |
| tojuvone | would somebody have a word of wisdom about having mark-host-down visible in corresponding server instances: https://review.openstack.org/192246 | 08:56 |
| *** xyhuang has joined #openstack-nova | 08:59 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: libvirt: add get_cpu_frequency method https://review.openstack.org/197829 | 09:00 |
| *** ijw has joined #openstack-nova | 09:04 | |
| *** baoli has joined #openstack-nova | 09:04 | |
| *** baoli has quit IRC | 09:09 | |
| *** ijw has quit IRC | 09:09 | |
| *** jamielennox is now known as jamielennox|away | 09:11 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: api：pass filters to method service_get_all https://review.openstack.org/198512 | 09:13 |
| *** subscope has quit IRC | 09:14 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Added server tags support in nova-api https://review.openstack.org/128940 | 09:15 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Added method exists to the Tag object https://review.openstack.org/178103 | 09:15 |
| *** subscope has joined #openstack-nova | 09:15 | |
| alex_xu | johnthetubaguy: yes, the extesion list is all about switch v2 to v2.1 smooth, after user switch to v2.1, user shouldn't depend on that anymore. | 09:17 |
| sahid | mikal: johnthetubaguy any chance to get you approval on this spec ? https://review.openstack.org/#/c/139688/ (the code is already ready) and mikal already gave ot the spec +2 | 09:17 |
| *** jamielennox|away is now known as jamielennox | 09:18 | |
| johnthetubaguy | sahid: I will try do one last review of the the exceptions today | 09:19 |
| ankit_ag | Hi all, Can someone please help reviewing https://review.openstack.org/#/c/194063/ | 09:19 |
| *** otter768 has joined #openstack-nova | 09:19 | |
| openstackgerrit | Carlos Gonçalves proposed openstack/python-novaclient: Support forcing service down https://review.openstack.org/185849 | 09:20 |
| sahid | johnthetubaguy: thank you | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce GuestBlock to wrap around Block API https://review.openstack.org/190210 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: update swap volume to use Guest https://review.openstack.org/190211 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: rename GuestVCPUInfo to VCPUInfo https://review.openstack.org/194133 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: return whether a domain is persistent https://review.openstack.org/190962 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make unit tests concise by setup guest object https://review.openstack.org/190963 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to wait for block device job https://review.openstack.org/190214 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: update live snapshot to use Guest object https://review.openstack.org/190212 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: update volume delete snapshot to use Guest https://review.openstack.org/190213 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: save the memory state of guest https://review.openstack.org/196991 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make host return guest information https://review.openstack.org/191046 | 09:21 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make host return power state of guest https://review.openstack.org/191047 | 09:21 |
| sahid | johnthetubaguy: just a question, you gave to this spec +2 | 09:21 |
| sahid | https://review.openstack.org/#/c/165838/ | 09:21 |
| sahid | so now i need to find an other core to approve it before the end of the day? | 09:22 |
| *** sambetts has joined #openstack-nova | 09:24 | |
| *** otter768 has quit IRC | 09:24 | |
| johnthetubaguy | sahid: ideally, yes | 09:28 |
| sahid | johnthetubaguy: hum ack | 09:28 |
| rdopiera | eglynn-officeafk: is there anything in particular I should be reading or preparing, while I wait for the token? | 09:31 |
| *** lpetrut has joined #openstack-nova | 09:34 | |
| *** sb has quit IRC | 09:35 | |
| *** strictlyb has joined #openstack-nova | 09:36 | |
| *** yingjun has quit IRC | 09:36 | |
| *** ociuhandu has joined #openstack-nova | 09:46 | |
| *** lyanchih has quit IRC | 09:47 | |
| *** lyanchih has joined #openstack-nova | 09:49 | |
| *** yassine_ has quit IRC | 09:55 | |
| *** ildikov has quit IRC | 09:59 | |
| *** eglynn has joined #openstack-nova | 10:02 | |
| *** mutoulbj has quit IRC | 10:03 | |
| *** shuquan has quit IRC | 10:04 | |
| *** ijw has joined #openstack-nova | 10:04 | |
| *** shyu has joined #openstack-nova | 10:05 | |
| *** Kennan has quit IRC | 10:05 | |
| *** chaochin has quit IRC | 10:07 | |
| *** park__heijlong has quit IRC | 10:07 | |
| *** lan_ has quit IRC | 10:07 | |
| *** ijw has quit IRC | 10:10 | |
| *** tojuvone has quit IRC | 10:10 | |
| *** kaisers has joined #openstack-nova | 10:11 | |
| *** kaisers has quit IRC | 10:11 | |
| *** kaisers has joined #openstack-nova | 10:12 | |
| *** ildikov has joined #openstack-nova | 10:14 | |
| *** lan_ has joined #openstack-nova | 10:19 | |
| *** park__heijlong has joined #openstack-nova | 10:20 | |
| *** chaochin has joined #openstack-nova | 10:20 | |
| *** Kennan has joined #openstack-nova | 10:21 | |
| *** marzif_ has joined #openstack-nova | 10:33 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: libvirt: Always default device names at boot https://review.openstack.org/189632 | 10:34 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: virt: add get_device_name_for_instance to the base driver class https://review.openstack.org/180637 | 10:34 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: libvirt: implement get_device_name_for_instance https://review.openstack.org/180638 | 10:34 |
| *** mpaolino has quit IRC | 10:35 | |
| *** baoli has joined #openstack-nova | 10:36 | |
| *** lan__ has joined #openstack-nova | 10:37 | |
| *** park_heijlong has joined #openstack-nova | 10:37 | |
| *** mpaolino has joined #openstack-nova | 10:37 | |
| *** pixelbeat_ has joined #openstack-nova | 10:39 | |
| *** Kennan2 has joined #openstack-nova | 10:39 | |
| *** lan_ has quit IRC | 10:40 | |
| *** chaochin has quit IRC | 10:40 | |
| *** chaochin has joined #openstack-nova | 10:41 | |
| *** Kennan has quit IRC | 10:41 | |
| *** park__heijlong has quit IRC | 10:41 | |
| *** baoli has quit IRC | 10:41 | |
| *** Longgeek_ has quit IRC | 10:41 | |
| *** Longgeek has joined #openstack-nova | 10:42 | |
| *** pradipta has quit IRC | 10:44 | |
| *** mutoulbj has joined #openstack-nova | 10:46 | |
| *** dims_ has joined #openstack-nova | 10:47 | |
| *** Kennan2 is now known as Kennan | 10:48 | |
| *** dims_ has quit IRC | 10:52 | |
| *** haypo has quit IRC | 10:55 | |
| *** ociuhandu has quit IRC | 10:58 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: ensure that get_info raises the correct exception https://review.openstack.org/191558 | 11:01 |
| openstackgerrit | garyk proposed openstack/nova: VMware: treat deletion exception with attached volumes https://review.openstack.org/198674 | 11:01 |
| *** jcoufal has joined #openstack-nova | 11:01 | |
| *** tobe has quit IRC | 11:01 | |
| *** sudipto has quit IRC | 11:04 | |
| *** ijw has joined #openstack-nova | 11:04 | |
| *** ijw has quit IRC | 11:09 | |
| *** yassine_ has joined #openstack-nova | 11:11 | |
| *** yassine_ has quit IRC | 11:11 | |
| *** shuquan has joined #openstack-nova | 11:11 | |
| *** subscope has quit IRC | 11:12 | |
| *** shuquan has quit IRC | 11:16 | |
| *** Jeffrey4l has quit IRC | 11:17 | |
| *** derekh has joined #openstack-nova | 11:18 | |
| *** otter768 has joined #openstack-nova | 11:20 | |
| *** Marga_ has quit IRC | 11:24 | |
| *** otter768 has quit IRC | 11:25 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Fix checks of resize down of instance created with volume https://review.openstack.org/186643 | 11:27 |
| *** deepthi has quit IRC | 11:29 | |
| *** Murali has quit IRC | 11:30 | |
| *** ildikov has quit IRC | 11:31 | |
| *** IanGovett has joined #openstack-nova | 11:33 | |
| andreykurilin | dansmith: hi! | 11:34 |
| johnthetubaguy | alaski: I would love to catch up with you about this spec at somepoint please: https://review.openstack.org/#/c/135387/13/specs/liberty/approved/improve-unshelve-performance.rst | 11:35 |
| * johnthetubaguy goes for some lunch... | 11:35 | |
| *** subscope has joined #openstack-nova | 11:36 | |
| *** baoli has joined #openstack-nova | 11:37 | |
| *** ociuhandu has joined #openstack-nova | 11:37 | |
| *** Marga_ has joined #openstack-nova | 11:38 | |
| *** bfic has joined #openstack-nova | 11:38 | |
| *** bfic has quit IRC | 11:39 | |
| *** baoli_ has joined #openstack-nova | 11:40 | |
| *** bfic has joined #openstack-nova | 11:40 | |
| *** baoli has quit IRC | 11:42 | |
| *** thorst has joined #openstack-nova | 11:42 | |
| *** ildikov has joined #openstack-nova | 11:44 | |
| *** baoli_ has quit IRC | 11:45 | |
| *** stefan_amann has joined #openstack-nova | 11:48 | |
| *** Kevin_Zheng has quit IRC | 11:50 | |
| *** moshele has quit IRC | 11:51 | |
| *** baoli has joined #openstack-nova | 11:51 | |
| *** setuid has quit IRC | 11:54 | |
| stefan_amann | johnthetubaguy: John, is there anything I need to do for https://review.openstack.org/#/c/194163/? You seem to be in agreement with Matt that we only need a bugfix, not a spec. I released the code changes under https://review.openstack.org/#/c/192556/ some time back. Is there anything else I need to do (i.e., change the commit message, open a bug, ...)? | 11:55 |
| *** ajayaa has joined #openstack-nova | 11:55 | |
| *** baoli_ has joined #openstack-nova | 11:55 | |
| *** baoli has quit IRC | 11:58 | |
| *** bradjones has joined #openstack-nova | 12:01 | |
| *** bradjones has quit IRC | 12:01 | |
| *** bradjones has joined #openstack-nova | 12:01 | |
| *** setuid has joined #openstack-nova | 12:01 | |
| bauzas | stefan_amann: john is afk for the moment, but it seems the consensus is just that you need to provide a spec | 12:02 |
| bauzas | ergh | 12:02 |
| bauzas | to *not* provide a spec | 12:02 |
| bauzas | and instead, just work on your implementation patch | 12:02 |
| *** piyanai has joined #openstack-nova | 12:02 | |
| stefan_amann | ok.. | 12:02 |
| bauzas | stefan_amann: just wait for johnthetubaguy to be back so he could remove his -2 from https://review.openstack.org/#/c/192556/ | 12:03 |
| stefan_amann | ok.. thanks! | 12:03 |
| bauzas | stefan_amann: in the meantime, you could just create either a spec-less blueprint (ie. only a Launchpad BP) or a Launchpad bug | 12:04 |
| *** dims_ has joined #openstack-nova | 12:04 | |
| *** abhishekk has quit IRC | 12:04 | |
| *** ijw has joined #openstack-nova | 12:04 | |
| bauzas | stefan_amann: and change the commit msg to provide the linkl | 12:04 |
| stefan_amann | ok.. will check later. Am in a meeting right now. Thanks | 12:05 |
| *** pask81 has joined #openstack-nova | 12:06 | |
| *** dims_ has quit IRC | 12:08 | |
| *** bkopilov has quit IRC | 12:09 | |
| *** dave-mccowan has joined #openstack-nova | 12:10 | |
| *** ijw has quit IRC | 12:10 | |
| *** lucasagomes is now known as lucas-hungry | 12:11 | |
| *** leakypipes is now known as jaypipes | 12:12 | |
| * jaypipes caffeinates | 12:12 | |
| *** haypo has joined #openstack-nova | 12:13 | |
| *** raildo has joined #openstack-nova | 12:15 | |
| *** Nikolay_St has quit IRC | 12:16 | |
| johnthetubaguy | stefan_amann: can you please remove the blueprint from the commit message, and link it to a bug that has all the details, and I can remove the -2 | 12:21 |
| *** vladikr has joined #openstack-nova | 12:21 | |
| rgerganov | jaypipes: hi, could you please bless https://review.openstack.org/#/c/148509/ ? | 12:21 |
| *** ildikov has quit IRC | 12:22 | |
| *** moshele has joined #openstack-nova | 12:22 | |
| rgerganov | jaypipes: thanks a lot! | 12:23 |
| openstackgerrit | John Garbutt proposed openstack/nova: Support host type specific block volume attachment https://review.openstack.org/192556 | 12:23 |
| jaypipes | rgerganov: done :) | 12:23 |
| *** edmondsw has joined #openstack-nova | 12:24 | |
| *** jcoufal has quit IRC | 12:24 | |
| *** shyu has quit IRC | 12:26 | |
| *** jcoufal has joined #openstack-nova | 12:27 | |
| *** aysyd has joined #openstack-nova | 12:29 | |
| *** xyhuang has quit IRC | 12:31 | |
| *** lyanchih has quit IRC | 12:31 | |
| sahid | dansmith: any chance you take a look at this https://review.openstack.org/#/c/165838/ it's the last day for approval and you already reviewed it | 12:35 |
| sahid | some comments has been done but they wll be discussed on the code reviews, then i will probably update it but should be great to have it approved if the general idea seems good for you :) | 12:36 |
| *** jasondotstar has joined #openstack-nova | 12:38 | |
| *** rfolco has joined #openstack-nova | 12:38 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Fix checks of resize down of instance created with volume https://review.openstack.org/186643 | 12:38 |
| *** ajayaa has quit IRC | 12:38 | |
| *** Murali has joined #openstack-nova | 12:39 | |
| *** ildikov has joined #openstack-nova | 12:39 | |
| openstackgerrit | garyk proposed openstack/nova: Flavors: use list comprehension for get_all_flavors https://review.openstack.org/198702 | 12:43 |
| *** pask81 has quit IRC | 12:43 | |
| *** mjturek1 has quit IRC | 12:44 | |
| stefan_amann | johnthetubaguy: I will do. Thanks | 12:46 |
| *** jecarey has quit IRC | 12:46 | |
| sahid | johnthetubaguy: thx for your review on rt ! | 12:47 |
| sahid | i'm sure jaypipes will be our second +2+W :p | 12:47 |
| sahid | on https://review.openstack.org/#/c/139688/ | 12:47 |
| *** mjturek1 has joined #openstack-nova | 12:48 | |
| *** jcoufal has quit IRC | 12:49 | |
| *** exploreshaifali has quit IRC | 12:49 | |
| *** exploreshaifali has joined #openstack-nova | 12:51 | |
| *** ankit_ag has quit IRC | 12:51 | |
| *** pask81 has joined #openstack-nova | 12:53 | |
| *** ZZelle has quit IRC | 12:53 | |
| jaypipes | sahid: I am not a nova-specs-core, so I can't +2 any spec. will add comments, though. | 12:53 |
| *** dims_ has joined #openstack-nova | 12:55 | |
| *** ijuwang has joined #openstack-nova | 12:56 | |
| dansmith | sahid: okay, it's first in the tab queue.. right behind coffee | 12:57 |
| dansmith | andreykurilin: o/ | 12:57 |
| sahid | jaypipes: ah... in all cases your reviews is still always valuable. thanks | 12:57 |
| sahid | s/is/are | 12:57 |
| *** dims__ has joined #openstack-nova | 12:57 | |
| *** exploreshaifali has quit IRC | 13:00 | |
| *** dims___ has joined #openstack-nova | 13:00 | |
| *** dims_ has quit IRC | 13:00 | |
| andreykurilin | dansmith: hi! First of all, thank you for +2 at my spec:) | 13:01 |
| *** dims_ has joined #openstack-nova | 13:01 | |
| *** doude has quit IRC | 13:01 | |
| andreykurilin | dansmith: What about +2 at my patches? :) | 13:02 |
| dansmith | andreykurilin: ping me later today and I can look | 13:02 |
| *** doude has joined #openstack-nova | 13:02 | |
| *** doude has quit IRC | 13:02 | |
| *** doude has joined #openstack-nova | 13:02 | |
| dansmith | andreykurilin: I have a bunch of things stacked up from the holiday weekend to work through first | 13:02 |
| andreykurilin | dansmith: ok, thanks | 13:02 |
| *** mdrabe has joined #openstack-nova | 13:03 | |
| *** dims__ has quit IRC | 13:03 | |
| *** ijw has joined #openstack-nova | 13:04 | |
| *** yingjun has joined #openstack-nova | 13:05 | |
| *** dims___ has quit IRC | 13:05 | |
| *** lyarwood has quit IRC | 13:05 | |
| *** lyarwood has joined #openstack-nova | 13:05 | |
| *** rlrossit has joined #openstack-nova | 13:05 | |
| *** nihilifer has quit IRC | 13:05 | |
| lxsli | bauzas: sched meeting? in alt? | 13:06 |
| *** erlon has joined #openstack-nova | 13:06 | |
| *** n0ano has joined #openstack-nova | 13:06 | |
| kashyap | Can anyone take a look at this useful Kilo backport? -- https://review.openstack.org/#/c/193497/2 | 13:06 |
| bauzas | lxsli: meh, I have 1400UTC ? | 13:06 |
| *** dims__ has joined #openstack-nova | 13:06 | |
| bauzas | lxsli: which is in one hour IMHO | 13:07 |
| lxsli | bauzas: right yes, in an hour then :) | 13:07 |
| lxsli | jaypipes: edleafe ^^ | 13:07 |
| bauzas | lxsli: I know you're very impatient | 13:07 |
| *** ameade_vacation is now known as ameade | 13:08 | |
| *** Nikolay_St has joined #openstack-nova | 13:08 | |
| *** yassine_ has joined #openstack-nova | 13:08 | |
| lxsli | bauzas: Just can't wait | 13:08 |
| *** nelsnelson has joined #openstack-nova | 13:08 | |
| *** mmedvede_ has joined #openstack-nova | 13:09 | |
| jaypipes | lxsli, edleafe: oops, sorry. | 13:09 |
| bauzas | http://eavesdrop.openstack.org/meetings/nova_scheduler/2015/nova_scheduler.2015-06-30-14.59.html | 13:09 |
| bauzas | baaam | 13:09 |
| *** xyang1 has joined #openstack-nova | 13:09 | |
| *** dims_ has quit IRC | 13:09 | |
| bauzas | one hour more to fix my 500+ unittests failing due to ReqSpec, yay \o/ | 13:09 |
| *** ijw has quit IRC | 13:10 | |
| jaypipes | lol :) | 13:10 |
| *** lucas-hungry is now known as lucasagomes | 13:10 | |
| * bauzas caffeinates | 13:10 | |
| * bauzas grumbles something about dictionaries and keys | 13:10 | |
| *** dims__ has quit IRC | 13:11 | |
| *** ivasev has joined #openstack-nova | 13:12 | |
| *** irenab has quit IRC | 13:13 | |
| *** ferest has joined #openstack-nova | 13:17 | |
| *** rotbeard has quit IRC | 13:18 | |
| *** Nikolay_St has quit IRC | 13:19 | |
| *** ferest has quit IRC | 13:20 | |
| *** _gryf is now known as _gryf|afk | 13:21 | |
| *** otter768 has joined #openstack-nova | 13:21 | |
| *** otter768 has quit IRC | 13:26 | |
| sahid | dansmith: whaoo thanks | 13:27 |
| *** jskinner has joined #openstack-nova | 13:27 | |
| dansmith | s/whaoo/wahoo/ :) | 13:27 |
| sahid | even for something like that :/ | 13:28 |
| sahid | ;) | 13:28 |
| *** sudipto has joined #openstack-nova | 13:28 | |
| dansmith | andreykurilin: okay, link? | 13:28 |
| andreykurilin | dansmith: https://review.openstack.org/#/c/152569/ - this is the first patch in the chain | 13:28 |
| dansmith | oh, that | 13:29 |
| dansmith | I'm probably not a good person to review that stuff | 13:29 |
| andreykurilin | :( | 13:29 |
| dansmith | andreykurilin: you should ping melwitt when she gets in | 13:29 |
| dansmith | but I bet she'll jump on it | 13:29 |
| andreykurilin | dansmith: thank you anyway | 13:30 |
| *** bkopilov has joined #openstack-nova | 13:30 | |
| andreykurilin | melwitt: ping. I need your opinion:) | 13:30 |
| dansmith | she is more of real west coaster, so probably not around for a couple of hours, FYI | 13:30 |
| andreykurilin | dansmith: got it. thanks | 13:31 |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: rebuild: fix rebuild of server with volume attached https://review.openstack.org/176891 | 13:31 |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: rebuild: make sure server is shut down before volumes are detached https://review.openstack.org/176892 | 13:31 |
| *** browne has joined #openstack-nova | 13:33 | |
| jaypipes | dansmith, alaski: https://review.openstack.org/#/c/196180 could use your reviews if you get a chance. thx in advance. | 13:34 |
| *** claudiub has joined #openstack-nova | 13:34 | |
| openstackgerrit | PrashantUpadhyay proposed openstack/nova: Check for volume size https://review.openstack.org/198714 | 13:35 |
| *** pbandzi has quit IRC | 13:35 | |
| *** ayoung has joined #openstack-nova | 13:35 | |
| *** exploreshaifali has joined #openstack-nova | 13:37 | |
| dansmith | jaypipes: found a typo, but +2 otherwise | 13:37 |
| jaypipes | dansmith: cheers | 13:37 |
| dansmith | jaypipes: looks like I can review the next patch aside from a test you need to re-add, yes? | 13:38 |
| jaypipes | dansmith: yeah, gimme one minute. | 13:38 |
| *** claudiub has quit IRC | 13:38 | |
| alaski | will look in just a bit | 13:38 |
| dansmith | okay | 13:38 |
| *** mriedem has joined #openstack-nova | 13:39 | |
| ayoung | jaypipes, am I correct in understanding that nova-compute, that makes the calls to other services like Glance etc, runs on the the hypervisor machines? | 13:39 |
| mriedem | ayoung: yes | 13:39 |
| ayoung | mriedem, and thus the hypervisor actually has Keystone tokens on it, right? | 13:40 |
| *** eharney has joined #openstack-nova | 13:40 | |
| *** zzzeek has joined #openstack-nova | 13:40 | |
| mriedem | ayoung: well, things like the vcenter driver are different | 13:40 |
| *** promulo__ has quit IRC | 13:40 | |
| mriedem | nova-compute is running on a host talking to a remote vcenter cluster | 13:41 |
| ayoung | mriedem, right, because those act as concentrators | 13:41 |
| openstackgerrit | PrashantUpadhyay proposed openstack/nova: Check for volume size https://review.openstack.org/198714 | 13:41 |
| *** ajayaa has joined #openstack-nova | 13:41 | |
| *** whenry_ has quit IRC | 13:41 | |
| ayoung | mriedem, but for Xen and for libvirt, the expecation is that the compute host is the hypervbisor, and that is the dominant way that openstack is deployed? | 13:41 |
| *** apuimedo has joined #openstack-nova | 13:41 | |
| mriedem | ayoung: yeah | 13:42 |
| ayoung | mriedem, aside from the initial boot calls, and the teardown calls, what operations do we expect to talk to remote service? | 13:42 |
| garyk | ayoung: what is a concentrator? | 13:43 |
| dansmith | attach/detach of things | 13:43 |
| dansmith | periodic refresh of network stuff goes to neutron | 13:43 |
| ayoung | garyk, a single machine that exposes access to many machines...a facade | 13:43 |
| *** Longgeek has quit IRC | 13:43 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Rework monitor plugin interface and API https://review.openstack.org/196181 | 13:43 |
| garyk | ok | 13:43 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Use stevedore for loading monitor extensions https://review.openstack.org/196182 | 13:43 |
| jaypipes | dansmith: k, all done. | 13:43 |
| ayoung | dansmith, is there the same "periodic refresh" to Cinder or any other service, or is that unique to Neutron? | 13:44 |
| dansmith | jaypipes: except the typo right? | 13:44 |
| jaypipes | dansmith: crap. sec. | 13:44 |
| dansmith | ayoung: just neutron, but when we attach/detach volumes we hit cinder from the compute nodes | 13:44 |
| ayoung | dansmith, how about snapshot? Is that kicked off by Nova? | 13:45 |
| garyk | ayoung: yes | 13:45 |
| dansmith | yup | 13:45 |
| ayoung | garyk, dansmith mriedem Are there any other cross-service calls you can think of ? | 13:47 |
| dansmith | lots of glance interaction | 13:47 |
| ayoung | Notifications don't need a token, so there should be no direct Ceilometer calls | 13:47 |
| dansmith | nova talks to barbican | 13:47 |
| ayoung | dansmith, thanks | 13:47 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Adds MonitorMetric object https://review.openstack.org/196180 | 13:47 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Rework monitor plugin interface and API https://review.openstack.org/196181 | 13:47 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Use stevedore for loading monitor extensions https://review.openstack.org/196182 | 13:47 |
| garyk | ayoung: nova talks with galnce, cidner and neutron at the moment | 13:47 |
| jaypipes | dansmith: ok, NOW done. | 13:47 |
| kashyap | mriedem: Nag-round-3: When you get a moment :-) -- https://review.openstack.org/#/c/193497/2 -- libvirt: handle code=38 + sigkill (ebusy) in destroy() | 13:48 |
| garyk | dansmith: out of interst where does it talk with barbican? | 13:48 |
| dansmith | garyk: encrypted LVM, right? | 13:48 |
| garyk | dansmith: is that via a service? | 13:48 |
| ayoung | dansmith, during boot, Nova fetches from glance. During Snapshot, Nova posts to glance. When else does Nova talk to glance? | 13:48 |
| *** Marga_ has quit IRC | 13:48 | |
| dansmith | ayoung: during resize, probably during image cache management runs.. lots of times | 13:49 |
| alex_xu | jaypipes: maybe my comment's problem, should we add 'test_get_host_metrics_one_failed' back also in https://review.openstack.org/#/c/196181/8/nova/tests/unit/compute/test_resource_tracker.py | 13:49 |
| dansmith | ayoung: why are you asking? | 13:49 |
| jaypipes | alex_xu: no. it's the same code path as the exception one. | 13:49 |
| ayoung | dansmith, I'm trying to identify the cases where a token needs to talk to multiple services. Its part of the dynamic policy work. | 13:49 |
| mriedem | garyk: nova.volume.encryptors.base talks to nova.keymgr which has a barbican backend | 13:49 |
| mriedem | which talks to barbicanclient | 13:49 |
| garyk | ok, thanks | 13:50 |
| dansmith | ayoung: okay, but presumably you'll need to do some mapping of the actual calls and not just trusting us to remember all the places, right? | 13:50 |
| ayoung | Ideally, we would split up the permissions on a token to only have the access required for a specific use case. I want to analyze how practical/painful it would be to do this. | 13:50 |
| ayoung | dansmith, 100% | 13:50 |
| dansmith | okay | 13:50 |
| ayoung | dansmith, right now, we are in the design phase. I just don't want to lie when briefing other developers | 13:50 |
| garyk | ayoung: one of the points to take into account is under stress - we have seen tons of failures when booting instances and tokens expiring | 13:51 |
| ayoung | dansmith, I had the idea of running a demo cluster with policy in the "let everything through, but log what would fail" mode to check this in an automated way...but I'm not there yet | 13:51 |
| garyk | there was work in juno to cache the tokesn but that did not solve things entirely | 13:51 |
| *** marun has joined #openstack-nova | 13:51 | |
| *** dboik has joined #openstack-nova | 13:52 | |
| ayoung | garyk, yeah, that is one thing I would like to address. It has to do with long lived processes. Downloads of images taking a while, as I recall? | 13:52 |
| dansmith | ayoung: yeah, well, that's good, but you'll miss things that don't always happen, like I think the ceph driver ends up talking to glance more often than regular storage backends, etc | 13:52 |
| alex_xu | jaypipes: yea, it's ture. but it can ensure the correct monitor's metrics was collected. | 13:53 |
| *** marun has left #openstack-nova | 13:53 | |
| ayoung | dansmith, right. I also want to make it easier for service users to get tokens on behalf of users to perform long lived tasks like these, but that is also more secure if those tokens are very limited in what they can perform. | 13:53 |
| alex_xu | jaypipes: but it's fine for me, not very critical. | 13:53 |
| garyk | ayoung: at times there are a number of tokens in play - an admin one and a user one - for example when creating a neutron port | 13:54 |
| ayoung | garyk, that is the "Service Token + User Token" use case...we only want things like that coming from Nova? | 13:55 |
| openstackgerrit | Eric Brown proposed openstack/nova: Replace openssl calls with cryptography lib https://review.openstack.org/198246 | 13:55 |
| *** irenab has joined #openstack-nova | 13:55 | |
| garyk | ayoung: not sure i understand the question? | 13:56 |
| *** r-daneel has joined #openstack-nova | 13:56 | |
| ayoung | garyk, just confirming my assumption as far as what you mean. | 13:56 |
| garyk | in addition to that each time an instance is spun up there are the token used for the metadata service - that is this is done between the neutron agents and the service to determine the instance id | 13:56 |
| openstackgerrit | Stefan Amann proposed openstack/nova: Support host type specific block volume attachment https://review.openstack.org/192556 | 13:56 |
| garyk | ayoung: ok. understand. then you are correct | 13:56 |
| *** Piet has quit IRC | 13:57 | |
| *** lyarwood has quit IRC | 13:57 | |
| *** lyarwood has joined #openstack-nova | 13:58 | |
| *** irenab has quit IRC | 13:58 | |
| jaypipes | johnthetubaguy: any chance we can get https://review.openstack.org/#/c/179224/ approved today? It's bauzas' spec on ensuring live migration gets destination hosts checked. | 13:58 |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Change List objects to use obj_relationships https://review.openstack.org/198730 | 13:58 |
| *** Marga_ has joined #openstack-nova | 13:58 | |
| *** irenab has joined #openstack-nova | 13:58 | |
| bauzas | jaypipes: I'm not that worried, it's a prio spec | 13:58 |
| *** tongli has joined #openstack-nova | 13:59 | |
| bauzas | jaypipes: so we can just review the exceptions today | 13:59 |
| bauzas | jaypipes: and then discuss on the prio specs by tomorrow or later | 13:59 |
| bauzas | jaypipes: but thanks anyway for the lights :) | 13:59 |
| jaypipes | yuyp | 14:00 |
| sahid | dansmith: any chance you look at this too https://review.openstack.org/#/c/139688/ this has been already +2ed by mikal then i fixed some typos then johnthetubaguy +2ed it - im scary we miss the feature freeze by today | 14:00 |
| sahid | (promise the last one) | 14:00 |
| *** xyhuang has joined #openstack-nova | 14:00 | |
| *** ctrath has joined #openstack-nova | 14:01 | |
| sahid | s/scary/scared | 14:01 |
| *** jecarey_ has joined #openstack-nova | 14:02 | |
| *** amotoki has joined #openstack-nova | 14:04 | |
| *** ijw has joined #openstack-nova | 14:04 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Make instance tags soft deletable https://review.openstack.org/198733 | 14:04 |
| *** apmelton has joined #openstack-nova | 14:04 | |
| *** humble_ has quit IRC | 14:06 | |
| * mriedem envisions the day when spec freeze and FFE is no longer a thing | 14:07 | |
| *** r-daneel has quit IRC | 14:07 | |
| mriedem | there are a buttload of cinder cores +1 on this https://review.openstack.org/#/c/197713/ rootwrap fix to make some other cinder backends (not just iscsi) work with encrypted volumes | 14:08 |
| garyk | mriedem: it is just a matter of everyone aggreeing and it could even be today :) | 14:08 |
| *** amotoki has quit IRC | 14:08 | |
| mriedem | by definition, buttload >= 5 | 14:08 |
| *** ijw has quit IRC | 14:09 | |
| *** Murali has quit IRC | 14:11 | |
| mriedem | kashyap: +2 | 14:17 |
| *** logan2 has quit IRC | 14:17 | |
| kashyap | mriedem: Thanks. Note to self: Wait for the +W this week, or pester again here if not. :-) | 14:19 |
| mriedem | kashyap: i lobbied in the -stable channel | 14:19 |
| kashyap | markus_z: Hmm, I think your serial console spec was harmless enough to be let in for Lemming^W^W^W^WLiberty. But oh well. . . | 14:19 |
| *** Piet has joined #openstack-nova | 14:20 | |
| kashyap | mriedem: Ah, thanks. | 14:20 |
| *** r-daneel has joined #openstack-nova | 14:20 | |
| markus_z | kashyap: Yes, I think so too. It started with a specless blueprint. I'm unsure how to resolve this. | 14:21 |
| garyk | kashyap: done | 14:21 |
| openstackgerrit | Deepak C Shetty (deepakcs) proposed openstack/nova: Nova part of volume snapshot improvements https://review.openstack.org/172813 | 14:21 |
| openstackgerrit | Deepak C Shetty (deepakcs) proposed openstack/nova: objects: Add support for OnlineSnapshotStatus object https://review.openstack.org/194538 | 14:21 |
| kashyap | garyk: Nice. | 14:22 |
| *** cfriesen has joined #openstack-nova | 14:22 | |
| kashyap | markus_z: I don't think much can be done, as I read in an email signature: "error compiling committee.c: too many arguments to function" :-) | 14:22 |
| *** tonytan4ever has joined #openstack-nova | 14:24 | |
| markus_z | kashyap: Did I understand it wrong, that some blueprints don't need a spec? | 14:24 |
| kashyap | markus_z: I'm not quite sure where people draw the line, I think it's mostly "apply common sense". I can't find the wiki page that documents some guidelines. | 14:25 |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Change List objects to use obj_relationships https://review.openstack.org/198730 | 14:25 |
| ndipanov | harlowja, would appreciate a comment on this https://review.openstack.org/#/c/197668/ | 14:26 |
| jaypipes | sahid: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:refactor-libvirt,n,z reviewed. | 14:26 |
| jaypipes | cores: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:refactor-libvirt,n,z looks to be pretty good, IMHO. would be cool to have some reviews on it. it's nice cleanup code. | 14:27 |
| sahid | jaypipes: wahoo thank you for these reviews ! | 14:27 |
| sahid | let me address your request | 14:27 |
| markus_z | kashyap: The README from the nova-specs repo: https://github.com/openstack/nova-specs "Trivial specifications" contains it. | 14:28 |
| mgoddard | hi, is there anyone with neutron API plugin knowledge who could comment on https://review.openstack.org/#/c/194597/? Thanks | 14:28 |
| jaypipes | sahid: np | 14:28 |
| mgoddard | sorry, bad URL. Should have been https://review.openstack.org/#/c/194597 | 14:30 |
| *** jeblair_ is now known as jeblair | 14:30 | |
| sudipto | dansmith, I am working a metric value object on top of jaypipes 's changes like here: http://paste.openstack.org/show/348700/ | 14:31 |
| kashyap | markus_z: Who asked to create a spec for that? | 14:31 |
| sudipto | dansmith, the approach we are taking here is - the monitor value can be expressed either as a simple Integer Value or a dict - that represents the NUMA topology like {0:10, 1:15} etc | 14:32 |
| *** marzif_ has quit IRC | 14:32 | |
| *** nkrinner has quit IRC | 14:33 | |
| markus_z | johnthetubaguy: bp libvirt-log-file-and-serial-console was approved as spec-less blueprint but you asked for a spec too. The spec got hit by spec freeze. The code changes are 24LoC. Maybe you can have a second look? | 14:33 |
| *** marzif_ has joined #openstack-nova | 14:33 | |
| sudipto | dansmith, and we want to say - that either of this fields should be set for it work - however, this is what happens: http://paste.openstack.org/show/348697/ | 14:33 |
| kashyap | markus_z: Your BP satisfies the first 2 sentences in the README above (i.e. no need for a spec, fwiw). But, whatever. . . | 14:33 |
| dansmith | sudipto: didn't jaypipes define the value field as just a string (on purpose)? | 14:33 |
| johnthetubaguy | markus_z: it got approved as a spec less, interesting, you got a link to the blueprint there? | 14:33 |
| *** beagles is now known as beagles_brb | 14:34 | |
| markus_z | johnthetubaguy: https://blueprints.launchpad.net/nova/+spec/libvirt-log-file-and-serial-console | 14:34 |
| *** bfic has left #openstack-nova | 14:34 | |
| johnthetubaguy | markus_z: sorry, yeah, didn't spot you gave me most of that already | 14:34 |
| kashyap | mriedem: -stable isn't much active at all, is it? Saying this after lurking for only a couple of weeks, so I may be mostly wrong. | 14:34 |
| mriedem | kashyap: it's a broadcast channel | 14:34 |
| johnthetubaguy | markus_z: so you should be good to go, as we approve the blueprint, because of that you don't need the spec | 14:35 |
| sudipto | dansmith, you mean the MonitorMetric value field? That's set as an Integer right? | 14:35 |
| *** stefan_amann has quit IRC | 14:35 | |
| markus_z | johnthetubaguy: Shall I abandon the spec and go on with the spec-less bp? | 14:35 |
| dansmith | sudipto: ah, yeah, just looked | 14:36 |
| *** stefan_amann has joined #openstack-nova | 14:36 | |
| johnthetubaguy | markus_z: yes, that what we want to happen when we approve the blueprint without a spec | 14:36 |
| johnthetubaguy | markus_z: I am trying to get this all written down, so it much clearer! | 14:36 |
| johnthetubaguy | s/it/its/ | 14:36 |
| markus_z | johnthetubaguy: Okay, cool, thank you! | 14:37 |
| johnthetubaguy | markus_z: no problems | 14:37 |
| *** moshele has quit IRC | 14:38 | |
| johnthetubaguy | markus_z: is all the code up for review now? | 14:38 |
| kashyap | mriedem: Hmm, sometimes it's hard to read you. Whether you're telling the truth or just being flippant :-) | 14:38 |
| sudipto | dansmith, basically - we are being forced to initialize both the fields per that unit tests file i shared...else gives an importerror for the uninitialized field. So I was wondering, how we could make the monitors just initialize either of these fields (integer_value OR numa_nodes_mem_bw_value) | 14:38 |
| markus_z | johnthetubaguy: Yes, it's https://review.openstack.org/#/c/188058/ | 14:38 |
| johnthetubaguy | markus_z: thats what I though, can you please change the blueprint state to NeedsCodeReview so everyone knows you have all your code up now | 14:39 |
| sudipto | dansmith, have tried setting default=None for both the fields - but that didn't work well... :( | 14:39 |
| dansmith | sudipto: okay, well, I'm focused on something right now, so I can't really concentrate at the moment | 14:39 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: removed unused method _get_default_deleted_value https://review.openstack.org/198746 | 14:39 |
| markus_z | johnthetubaguy: Done (I thought I did that already, but I must have missed that, sorry) | 14:40 |
| sudipto | dansmith, ok np. | 14:40 |
| johnthetubaguy | markus_z: no worries, thats cool, thank you! | 14:40 |
| mriedem | kashyap: it's a cast kind of channel as far as i've seen - i show up and dump a message and maybe someone will read it hours later | 14:41 |
| *** logan2 has joined #openstack-nova | 14:41 | |
| kashyap | mriedem: Yep, noted. | 14:43 |
| *** Longgeek has joined #openstack-nova | 14:43 | |
| *** otter768 has joined #openstack-nova | 14:45 | |
| *** Longgeek has quit IRC | 14:49 | |
| kashyap | johnthetubaguy: Cool - you unblocked that serial console spec from markus_z. I was confused too, why it needed a spec. | 14:49 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Replace stubs with mocks for test_dmcrypt https://review.openstack.org/198752 | 14:50 |
| *** moshele has joined #openstack-nova | 14:50 | |
| johnthetubaguy | kashyap: it didn't need one, it got approved a little while back, as the blueprint was raised as a specless blueprint in the nova-meeting, just seems the spec review still happened, for some reason | 14:50 |
| kashyap | johnthetubaguy: Yep, saw your comment there. | 14:51 |
| openstackgerrit | jichenjc proposed openstack/nova: Catch invalid id input in service_delete https://review.openstack.org/198753 | 14:51 |
| johnthetubaguy | cool | 14:51 |
| *** thangp has joined #openstack-nova | 14:52 | |
| *** xyhuang has quit IRC | 14:52 | |
| *** mgoddard has quit IRC | 14:53 | |
| *** xyhuang has joined #openstack-nova | 14:53 | |
| kashyap | garyk: Thanks for +Wing that change. | 14:53 |
| *** mgoddard has joined #openstack-nova | 14:53 | |
| bauzas | does someone know how I can git blame with git.o.o ? | 14:55 |
| bauzas | git.o.o is pretty cool for recursing over some commits, but now I need to see something | 14:56 |
| *** moshele has quit IRC | 14:57 | |
| *** artom has joined #openstack-nova | 14:58 | |
| bauzas | nvm, I used github for that... :( | 14:58 |
| *** cbader has joined #openstack-nova | 14:58 | |
| *** dguitarbite has quit IRC | 14:58 | |
| *** moshele has joined #openstack-nova | 14:59 | |
| jroll | so I'm playing with cells+ironic in devstack, and cells aren't seeing the baremetal flavor (among others). the query seems fine... thoughts? http://paste.openstack.org/show/348770/ | 14:59 |
| *** Rahul has joined #openstack-nova | 15:00 | |
| *** Rahul has left #openstack-nova | 15:00 | |
| *** anotherRahul has joined #openstack-nova | 15:01 | |
| *** mtanino has joined #openstack-nova | 15:02 | |
| *** _gryf|afk is now known as _gryf | 15:02 | |
| johnthetubaguy | jroll: you need to manually copy the flavors between your cells, assuming its cells v1 | 15:02 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Split instance resize operation into 2 phases https://review.openstack.org/195088 | 15:03 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Create class hierarchy for tasks in conductor https://review.openstack.org/184495 | 15:03 |
| _gryf | bauzas, I think that you cannot do this on git.o.o, unless it's explicitly enabled in config (it's looks like webgit to me) | 15:03 |
| *** garyk has quit IRC | 15:03 | |
| johnthetubaguy | jroll: I kinda though we only needed the flavor in the API / top cell now, but that might be driver specific I guess | 15:03 |
| jroll | johnthetubaguy: OH. right. | 15:03 |
| jroll | it's the child cell missing it | 15:04 |
| *** ijw has joined #openstack-nova | 15:04 | |
| johnthetubaguy | jroll: long term flavor is moving to the new API cell, so the child cell will not have its own flavor | 15:04 |
| johnthetubaguy | jroll: at least thats the cells v2 plan, roughly | 15:04 |
| *** mdrabe has quit IRC | 15:04 | |
| jroll | johnthetubaguy: yeah. also TIL devstack uses nova_api, nova_cell databases. not just nova. | 15:04 |
| *** mmedvede_ has quit IRC | 15:04 | |
| johnthetubaguy | s/API cell/API cell DB/ | 15:04 |
| jroll | thank you sir | 15:05 |
| *** lyanchih has joined #openstack-nova | 15:05 | |
| johnthetubaguy | jroll: no worries, that one bit me when I did cells and devstack a couple of months back | 15:05 |
| jroll | yeah, I can imagine | 15:06 |
| dansmith | hmm, anyone seeing an inability to run the pep8 tox target? | 15:07 |
| dansmith | ValueError: ('Expected version spec in', "MySQL-python;python_version=='2.7'", 'at', ";python_version=='2.7'") | 15:08 |
| mriedem | dansmith: there was something about that in the ML from what i remember,l | 15:08 |
| mriedem | dansmith: one of lifeless' many pbr/requirements changes threads | 15:08 |
| dansmith | hmm, I'm not seeing it | 15:09 |
| dansmith | oh | 15:09 |
| *** dims_ has joined #openstack-nova | 15:09 | |
| mriedem | https://bugs.launchpad.net/devstack/+bug/1468808 looks related | 15:09 |
| openstack | Launchpad bug 1468808 in devstack "stack.sh downgrades pbr if old dependencies are present" [Undecided,Confirmed] | 15:09 |
| *** agireud has joined #openstack-nova | 15:09 | |
| *** ijw has quit IRC | 15:09 | |
| mriedem | dansmith: do you have latest pip? | 15:10 |
| dansmith | using distro pip, so prolly not | 15:10 |
| *** yingjun has quit IRC | 15:11 | |
| dansmith | doesn't help | 15:11 |
| *** logan2 has quit IRC | 15:11 | |
| *** amotoki has joined #openstack-nova | 15:12 | |
| *** karimb has quit IRC | 15:13 | |
| *** markvoelker has quit IRC | 15:14 | |
| *** haomaiwang has quit IRC | 15:14 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Convert test_compute_utils to use Instance object https://review.openstack.org/196912 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Refactor test_servers to use instance objects https://review.openstack.org/197770 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Remove legacy flavor compatibility code from Instance https://review.openstack.org/197771 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Decouple instance object tests from the api fakes module https://review.openstack.org/197769 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Remove flavor migration from db_api and nova-manage https://review.openstack.org/196911 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix test_security_groups to use Instance object properly https://review.openstack.org/198765 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix test_extended_server_attributes for proper Instance object usage https://review.openstack.org/198766 | 15:14 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix Instance object usage in test_extended_ips tests https://review.openstack.org/198767 | 15:14 |
| *** haomaiw__ has joined #openstack-nova | 15:15 | |
| *** beagles_brb is now known as beagles | 15:15 | |
| dansmith | I'll just let jenkins run pep8 I guess | 15:16 |
| anotherRahul | Hi, I am trying to get my bearings around the latest for tagging support, and came across this https://blueprints.launchpad.net/nova/+spec/tag-instances ... it seems to be slated for Liberty. Question I have is will the enhancements also include updated Horizon screens to display/edit the VM instance tags please. | 15:19 |
| *** logan2 has joined #openstack-nova | 15:19 | |
| *** mdrabe has joined #openstack-nova | 15:19 | |
| *** andreykurilin__ has joined #openstack-nova | 15:20 | |
| *** agireud has quit IRC | 15:21 | |
| *** Nikolay_St has joined #openstack-nova | 15:21 | |
| *** ameade has quit IRC | 15:22 | |
| *** ameade_vacation has joined #openstack-nova | 15:22 | |
| *** ameade_vacation is now known as ameade | 15:23 | |
| *** agireud has joined #openstack-nova | 15:23 | |
| *** belmoreira has quit IRC | 15:27 | |
| *** lpetrut has quit IRC | 15:27 | |
| mriedem | anotherRahul: horizon changes are completely separate | 15:27 |
| *** tsekiyama has joined #openstack-nova | 15:27 | |
| mriedem | anotherRahul: the instance tags changes have to be merged in nova first and then someone would need to make changes to horizon to use them, which i'm assuming would require a separate blueprint in the horizon project | 15:27 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Add logging for dm-crypt error conditions https://review.openstack.org/198774 | 15:32 |
| *** Nikolay_St has quit IRC | 15:33 | |
| *** hemnafk is now known as hemna | 15:33 | |
| *** wverdugo500 has joined #openstack-nova | 15:34 | |
| *** xyhuang has quit IRC | 15:34 | |
| jroll | johnthetubaguy: looks like key-pairs not supported in cells either? can't find a bug about it but it's exploding | 15:34 |
| johnthetubaguy | jroll: thats news to me, we certainly have it working with XenAPI, AFAIK | 15:36 |
| jroll | johnthetubaguy: hrm, blowing up in the scheduler, I'll dig on it | 15:37 |
| *** annegentle has joined #openstack-nova | 15:37 | |
| jroll | had to sync the keys manually | 15:37 |
| johnthetubaguy | hmm :S | 15:37 |
| jroll | oh, sorry, compute manager. scheduler is just rescheduling | 15:38 |
| *** exploreshaifali has quit IRC | 15:38 | |
| openstackgerrit | jichenjc proposed openstack/nova: Catch invalid id input in service_delete https://review.openstack.org/198753 | 15:39 |
| *** moshele has quit IRC | 15:39 | |
| johnthetubaguy | jroll: you got a stack trace / log? | 15:39 |
| rlrossit | dansmith: I gave a crack at fixing over all of the objects to use obj_relationships. Kind of unsure how we want to do the change for child_versions in ObjectListBase that we're going to port over to o.vo. I kind of cobbled together stuff already in o.vo (https://review.openstack.org/#/c/198730/) | 15:40 |
| dansmith | rlrossit: thanks, I'll take a look in a bit | 15:41 |
| jroll | johnthetubaguy: http://paste.openstack.org/show/348866/ | 15:41 |
| jroll | johnthetubaguy: I think the tl;dr is keypairs not syncing | 15:42 |
| *** thedodd has joined #openstack-nova | 15:42 | |
| johnthetubaguy | jroll: that looks so familiar, hmmm | 15:42 |
| *** haypo has quit IRC | 15:43 | |
| bauzas | jroll: sorry, I'm on a call, but yeah keypairs are not supported by cells | 15:44 |
| jroll | bauzas: cool, thanks. | 15:45 |
| jroll | "cool" :P | 15:45 |
| johnthetubaguy | hmm, not sure how we have them working in our cloud then... | 15:45 |
| jroll | johnthetubaguy: ikr. | 15:45 |
| jroll | I can dig | 15:45 |
| jroll | it's not a huge deal | 15:45 |
| johnthetubaguy | jroll: depends if its a recent break or not | 15:46 |
| bauzas | jroll: see the issues there https://github.com/openstack/nova/blob/master/devstack/tempest-dsvm-cells-rc | 15:46 |
| jroll | bauzas: thank you sir | 15:46 |
| jroll | johnthetubaguy: doesn't look like a break, I don't see any sort of cells code around keypairs | 15:47 |
| bauzas | jroll: in particular https://github.com/openstack/nova/blob/master/devstack/tempest-dsvm-cells-rc#L61 is excluded from the Tempest run | 15:47 |
| *** zhiyan has quit IRC | 15:47 | |
| jroll | nod | 15:47 |
| *** yassine_ has quit IRC | 15:47 | |
| *** zhiyan has joined #openstack-nova | 15:47 | |
| jaypipes | dansmith: https://review.openstack.org/#/c/197911/ could use a once-over from ya. thx in advance. | 15:47 |
| johnthetubaguy | jroll: I can't quite tell the why from your stack trace right now, it seems to hide the details :( | 15:48 |
| * bauzas asks to Santa a nice Cells v2 for XMas | 15:48 | |
| jroll | johnthetubaguy: I really just think it's never synced to the child. and so compute can't see it. | 15:48 |
| *** Sukhdev has joined #openstack-nova | 15:48 | |
| *** haypo has joined #openstack-nova | 15:48 | |
| johnthetubaguy | jroll: from memory, as the API cell passes the object down to the compute, it usually works of the on boot injection, but looks like something bad is happing | 15:48 |
| johnthetubaguy | jroll: it looks like your method got a dict rather than an object for some reason | 15:49 |
| *** eglynn is now known as eglynn-afk | 15:49 | |
| bauzas | johnthetubaguy: well, the main problem is that most objects are created on the top cell and are not populated down to the child cells | 15:49 |
| *** burt has joined #openstack-nova | 15:49 | |
| johnthetubaguy | bauzas: agreed, but this used to work, somehow | 15:49 |
| *** armax has joined #openstack-nova | 15:50 | |
| bauzas | johnthetubaguy: we have some objects that do the sync when .save() but AFAIK keypairs are not | 15:50 |
| bauzas | johnthetubaguy: I wonder if it was a side effect of something being persisted on the child cell | 15:50 |
| jroll | johnthetubaguy: oh heh https://bugs.launchpad.net/nova/+bug/1443816 | 15:51 |
| openstack | Launchpad bug 1443816 in OpenStack Compute (nova) "cells: config drive doesn't work with cells when injecting an ssh key" [High,In progress] - Assigned to lyanchih (lyanchih) | 15:51 |
| johnthetubaguy | bauzas: keypairs are only ever stored in the API, by design, if my memory is working correctly, although its a bit fuzzy now | 15:51 |
| *** scheuran has quit IRC | 15:51 | |
| *** emagana has joined #openstack-nova | 15:51 | |
| johnthetubaguy | jroll: ah, yes, thats it | 15:51 |
| johnthetubaguy | jroll: I have a feeling we reverted that patch internally | 15:52 |
| jroll | johnthetubaguy: where do you think I found the bug? :) | 15:52 |
| johnthetubaguy | yeah | 15:52 |
| openstackgerrit | jichenjc proposed openstack/nova: Check flavor type before add tenant access https://review.openstack.org/124338 | 15:52 |
| bauzas | johnthetubaguy: well, isn't the child cell DB exactly the same as the top one ? | 15:52 |
| johnthetubaguy | bauzas: nope | 15:52 |
| bauzas | johnthetubaguy: I mean it's a db sync | 15:52 |
| bauzas | https://github.com/openstack-dev/devstack/blob/master/lib/nova#L623-L646 | 15:52 |
| *** patrickeast has joined #openstack-nova | 15:53 | |
| johnthetubaguy | bauzas: well they have the same schema, not the same data | 15:53 |
| bauzas | johnthetubaguy: that I agree :) | 15:53 |
| jroll | bauzas: the keypair in question was added after devstack run, anyway | 15:53 |
| jroll | one more thing - there's no api to put a flavor in a cell, right? I'll have to use raw sql in lib/ironic to fix that bit? | 15:54 |
| jroll | (for devstack runs) | 15:54 |
| bauzas | jroll: I remember the CERN folks doing some hacks for that | 15:54 |
| johnthetubaguy | jroll: the cell level API might do it for you | 15:54 |
| johnthetubaguy | jroll: don't think devstack does that | 15:54 |
| jroll | johnthetubaguy: does devstack run a... ok | 15:54 |
| jroll | :) | 15:54 |
| johnthetubaguy | jroll: I usually dump/restore the mysql table | 15:55 |
| jroll | johnthetubaguy: yep, was just going to put this in devstack: insert into nova_cell.instance_types select * from nova.instance_types where name='baremetal'; | 15:55 |
| johnthetubaguy | well, you need something a bit more general, but yeah | 15:56 |
| jroll | well, this is in lib/ironic | 15:56 |
| jroll | we create a baremetal flavor | 15:56 |
| *** Marga_ has quit IRC | 15:57 | |
| jroll | this is specifically to fix ironic flavors in devstack+cells+ironic https://bugs.launchpad.net/ironic/+bug/1471854 | 15:57 |
| openstack | Launchpad bug 1471854 in Ironic "Ironic with nova-cells doesn't work out of the box in devstack" [Low,New] | 15:57 |
| jroll | and iterate from there | 15:57 |
| bauzas | jroll: https://github.com/cernops/nova-flavor-create | 15:58 |
| jroll | bauzas: cool | 15:58 |
| *** Longgeek has joined #openstack-nova | 16:00 | |
| *** prashup has joined #openstack-nova | 16:00 | |
| *** thorst is now known as thorst_afk | 16:02 | |
| *** jskinner has quit IRC | 16:03 | |
| *** wverdugo500 has left #openstack-nova | 16:04 | |
| *** rdopiera has quit IRC | 16:04 | |
| *** jskinner has joined #openstack-nova | 16:04 | |
| *** ijw has joined #openstack-nova | 16:04 | |
| *** rwsu has joined #openstack-nova | 16:04 | |
| *** Marga_ has joined #openstack-nova | 16:05 | |
| *** Longgeek has quit IRC | 16:05 | |
| *** pask81 has quit IRC | 16:07 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce GuestBlock to wrap around Block API https://review.openstack.org/190210 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: update swap volume to use Guest https://review.openstack.org/190211 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: rename GuestVCPUInfo to VCPUInfo https://review.openstack.org/194133 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make unit tests concise by setup guest object https://review.openstack.org/190963 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: introduce method to wait for block device job https://review.openstack.org/190214 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: update live snapshot to use Guest object https://review.openstack.org/190212 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: update volume delete snapshot to use Guest https://review.openstack.org/190213 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: save the memory state of guest https://review.openstack.org/196991 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make host return guest information https://review.openstack.org/191046 | 16:09 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: make host return power state of guest https://review.openstack.org/191047 | 16:09 |
| *** ijw has quit IRC | 16:09 | |
| *** markvoelker has joined #openstack-nova | 16:10 | |
| *** stefan_amann has quit IRC | 16:11 | |
| *** mpavone has quit IRC | 16:11 | |
| *** sahid has quit IRC | 16:12 | |
| prashup | @jichen | 16:12 |
| prashup | can you please tell me how my change is backward incompatible; | 16:13 |
| openstackgerrit | greghaynes proposed openstack/nova: DO NOT MERGE: testing experimental job https://review.openstack.org/198795 | 16:13 |
| jaypipes | sdague: https://review.openstack.org/#/c/124338/ needs your final review/approval. thx in advance! | 16:14 |
| jaypipes | johnthetubaguy: have you chatted with jroll and devananda about https://review.openstack.org/#/c/194453/? | 16:15 |
| johnthetubaguy | jaypipes: thats the first I have seen of that | 16:15 |
| jroll | :) | 16:16 |
| *** Marga_ has quit IRC | 16:16 | |
| jaypipes | johnthetubaguy: to summarize, this is an important patch/proposal that *begins* to align Ironic's worldview with that of nova-compute. | 16:16 |
| ctrath | mriedem: I still don't have CLA, so I can't submit that fix for bug 1186354 for review, but it's ready | 16:16 |
| openstack | bug 1186354 in OpenStack Compute (nova) "Limits API doesn't work with Neutron" [Medium,Confirmed] https://launchpad.net/bugs/1186354 - Assigned to Cale Rath (ctrath) | 16:16 |
| *** Marga_ has joined #openstack-nova | 16:16 | |
| jaypipes | johnthetubaguy: it's also pretty low-risk, since it does not have any impact outside of the Ironic driver. | 16:16 |
| johnthetubaguy | jroll: jaypipes: this is the plan we spoke about at the summit right? | 16:17 |
| mriedem | ctrath: ok | 16:17 |
| ctrath | mriedem: looking for something else if you have suggestions | 16:17 |
| jaypipes | johnthetubaguy: yup. | 16:17 |
| jroll | johnthetubaguy: indeed | 16:17 |
| jaypipes | johnthetubaguy: and that we've refined over a couple chats. | 16:17 |
| mriedem | ctrath: code reviews :) | 16:17 |
| ctrath | mriedem: will do my best :-) | 16:17 |
| prashup | @johnthetubaguy can you review my new patch. i have changed it according to review comments in previous patch | 16:18 |
| johnthetubaguy | jaypipes jroll: so I have made a feature freeze exception request, by adding it here: https://etherpad.openstack.org/p/liberty-spec-freeze-exceptions | 16:18 |
| prashup | https://review.openstack.org/#/c/198714/2 | 16:18 |
| jaypipes | johnthetubaguy: ty sir. | 16:18 |
| jaypipes | johnthetubaguy: I will sponsor this. | 16:18 |
| johnthetubaguy | now its way late, but agreed its important | 16:18 |
| jroll | johnthetubaguy: ty | 16:19 |
| prashup | @johnthetubaguy ^^^ :o | 16:19 |
| prashup | johnthetubaguy: ^^^ :o | 16:19 |
| mriedem | prashup: geez, i think john gets it | 16:19 |
| mriedem | prashup: i added ndipanov since he's working on lots of related bdm things | 16:20 |
| prashup | yeah ; i know it. but i used the wrong @ there | 16:20 |
| mriedem | prashup: @ does nothing here, this isn't twitter | 16:20 |
| johnthetubaguy | yeah, my client gets all of those | 16:20 |
| kashyap | mriedem: lol | 16:20 |
| prashup | thats why corrected it!!! | 16:20 |
| kashyap | prashup: He means | 16:21 |
| mriedem | prashup: when you call out someone's nick, they get pinged | 16:21 |
| kashyap | No need to correct it. | 16:21 |
| johnthetubaguy | prashup: which spec, do you have a link? | 16:21 |
| mriedem | regardless | 16:21 |
| prashup | oh. | 16:21 |
| prashup | i dint know it | 16:21 |
| ndipanov | mriedem, looking | 16:21 |
| * jaypipes goes to review dansmith's robustify-evacuate patch series... | 16:21 | |
| mriedem | johnthetubaguy: it's not a spec, it's a bug fix that you were on before | 16:21 |
| *** jaypipes is now known as jay-deepthought | 16:21 | |
| prashup | i posted the link to previous patch in the comments | 16:21 |
| sdague | jay-deepthought: so, we should fix the commit message first | 16:22 |
| openstackgerrit | Sean Dague proposed openstack/nova: Check flavor type before add tenant access https://review.openstack.org/124338 | 16:22 |
| dansmith | \o/ | 16:22 |
| ndipanov | jay-deepthought, I had a -1 on the first patch tehre beacuase I am doing something with formalizing states | 16:22 |
| jay-deepthought | sdague: done. thx. | 16:23 |
| dansmith | ndipanov: did you see my reply to that? | 16:23 |
| ndipanov | dansmith, no :) | 16:23 |
| dansmith | I want you to move your -1 to the patch that adds lots of states | 16:24 |
| dansmith | and off the patch that fixes the data loss issue | 16:24 |
| dansmith | please :) | 16:24 |
| ndipanov | yes can do | 16:24 |
| ndipanov | let me look at it now | 16:24 |
| *** lpetrut has joined #openstack-nova | 16:24 | |
| *** exploreshaifali has joined #openstack-nova | 16:24 | |
| jogo | mriedem: so our 3rd party CI for nova is really not doing well | 16:25 |
| jogo | VMWare is down, hyperv doesn't work and citrix doesn't report in the right format | 16:25 |
| openstackgerrit | PrashantUpadhyay proposed openstack/nova: Check for volume size https://review.openstack.org/198804 | 16:25 |
| jogo | johnthetubaguy: ^ | 16:25 |
| *** otter768 has quit IRC | 16:26 | |
| *** jwcroppe_ has joined #openstack-nova | 16:27 | |
| johnthetubaguy | jogo: oh, you mean the XenServer one not the XenProject one I guess, I don't think they have that on zuul yet, BobBall has the details | 16:28 |
| johnthetubaguy | jogo: I want to dig deep into that when I get to the feature classification stuff | 16:28 |
| johnthetubaguy | jogo: I saw you had a nice web page that starts to track that now :) I think BobBall had something up as well, somewhere | 16:29 |
| jogo | Citrix XenServer I mean | 16:29 |
| jogo | XenProject CI actually works pretty well | 16:30 |
| johnthetubaguy | jogo: I think we need some trending of the numbers, somehow, I kinda want the same for review stats too, but not really had chance to sit down and think about that yet | 16:30 |
| *** haypo has left #openstack-nova | 16:30 | |
| mriedem | jogo: honestly i only really care about those when looking at virt drivers and i want to see that the related virt driver CI is passing on the change | 16:30 |
| *** jichen has quit IRC | 16:30 | |
| johnthetubaguy | jogo: yeah, thats started with zuul, I think the plan is to move the XenServer one onto zuul at some point, not not 100% sure | 16:30 |
| *** jwcroppe has quit IRC | 16:30 | |
| mriedem | one of the xen ones frequently reports success immediately after the change is posted | 16:31 |
| *** ociuhandu has quit IRC | 16:31 | |
| johnthetubaguy | mriedem: thats the zuul bug | 16:31 |
| johnthetubaguy | mriedem: it decides there are no tests to run, so votes +1 | 16:31 |
| *** lyanchih has quit IRC | 16:31 | |
| * jogo goes to his real job | 16:32 | |
| *** jwcroppe_ has quit IRC | 16:32 | |
| johnthetubaguy | but yeah, that with skew the stats someone | 16:32 |
| johnthetubaguy | jogo: appreciate the heads up though, I need to dig into that more soon | 16:32 |
| johnthetubaguy | s/someone/somewhat/ | 16:32 |
| johnthetubaguy | jroll: that resource tracker stuff doesn't feel quite right, as you might lock the same resource twice, I guess? | 16:34 |
| *** davideagnello has joined #openstack-nova | 16:35 | |
| jroll | johnthetubaguy: hmm. you may have a point. | 16:35 |
| johnthetubaguy | jroll: its totally fixable though | 16:35 |
| jroll | johnthetubaguy: would claims not handle that? | 16:36 |
| jroll | yeah | 16:36 |
| johnthetubaguy | jroll: I think you just need to configure each nova-compute node with an index, so it can pick a unique subset of nodes to expose, or something like that? | 16:36 |
| johnthetubaguy | jroll: hmm, although that could create other issues | 16:36 |
| jroll | johnthetubaguy: yeah, we were hoping not to shard ironic nodes to specific computes | 16:37 |
| openstackgerrit | Andrew Bogott proposed openstack/nova: Include project_id in instance metadata. https://review.openstack.org/197185 | 16:37 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make evacuate update the migration object along the way https://review.openstack.org/194373 | 16:37 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add some notifications to the evacuate path https://review.openstack.org/194372 | 16:37 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make evacuate leave a record for the source compute host to process https://review.openstack.org/183354 | 16:37 |
| *** davideagnello has quit IRC | 16:37 | |
| *** andrewbogott has joined #openstack-nova | 16:37 | |
| johnthetubaguy | jroll: I was thinking of hashing to distribute them, which does reduce a lot of the issues | 16:38 |
| jay-deepthought | ndipanov: hey, so evacuate() is not a move operation... | 16:39 |
| jroll | johnthetubaguy: yeah, trying to avoid that if possible | 16:39 |
| jay-deepthought | ndipanov: it is a delete-and-rebuild operation. | 16:39 |
| *** packet has joined #openstack-nova | 16:39 | |
| ndipanov | jay-deepthought, hmmm | 16:39 |
| ndipanov | well from the pov of resource tracking we still wanna claim no? | 16:40 |
| jay-deepthought | ndipanov: yes, for sure, but it should just be handled via instance_claim()'s normal approach. | 16:40 |
| dansmith | how is it not a move? | 16:40 |
| dansmith | it's not moving data, | 16:40 |
| dansmith | but it's moving where the instance is running | 16:40 |
| dansmith | seems like equivalent semantics to cold migrate | 16:40 |
| ndipanov | it could be... | 16:40 |
| ndipanov | using only instance_claim yes | 16:41 |
| jay-deepthought | dansmith: no, since cold migrate is snapshot'ing and moving the workload. evacuate is rebuild'ing from the base image. | 16:41 |
| *** kun_huang has left #openstack-nova | 16:41 | |
| *** tjones1 has joined #openstack-nova | 16:41 | |
| *** tongli has quit IRC | 16:42 | |
| *** eharney has quit IRC | 16:42 | |
| dansmith | jay-deepthought: I don't see the difference from the RT perspective | 16:42 |
| *** tongli has joined #openstack-nova | 16:42 | |
| ndipanov | dansmith, there is not a lot | 16:42 |
| ndipanov | and also | 16:42 |
| ndipanov | jay-deepthought, it is very useful from the pov of automation code outside of nova to treat this as a move | 16:43 |
| dansmith | jay-deepthought: maybe you mean because you're holding resources on both nodes for a short period of time? | 16:43 |
| dansmith | in the cold migrate case, that is | 16:43 |
| jroll | johnthetubaguy: if you'd like to leave note sin the spec review that would be super helpful, I'll think on this | 16:43 |
| dansmith | that's kinda what I was getting at with my comment about having an intermediate claim that was almost like boot, where you don't hold both | 16:43 |
| *** davideagnello has joined #openstack-nova | 16:44 | |
| ndipanov | dansmith, that's incance claim than | 16:44 |
| ndipanov | then * | 16:44 |
| jay-deepthought | right. | 16:44 |
| johnthetubaguy | jroll: I added a +0 vote with that note, let me know if you need that claifying | 16:44 |
| ndipanov | I think it makes sense to have the migration records | 16:44 |
| jroll | thanks johnthetubaguy | 16:44 |
| ndipanov | and once we have those, it's only natural to use a move claim | 16:45 |
| dansmith | ndipanov: right, that was my point | 16:45 |
| jay-deepthought | dansmith: I guess what is different to me (from a resource tracking perspective) is that in the evacuate() case, the removal of the old resources (i.e. the ones on the dead node) are not connected to the *addition* of the resources on the new node that was rebuilt on. | 16:45 |
| dansmith | jay-deepthought: okay, not sure I see a huge distinction, but that's cool | 16:45 |
| jay-deepthought | dansmith: it's identical to the case when a user terminates an instance on one node and then builds an instance of the same type on another node. | 16:46 |
| jay-deepthought | dansmith: we don't connect those two resource tracking activities together. | 16:46 |
| dansmith | well, | 16:46 |
| jay-deepthought | dansmith: which is why I don't think we should in the case of evacudate() | 16:46 |
| dansmith | you mean because the claim would affect quota? | 16:46 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: handle rescue flag first in blockinfo.get_disk_mapping https://review.openstack.org/198812 | 16:47 |
| ayoung | dansmith, mriedem if I perform an administrative Nova action, like setting quota or something with cells, is that still executed by nova-compute-maybe-running-on-a-hypervisor? | 16:47 |
| jay-deepthought | dansmith: wasn't even thinking about quota, honestly. | 16:47 |
| *** tongli has quit IRC | 16:47 | |
| mriedem | ayoung: changing quota limits for a project/user? | 16:47 |
| dansmith | jay-deepthought: okay, then not sure why it matters, and it's technically the same instance (especially for pci devices, IPs, etc) so it seems right to do a move-like claim | 16:47 |
| jay-deepthought | ayoung: depends. quota change actions are executed against the top-level cell database. | 16:47 |
| dansmith | jay-deepthought: but anyway, I'm not an expert here at all | 16:47 |
| jay-deepthought | dansmith: hmm, I suppose that point might change my mind on this... hadn't considered that. | 16:48 |
| mriedem | ayoung: changing quota limits doesn't have anything to do with nova-compute as far as i know, it's a direct to db api change | 16:48 |
| ayoung | mriedem, sure...the actual call is not important...just trying to determine if there are a class of calls that come through the API that are not executed by being picked up by the compute nodes | 16:48 |
| *** ildikov has quit IRC | 16:48 | |
| openstackgerrit | Chinmaya Bharadwaj proposed openstack/nova: VMware: Fixes instance creation failure using block device mapping https://review.openstack.org/197192 | 16:49 |
| *** tonytan4ever has quit IRC | 16:50 | |
| *** annegentle has quit IRC | 16:50 | |
| *** lyarwood has quit IRC | 16:50 | |
| *** jbernard has quit IRC | 16:50 | |
| *** ondergetekende_ has quit IRC | 16:50 | |
| *** ondergetekende has joined #openstack-nova | 16:50 | |
| *** jbernard has joined #openstack-nova | 16:51 | |
| *** leseb has quit IRC | 16:53 | |
| *** annegentle has joined #openstack-nova | 16:54 | |
| andrewbogott | If anyone has a moment, I would enjoy a review of https://review.openstack.org/#/c/197185/. It’s quite simple. | 16:54 |
| ayoung | andrewbogott, interesting... | 16:55 |
| *** leseb has joined #openstack-nova | 16:55 | |
| *** lyarwood has joined #openstack-nova | 16:55 | |
| *** eharney has joined #openstack-nova | 16:57 | |
| andrewbogott | ayoung: I just now rebased it, hence the lack of CI votes. | 16:57 |
| ayoung | andrewbogott, what use case drove that? | 16:57 |
| *** pixelbeat_ has quit IRC | 16:58 | |
| *** lucasagomes is now known as lucas-dinner | 16:58 | |
| andrewbogott | ayoung: there’s a bug attached which links to an email thread. Instances in my cloud do set up ssh keys automatically based on project, among other things. | 16:59 |
| andrewbogott | They also use naming like instancename.projectname.tld | 16:59 |
| andrewbogott | Also I do salt management of instances using project-based wildcards | 17:01 |
| andrewbogott | etc. etc. :) | 17:01 |
| *** Longgeek has joined #openstack-nova | 17:01 | |
| *** derekh has quit IRC | 17:02 | |
| *** ijw has joined #openstack-nova | 17:04 | |
| *** amotoki has quit IRC | 17:04 | |
| *** ijw has quit IRC | 17:05 | |
| mriedem | andrewbogott: that would certainly muck with any chance of having a tenant ownership change feature in nova, which has been proposed and operators said they wanted it | 17:05 |
| *** ijw has joined #openstack-nova | 17:05 | |
| andrewbogott | mriedem: the patch would muck with it, or my use case would? | 17:05 |
| mriedem | the patch | 17:06 |
| *** doude has quit IRC | 17:06 | |
| bauzas | lxsli: jay-deepthought: still around ? | 17:06 |
| jay-deepthought | bauzas: yup. | 17:06 |
| andrewbogott | mriedem: how so? If an instance moves to a different tenant then the metadata value would update accordingly | 17:06 |
| *** Longgeek has quit IRC | 17:06 | |
| *** doude has joined #openstack-nova | 17:06 | |
| *** doude has quit IRC | 17:06 | |
| *** doude has joined #openstack-nova | 17:06 | |
| andrewbogott | unless the instance object itself has an out-of-date project_id which seems like its own problem. | 17:06 |
| *** ssurana has joined #openstack-nova | 17:06 | |
| bauzas | https://review.openstack.org/#/c/145528/25jay-deepthought: thanks for reviewing | 17:06 |
| bauzas | argh | 17:06 |
| bauzas | jay-deepthought: thanks for reviewing https://review.openstack.org/#/c/145528/25 | 17:06 |
| bauzas | (better) | 17:06 |
| *** thorst_afk is now known as thorst | 17:07 | |
| bauzas | jay-deepthought: food for thoughts on L40 at https://review.openstack.org/#/c/145528/25/nova/objects/request_spec.py,cm | 17:07 |
| mriedem | andrewbogott: yeah nvm, this might be fine | 17:08 |
| bauzas | jay-deepthought: for the moment, the scheduler provides a way to have multiple instances checked | 17:08 |
| mriedem | andrewbogott: i was thinking config drive writing data to the disk on the vm and then that is stale | 17:08 |
| bauzas | jay-deepthought: all of that is using num_instances | 17:08 |
| andrewbogott | yeah, config drive wouldn’t work | 17:08 |
| bauzas | jay-deepthought: min-count and max-count are just API fields, but at the bottom, it goes to the sched using num_instances | 17:09 |
| *** piyanai has quit IRC | 17:09 | |
| bauzas | jay-deepthought: I know there is a debate on dumping the min/max count and multiple instances query for Nova using Heat or something else | 17:09 |
| bauzas | jay-deepthought: but until this is merged, I don't see why we should remove this num_instances field | 17:10 |
| *** tongli has joined #openstack-nova | 17:10 | |
| *** markus_z has quit IRC | 17:10 | |
| bauzas | jay-deepthought: of course, once someone is working on removing the possibility to have multiple creates, he will deprecate num_instances, for sure | 17:10 |
| jay-deepthought | bauzas: and how is the min/max count handled in the request spec object? | 17:11 |
| bauzas | jay-deepthought: there is no min/max count | 17:12 |
| bauzas | jay-deepthought: I mean for the scheduler | 17:12 |
| ayoung | andrewbogott, so...does that lead to fetching Keystone tokens, or just other uses for the project_id? | 17:13 |
| jay-deepthought | bauzas: k, so it's handled externally to the scheduler, then? | 17:13 |
| bauzas | jay-deepthought: yup | 17:13 |
| jay-deepthought | k | 17:13 |
| bauzas | jay-deepthought: lemme show you that | 17:13 |
| jay-deepthought | bauzas: alright, then feel free to ignore that particular comment. | 17:13 |
| andrewbogott | ayoung: well… I’m using some ancient keystone code where the projects are stored in ldap. So it happens that for me project_name == project_id | 17:13 |
| andrewbogott | when the ldap backend is deprecated then I’ll probably have to hit the keystone api. | 17:14 |
| andrewbogott | It would be nice to have project_name in metadata but that would require a nova metadata request to hit keystone which seems… unlikely. | 17:14 |
| bauzas | jay-deepthought: FYI https://github.com/openstack/nova/blob/master/nova/compute/api.py#L927-L928 | 17:15 |
| bauzas | jay-deepthought: so it retunrs a number of instances | 17:15 |
| jay-deepthought | bauzas: lol, ok, thanks :) | 17:15 |
| bauzas | jay-deepthought: np | 17:15 |
| bauzas | (which makes sense, because we need to check quotas first :) | 17:16 |
| andrewbogott | ayoung: (unless nova querying keystone is already a solved problem?) | 17:16 |
| bauzas | well, provided quotas are working :) | 17:16 |
| bauzas | :p | 17:16 |
| jay-deepthought | heh, yeah | 17:16 |
| mriedem | andrewbogott: comments inline | 17:17 |
| andrewbogott | thanks | 17:17 |
| *** promulo has joined #openstack-nova | 17:17 | |
| ayoung | andrewbogott, not yet. But there are more issues there. I like "project_id == project_name" but it does not really namespace things for the Hierarchical case | 17:17 |
| mriedem | alaski: johnthetubaguy: i'm assuming we don't need a blueprint for trivial changes to the metadata service for a new api version? https://review.openstack.org/#/c/197185/ - i don't really know though since we require blueprints for other changes to the rest API for new microversions | 17:17 |
| ayoung | andrewbogott, so, the AWS approach is to create an "entity" and use that to query other things via Credentials. The Keystone equivalent is a user, and creating a user in a service specific domain is OK...with caveats of course | 17:18 |
| johnthetubaguy | mriedem: hmm, good question, it is a public REST API I guess :( | 17:19 |
| andrewbogott | ayoung: yeah, I’m aware that my setup is going to break in various ways a few releases down the line. Still, getting the project id is half the battle. | 17:19 |
| ayoung | andrewbogott, in general, though, I think this is step in the right direction | 17:19 |
| andrewbogott | ayoung: it happens that everything about projects and instances therein is public knowledge in my setup. So probably I’ll fidget with my policies and make a universal read-only role and a special user with public credentials. | 17:20 |
| andrewbogott | I need that for other things anyway, I can just embed that account on new instances and they can ask keystone to resolve project names. | 17:20 |
| *** logan2 has quit IRC | 17:21 | |
| harlowja | ndipanov ok commented on https://review.openstack.org/#/c/197668/ :) | 17:22 |
| ndipanov | harlowja, just replied | 17:23 |
| ndipanov | :) | 17:23 |
| harlowja | :) | 17:23 |
| ndipanov | but yes I do need to add docstrings | 17:23 |
| *** nihilifer has joined #openstack-nova | 17:24 | |
| ndipanov | fixing tests that moving to this broke right now, so after I push that patch it might be more clear from the usage in the code why I did some things | 17:24 |
| harlowja | ndipanov the state-space stuff is just a thing that makes state-machiens then? | 17:24 |
| ndipanov | harlowja, yes - since we use them in several services | 17:24 |
| ndipanov | + declaring everything on one line is more concise imho | 17:25 |
| harlowja | k, def seems like something that could live up in automaton if its just that | 17:25 |
| harlowja | one line, 2 lines... meh | 17:25 |
| *** nihilifer has quit IRC | 17:28 | |
| *** marzif_ has quit IRC | 17:28 | |
| ndipanov | harlowja, well it's not only 2 lines it's one additional line per transition - I think this https://review.openstack.org/#/c/197669/2/nova/compute/migration_states.py is much clearer than the ironic example where you call add_transition that you have to explicitly name (not relevant for us) | 17:29 |
| *** iamjarvo has joined #openstack-nova | 17:29 | |
| harlowja | to me its the equivalent of adjanceny lists ( https://en.wikipedia.org/wiki/Adjacency_list ) in graphs vs an edge list | 17:29 |
| ndipanov | it is basically an adecency list yes | 17:29 |
| harlowja | both are valid, and both do the same thing in the end | 17:29 |
| ndipanov | now this may not be relevatn to all use cases | 17:29 |
| ndipanov | but for Nova | 17:30 |
| harlowja | why though wouldn't explicitly naming be a good thing in nova? | 17:30 |
| ndipanov | where we really only have strings now and would like to have some way to formailize it but not use any other niceties of automaton hence a wrapper | 17:30 |
| ndipanov | explicitly naming transitions (edges) | 17:30 |
| harlowja | k | 17:31 |
| ndipanov | well we really have no semantics for them | 17:31 |
| andrewbogott | mriedem: thanks for the comments. I have some meetings now but will follow up in a few hours. | 17:31 |
| ndipanov | for now | 17:31 |
| harlowja | ndipanov gotcha | 17:31 |
| harlowja | k | 17:31 |
| harlowja | ndipanov ya, typically until u know what to call those it becomes "on_%s" % target_state | 17:31 |
| ndipanov | plus the potential renaming thing with old names potentially existing in the db | 17:31 |
| harlowja | ya, aliasing or something | 17:32 |
| ndipanov | harlowja, so really just a bit of code to bridge the generality of automaton and the specific states tracking we have in nova | 17:33 |
| *** tonytan4ever has joined #openstack-nova | 17:33 | |
| harlowja | ya, if the state-space is just another way of making a state-machine, seems like it could just be generalized | 17:34 |
| harlowja | and put in automaton somewhere | 17:34 |
| ndipanov | well we will have several spaces | 17:34 |
| ndipanov | migration states | 17:34 |
| ndipanov | and instance states | 17:34 |
| harlowja | ya, i get that | 17:34 |
| harlowja | those are just different tables to me :-P | 17:34 |
| ndipanov | (some other maybe) | 17:34 |
| harlowja | like @ https://github.com/openstack/taskflow/blob/master/taskflow/states.py#L87 :-P | 17:34 |
| harlowja | ^ which will get converted to individual machiens to | 17:34 |
| harlowja | thats just another way of representing your state-space stuff (imho) | 17:35 |
| ndipanov | yes | 17:35 |
| ndipanov | with a slightly more python hacking in my case :) | 17:35 |
| harlowja | ya | 17:35 |
| ndipanov | but same thing | 17:35 |
| harlowja | agreed | 17:36 |
| *** mpaolino has quit IRC | 17:36 | |
| harlowja | sooo that makes me wonder if the python hacking is 'hacking' or 'useful' idk | 17:36 |
| ndipanov | well it's pretty standard python stuff really - so I wasn't too worried :) | 17:37 |
| ndipanov | we have way cleverer code than that :) | 17:37 |
| ndipanov | 'cleverer' | 17:37 |
| harlowja | ^ not sure thats something to brag about :-P | 17:37 |
| harlowja | 'clever' is a dangerous word (but i get what u mean, ha) | 17:38 |
| jay-deepthought | dansmith: I'm stopping at https://review.openstack.org/#/c/183200/, since you need to rebase it and the microversion has changed since you originally submitted. | 17:39 |
| *** jay-deepthought is now known as jaypipes | 17:40 | |
| *** mutoulbj has quit IRC | 17:42 | |
| ndipanov | jaypipes, I see you read this too https://review.openstack.org/#/c/197668/ - seen my replies? | 17:42 |
| *** mutoulbj has joined #openstack-nova | 17:43 | |
| jaypipes | ndipanov: yeah. I still don't think that patch makes anything more clear, though :( Perhaps it's down to our style of programmin. | 17:44 |
| ayoung | Vek, http://git.openstack.org/cgit/openstack/nova/tree/nova/policy.py#n120 that can go away now, right? The referenced bug https://bugs.launchpad.net/openstack-common/+bug/1039132 is closed, no need for an explicit check in nova policy? | 17:46 |
| openstack | Launchpad bug 1039132 in openstack-common "Policy checking fails against non-strings" [Undecided,Fix released] - Assigned to Vish Ishaya (vishvananda) | 17:46 |
| alaski | mriedem: is Nova the only consumer of the metadata api? (we don't deploy it so I'm not too familiar) | 17:47 |
| ndipanov | jaypipes, I think that is definitely possible especially since I tried to work towards a nice declarative syntax | 17:48 |
| *** Sukhdev has quit IRC | 17:48 | |
| ayoung | $ find nova -name \*py | xargs grep is_admin | grep -v test | wc -l | 17:48 |
| ayoung | 50 | 17:48 |
| ayoung | wow...that is a lot of embedded policy checks | 17:48 |
| *** agireud has quit IRC | 17:48 | |
| alaski | mriedem: if so I would be less concerned about versioning for backwards compatible changes. as long as it doesn't affect upgrades, and/or fails cleanly if newer code doesn't find an expected field in the metadata api | 17:48 |
| ndipanov | but tell me this: https://review.openstack.org/#/c/197669/2/nova/compute/migration_states.py is not nicer than: | 17:48 |
| *** Sukhdev has joined #openstack-nova | 17:48 | |
| ndipanov | https://github.com/openstack/ironic/blob/1beb8ca607eb8bd77c4b4995cf5b4f007688a5c7/ironic/common/fsm.py | 17:49 |
| ndipanov | jaypipes, ^ :) | 17:49 |
| *** anotherRahul has quit IRC | 17:49 | |
| ndipanov | errr this I mean https://github.com/openstack/ironic/blob/master/ironic/common/states.py | 17:49 |
| johnthetubaguy | alaski: isn't the guest VM the main consumer of the metadata service API, not that API is used to generate config drive that is accessed by the guest | 17:49 |
| jaypipes | ndipanov: yeah, I don't like that either. | 17:49 |
| johnthetubaguy | alaski: s/not/note/ | 17:49 |
| * bauzas mutters something about how he hates TrustedFilter | 17:49 | |
| ndipanov | jaypipes, you're a tough man to please :) | 17:50 |
| ctrath | rlrossit: I'm looking at https://review.openstack.org/#/c/198730/ . Do you know if there's a document that describes how obj_relationships works? | 17:50 |
| *** annashen has joined #openstack-nova | 17:50 | |
| *** logan2 has joined #openstack-nova | 17:50 | |
| alaski | johnthetubaguy: gotcha. then perhaps a versioning discussion is needed. it would seem reasonable to be able to add fields without issue, but I don't know the consumers well enough to know if that would be fine in practice | 17:50 |
| jaypipes | ndipanov: :) yes, I've been told... | 17:51 |
| johnthetubaguy | alaski: yeah, I am not sure either really. the patch does add a new version in the usual way, but thats actually a bit broken really (since we release every commit) | 17:51 |
| mriedem | alaski: i think whatever is running on the guest vm is the consumer of the metadata service api | 17:52 |
| mriedem | http://docs.openstack.org/admin-guide-cloud/content/section_metadata-service.html | 17:52 |
| alaski | johnthetubaguy: oh, I see. I didn't realize there was rudimentary versioning there already | 17:52 |
| mriedem | alaski: so app dev i guess | 17:52 |
| *** jwcroppe has joined #openstack-nova | 17:53 | |
| *** jasondotstar has quit IRC | 17:56 | |
| *** _diana_ has joined #openstack-nova | 17:57 | |
| *** jskinner has quit IRC | 17:58 | |
| *** agireud has joined #openstack-nova | 17:58 | |
| *** jskinner has joined #openstack-nova | 17:58 | |
| *** agireud has quit IRC | 17:59 | |
| harlowja | jaypipes ndipanov i'm fine with an adjancy_list way of representing stuff, both are the same imho (and one may be nicer than the other...) | 17:59 |
| harlowja | based on your personal preferences, they both do the same thing in the end :-P | 17:59 |
| harlowja | make the following in automaton | 18:00 |
| harlowja | __init__.py | 18:00 |
| harlowja | readers/ | 18:00 |
| harlowja | adjancy_list.py | 18:00 |
| harlowja | and there u go, magic | 18:00 |
| harlowja | put whatever 'reader' in there that makes u feel happy :-P | 18:01 |
| *** aysyd has quit IRC | 18:02 | |
| alaski | mriedem: I would lean towards a spec, or at the very least a blueprint to bring up at the Nova meeting. something more visible than just two core reviewers at least | 18:02 |
| mriedem | alaski: yeah, tend to agree | 18:03 |
| *** Kennan2 has joined #openstack-nova | 18:03 | |
| *** aysyd has joined #openstack-nova | 18:03 | |
| mriedem | alaski: want to say that in here? https://review.openstack.org/#/c/197185/ | 18:03 |
| *** zhenguo has quit IRC | 18:03 | |
| mriedem | or hte ML | 18:03 |
| *** Kennan has quit IRC | 18:04 | |
| *** Piet has quit IRC | 18:04 | |
| openstackgerrit | Zachary Sais proposed openstack/nova: Move existing V2 to legacy-v2 https://review.openstack.org/193725 | 18:04 |
| *** yamahata has joined #openstack-nova | 18:05 | |
| ndipanov | harlowja, will do! | 18:05 |
| harlowja | ndipanov sweet | 18:05 |
| ndipanov | as soon as I get these tests passing | 18:05 |
| harlowja | :) | 18:06 |
| *** andrewbogott is now known as andrewbogott_afk | 18:06 | |
| harlowja | then u can use https://github.com/openstack/automaton/blob/master/automaton/converters/pydot.py#L26 to make a pretty diagram to, lol | 18:06 |
| harlowja | like @ http://docs.openstack.org/developer/ironic/dev/states.html | 18:06 |
| harlowja | pretty diagrams ++ | 18:06 |
| harlowja | lol | 18:06 |
| johnthetubaguy | mriedem: alaski: applied procedural -2 after your comments, trying to avoid them, but it feels like the correct thing at this point | 18:07 |
| *** med_` is now known as med_ | 18:07 | |
| *** med_ has joined #openstack-nova | 18:07 | |
| *** jasondotstar has joined #openstack-nova | 18:08 | |
| ndipanov | harlowja, I actually think that's a very nice feature and plan to use it in docs | 18:08 |
| openstackgerrit | Darragh O'Reilly proposed openstack/nova: Ignore bridge already exists error when creating bridge https://review.openstack.org/198833 | 18:08 |
| harlowja | ndipanov agreed | 18:08 |
| ndipanov | that I also plan to write :) | 18:08 |
| harlowja | nice | 18:08 |
| harlowja | ndipanov https://github.com/openstack/ironic/blob/master/tox.ini#L22 or similar is what these projects have | 18:09 |
| harlowja | a little helper script that just makes those diagrams | 18:09 |
| ndipanov | wow that also exists | 18:10 |
| ndipanov | cool | 18:10 |
| harlowja | ya, i need to shrink that script down to something smaller (since the converter code is now common) | 18:11 |
| *** otter768 has joined #openstack-nova | 18:11 | |
| rlrossit | ctrath: you would have to ask dansmith if/where doc is on obj_relationships, but most of the base code lives in an oslo project now (https://github.com/openstack/oslo.versionedobjects) | 18:11 |
| ctrath | rolrossit.. thanks, I'll start there | 18:13 |
| ctrath | rlrossit: ^ | 18:13 |
| *** agireud_ has joined #openstack-nova | 18:13 | |
| mriedem | ctrath: rlrossit: https://github.com/openstack/oslo.versionedobjects/blob/master/oslo_versionedobjects/base.py#L261 | 18:13 |
| mriedem | see the giant docstring | 18:13 |
| rlrossit | mriedem: if we roll eventlet.spawn() in with https://bugs.launchpad.net/nova/+bug/1468513 that exposes a lot of places that are using eventlet.spawn()... should those be changed over to use nova.utils.spawn()? | 18:14 |
| openstack | Launchpad bug 1468513 in OpenStack Compute (nova) "hacking check needed for using greenthread.spawn()" [Low,In progress] - Assigned to Ryan Rossiter (rlrossit) | 18:14 |
| mriedem | rlrossit: i'd make it a separate change | 18:15 |
| mriedem | handle greenthread.spawn() as it's own hacking check | 18:15 |
| ctrath | mriedem: yeah, I read that multiple times. I'm wondering if the 'value', subobject_version, maps to the VERSION field in the subobject class | 18:15 |
| *** lpetrut has quit IRC | 18:16 | |
| rlrossit | mriedem: cool. you want me to put up a change for eventlet.spawn() then too? | 18:17 |
| mriedem | ctrath: yes, the versions in the obj_relationships mapping are always corresponding to the VERSION attribute in the objects | 18:18 |
| mriedem | rlrossit: i'd hold off on eventlet.spawn | 18:18 |
| ctrath | for instance, floating_ip.py has obj_relationships.fixed_ip that only go up to 1.6. The current FixedIP.VERSION is 1.10 | 18:18 |
| rlrossit | alrighty | 18:18 |
| rlrossit | ctrath: that was part of the can of worms I opened when working on this | 18:18 |
| ctrath | rlrossit: so, what gives? | 18:18 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: update docstring in blockinfo module for disk_info https://review.openstack.org/198836 | 18:19 |
| ctrath | rlrossit: do new mappings need added? | 18:19 |
| rlrossit | I wasn't sure if we wanted to do that in this patch or a separate one | 18:19 |
| *** dims_ has quit IRC | 18:20 | |
| *** pixelbeat_ has joined #openstack-nova | 18:24 | |
| *** lpetrut has joined #openstack-nova | 18:26 | |
| *** ajayaa has quit IRC | 18:26 | |
| ctrath | rlrossit: do you know if there is another bug open for the above issue? | 18:27 |
| *** SourabhP has joined #openstack-nova | 18:27 | |
| *** andrewbogott_afk is now known as andrewbogott | 18:27 | |
| *** annashen has quit IRC | 18:28 | |
| *** thangp is now known as thangp_afk | 18:29 | |
| *** tjones1 has quit IRC | 18:33 | |
| *** sudipto has quit IRC | 18:33 | |
| *** Sukhdev has quit IRC | 18:34 | |
| rlrossit | ctrath: there isn't a separate bug for it, but dansmith might want me to handle it in the obj_relationships change. | 18:34 |
| ctrath | rlrossit: sounds good. I added a comment to the bug. ty | 18:34 |
| *** lpetrut has quit IRC | 18:36 | |
| *** annashen has joined #openstack-nova | 18:38 | |
| *** annashen has quit IRC | 18:40 | |
| *** browne has quit IRC | 18:41 | |
| *** jdurgin has joined #openstack-nova | 18:45 | |
| *** tjones1 has joined #openstack-nova | 18:47 | |
| *** user_ has joined #openstack-nova | 18:48 | |
| *** user_ has quit IRC | 18:48 | |
| *** baoli_ has quit IRC | 18:49 | |
| *** baoli has joined #openstack-nova | 18:50 | |
| *** _diana_ has quit IRC | 18:51 | |
| *** _diana_ has joined #openstack-nova | 18:52 | |
| *** aysyd has quit IRC | 18:56 | |
| *** piyanai has joined #openstack-nova | 18:57 | |
| *** annashen has joined #openstack-nova | 18:57 | |
| *** prashup has quit IRC | 18:57 | |
| *** agireud_ has quit IRC | 18:58 | |
| *** _diana_ has quit IRC | 18:59 | |
| *** jskinner has quit IRC | 18:59 | |
| *** jskinner has joined #openstack-nova | 19:00 | |
| *** aysyd has joined #openstack-nova | 19:00 | |
| *** jichen has joined #openstack-nova | 19:01 | |
| *** tongli has quit IRC | 19:01 | |
| *** _diana_ has joined #openstack-nova | 19:01 | |
| *** tongli has joined #openstack-nova | 19:02 | |
| *** agireud has joined #openstack-nova | 19:02 | |
| *** signed8bit has joined #openstack-nova | 19:02 | |
| *** Longgeek has joined #openstack-nova | 19:02 | |
| _gryf | hey guys, is there anything special to do to make XenProject CI to pass? | 19:05 |
| dansmith | rlrossit: okay, sorry, just got off an epic call | 19:06 |
| bauzas | _gryf: try a xen:recheck | 19:06 |
| *** tongli has quit IRC | 19:06 | |
| bauzas | hum | 19:06 |
| mriedem | _gryf: assuming you aren't making changes related to xen | 19:06 |
| bauzas | 'xen: recheck' (with a space) | 19:06 |
| mriedem | recheck is fine | 19:06 |
| _gryf | bauzas, thanks | 19:06 |
| rlrossit | dansmith: was there like rock music going on during it? | 19:07 |
| *** nihilifer has joined #openstack-nova | 19:07 | |
| dansmith | rlrossit: no, dreary end-of-the-world music | 19:07 |
| mriedem | emo | 19:07 |
| *** tongli has joined #openstack-nova | 19:07 | |
| dansmith | rlrossit: so, I see stuff in the backlog, but ... tell me again what you're looking for? | 19:08 |
| dansmith | ah, I have this patch open | 19:08 |
| dansmith | ctrath: we wouldn't check these sorts of things in pep8 generally | 19:09 |
| dansmith | ctrath: we have a trap in place to make sure that they get updated, but it was missing list objects until rlrossit fixed it last week | 19:09 |
| ctrath | dansmith: ok. sounds good… thanks. | 19:10 |
| rlrossit | dansmith: but even with my change it wasn't getting caught... | 19:10 |
| dansmith | rlrossit: it will only get the new changes | 19:10 |
| dansmith | rlrossit: it doesn't check full history | 19:10 |
| dansmith | because it can't | 19:10 |
| ctrath | dansmith: but, in my example that I provided in the bug, the latest must not be checked | 19:11 |
| rlrossit | was just about to say the same thing ctrath | 19:11 |
| *** tongli has quit IRC | 19:12 | |
| dansmith | okay, let me look again | 19:12 |
| *** browne has joined #openstack-nova | 19:14 | |
| *** piyanai has quit IRC | 19:17 | |
| *** _diana_ has quit IRC | 19:18 | |
| *** nihilifer has quit IRC | 19:18 | |
| mriedem | jaypipes: bauzas: when you were going through https://review.openstack.org/#/c/76234/ did DiskBus types of uml or lxc come up? | 19:19 |
| mriedem | since those are checked here http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/blockinfo.py#n112 | 19:19 |
| *** piyanai has joined #openstack-nova | 19:19 | |
| bauzas | mriedem: well, honestly, we probably missed that | 19:20 |
| bauzas | mriedem: just because it's awfully difficult to see if the object provides all the needed fields | 19:20 |
| bauzas | mriedem: for example, I had to provide an update for the object here | 19:20 |
| jaypipes | mriedem: that's a great catch. | 19:20 |
| bauzas | mriedem: https://review.openstack.org/#/c/197911/1 | 19:21 |
| mriedem | jaypipes: bauzas: np, it's big and hairy, i just noticed it, so was wondering if there was a reason it was left out (if intentional) | 19:21 |
| mriedem | if not, i'll push a change to add those fields to the enum | 19:21 |
| bauzas | mriedem: yeah, honestly I dunno | 19:21 |
| bauzas | mriedem: danpb is on vacation for the next 2 weeks | 19:21 |
| bauzas | mriedem: so, maybe a change would be cool for discussing about that | 19:22 |
| mriedem | i was going to say, i was going to ask danpb but i'll just add him to the review | 19:22 |
| mriedem | not sure how the version compat is going to work with adding the new field, but i'll see | 19:22 |
| mriedem | *new enum types | 19:22 |
| *** iamjarvo has quit IRC | 19:22 | |
| jaypipes | mriedem: feel free. | 19:22 |
| bauzas | mriedem: cool | 19:22 |
| bauzas | mriedem: not sure the object is really useds | 19:23 |
| *** _diana_ has joined #openstack-nova | 19:23 | |
| *** piyanai has quit IRC | 19:24 | |
| mriedem | bauzas: well, that's why i was looking, because https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bdm-enum-fields,n,z | 19:24 |
| mriedem | bdm has the disk_bus field also | 19:24 |
| bauzas | mmmm | 19:25 |
| bauzas | mriedem: okay, just provide the new enum fields and we'll see how to have a compat in the change | 19:25 |
| *** iamjarvo has joined #openstack-nova | 19:26 | |
| *** iamjarvo has quit IRC | 19:26 | |
| bauzas | mriedem: honestly, I don't see it really needed to be worked on | 19:26 |
| *** iamjarvo has joined #openstack-nova | 19:26 | |
| bauzas | mriedem: we're just adding 2 options to an enum, I don't see a problem with an RPC interface there | 19:26 |
| mriedem | bauzas: well, if i change the BlockDeviceMapping.disk_bus field to use the DiskBusField type, it's an enum and the enum should be correct for how the code is used | 19:27 |
| bauzas | that I agree :) | 19:27 |
| mriedem | also, the v2.1 api schema checks this stuff and right now it let's in any string - if we have the ALL tuple correct in our enum field objects, we can use those in the v2.1 api schema | 19:27 |
| *** piyanai has joined #openstack-nova | 19:27 | |
| *** Longgeek has quit IRC | 19:27 | |
| mriedem | which will avoid the ValueError and resulting 500 in the code later :) | 19:28 |
| bauzas | mriedem: well, I have the same problem with the reqspec BP | 19:28 |
| bauzas | mriedem: where the API validation can be different from the object coercing | 19:28 |
| bauzas | mriedem: sure | 19:28 |
| *** mutoulbj has quit IRC | 19:30 | |
| *** dims_ has joined #openstack-nova | 19:36 | |
| dansmith | rlrossit: ctrath: I see what you're saying | 19:36 |
| dansmith | it's not detecting the incomplete mapping in obj_relationships on floating | 19:37 |
| *** annashen has quit IRC | 19:37 | |
| dansmith | but nothing is complaining because the static version in test_objects is right | 19:37 |
| ctrath | dansmith: great. I'm assuming this should warrant another bug? I'm new to this, so just wondering | 19:37 |
| dansmith | ctrath: well, there should certainly be a bug that floating is wrong.. not sure if that's filed already | 19:37 |
| *** lpetrut has joined #openstack-nova | 19:38 | |
| *** annashen has joined #openstack-nova | 19:38 | |
| dansmith | ctrath: there's also a bug that we either need a test (or have one that is broken) to detect that floating is wrong | 19:38 |
| rlrossit | dansmith: I've been wondering if there is a way to not have the map in test_objects... is there a way we can take the "live" relationships of the objects? | 19:38 |
| dansmith | the fix for both is probably the same (write or fix a test, and then fix the broken mapping) | 19:38 |
| *** tongli has joined #openstack-nova | 19:38 | |
| *** whenry_ has joined #openstack-nova | 19:39 | |
| dansmith | rlrossit: that's the test I'm talking about writing (or fixing) .. just making sure the models are internally consistent | 19:39 |
| *** Sukhdev has joined #openstack-nova | 19:40 | |
| dansmith | rlrossit: the hash test is required so that we can tell when something has changed. I'm not sure we can get the relationships test to work without the static model and still detect changes, but if so, feel free to write/prove it | 19:40 |
| dansmith | the static bit is mostly there as a trap for a developer | 19:40 |
| dansmith | like, double-entry accounting | 19:40 |
| dansmith | you change something here, you have to change it over there | 19:40 |
| dansmith | in doing so, if you forget to change it in the tests, then we throw an error and say "Update me, but first, make sure you did A, B, and C steps" | 19:41 |
| *** whenry_ has quit IRC | 19:41 | |
| dansmith | before these tests, | 19:41 |
| dansmith | people would bump one version and not the other, not realize they needed to do both, | 19:41 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: Introducing new forced_down field for a Service object https://review.openstack.org/185280 | 19:41 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: New nova API call to mark nova-compute down https://review.openstack.org/184086 | 19:41 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: Service group drivers forced_down flag utilization https://review.openstack.org/197063 | 19:41 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: Refactoring of the os-services module https://review.openstack.org/185279 | 19:41 |
| dansmith | reviewers wouldn't notice, etc | 19:41 |
| dansmith | which is how the floating/fixed bug got into the tree.. there wasn't a test alerting folks to the need to do the bump | 19:41 |
| dansmith | for whatever is causing that to not be right | 19:41 |
| *** annashen has quit IRC | 19:42 | |
| rlrossit | so it's a further improvement on mriedem's change here https://review.openstack.org/#/c/197684/ ? | 19:43 |
| *** angdraug has joined #openstack-nova | 19:43 | |
| rlrossit | actually changing the test instead of just changing the log message | 19:43 |
| *** tongli has quit IRC | 19:44 | |
| dansmith | if it's not catching something, yeah | 19:45 |
| dansmith | rlrossit: maybe tweak some versions locally and figure out a pattern | 19:45 |
| dansmith | rlrossit: historically this has worked fairly well, except for the list object problem | 19:46 |
| dansmith | rlrossit: so either we totally broke it somehow, or there is a pattern | 19:46 |
| dansmith | so maybe bump some version on another object, see if it's detected, | 19:46 |
| *** _diana_ has quit IRC | 19:46 | |
| dansmith | and then try to figure out why the current broken fixed/floating thing isn't being checked | 19:46 |
| dansmith | also, maybe look at the existing (older) child_versions tests | 19:46 |
| *** Nikolay_St has joined #openstack-nova | 19:46 | |
| dansmith | maybe there is a test case that is checking child_versions and not obj_relationships | 19:46 |
| *** andreykurilin__ has quit IRC | 19:47 | |
| rlrossit | maybe I should rebase my fixed IP change on my obj_relationships change because I bump fixed IP in there but don't get caught. | 19:47 |
| *** haypo has joined #openstack-nova | 19:48 | |
| openstackgerrit | jichenjc proposed openstack/nova: Check flavor type before add tenant access https://review.openstack.org/124338 | 19:48 |
| *** Piet has joined #openstack-nova | 19:48 | |
| *** andreykurilin__ has joined #openstack-nova | 19:49 | |
| _gryf | dansmith, hey | 19:49 |
| *** _diana_ has joined #openstack-nova | 19:49 | |
| dansmith | _gryf: hi | 19:49 |
| _gryf | dansmith, could you please take a quick look on my review? | 19:50 |
| dansmith | which one? | 19:50 |
| *** haypo has left #openstack-nova | 19:50 | |
| _gryf | dansmith, those four I've uploaded just now (starting from the https://review.openstack.org/185279) | 19:51 |
| dansmith | you said "quick" and "review" .. now it's all four? :) | 19:51 |
| dansmith | I need to get back to those, but not sure it's going to happen today | 19:52 |
| * dansmith was up super early and is fading fast | 19:52 | |
| *** Marga_ has quit IRC | 19:52 | |
| * _gryf duck ;) | 19:52 | |
| _gryf | dansmith, okay | 19:52 |
| *** _gryf is now known as _gryf|afk | 19:55 | |
| *** _diana_ has quit IRC | 19:55 | |
| *** whenry_ has joined #openstack-nova | 19:56 | |
| *** _diana_ has joined #openstack-nova | 20:00 | |
| *** jasondotstar has quit IRC | 20:00 | |
| *** Marga_ has joined #openstack-nova | 20:03 | |
| *** _diana_ has quit IRC | 20:06 | |
| *** Marga_ has quit IRC | 20:06 | |
| *** annashen has joined #openstack-nova | 20:07 | |
| *** Marga_ has joined #openstack-nova | 20:07 | |
| *** _diana_ has joined #openstack-nova | 20:07 | |
| *** yamahata has quit IRC | 20:08 | |
| *** yamahata has joined #openstack-nova | 20:08 | |
| *** _diana_ has quit IRC | 20:08 | |
| *** exploreshaifali has quit IRC | 20:10 | |
| *** promulo__ has joined #openstack-nova | 20:15 | |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Add test case for suspend https://review.openstack.org/198880 | 20:16 |
| *** promulo has quit IRC | 20:16 | |
| *** dims_ has quit IRC | 20:17 | |
| *** Mike_D_laptop has joined #openstack-nova | 20:19 | |
| *** piyanai has quit IRC | 20:20 | |
| *** piyanai has joined #openstack-nova | 20:22 | |
| *** Longgeek has joined #openstack-nova | 20:24 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify filters for using RequestSpec object https://review.openstack.org/197912 | 20:25 |
| *** packet has quit IRC | 20:25 | |
| *** annegentle has quit IRC | 20:27 | |
| openstackgerrit | Marian Horban proposed openstack/nova: Switch to oslo.service library https://review.openstack.org/192900 | 20:28 |
| lifeless | dansmith: virtualenv vendors pip | 20:30 |
| dansmith | lifeless: eh? | 20:30 |
| lifeless | dansmith: tox -epep8 depends on a recent virtualenv - e.g. 13.1.0 (preferred, it has pip 7.1) | 20:30 |
| lifeless | dansmith: it also depends on a recent pip for the sdist stage which runs outside the virtualenvs | 20:31 |
| lifeless | dansmith: w.r.t. 03:08 < dansmith> ValueError: ('Expected version spec in', "MySQL-python;python_version=='2.7'", 'at', ";python_version=='2.7'") | 20:31 |
| dansmith | I have pip 7.1 ... are you saying I need a newer virtualenv too? | 20:31 |
| *** nihilifer has joined #openstack-nova | 20:32 | |
| *** Marga__ has joined #openstack-nova | 20:32 | |
| *** Marga__ has quit IRC | 20:32 | |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Add unit tests for methods https://review.openstack.org/198880 | 20:32 |
| *** Marga_ has quit IRC | 20:33 | |
| *** Marga__ has joined #openstack-nova | 20:33 | |
| *** ZZelle has joined #openstack-nova | 20:36 | |
| *** Marga__ has quit IRC | 20:38 | |
| *** ZZelle has quit IRC | 20:38 | |
| *** vladikr has quit IRC | 20:38 | |
| *** Marga_ has joined #openstack-nova | 20:38 | |
| dansmith | lifeless: ^ | 20:39 |
| lifeless | dansmith: thats my first guess based on the backscroll yes. | 20:39 |
| lifeless | dansmith: tox -epep8 makes a new venv; that has the bundled pip from whatever release of virtualenv you have. | 20:39 |
| openstackgerrit | Eric Brown proposed openstack/nova: Replace openssl calls with cryptography lib https://review.openstack.org/198246 | 20:39 |
| lifeless | dansmith: that will then fail on the ';' if you have e.g. pip 1.5.4 or something inside the virtualenv | 20:39 |
| * dansmith tries | 20:40 | |
| lifeless | dansmith: easy enough to validate, after it fails do . .tox/pep8/bin/activate, then pip --version | 20:40 |
| dansmith | I just pip-installed virtualenv | 20:40 |
| dansmith | I think it's taking longer than it was, so maybe good | 20:40 |
| lifeless | :) | 20:40 |
| lifeless | My long term goal is to get to a point where all these things can be expressed as metadata | 20:41 |
| lifeless | so you get useful errors | 20:41 |
| lifeless | but we're a ways out from that | 20:41 |
| dansmith | useful errors would be good | 20:41 |
| dansmith | shouting your name flanked with expletives didn't seem to fix it | 20:42 |
| lifeless | clearly I'm not scary enough yet. I shall work on it. | 20:42 |
| *** jichen has quit IRC | 20:44 | |
| *** tongli has joined #openstack-nova | 20:44 | |
| *** bradjones is now known as bradjones|away | 20:44 | |
| *** ZZelle has joined #openstack-nova | 20:45 | |
| *** ildikov has joined #openstack-nova | 20:45 | |
| *** ivasev has quit IRC | 20:46 | |
| *** tongli has quit IRC | 20:48 | |
| clarkb | I wonder if virtualenv could stop bundling pip and setuptools | 20:49 |
| *** tongli has joined #openstack-nova | 20:49 | |
| *** Longgeek has quit IRC | 20:50 | |
| dansmith | yeah, that seems super weird to me | 20:50 |
| *** Marga_ has quit IRC | 20:50 | |
| dansmith | lifeless: that worked, thanks | 20:50 |
| *** thedodd has quit IRC | 20:51 | |
| *** Mike_D_laptop has quit IRC | 20:52 | |
| rlrossit | dansmith: figured it out https://bugs.launchpad.net/nova/+bug/1471978 | 20:56 |
| openstack | Launchpad bug 1471978 in OpenStack Compute (nova) "test_relationships() uses subobject version instead of relationship version" [Undecided,New] | 20:56 |
| rlrossit | the subobject version that test_relationships was testing against is always the most recent version of the subobject, not what's being held in obj_relationships | 20:57 |
| dansmith | rlrossit: well, hmm | 20:58 |
| *** piyanai has quit IRC | 20:58 | |
| rlrossit | I got a fix to that bug in a quick-and-dirty branch right now and I figured out what needs to happen with it | 20:58 |
| dansmith | I guess I'll wait and see the code.. it needs to look at the current version of the object to build the current tree, so it can diff it against stored | 20:58 |
| dansmith | but if you think that's right, I'll try to break it when you get a patch up and see | 20:59 |
| rlrossit | isn't the "current" version what is held in the static tree in test_objects? | 20:59 |
| dansmith | but if so, excellent :) | 20:59 |
| dansmith | no | 20:59 |
| dansmith | current is the calculated version based on the objects we're looking at | 20:59 |
| dansmith | stored is the one we manually persist | 20:59 |
| *** amotoki has joined #openstack-nova | 20:59 | |
| rlrossit | ok yeah, that's what my patch does | 21:00 |
| dansmith | maybe I'm misreading which you mean when you say "obj_relationships" | 21:00 |
| openstackgerrit | Marian Horban proposed openstack/nova: libvirt:Rsync remote FS driver was added https://review.openstack.org/146887 | 21:00 |
| *** thangp_afk is now known as thangp | 21:00 | |
| dansmith | since the mapping and the per-object one are called the same | 21:00 |
| *** eharney has quit IRC | 21:01 | |
| *** ociuhandu has joined #openstack-nova | 21:01 | |
| *** annashen has quit IRC | 21:05 | |
| *** rfolco has quit IRC | 21:05 | |
| *** annegentle has joined #openstack-nova | 21:05 | |
| *** annashen has joined #openstack-nova | 21:06 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: block storage support https://review.openstack.org/184652 | 21:07 |
| *** thorst has quit IRC | 21:07 | |
| jroll | alaski: ^ pewpew, that was a fun adventure into context and objects. | 21:07 |
| *** lpetrut has quit IRC | 21:08 | |
| cfriesen | so would anyone object to a patch that bumps MAX_FUNC to 0xff (in nova/pci/devspec.py) The current value of 0x7 seems to severely limit the number of VFs per PF. | 21:09 |
| *** dims_ has joined #openstack-nova | 21:09 | |
| *** annashen has quit IRC | 21:10 | |
| cfriesen | jaypipes: bauzas: any comments on this? ^ | 21:13 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix test_security_groups to use Instance object properly https://review.openstack.org/198765 | 21:13 |
| openstackgerrit | Dan Smith proposed openstack/nova: Remove legacy flavor compatibility code from Instance https://review.openstack.org/197771 | 21:13 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix Instance object usage in test_extended_ips tests https://review.openstack.org/198767 | 21:13 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix test_extended_server_attributes for proper Instance object usage https://review.openstack.org/198766 | 21:13 |
| *** dims_ has quit IRC | 21:14 | |
| *** annashen has joined #openstack-nova | 21:14 | |
| *** thangp has quit IRC | 21:14 | |
| bauzas | cfriesen: too late for me, my brain is fried :) | 21:14 |
| alaski | jroll: heh. we found another fun issue along the same lines with info_cache | 21:15 |
| jroll | alaski: yeah, that was unexpected :/ | 21:15 |
| openstackgerrit | Joel Coffman proposed openstack/nova: libvirt: Disconnect dm-crypt on instance suspend/stop https://review.openstack.org/141485 | 21:16 |
| *** dims has joined #openstack-nova | 21:16 | |
| bauzas | cfriesen: honestly wondering if it could be a conf opt | 21:16 |
| *** aysyd has quit IRC | 21:17 | |
| jroll | alaski: on the plus side, I found devstack+ironic+cells mostly works | 21:17 |
| alaski | jroll: awesome | 21:18 |
| *** piyanai has joined #openstack-nova | 21:18 | |
| *** ctrath has quit IRC | 21:19 | |
| melwitt | alaski: I'm going to update the db connection switching spec and I was wondering, will we be storing the cell db connection strings somewhere? API DB in cell mappings? just thinking about it since we don't want to use that in RequestContext | 21:20 |
| *** neiljerram_bb has joined #openstack-nova | 21:20 | |
| *** artom has quit IRC | 21:20 | |
| alaski | melwitt: yes, it's in the cell_mappings table | 21:21 |
| melwitt | alaski: cool, thanks | 21:21 |
| flwang | mriedem: ping | 21:22 |
| flwang | mriedem: dansmith: could you revisit the nova backup spec https://review.openstack.org/190919 ? thanks | 21:23 |
| *** ociuhandu has quit IRC | 21:25 | |
| *** jasondotstar has joined #openstack-nova | 21:25 | |
| melwitt | alaski: I'm wondering whether I have to be specific about what key we'll store in the RequestContext's db_connection field right now. first simple idea would be a hash of the connection string but I don't know if that would be considered not secure enough. thinking being that we'll need a way to check whether we've created the wanted EngineFacade before or not | 21:26 |
| *** piyanai has quit IRC | 21:27 | |
| alaski | dansmith: question for you on https://review.openstack.org/#/c/183354/ | 21:27 |
| alaski | melwitt: I don't think you need to be real specific at this point, since it might change during the code review. Just specific enough to explain the basic flow of operations | 21:29 |
| dansmith | alaski: well, it's not completed until it's cleaned up from the source host, or at least that's what I was going for | 21:29 |
| dansmith | alaski: it could be completed-awaiting-cleanup or something | 21:29 |
| dansmith | alaski: but yeah, if the evacuation fails (even now) then the instance is kinda homeless until it's re-evacuated or rebuilt to try again | 21:30 |
| dansmith | alaski: unless we have something babysitting that process, I don't think we want anything more complicated though | 21:30 |
| alaski | dansmith: yeah. there's really two statues we care about | 21:30 |
| alaski | *statuses | 21:31 |
| *** fawadkhaliq has joined #openstack-nova | 21:31 | |
| dansmith | alaski: right, so you want them to be called something else? | 21:31 |
| alaski | I'm not sure yet | 21:31 |
| alaski | here's my other concern | 21:31 |
| alaski | sometimes we lose a host and don't bring it back up, so there may never be a cleanup | 21:32 |
| dansmith | sure, there's nothing really wrong with that | 21:32 |
| dansmith | alaski: if we had a nova-manage remove-host $host command, then it could remove those tombstones | 21:33 |
| dansmith | but we don't, right? | 21:33 |
| alaski | not afaik | 21:33 |
| dansmith | right, so there's lots of residue left from that situation right now, AFAIK | 21:33 |
| rlrossit | dansmith: my patch is finding all of the different versioning right now and it's really annoying because I just want to get this pushed up :/ | 21:33 |
| dansmith | alaski: leaving the migration tombstone won't prevent further evacuations | 21:33 |
| dansmith | rlrossit: all of the different versioning? | 21:33 |
| rlrossit | dansmith: the static tree in test_objects is different from a few of the object relationships in some of classes (FixedIP and Instance are the main culprits) | 21:34 |
| dansmith | rlrossit: well, if there really are broken links and you're finding them then that's good :) | 21:35 |
| lifeless | clarkb: not really | 21:35 |
| alaski | dansmith: I just checked other migrations, and they finish with 'confirmed'. my concern was having an overloaded status that meant different things based on migration type. but that's not the case so I retract my concern | 21:35 |
| lifeless | clarkb: without pip and setuptools (and wheel now) inside virtualenv, a new virtualenv has no way to install anything | 21:35 |
| *** jecarey_ has quit IRC | 21:36 | |
| lifeless | clarkb: and we don't have a sanely copy-in-able version of those tools outside the environment | 21:36 |
| dansmith | alaski: well, confirmed would be fine for this if you prefer it, but it wouldn't make sense for live migration I don't think | 21:36 |
| dansmith | alaski: so at least that is going to be different | 21:36 |
| *** iamjarvo has quit IRC | 21:36 | |
| dansmith | alaski: I was just thinking that the evacuation is still technically in progress until we get the ack that it's cleaned up | 21:36 |
| alaski | dansmith: I don't want this to be confirmed :) | 21:36 |
| dansmith | alaski: especially because we've left customer data on the sending host until that cleanup | 21:36 |
| dansmith | okay | 21:36 |
| alaski | I was expecting that the movement of the instance would affect the status of the migration, not the cleanup. though that does need to be indicated somehow | 21:37 |
| dansmith | alaski: there is a later patch that does that | 21:38 |
| dansmith | (If I get what you're saying) | 21:38 |
| alaski | dansmith: okay, I'll look. I'm also saying that if cleanup is going to affect status, I want it to be distinct from a movement status | 21:39 |
| dansmith | oh | 21:39 |
| dansmith | alaski: if we do that we're going to need another field in the DB :( | 21:39 |
| *** Sukhdev has quit IRC | 21:40 | |
| alaski | dansmith: yeah. I'm fine with completed for now, since it is distince from confirmed | 21:40 |
| alaski | *distinct | 21:40 |
| dansmith | alaski: okay, well, this has me thinking about maybe changing the later patch | 21:40 |
| *** promulo has joined #openstack-nova | 21:41 | |
| dansmith | alaski: I think this patch is internally and externally consistent (and plugs the gaping hole), | 21:41 |
| dansmith | but I think maybe I want to tweak some things about the later patch based on this | 21:41 |
| *** promulo__ has quit IRC | 21:41 | |
| *** radek__ has quit IRC | 21:41 | |
| dansmith | so if you think we should do another field for ack'd or something, we could do that in the (or a) later patch and make resize/migrate also set that field | 21:41 |
| *** eglynn-afk has quit IRC | 21:42 | |
| *** Sukhdev has joined #openstack-nova | 21:42 | |
| dansmith | because the running status of where and how it's being moved is probably not the right thing to be checking for whether we've acked/handled/whatever the cleanup | 21:43 |
| alaski | yeah. I think that would be good, but could be added later | 21:44 |
| dansmith | yeah | 21:45 |
| alaski | does any of that affect what you want to do in this review? | 21:45 |
| *** dboik has quit IRC | 21:46 | |
| dansmith | I dunno, but since this works to defeat the data-loss issue, I'd /like/ to get this merged.. and refine the other bits in subsequent patches | 21:46 |
| *** ianbrown has joined #openstack-nova | 21:46 | |
| *** ianbrown has quit IRC | 21:46 | |
| dansmith | but if you want to wait until I come up with a plan, that's okay too | 21:46 |
| *** Longgeek has joined #openstack-nova | 21:47 | |
| *** yamahata has quit IRC | 21:49 | |
| alaski | no this is fine. you had mentioned changing later patches, just making sure you didn't have second thoughts here before I +w | 21:50 |
| dansmith | no, I think this addresses the issue for the moment and doesn't conflict with what I'm thinking about doing | 21:50 |
| dansmith | the other one has a -1 on it anyway for other tweaks | 21:51 |
| *** ijw has quit IRC | 21:51 | |
| dansmith | I'll get going on this and that in the morning | 21:51 |
| dansmith | this spec has really had bad scope creep, | 21:51 |
| dansmith | but since I named it "robustify evacuation" everything seems to fall into the robustification part | 21:51 |
| dansmith | next time I will call it "make-evacuate-better-in-this-very-specific-way" :P | 21:52 |
| *** fawadkhaliq has quit IRC | 21:52 | |
| alaski | heh, I was just typing that but more slowly | 21:52 |
| dansmith | heh | 21:52 |
| dansmith | also, ndipanov wants to define an actual state machine for these things, and start with the migration object | 21:53 |
| dansmith | which seems like something you'd like | 21:53 |
| dansmith | such that we always end up in the same terminal state or something | 21:53 |
| *** dannywilson has joined #openstack-nova | 21:53 | |
| alaski | that sounds great | 21:53 |
| dansmith | so maybe we should leave status as free-form-ish and introduce a new one called "state" that has to follow the FSM | 21:53 |
| dansmith | and do it that way | 21:53 |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Change List objects to use obj_relationships https://review.openstack.org/198730 | 21:54 |
| alaski | yeah, using status for that seems like an exercise in frustration | 21:54 |
| dansmith | yeah | 21:55 |
| dansmith | well, his reason for wanting to do it was the fact that we have a lot of fuzziness there | 21:55 |
| *** neiljerram_bb has quit IRC | 21:55 | |
| alaski | if we keep adding more strict and well defined things to Nova I'm not sure how we'll cope with the reduction in bugs | 21:55 |
| *** eglynn-afk has joined #openstack-nova | 21:55 | |
| dansmith | I'll punch a hole with a rusty nail and let it get infected | 21:56 |
| dansmith | that'll fill the void | 21:56 |
| openstackgerrit | melanie witt proposed openstack/nova-specs: Database connection switching for cells https://review.openstack.org/194935 | 21:57 |
| alaski | dansmith: heh | 21:57 |
| *** BadCub is now known as BadCub-away | 21:58 | |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Fix test_relationships() for subobject versions https://review.openstack.org/198904 | 21:58 |
| rlrossit | dansmith: ^ I got it up, but I need to head out. Take a look if you want, or we can discuss tomorrow | 21:59 |
| rlrossit | the latter lets you be lazy about it :) | 21:59 |
| *** nihilifer has quit IRC | 21:59 | |
| alaski | dansmith: oh. just remembered that there's a deprecated_for_removal parameter you can use for https://review.openstack.org/#/c/183354/10/nova/utils.py,cm | 22:00 |
| alaski | dansmith: http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/ironic/driver.py#n85 | 22:00 |
| *** Sukhdev has quit IRC | 22:01 | |
| *** Sukhdev_ has joined #openstack-nova | 22:02 | |
| *** rlrossit has left #openstack-nova | 22:02 | |
| *** Kennan has joined #openstack-nova | 22:02 | |
| dansmith | oh, I just saw that in another context | 22:03 |
| dansmith | didn't realize | 22:03 |
| dansmith | I'll change | 22:03 |
| *** Kennan2 has quit IRC | 22:03 | |
| *** _diana_ has joined #openstack-nova | 22:03 | |
| *** burt has quit IRC | 22:04 | |
| *** tonytan4ever has quit IRC | 22:04 | |
| dansmith | alaski: oh, you already +Wd | 22:04 |
| dansmith | alaski: mind if I sneak it in after? | 22:05 |
| alaski | dansmith: that's fine, it's a minor thing | 22:05 |
| alaski | dansmith: just would be good to add it | 22:05 |
| dansmith | also, what does it give us? some sort of warning in the sample output or something? | 22:05 |
| mriedem | warning in the logs when it's used | 22:05 |
| mriedem | on startup | 22:05 |
| mriedem | so you want to use it | 22:06 |
| *** Longgeek has quit IRC | 22:06 | |
| dansmith | oh sweet | 22:06 |
| *** ildikov has quit IRC | 22:06 | |
| *** gibi has quit IRC | 22:06 | |
| *** Sukhdev_ has quit IRC | 22:06 | |
| mriedem | https://github.com/openstack/oslo.config/blob/master/oslo_config/cfg.py#L747 | 22:07 |
| *** mdrabe has quit IRC | 22:09 | |
| dansmith | hmm, that change is failing postgres | 22:09 |
| *** Sukhdev has joined #openstack-nova | 22:10 | |
| dansmith | oh ffs | 22:10 |
| dansmith | how is mysql ignoring this? | 22:11 |
| *** yamahata has joined #openstack-nova | 22:12 | |
| dansmith | alaski: well, good news is I'll be able to get that deprecated flag in :D | 22:16 |
| alaski | huzzah | 22:16 |
| dansmith | the db model enum had "evacuation" and the object/code had "evacuate" | 22:17 |
| dansmith | and mysql seems to not care at all :/ | 22:17 |
| *** jasondotstar has quit IRC | 22:17 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Make evacuate update the migration object along the way https://review.openstack.org/194373 | 22:20 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add some notifications to the evacuate path https://review.openstack.org/194372 | 22:20 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make evacuate leave a record for the source compute host to process https://review.openstack.org/183354 | 22:20 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix incorrect enum in Migration object https://review.openstack.org/198907 | 22:20 |
| dansmith | mriedem: that last one is trivial to review, by the way :P | 22:20 |
| dansmith | free review stats | 22:20 |
| mriedem | gotta run | 22:21 |
| mriedem | i'll find a reason to -1 tomorrow | 22:21 |
| *** mriedem is now known as mriedem_away | 22:23 | |
| *** signed8__ has joined #openstack-nova | 22:24 | |
| dansmith | good luck, because it's perfect | 22:24 |
| *** signed8bit has quit IRC | 22:25 | |
| *** zzzeek has quit IRC | 22:25 | |
| alaski | dansmith: well that's annoying. you could update the db model in that review as well | 22:28 |
| dansmith | alaski: oh oops | 22:28 |
| dansmith | and here I said it was perfect :D | 22:28 |
| alaski | it's all in how you define it | 22:29 |
| dansmith | heh | 22:29 |
| *** hogepodge has quit IRC | 22:30 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Make evacuate update the migration object along the way https://review.openstack.org/194373 | 22:31 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add some notifications to the evacuate path https://review.openstack.org/194372 | 22:31 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix incorrect enum in Migration object and DB model https://review.openstack.org/198907 | 22:31 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make evacuate leave a record for the source compute host to process https://review.openstack.org/183354 | 22:31 |
| dansmith | melwitt: fixed now :) | 22:33 |
| *** tongli has quit IRC | 22:33 | |
| *** tongli has joined #openstack-nova | 22:34 | |
| * melwitt looks | 22:35 | |
| *** eglynn-afk has quit IRC | 22:35 | |
| *** _diana_ has quit IRC | 22:35 | |
| *** tongli has quit IRC | 22:37 | |
| *** jasondotstar has joined #openstack-nova | 22:37 | |
| *** andreykurilin__ has quit IRC | 22:37 | |
| *** tongli has joined #openstack-nova | 22:37 | |
| *** ijuwang has quit IRC | 22:37 | |
| *** annegentle has quit IRC | 22:40 | |
| melwitt | dansmith: maybe this is dumb but where is the schema actually defined? I thought it was by the models. or are you saying that since the db migration added it as "evacuation" the model should have been based on that? | 22:40 |
| dansmith | melwitt: the migration defines the schema, | 22:40 |
| dansmith | melwitt: and what is in the migration is what is in everyone | 22:40 |
| dansmith | 's actual DB that has deployed master | 22:41 |
| dansmith | so it's the right version, unless we change it, which would be another migration (and silly) | 22:41 |
| melwitt | dansmith: got it | 22:41 |
| *** signed8__ is now known as signed8bit_ZZZzz | 22:41 | |
| *** tongli has quit IRC | 22:42 | |
| *** annegentle has joined #openstack-nova | 22:44 | |
| *** lucas-dinner has quit IRC | 22:45 | |
| alaski | there is a test to compare models to migration results, but it must not handle enums | 22:46 |
| *** hogepodge has joined #openstack-nova | 22:46 | |
| *** andreykurilin__ has joined #openstack-nova | 22:46 | |
| andreykurilin__ | melwitt: hi | 22:47 |
| melwitt | andreykurilin__: hi | 22:47 |
| *** edmondsw has quit IRC | 22:47 | |
| dansmith | alaski: yeah, apparently not | 22:48 |
| *** tongli has joined #openstack-nova | 22:48 | |
| andreykurilin__ | melwitt: I need review for my patches related to support microversion in novaclient and dansmith recommended you. Can you help me? | 22:49 |
| melwitt | andreykurilin__: yes, I've been intending to review those. I'm not very expert about the microversioning but my hope is the specs and existing documents will be enough for me to review them meaningfully | 22:51 |
| openstackgerrit | Tracy Jones proposed openstack/nova: Cleanup of compute api reboot method https://review.openstack.org/127427 | 22:51 |
| melwitt | andreykurilin__: so I'll go through them this week | 22:52 |
| andreykurilin__ | melwitt: got it, thanks. | 22:53 |
| *** tongli has quit IRC | 22:53 | |
| *** browne has quit IRC | 22:54 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: block storage support https://review.openstack.org/184652 | 22:54 |
| andreykurilin__ | melwitt: I'll try to be online 27/7, so if you will have some questions - just ping me:) | 22:54 |
| *** piyanai has joined #openstack-nova | 22:54 | |
| *** dims has quit IRC | 22:54 | |
| *** tongli has joined #openstack-nova | 22:57 | |
| melwitt | andreykurilin__: no worries on that, I'll put comments on the review if I have questions and I get alerts whenever there are responses | 22:59 |
| *** amotoki has quit IRC | 22:59 | |
| *** jskinner has quit IRC | 23:01 | |
| *** asettle has joined #openstack-nova | 23:02 | |
| *** annegentle has quit IRC | 23:02 | |
| *** tongli has quit IRC | 23:02 | |
| andreykurilin__ | melwitt: I worry a bit, because first patch-set of first patch in the chain was published at February 3 :( | 23:04 |
| melwitt | andreykurilin__: I understand. back then I think there wasn't enough documentation of microversioning in general, for me to be able to do a proper review. in those cases we were waiting on API experts like oomichi and alex_xu. now, I think there is enough documentation explaining about how a client should use microversions, so I can review the code based on that | 23:07 |
| melwitt | andreykurilin__: thanks for being patient and doing the hard work. I want to get them merged soon, we need the support in novaclient | 23:08 |
| andreykurilin__ | melwitt: yeah... OpenStack documentation can be a bottleneck, sometimes:) | 23:09 |
| andreykurilin__ | melwitt: thank you for feedback. Now I can stop disturb you:) Have a nice day | 23:11 |
| *** jamielennox is now known as jamielennox|away | 23:11 | |
| *** piyanai_ has joined #openstack-nova | 23:12 | |
| melwitt | andreykurilin__: thanks for the reminder on the patches, they're on my list | 23:12 |
| *** Piet has quit IRC | 23:13 | |
| *** achanda has joined #openstack-nova | 23:14 | |
| *** piyanai has quit IRC | 23:15 | |
| *** piyanai_ is now known as piyanai | 23:15 | |
| *** piyanai has quit IRC | 23:16 | |
| *** annegentle has joined #openstack-nova | 23:18 | |
| *** tongli has joined #openstack-nova | 23:18 | |
| *** jaypipes has quit IRC | 23:18 | |
| tjones1 | dansmith: i added those tests you were asking about https://review.openstack.org/#/c/127427/ | 23:19 |
| *** annegentle has quit IRC | 23:20 | |
| *** jamielennox|away is now known as jamielennox | 23:22 | |
| *** tongli has quit IRC | 23:23 | |
| *** slong has joined #openstack-nova | 23:23 | |
| *** thorst has joined #openstack-nova | 23:24 | |
| *** ZZelle has quit IRC | 23:26 | |
| *** thorst has quit IRC | 23:30 | |
| *** achanda has quit IRC | 23:30 | |
| *** Longgeek has joined #openstack-nova | 23:32 | |
| *** agireud has quit IRC | 23:34 | |
| *** asettle has quit IRC | 23:36 | |
| gmann | johnthetubaguy: Thanks, sounds good to me now | 23:37 |
| *** RuiChen has quit IRC | 23:38 | |
| *** browne has joined #openstack-nova | 23:38 | |
| *** piyanai has joined #openstack-nova | 23:38 | |
| *** RuiChen has joined #openstack-nova | 23:40 | |
| *** emagana has quit IRC | 23:41 | |
| *** angdraug has quit IRC | 23:45 | |
| *** achanda has joined #openstack-nova | 23:45 | |
| *** annashen has quit IRC | 23:45 | |
| *** tjones1 has quit IRC | 23:46 | |
| *** annashen has joined #openstack-nova | 23:46 | |
| *** Longgeek has quit IRC | 23:50 | |
| *** annashen has quit IRC | 23:51 | |
| *** achanda has quit IRC | 23:51 | |
| *** zzzeek has joined #openstack-nova | 23:55 | |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova: Ironic: block storage support https://review.openstack.org/184652 | 23:56 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Merge availability_zone tests between v2 and v2.1 https://review.openstack.org/171017 | 23:58 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Merge server password tests between v2 and v2.1 https://review.openstack.org/197435 | 23:58 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Modified testscenario for micro version 2.4 https://review.openstack.org/197394 | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!