*** Jeffrey4l has joined #openstack-nova | 00:00 | |
*** haomaiwang has quit IRC | 00:01 | |
*** haomaiwang has joined #openstack-nova | 00:01 | |
*** Sukhdev has quit IRC | 00:03 | |
*** rk4n has quit IRC | 00:04 | |
*** mylu has joined #openstack-nova | 00:06 | |
*** yamahata has joined #openstack-nova | 00:07 | |
*** y_sawai has quit IRC | 00:07 | |
*** john51 has quit IRC | 00:10 | |
*** diogogmt has quit IRC | 00:10 | |
*** john51 has joined #openstack-nova | 00:11 | |
*** yonglihe has quit IRC | 00:12 | |
*** sdague has quit IRC | 00:15 | |
*** eil397 has joined #openstack-nova | 00:16 | |
*** annasort has joined #openstack-nova | 00:18 | |
*** ljxiash has joined #openstack-nova | 00:18 | |
*** nic has quit IRC | 00:21 | |
*** ljxiash has quit IRC | 00:23 | |
*** zenoway has joined #openstack-nova | 00:23 | |
*** Sukhdev has joined #openstack-nova | 00:24 | |
*** ivase_ has quit IRC | 00:24 | |
*** ivase has joined #openstack-nova | 00:25 | |
*** Sukhdev has quit IRC | 00:27 | |
*** zenoway has quit IRC | 00:28 | |
openstackgerrit | ShaoHe Feng proposed openstack/nova: Add new APIs and deprecate old API for migrations https://review.openstack.org/258771 | 00:29 |
---|---|---|
*** ivase_ has joined #openstack-nova | 00:31 | |
*** qeelee has joined #openstack-nova | 00:31 | |
*** ivase has quit IRC | 00:31 | |
*** bjornar__ has quit IRC | 00:32 | |
*** Sukhdev has joined #openstack-nova | 00:35 | |
*** ivase has joined #openstack-nova | 00:36 | |
*** ivase_ has quit IRC | 00:37 | |
openstackgerrit | Jay Pipes proposed openstack/nova: Use new inventory schema in compute_node_get_all() https://review.openstack.org/284484 | 00:38 |
jaypipes | bauzas, dansmith, alaski: ^^ | 00:38 |
*** mylu has quit IRC | 00:42 | |
*** ivase_ has joined #openstack-nova | 00:42 | |
*** ivase has quit IRC | 00:43 | |
*** Sukhdev has quit IRC | 00:44 | |
*** zz_dimtruck is now known as dimtruck | 00:45 | |
*** zenoway has joined #openstack-nova | 00:48 | |
*** ivase has joined #openstack-nova | 00:48 | |
*** ildikov has quit IRC | 00:48 | |
*** y_sawai has joined #openstack-nova | 00:48 | |
*** ivase_ has quit IRC | 00:49 | |
*** y_sawai_ has joined #openstack-nova | 00:51 | |
*** ildikov has joined #openstack-nova | 00:52 | |
*** zenoway has quit IRC | 00:52 | |
*** y_sawai has quit IRC | 00:53 | |
*** ivase_ has joined #openstack-nova | 00:54 | |
*** mylu has joined #openstack-nova | 00:54 | |
*** ivase has quit IRC | 00:55 | |
*** y_sawai_ has quit IRC | 00:56 | |
*** delatte has joined #openstack-nova | 00:58 | |
*** ivase has joined #openstack-nova | 01:00 | |
*** dimtruck is now known as zz_dimtruck | 01:00 | |
*** ivase_ has quit IRC | 01:00 | |
*** delattec has quit IRC | 01:00 | |
*** haomaiwang has quit IRC | 01:01 | |
*** Sukhdev has joined #openstack-nova | 01:01 | |
*** haomaiwang has joined #openstack-nova | 01:01 | |
*** eil397 has quit IRC | 01:03 | |
*** ctrath has joined #openstack-nova | 01:03 | |
*** ivase_ has joined #openstack-nova | 01:05 | |
*** ivase has quit IRC | 01:06 | |
*** jdillaman has quit IRC | 01:08 | |
*** Sukhdev has quit IRC | 01:10 | |
*** ivase has joined #openstack-nova | 01:11 | |
*** ivase_ has quit IRC | 01:12 | |
*** mylu has quit IRC | 01:13 | |
*** y_sawai has joined #openstack-nova | 01:14 | |
*** ivase_ has joined #openstack-nova | 01:17 | |
*** aorourke is now known as aorourke-afk | 01:17 | |
openstackgerrit | Sean Dague proposed openstack/nova: complete the removal of api_version from rest client parameters https://review.openstack.org/284257 | 01:18 |
*** ivase has quit IRC | 01:18 | |
*** EinstCrazy has joined #openstack-nova | 01:20 | |
*** ducttape_ has joined #openstack-nova | 01:20 | |
*** jdillaman has joined #openstack-nova | 01:21 | |
openstackgerrit | Sean Dague proposed openstack/nova: fix usage of opportunistic test cases with enginefacade https://review.openstack.org/283728 | 01:22 |
openstackgerrit | Sean Dague proposed openstack/nova: add regression test for bug #1538011 https://review.openstack.org/283364 | 01:22 |
openstack | bug 1538011 in OpenStack Compute (nova) "Aggregate metadata keys are not case sensitive" [High,In progress] https://launchpad.net/bugs/1538011 - Assigned to Augustina Ragwitz (auggy) | 01:22 |
openstackgerrit | Sean Dague proposed openstack/nova: add regression test for bug #1541691 https://review.openstack.org/283261 | 01:22 |
openstack | bug 1541691 in OpenStack Compute (nova) "server boot with leading and trailing white spaces in name, displays weird error message" [High,In progress] https://launchpad.net/bugs/1541691 - Assigned to Alex Xu (xuhj) | 01:22 |
*** haomaiwang has quit IRC | 01:22 | |
*** ivase has joined #openstack-nova | 01:23 | |
openstackgerrit | Sean Dague proposed openstack/nova: reduce pep8 requirements to just hacking https://review.openstack.org/267097 | 01:23 |
*** ivase_ has quit IRC | 01:23 | |
*** jaypipes has quit IRC | 01:24 | |
*** vladikr has quit IRC | 01:27 | |
*** ivase_ has joined #openstack-nova | 01:28 | |
*** thorst has joined #openstack-nova | 01:28 | |
*** ivase has quit IRC | 01:29 | |
*** ducttape_ has quit IRC | 01:30 | |
*** mtanino has quit IRC | 01:32 | |
*** zhangjn has quit IRC | 01:32 | |
*** abitha has quit IRC | 01:32 | |
*** Guest35189 has quit IRC | 01:32 | |
*** zhangyufei has joined #openstack-nova | 01:33 | |
*** mutiwu has joined #openstack-nova | 01:33 | |
*** thorst has quit IRC | 01:34 | |
*** mutiwu is now known as Guest3981 | 01:34 | |
*** ivase has joined #openstack-nova | 01:34 | |
*** thorst has joined #openstack-nova | 01:34 | |
*** ivase_ has quit IRC | 01:35 | |
*** edtubill has joined #openstack-nova | 01:36 | |
*** zhangjn has joined #openstack-nova | 01:37 | |
*** ctrath1 has joined #openstack-nova | 01:37 | |
*** gongysh has joined #openstack-nova | 01:38 | |
*** thorst has quit IRC | 01:39 | |
*** ivase_ has joined #openstack-nova | 01:40 | |
*** ctrath has quit IRC | 01:41 | |
*** ivase has quit IRC | 01:41 | |
*** ljxiash has joined #openstack-nova | 01:44 | |
*** vladikr has joined #openstack-nova | 01:44 | |
*** zzzeek has quit IRC | 01:45 | |
*** zzzeek has joined #openstack-nova | 01:45 | |
*** ivase has joined #openstack-nova | 01:46 | |
*** ivase_ has quit IRC | 01:46 | |
*** smatzek__ has joined #openstack-nova | 01:47 | |
*** tangchen_ has joined #openstack-nova | 01:47 | |
*** zzzeek has quit IRC | 01:47 | |
tangchen_ | Hi alex_xu, would you please help to take a look at this patch ? https://review.openstack.org/#/c/280120/ | 01:47 |
*** Guest75 has quit IRC | 01:47 | |
*** Guest3981 has quit IRC | 01:48 | |
*** mutiwu_ has joined #openstack-nova | 01:48 | |
*** zzzeek has joined #openstack-nova | 01:50 | |
*** ivase_ has joined #openstack-nova | 01:51 | |
*** dims has quit IRC | 01:52 | |
*** zzzeek has quit IRC | 01:52 | |
*** ivase has quit IRC | 01:52 | |
*** zzzeek has joined #openstack-nova | 01:54 | |
*** tangchen_ has quit IRC | 01:57 | |
*** ivase has joined #openstack-nova | 01:57 | |
*** ivase_ has quit IRC | 01:58 | |
*** haomaiwang has joined #openstack-nova | 01:59 | |
*** haomaiwang has quit IRC | 02:01 | |
*** haomaiwang has joined #openstack-nova | 02:01 | |
*** zhangyufei has quit IRC | 02:01 | |
*** ijw has quit IRC | 02:02 | |
*** zenoway has joined #openstack-nova | 02:02 | |
*** ducttape_ has joined #openstack-nova | 02:03 | |
*** ivase_ has joined #openstack-nova | 02:03 | |
*** ivase has quit IRC | 02:04 | |
*** Sukhdev has joined #openstack-nova | 02:05 | |
*** diogogmt has joined #openstack-nova | 02:05 | |
*** Guest75 has joined #openstack-nova | 02:06 | |
*** zenoway has quit IRC | 02:08 | |
*** Sukhdev has quit IRC | 02:09 | |
*** ivase has joined #openstack-nova | 02:09 | |
*** mutiwu_ has quit IRC | 02:09 | |
*** ivase_ has quit IRC | 02:10 | |
*** lyanchih has joined #openstack-nova | 02:11 | |
openstackgerrit | ShaoHe Feng proposed openstack/nova: Add new APIs and deprecate old API for migrations https://review.openstack.org/258771 | 02:12 |
*** andreykurilin__ has quit IRC | 02:13 | |
*** tangchen has joined #openstack-nova | 02:15 | |
*** ivase_ has joined #openstack-nova | 02:15 | |
*** tangchen has quit IRC | 02:15 | |
*** ivase has quit IRC | 02:15 | |
*** erlon has quit IRC | 02:16 | |
claudiub | mikal: hello. Sorry to bug you, but if you have one free moment to review a 13 LOC patch, it would be awesome. :D https://review.openstack.org/#/c/258614/5 | 02:17 |
*** gongysh has quit IRC | 02:18 | |
*** penick has quit IRC | 02:19 | |
*** ivase has joined #openstack-nova | 02:21 | |
*** ivase_ has quit IRC | 02:21 | |
*** Guest75 has quit IRC | 02:22 | |
*** vilobhmm11 has quit IRC | 02:24 | |
*** hoangcx has joined #openstack-nova | 02:25 | |
*** vilobhmm11 has joined #openstack-nova | 02:25 | |
*** houming has joined #openstack-nova | 02:26 | |
*** ivase_ has joined #openstack-nova | 02:26 | |
*** ivase has quit IRC | 02:27 | |
*** Bjoern_ has joined #openstack-nova | 02:27 | |
*** armax has joined #openstack-nova | 02:28 | |
*** amotoki has joined #openstack-nova | 02:29 | |
*** zenoway has joined #openstack-nova | 02:30 | |
*** shaohe_feng_ has quit IRC | 02:30 | |
*** zz_dimtruck is now known as dimtruck | 02:32 | |
*** ivase has joined #openstack-nova | 02:32 | |
*** ivase_ has quit IRC | 02:33 | |
*** vilobhmm11 has quit IRC | 02:35 | |
*** zenoway has quit IRC | 02:35 | |
*** vilobhmm11 has joined #openstack-nova | 02:36 | |
*** Guest75 has joined #openstack-nova | 02:36 | |
*** vilobhmm11 has quit IRC | 02:37 | |
*** tojuvone_ has quit IRC | 02:37 | |
*** ivase_ has joined #openstack-nova | 02:38 | |
*** ivase has quit IRC | 02:38 | |
mikal | claudiub: I'm in a meeting, but I'll try to get to it today | 02:39 |
claudiub | sure, thanks. :) | 02:39 |
*** zhangyufei has joined #openstack-nova | 02:42 | |
*** ducttape_ has quit IRC | 02:43 | |
*** ducttape_ has joined #openstack-nova | 02:43 | |
*** ivase has joined #openstack-nova | 02:43 | |
*** ivase_ has quit IRC | 02:44 | |
*** tojuvone has joined #openstack-nova | 02:44 | |
*** yamahata has quit IRC | 02:46 | |
*** links has joined #openstack-nova | 02:47 | |
*** smatzek__ has quit IRC | 02:49 | |
*** ivase_ has joined #openstack-nova | 02:49 | |
*** ivase has quit IRC | 02:51 | |
*** apoorvad has quit IRC | 02:52 | |
*** apoorvad has joined #openstack-nova | 02:53 | |
*** zenoway has joined #openstack-nova | 02:54 | |
*** ivase has joined #openstack-nova | 02:55 | |
*** ivase_ has quit IRC | 02:55 | |
*** gongysh has joined #openstack-nova | 02:56 | |
*** apoorvad has quit IRC | 02:57 | |
*** y_sawai has quit IRC | 02:57 | |
*** y_sawai has joined #openstack-nova | 02:58 | |
*** edtubill has quit IRC | 02:59 | |
*** zenoway has quit IRC | 02:59 | |
*** haomaiwang has quit IRC | 03:01 | |
*** ivase_ has joined #openstack-nova | 03:01 | |
*** haomaiwang has joined #openstack-nova | 03:01 | |
*** ivase has quit IRC | 03:02 | |
*** gcb has quit IRC | 03:02 | |
*** lyanchih has quit IRC | 03:03 | |
*** gcb has joined #openstack-nova | 03:03 | |
*** liusheng has left #openstack-nova | 03:04 | |
*** boris-42 has quit IRC | 03:04 | |
*** ctrath1 has quit IRC | 03:06 | |
*** ducttape_ has quit IRC | 03:06 | |
*** ivase has joined #openstack-nova | 03:07 | |
*** edmondsw has quit IRC | 03:07 | |
*** gcb has quit IRC | 03:07 | |
*** ivase_ has quit IRC | 03:07 | |
*** gcb has joined #openstack-nova | 03:08 | |
*** lyanchih has joined #openstack-nova | 03:10 | |
*** ducttape_ has joined #openstack-nova | 03:12 | |
*** ivase_ has joined #openstack-nova | 03:12 | |
*** ivase has quit IRC | 03:13 | |
*** Guest75 has quit IRC | 03:16 | |
*** lyanchih has quit IRC | 03:16 | |
*** ivase has joined #openstack-nova | 03:18 | |
*** zenoway has joined #openstack-nova | 03:18 | |
*** ivase_ has quit IRC | 03:19 | |
*** amotoki has quit IRC | 03:19 | |
*** lyanchih has joined #openstack-nova | 03:23 | |
*** zenoway has quit IRC | 03:23 | |
*** diogogmt has quit IRC | 03:23 | |
*** ivase_ has joined #openstack-nova | 03:25 | |
*** ivase has quit IRC | 03:25 | |
*** diogogmt has joined #openstack-nova | 03:26 | |
*** ducttape_ has quit IRC | 03:26 | |
*** salv-orlando has joined #openstack-nova | 03:27 | |
*** salv-orlando has quit IRC | 03:28 | |
*** amotoki has joined #openstack-nova | 03:30 | |
*** ivase has joined #openstack-nova | 03:30 | |
*** salv-orl_ has quit IRC | 03:30 | |
*** baoli_ has quit IRC | 03:31 | |
*** ivase_ has quit IRC | 03:31 | |
*** links has quit IRC | 03:34 | |
*** amotoki has quit IRC | 03:35 | |
*** ivase_ has joined #openstack-nova | 03:36 | |
*** ivase has quit IRC | 03:36 | |
*** ljxiash has quit IRC | 03:38 | |
*** ljxiash has joined #openstack-nova | 03:38 | |
*** vladikr has quit IRC | 03:38 | |
*** baoli has joined #openstack-nova | 03:40 | |
*** amotoki has joined #openstack-nova | 03:40 | |
*** ducttape_ has joined #openstack-nova | 03:41 | |
*** ivase has joined #openstack-nova | 03:41 | |
*** ivase_ has quit IRC | 03:42 | |
*** baoli has quit IRC | 03:42 | |
*** zenoway has joined #openstack-nova | 03:43 | |
*** baoli has joined #openstack-nova | 03:43 | |
*** ljxiash has quit IRC | 03:43 | |
*** toure_ is now known as toure | 03:43 | |
*** diogogmt has quit IRC | 03:43 | |
*** y_sawai has quit IRC | 03:45 | |
*** y_sawai has joined #openstack-nova | 03:45 | |
*** zenoway has quit IRC | 03:47 | |
*** baoli has quit IRC | 03:48 | |
*** Bjoern_ has quit IRC | 03:51 | |
*** vladikr has joined #openstack-nova | 03:51 | |
*** ducttape_ has quit IRC | 03:53 | |
*** mingyu has quit IRC | 03:53 | |
*** mingyu has joined #openstack-nova | 03:53 | |
*** claudiub has quit IRC | 03:54 | |
*** ducttape_ has joined #openstack-nova | 03:56 | |
*** zenoway has joined #openstack-nova | 03:56 | |
*** qeelee has quit IRC | 04:00 | |
*** zenoway has quit IRC | 04:00 | |
*** haomaiwang has quit IRC | 04:01 | |
*** amotoki has quit IRC | 04:01 | |
*** haomaiwa_ has joined #openstack-nova | 04:01 | |
*** hoangcx has quit IRC | 04:02 | |
*** mingyu has quit IRC | 04:02 | |
*** hoangcx has joined #openstack-nova | 04:03 | |
*** mingyu has joined #openstack-nova | 04:04 | |
*** mubirru has joined #openstack-nova | 04:05 | |
*** shyama has joined #openstack-nova | 04:10 | |
*** zenoway has joined #openstack-nova | 04:10 | |
*** amotoki has joined #openstack-nova | 04:13 | |
*** zenoway has quit IRC | 04:15 | |
*** ducttape_ has quit IRC | 04:24 | |
openstackgerrit | Dongcan Ye proposed openstack/nova: VMware: Fix getting wrong image cahe folder https://review.openstack.org/284529 | 04:25 |
*** gcb has quit IRC | 04:28 | |
*** gcb has joined #openstack-nova | 04:28 | |
*** ljxiash has joined #openstack-nova | 04:28 | |
*** salv-orlando has joined #openstack-nova | 04:29 | |
*** links has joined #openstack-nova | 04:29 | |
*** ducttape_ has joined #openstack-nova | 04:29 | |
*** haomaiwa_ has quit IRC | 04:30 | |
*** _RuiChen has quit IRC | 04:32 | |
*** RuiChen has joined #openstack-nova | 04:33 | |
*** ljxiash has quit IRC | 04:34 | |
*** ljxiash has joined #openstack-nova | 04:34 | |
*** achanda has quit IRC | 04:35 | |
*** zenoway has joined #openstack-nova | 04:36 | |
*** ducttape_ has quit IRC | 04:37 | |
*** lyanchih has quit IRC | 04:37 | |
*** ljxiash has quit IRC | 04:39 | |
*** salv-orlando has quit IRC | 04:39 | |
*** haomaiwa_ has joined #openstack-nova | 04:39 | |
*** Marga__ has joined #openstack-nova | 04:40 | |
*** zenoway has quit IRC | 04:41 | |
*** Marga__ has quit IRC | 04:44 | |
*** tangchen has joined #openstack-nova | 04:44 | |
*** Marga_ has quit IRC | 04:44 | |
*** mingyu has quit IRC | 04:49 | |
*** mingyu has joined #openstack-nova | 04:51 | |
*** armax has quit IRC | 04:51 | |
*** mingyu has quit IRC | 04:52 | |
*** y_sawai_ has joined #openstack-nova | 04:53 | |
*** y_sawai has quit IRC | 04:54 | |
*** achanda has joined #openstack-nova | 04:55 | |
*** Marga_ has joined #openstack-nova | 04:56 | |
*** qeelee has joined #openstack-nova | 04:59 | |
*** Marga_ has quit IRC | 05:00 | |
*** haomaiwa_ has quit IRC | 05:01 | |
*** zenoway has joined #openstack-nova | 05:01 | |
*** Marga_ has joined #openstack-nova | 05:01 | |
*** haomaiwang has joined #openstack-nova | 05:01 | |
*** Marga_ has quit IRC | 05:01 | |
*** Marga_ has joined #openstack-nova | 05:01 | |
*** Marga_ has quit IRC | 05:02 | |
*** Marga_ has joined #openstack-nova | 05:02 | |
*** y_sawai_ has quit IRC | 05:02 | |
*** dimtruck is now known as zz_dimtruck | 05:03 | |
*** y_sawai has joined #openstack-nova | 05:04 | |
*** garyk has joined #openstack-nova | 05:05 | |
*** zenoway has quit IRC | 05:05 | |
*** yonglihe has joined #openstack-nova | 05:05 | |
*** qeelee has quit IRC | 05:10 | |
*** belmoreira has joined #openstack-nova | 05:12 | |
*** ljxiash has joined #openstack-nova | 05:14 | |
*** Fdaisuke has quit IRC | 05:15 | |
*** mingyu has joined #openstack-nova | 05:16 | |
*** mingyu has quit IRC | 05:16 | |
*** Fdaisuke has joined #openstack-nova | 05:17 | |
*** shyama has quit IRC | 05:24 | |
*** shyama_ has joined #openstack-nova | 05:24 | |
*** shyama_ is now known as shyama | 05:24 | |
*** zenoway has joined #openstack-nova | 05:25 | |
*** lyanchih has joined #openstack-nova | 05:26 | |
*** xiexs has quit IRC | 05:28 | |
*** zenoway has quit IRC | 05:29 | |
*** mingyu has joined #openstack-nova | 05:30 | |
*** mingyu has quit IRC | 05:31 | |
*** mingyu has joined #openstack-nova | 05:31 | |
*** tangchen has quit IRC | 05:31 | |
*** y_sawai_ has joined #openstack-nova | 05:32 | |
*** y_sawai has quit IRC | 05:33 | |
*** irenab has quit IRC | 05:43 | |
*** y_sawai_ has quit IRC | 05:44 | |
*** liverpooler has quit IRC | 05:44 | |
*** y_sawai has joined #openstack-nova | 05:45 | |
*** zenoway has joined #openstack-nova | 05:49 | |
*** qeelee has joined #openstack-nova | 05:49 | |
*** tangchen has joined #openstack-nova | 05:52 | |
*** dims has joined #openstack-nova | 05:52 | |
*** zenoway has quit IRC | 05:53 | |
openstackgerrit | ShaoHe Feng proposed openstack/nova: Add new APIs and deprecate old API for migrations https://review.openstack.org/258771 | 05:57 |
*** nkrinner has joined #openstack-nova | 05:58 | |
*** salv-orlando has joined #openstack-nova | 05:58 | |
*** doug-fish has joined #openstack-nova | 06:00 | |
*** haomaiwang has quit IRC | 06:01 | |
*** haomaiwang has joined #openstack-nova | 06:01 | |
*** edtubill has joined #openstack-nova | 06:03 | |
*** gongysh has quit IRC | 06:03 | |
*** salv-orlando has quit IRC | 06:03 | |
*** ekuris has joined #openstack-nova | 06:03 | |
*** harlowja_at_home has quit IRC | 06:11 | |
*** achinm has joined #openstack-nova | 06:11 | |
*** harlowja_at_home has joined #openstack-nova | 06:13 | |
*** tangchen has quit IRC | 06:14 | |
*** zenoway has joined #openstack-nova | 06:15 | |
*** n0ano has quit IRC | 06:15 | |
*** y_sawai has quit IRC | 06:16 | |
*** y_sawai has joined #openstack-nova | 06:17 | |
*** gcb has quit IRC | 06:18 | |
*** zenoway has quit IRC | 06:21 | |
*** y_sawai has quit IRC | 06:25 | |
*** y_sawai has joined #openstack-nova | 06:26 | |
*** gcb has joined #openstack-nova | 06:30 | |
*** harlowja_at_home has quit IRC | 06:33 | |
*** harlowja_at_home has joined #openstack-nova | 06:33 | |
*** aagrawal has joined #openstack-nova | 06:37 | |
*** irenab has joined #openstack-nova | 06:37 | |
*** edtubill has quit IRC | 06:38 | |
*** tyagiprince has joined #openstack-nova | 06:41 | |
*** zenoway has joined #openstack-nova | 06:43 | |
*** dims has quit IRC | 06:43 | |
*** mrhillsman has quit IRC | 06:45 | |
*** achanda_ has joined #openstack-nova | 06:45 | |
*** dave-mccowan has quit IRC | 06:45 | |
*** achanda has quit IRC | 06:46 | |
*** zenoway has quit IRC | 06:47 | |
*** ljxiash has quit IRC | 06:47 | |
*** ivase_ has joined #openstack-nova | 06:48 | |
*** josecastroleon has joined #openstack-nova | 06:49 | |
*** ivase has quit IRC | 06:49 | |
*** fawadkhaliq has joined #openstack-nova | 06:50 | |
*** qeelee_ has joined #openstack-nova | 06:50 | |
*** qeelee has quit IRC | 06:50 | |
*** qeelee_ is now known as qeelee | 06:50 | |
*** hoangcx_ has joined #openstack-nova | 06:51 | |
*** zenoway has joined #openstack-nova | 06:52 | |
*** hoangcx has quit IRC | 06:52 | |
*** belmoreira has quit IRC | 06:53 | |
*** doug-fish has quit IRC | 06:54 | |
*** ivase has joined #openstack-nova | 06:54 | |
openstackgerrit | Chris Friesen proposed openstack/nova: check max_net_count against min_count when booting https://review.openstack.org/283861 | 06:54 |
*** harlowja_at_home has quit IRC | 06:54 | |
*** ljxiash has joined #openstack-nova | 06:55 | |
*** ivase_ has quit IRC | 06:55 | |
*** gongysh has joined #openstack-nova | 06:56 | |
*** zzzeek has quit IRC | 06:56 | |
*** zzzeek has joined #openstack-nova | 06:59 | |
*** ivase_ has joined #openstack-nova | 07:00 | |
*** haomaiwang has quit IRC | 07:01 | |
*** ivase has quit IRC | 07:01 | |
*** haomaiwang has joined #openstack-nova | 07:01 | |
*** salv-orlando has joined #openstack-nova | 07:02 | |
openstackgerrit | Bartosz Górski proposed openstack/python-novaclient: Adds missing internationalization for help message https://review.openstack.org/284458 | 07:05 |
*** codebauss has joined #openstack-nova | 07:06 | |
*** ivase has joined #openstack-nova | 07:06 | |
*** ivase_ has quit IRC | 07:07 | |
*** salv-orlando has quit IRC | 07:07 | |
*** codebauss is now known as mrhillsman | 07:07 | |
*** mjura has joined #openstack-nova | 07:10 | |
*** zzzeek has quit IRC | 07:10 | |
openstackgerrit | Feng Xi Yan proposed openstack/nova: <VMWARE> Add power state cache for host init and power state sync. https://review.openstack.org/242006 | 07:11 |
*** ivase_ has joined #openstack-nova | 07:12 | |
*** ivase has quit IRC | 07:13 | |
*** paul-carlton1 has joined #openstack-nova | 07:14 | |
*** tomoiaga has joined #openstack-nova | 07:16 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove APIRouterV3 https://review.openstack.org/284559 | 07:17 |
*** ivase has joined #openstack-nova | 07:17 | |
openstackgerrit | garyk proposed openstack/nova: VMware: remove deprecation warnings from oslo_versionedobjects https://review.openstack.org/284286 | 07:18 |
*** ivase_ has quit IRC | 07:18 | |
*** zenoway has quit IRC | 07:19 | |
*** lpetrut has joined #openstack-nova | 07:19 | |
*** zenoway has joined #openstack-nova | 07:19 | |
*** tomoiaga has quit IRC | 07:21 | |
*** ivase_ has joined #openstack-nova | 07:23 | |
*** zenoway has quit IRC | 07:24 | |
*** scheuran has joined #openstack-nova | 07:24 | |
*** ivase has quit IRC | 07:24 | |
*** zenoway has joined #openstack-nova | 07:24 | |
*** salv-orlando has joined #openstack-nova | 07:25 | |
*** salv-orl_ has joined #openstack-nova | 07:26 | |
*** salv-orlando has quit IRC | 07:26 | |
*** zhangyufei has quit IRC | 07:27 | |
*** mingyu has quit IRC | 07:28 | |
*** ivase_ has quit IRC | 07:29 | |
*** ivase has joined #openstack-nova | 07:30 | |
*** ociuhandu has joined #openstack-nova | 07:30 | |
*** ociuhandu has quit IRC | 07:31 | |
*** zhangyufei has joined #openstack-nova | 07:32 | |
*** ociuhandu has joined #openstack-nova | 07:32 | |
*** cfriesen has quit IRC | 07:32 | |
*** liverpooler has joined #openstack-nova | 07:32 | |
*** ivase_ has joined #openstack-nova | 07:35 | |
*** ljxiash_ has joined #openstack-nova | 07:35 | |
*** ljxiash has quit IRC | 07:36 | |
*** ivase has quit IRC | 07:36 | |
*** tyagiprince1 has joined #openstack-nova | 07:36 | |
*** tyagiprince has quit IRC | 07:38 | |
*** tyagiprince1 is now known as tyagiprince | 07:38 | |
*** ivase has joined #openstack-nova | 07:41 | |
*** ivase_ has quit IRC | 07:41 | |
*** zul has quit IRC | 07:41 | |
*** ivase_ has joined #openstack-nova | 07:46 | |
*** ivase has quit IRC | 07:47 | |
*** zul has joined #openstack-nova | 07:47 | |
*** zhangyufei has quit IRC | 07:48 | |
*** ociuhandu has quit IRC | 07:49 | |
*** mpavone has joined #openstack-nova | 07:49 | |
*** ljxiash_ has quit IRC | 07:49 | |
*** belmoreira has joined #openstack-nova | 07:51 | |
*** fawadkhaliq has quit IRC | 07:51 | |
*** ivase has joined #openstack-nova | 07:52 | |
*** ljxiash has joined #openstack-nova | 07:52 | |
*** ivase_ has quit IRC | 07:54 | |
*** moshele has joined #openstack-nova | 07:57 | |
*** belmoreira has quit IRC | 07:58 | |
*** ivase_ has joined #openstack-nova | 07:58 | |
*** ivase has quit IRC | 07:58 | |
*** belmoreira has joined #openstack-nova | 08:00 | |
*** haomaiwang has quit IRC | 08:01 | |
*** haomaiwa_ has joined #openstack-nova | 08:01 | |
*** ivase_ has quit IRC | 08:04 | |
*** ivase has joined #openstack-nova | 08:04 | |
*** rdopiera has joined #openstack-nova | 08:05 | |
*** belmoreira has quit IRC | 08:06 | |
*** matrohon has joined #openstack-nova | 08:08 | |
*** ivase_ has joined #openstack-nova | 08:09 | |
*** ivase has quit IRC | 08:10 | |
*** rk4n has joined #openstack-nova | 08:11 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove unused methods in nova/utils.py https://review.openstack.org/284580 | 08:13 |
*** lpetrut has quit IRC | 08:13 | |
*** ivase has joined #openstack-nova | 08:15 | |
*** ivase_ has quit IRC | 08:16 | |
*** erno has joined #openstack-nova | 08:17 | |
*** Oku_OS has quit IRC | 08:18 | |
*** browne has quit IRC | 08:19 | |
*** y_sawai has quit IRC | 08:19 | |
*** tyagiprince has quit IRC | 08:19 | |
*** rk4n has quit IRC | 08:20 | |
*** y_sawai has joined #openstack-nova | 08:20 | |
*** ivase_ has joined #openstack-nova | 08:21 | |
*** ivase has quit IRC | 08:21 | |
*** rk4n has joined #openstack-nova | 08:22 | |
*** jichen has joined #openstack-nova | 08:23 | |
*** y_sawai has quit IRC | 08:24 | |
*** ivase has joined #openstack-nova | 08:27 | |
*** ivase_ has quit IRC | 08:28 | |
*** lyanchih has quit IRC | 08:29 | |
openstackgerrit | Eli Qiao proposed openstack/nova: Allow block_migration and disk_over_commit to be None https://review.openstack.org/254102 | 08:31 |
openstackgerrit | Eli Qiao proposed openstack/nova: API: Improve os-migrateLive input parameters https://review.openstack.org/259319 | 08:31 |
*** ivase_ has joined #openstack-nova | 08:32 | |
*** ivase has quit IRC | 08:33 | |
*** PaulMurray has joined #openstack-nova | 08:34 | |
PaulMurray | good morning | 08:35 |
*** mingyu has joined #openstack-nova | 08:35 | |
*** ivase_ has quit IRC | 08:38 | |
*** ivase has joined #openstack-nova | 08:39 | |
*** lyanchih has joined #openstack-nova | 08:40 | |
bauzas | good morning Nova | 08:42 |
*** ivase_ has joined #openstack-nova | 08:44 | |
*** ivase has quit IRC | 08:44 | |
*** Mackie2k has joined #openstack-nova | 08:44 | |
*** rk4n has quit IRC | 08:47 | |
*** ivase has joined #openstack-nova | 08:50 | |
*** ivase_ has quit IRC | 08:50 | |
*** ihrachys has joined #openstack-nova | 08:51 | |
*** danpb has joined #openstack-nova | 08:51 | |
*** boris-42 has joined #openstack-nova | 08:52 | |
*** amotoki has quit IRC | 08:53 | |
*** tomoiaga has joined #openstack-nova | 08:53 | |
*** ivase has quit IRC | 08:56 | |
*** ivase_ has joined #openstack-nova | 08:56 | |
*** hoangcx_ has quit IRC | 08:59 | |
*** mnestratov has joined #openstack-nova | 09:00 | |
*** hoangcx has joined #openstack-nova | 09:00 | |
*** achanda_ has quit IRC | 09:00 | |
*** haomaiwa_ has quit IRC | 09:01 | |
*** haomaiwang has joined #openstack-nova | 09:01 | |
*** jistr has joined #openstack-nova | 09:01 | |
*** ivase has joined #openstack-nova | 09:01 | |
*** claudiub has joined #openstack-nova | 09:02 | |
*** ivase_ has quit IRC | 09:03 | |
*** achanda has joined #openstack-nova | 09:04 | |
bauzas | oh snap, I used a wrong channel... | 09:05 |
*** paul-carlton1 has quit IRC | 09:07 | |
*** ivase_ has joined #openstack-nova | 09:07 | |
*** ivase has quit IRC | 09:08 | |
*** mvk has joined #openstack-nova | 09:09 | |
*** yassine__ has joined #openstack-nova | 09:10 | |
*** dedery has joined #openstack-nova | 09:10 | |
*** amotoki has joined #openstack-nova | 09:10 | |
*** Oku_OS has joined #openstack-nova | 09:10 | |
*** samueldmq1 has joined #openstack-nova | 09:11 | |
*** ivase has joined #openstack-nova | 09:13 | |
*** ivase_ has quit IRC | 09:13 | |
PaulMurray | shaohe_feng, hi | 09:15 |
*** samueldmq1 has quit IRC | 09:15 | |
*** ildikov has quit IRC | 09:15 | |
PaulMurray | shaohe_feng, I'm going off line for half an hour, but will be back. Wanted to talk about the in progress migrations | 09:16 |
* PaulMurray AFK 30 mins | 09:16 | |
*** fawadkhaliq has joined #openstack-nova | 09:17 | |
*** achanda has quit IRC | 09:19 | |
*** ivase_ has joined #openstack-nova | 09:19 | |
*** ivase has quit IRC | 09:20 | |
*** claudiub|2 has joined #openstack-nova | 09:20 | |
*** ndipanov has joined #openstack-nova | 09:21 | |
*** achanda has joined #openstack-nova | 09:24 | |
*** claudiub has quit IRC | 09:24 | |
*** ivase has joined #openstack-nova | 09:24 | |
*** ivase_ has quit IRC | 09:25 | |
*** claudiub has joined #openstack-nova | 09:26 | |
*** lucas-hungry is now known as lucasagomes | 09:26 | |
*** paul-carlton1 has joined #openstack-nova | 09:26 | |
*** salv-orlando has joined #openstack-nova | 09:27 | |
*** sudipto has joined #openstack-nova | 09:27 | |
*** PaulMurray has quit IRC | 09:28 | |
*** claudiub|2 has quit IRC | 09:29 | |
*** fawadkhaliq has quit IRC | 09:29 | |
*** achanda has quit IRC | 09:29 | |
*** ivase has quit IRC | 09:30 | |
*** salv-orl_ has quit IRC | 09:30 | |
*** ivase has joined #openstack-nova | 09:30 | |
*** ivase_ has joined #openstack-nova | 09:36 | |
*** ivase has quit IRC | 09:37 | |
jed56 | hello nova team | 09:37 |
jed56 | can you tell me if it is possible to retrieve from "nova" if a particular "nova-compute " is configured with shared storage or block migration ? | 09:37 |
danpb | that's not something we expose right now | 09:38 |
danpb | in fact nova itself is pretty dumb and doesn't really know if it has shared storage or not | 09:38 |
jed56 | so if we want to migrate a VM we need to try first in shared storage and if it failed retry we block migration | 09:39 |
*** ljxiash has quit IRC | 09:40 | |
*** ljxiash has joined #openstack-nova | 09:40 | |
danpb | yeah its pretty sucky situation - basically we expect the cloud admin to remember/know whether they've used shared storage or not and pick the right migraiton option | 09:41 |
*** liverpooler has quit IRC | 09:41 | |
jed56 | do you know if we can have two configurations at the same time : - shared storage and block migration ? | 09:42 |
danpb | in the future we're planning to fix this so nova knows it has shared storage or not, and will automatically do block migration if it needs it | 09:42 |
*** liverpooler has joined #openstack-nova | 09:42 | |
*** ivase has joined #openstack-nova | 09:42 | |
*** mingyu has quit IRC | 09:42 | |
danpb | jed56: sure you can, but it makes life harder for you to predict whether you need block migration or not for any given vm | 09:42 |
danpb | so i'd tend to recommend you pick or the other right now | 09:42 |
*** salv-orlando has quit IRC | 09:43 | |
jed56 | hmm okay | 09:43 |
*** ivase_ has quit IRC | 09:43 | |
johnthetubaguy | jed56: have you seen this spec we are working on: http://specs.openstack.org/openstack/nova-specs/specs/mitaka/approved/making_live_migration_api_friendly.html | 09:43 |
johnthetubaguy | the idea is to add a block_migration = "auto" flag this release | 09:43 |
*** ljxiash_ has joined #openstack-nova | 09:44 | |
*** tyagiprince has joined #openstack-nova | 09:44 | |
jed56 | ahh this is a great news | 09:44 |
jed56 | :) | 09:44 |
*** efoley has joined #openstack-nova | 09:44 | |
johnthetubaguy | jed56: so the initial version might not do exactly what you want, depends on your deployment scenrio | 09:44 |
johnthetubaguy | but as danpb said, we are trying to fix up in Newton our tracking of shared storage, with generic-resource-pools spec | 09:45 |
*** ljxiash has quit IRC | 09:45 | |
*** Mackie2k has quit IRC | 09:45 | |
johnthetubaguy | jed56: for our admins, we only use block_migration=true, so this stops them having to remember the correct value (if the pass false thy just see a failed live-migration, becuase nova says false was a bad value, which is dumb...) | 09:46 |
*** lyanchih has quit IRC | 09:46 | |
johnthetubaguy | but if you have several pools of storage, and you don't pick your destination host, it can all go a bit funny right now, but baby steps forward here | 09:46 |
*** vincentfrancoise has joined #openstack-nova | 09:46 | |
*** ivase_ has joined #openstack-nova | 09:47 | |
jed56 | woah... | 09:48 |
pkholkin_ | dims: hi Dims, what was the question? | 09:48 |
*** ivase has quit IRC | 09:48 | |
*** haomaiw__ has joined #openstack-nova | 09:48 | |
*** haomaiwang has quit IRC | 09:49 | |
jed56 | thanks have a nice day | 09:49 |
*** PaulMurray has joined #openstack-nova | 09:50 | |
* PaulMurray is back | 09:51 | |
*** qeelee has quit IRC | 09:51 | |
*** grassy is now known as davechen_afk | 09:53 | |
*** ivase has joined #openstack-nova | 09:54 | |
*** sfinucan has joined #openstack-nova | 09:54 | |
johnthetubaguy | pkholkin_: I have a feeling he is in the west coast of the US this week | 09:54 |
*** ivase_ has quit IRC | 09:55 | |
johnthetubaguy | PaulMurray: hows live-migrate shaping up? | 09:55 |
pkholkin_ | johnthetubaguy: yes, John, I feel the same) | 09:55 |
pkholkin_ | we had some question, will wait | 09:55 |
pkholkin_ | he* | 09:55 |
PaulMurray | johnthetubaguy, need to get https://review.openstack.org/#/c/258771/32 sorted out today if possible | 09:56 |
PaulMurray | johnthetubaguy, that has abort blocked behind it | 09:56 |
PaulMurray | johnthetubaguy, then a chain of 4 patches for novaclient has just been unblocked by the first patch - needs approval | 09:57 |
johnthetubaguy | Thats is shaohe_feng's patch I guess, just replying to the ML post | 09:57 |
PaulMurray | yes, I was looking at that - but go aheaad - I think there is actually little to do there | 09:58 |
PaulMurray | BTW - the CI failures seem to be trouble in setup - is something going on at the moment? | 09:58 |
PaulMurray | lots of failures in the check queue | 09:58 |
*** ivase_ has joined #openstack-nova | 09:59 | |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Rework policies for virt-driver CPU thread pinning https://review.openstack.org/244198 | 10:00 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Move thread policy features between specs https://review.openstack.org/244857 | 10:00 |
*** ivase has quit IRC | 10:00 | |
johnthetubaguy | PaulMurray: good question, not sure, bauzas usually knows about those issues before I spot them | 10:00 |
*** derekh has joined #openstack-nova | 10:00 | |
*** haomaiw__ has quit IRC | 10:01 | |
*** ildikov has joined #openstack-nova | 10:01 | |
bauzas | johnthetubaguy: PaulMurray: I haven't seen any CI problem AFAIK for today | 10:01 |
bauzas | PaulMurray: link ? | 10:01 |
*** haomaiwang has joined #openstack-nova | 10:01 | |
*** mvk has quit IRC | 10:01 | |
*** lyanchih has joined #openstack-nova | 10:01 | |
* bauzas is just rushing to deliver the last Mitaka bits for check-dests (coldmigrate/resize and rebuild) | 10:02 | |
sfinucan | danpb: mikal: ndipanov: Appreciate that you're all busy today, but it would be awesome if ye could get this bp rework in today. It's technically a doc change, and a change in the admin guide is blocked by it. Please and thank you :) https://review.openstack.org/244857 | 10:02 |
ndipanov | sfinucan, I happen to be rather free today, but no +2 on specs... tough luck :) | 10:04 |
ndipanov | sfinucan, just kidding I am not free at all :( | 10:04 |
ndipanov | sfinucan, btw did you ever get a chance to try the shelve fixes? | 10:04 |
sfinucan | ndipanov: I'm doing that today, funnily enough :) | 10:04 |
*** ivase has joined #openstack-nova | 10:05 | |
sfinucan | spent the week triaging numa/pinning bugs | 10:05 |
*** ivase_ has quit IRC | 10:05 | |
*** efoley has quit IRC | 10:06 | |
johnthetubaguy | sfinucan: did you only drop those "s" bits since mikal added his +2? | 10:06 |
*** efoley_ has joined #openstack-nova | 10:06 | |
*** vincentfrancoise has left #openstack-nova | 10:06 | |
*** hoangcx has quit IRC | 10:06 | |
sfinucan | johnthetubaguy: That, and removed a dependency on a bug that's already been fixed by ndipanov | 10:06 |
*** EinstCrazy has quit IRC | 10:07 | |
* johnthetubaguy scrolls down | 10:07 | |
johnthetubaguy | ah yes | 10:07 |
*** daemontool has joined #openstack-nova | 10:07 | |
*** EinstCrazy has joined #openstack-nova | 10:07 | |
*** ljxiash has joined #openstack-nova | 10:08 | |
*** ljxiash_ has quit IRC | 10:10 | |
*** ivase has quit IRC | 10:11 | |
*** ivase has joined #openstack-nova | 10:11 | |
johnthetubaguy | sfinucan: so that should unblock the doc work, but could you do a follow up to that last one, we probably should add a redirect for that rename: https://review.openstack.org/#/c/244857 | 10:12 |
*** EinstCrazy has quit IRC | 10:12 | |
wznoinsk | hi all | 10:13 |
wznoinsk | could someone have a look at my https://review.openstack.org/#/c/280299/ , it's a fix do api-guide in nova repository, I'm wondering does anybody care or not (cause docs are planned to be somewhere else or else) | 10:14 |
*** rk4n has joined #openstack-nova | 10:16 | |
sfinucan | johnthetubaguy: Sure | 10:16 |
sfinucan | Or should I revert the file rename? | 10:16 |
*** ivase_ has joined #openstack-nova | 10:16 | |
johnthetubaguy | sfinucan: a redirect is fine, I quite like the rename | 10:17 |
johnthetubaguy | sfinucan: we have a file of redirects in place already, so should just be adding an extra one | 10:17 |
*** ivase has quit IRC | 10:18 | |
johnthetubaguy | normally used for the approved->implemented kind of thing | 10:18 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Add redirect for renamed CPU thread policy spec https://review.openstack.org/284632 | 10:18 |
sfinucan | johnthetubaguy: Sweet. Done | 10:19 |
*** qeelee has joined #openstack-nova | 10:19 | |
*** lyanchih has quit IRC | 10:19 | |
johnthetubaguy | sfinucan: sorry, I was meaning put it in this file: https://github.com/openstack/nova-specs/blob/master/specs/mitaka/redirects | 10:20 |
*** daemontool has quit IRC | 10:21 | |
johnthetubaguy | this one has examples in it already: https://github.com/openstack/nova-specs/blob/master/specs/liberty/redirects | 10:21 |
pkoniszewski | alex_xu: hi, there? | 10:21 |
*** ljxiash has quit IRC | 10:21 | |
*** ivase has joined #openstack-nova | 10:22 | |
*** ivase_ has quit IRC | 10:22 | |
PaulMurray | bauzas, this is the one with the failures: https://review.openstack.org/#/c/258771/32 don't want to waste your time, I was just asking because I saw a lot of red in the check queue this monring | 10:23 |
bauzas | PaulMurray: well, that's all the tempest tests failing | 10:24 |
bauzas | PaulMurray: in general, it's only one job which fails | 10:24 |
bauzas | PaulMurray: so my first thought would that be a PEBKAC | 10:24 |
*** daemontool has joined #openstack-nova | 10:24 | |
bauzas | checking logs tho | 10:25 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Add redirect for renamed CPU thread policy spec https://review.openstack.org/284632 | 10:25 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Add redirect for renamed CPU thread policy spec https://review.openstack.org/284632 | 10:25 |
*** tyagiprince1 has joined #openstack-nova | 10:26 | |
*** tyagiprince has quit IRC | 10:26 | |
*** tyagiprince1 is now known as tyagiprince | 10:26 | |
openstackgerrit | Andrey Pavlov proposed openstack/nova: remove EC2 legacy code https://review.openstack.org/266425 | 10:27 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Mark CPU thread pinning as implemented https://review.openstack.org/284633 | 10:27 |
*** ljxiash has joined #openstack-nova | 10:27 | |
openstackgerrit | Sergey Nikitin proposed openstack/nova: DO_NOT_MERGE https://review.openstack.org/284634 | 10:28 |
*** ivase_ has joined #openstack-nova | 10:28 | |
*** ivase has quit IRC | 10:28 | |
*** achanda has joined #openstack-nova | 10:29 | |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Mark CPU thread pinning as implemented https://review.openstack.org/284633 | 10:30 |
openstackgerrit | Stephen Finucane proposed openstack/nova-specs: Add redirect for renamed CPU thread policy spec https://review.openstack.org/284632 | 10:30 |
wznoinsk | alex_xu - would you ahve a moment to review https://review.openstack.org/#/c/280299/ ? | 10:31 |
sfinucan | johnthetubaguy: Third time lucky, I hope :fingerscrossed: | 10:31 |
*** qeelee has quit IRC | 10:31 | |
bauzas | PaulMurray: mmm, interesting http://graphite.openstack.org/render/?from=-200hours&height=500&until=now&width=800&bgcolor=ffffff&fgcolor=000000&yMax=100&yMin=0&vtitle=Failure%20Rate%20in%20Percent&title=Test%20failure%20rates%20over%20last%20200%20hours&&target=lineWidth%28color%28alias%28movingAverage%28asPercent%28stats.zuul.pipeline.check.job.gate-tempest-dsvm-cells.FAILURE,sum%28stats.zuul.pipeline.check.job.gate-tempest-dsv | 10:32 |
bauzas | I probably missed a big problem | 10:32 |
*** ljxiash has quit IRC | 10:32 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Use constant_time_compare from oslo.utils https://review.openstack.org/284635 | 10:33 |
*** ivase has joined #openstack-nova | 10:34 | |
*** achanda has quit IRC | 10:35 | |
*** ivase_ has quit IRC | 10:35 | |
*** mjura has quit IRC | 10:36 | |
*** lyanchih has joined #openstack-nova | 10:38 | |
*** zenoway has quit IRC | 10:39 | |
*** lpetrut has joined #openstack-nova | 10:39 | |
*** zenoway has joined #openstack-nova | 10:39 | |
*** ivase_ has joined #openstack-nova | 10:39 | |
*** ivase has quit IRC | 10:40 | |
PaulMurray | bauzas, do you know what this is: | 10:41 |
PaulMurray | 2016-02-25 06:22:46.660 | + /tmp/ansible/bin/ansible subnodes -f 5 -i /home/jenkins/workspace/gate-tempest-dsvm-full/inventory -m synchronize -a 'mode=pull src='\''/opt/stack/logs/'\'' dest='\''/opt/stack/logs/subnode-{{ host_counter }}'\''' | 10:41 |
PaulMurray | 2016-02-25 06:22:46.660 | ERROR! Specified hosts options do not match any hosts | 10:41 |
bauzas | PaulMurray: that's for collecting the logs AFAIK | 10:41 |
*** pcaruana has joined #openstack-nova | 10:42 | |
bauzas | but the change you pointed me had an issue with devstack not returning correctly | 10:42 |
bauzas | PaulMurray: http://logs.openstack.org/71/258771/32/check/gate-tempest-dsvm-full/30e1b7a/console.html#_2016-02-25_06_22_33_136 | 10:43 |
PaulMurray | bauzas, yes, saw that - been trying to find where it fails | 10:43 |
*** zenoway has quit IRC | 10:44 | |
*** cdent has joined #openstack-nova | 10:45 | |
*** ivase has joined #openstack-nova | 10:45 | |
*** ivase_ has quit IRC | 10:46 | |
bauzas | PaulMurray: that's a $100M question | 10:48 |
PaulMurray | bauzas, the only two patches in the check queue failing in the same way now | 10:48 |
PaulMurray | https://review.openstack.org/259319 | 10:48 |
PaulMurray | https://review.openstack.org/254102 | 10:49 |
*** belmoreira has joined #openstack-nova | 10:49 | |
bauzas | fun | 10:49 |
PaulMurray | I suspect this is happening to all the nova jobs | 10:49 |
bauzas | lemme see logstash | 10:49 |
* bauzas doing 3 things at the same time, so sorry about the log | 10:50 | |
bauzas | lag* | 10:50 |
*** tyagiprince has quit IRC | 10:50 | |
*** tyagiprince has joined #openstack-nova | 10:50 | |
bauzas | I don't remember but we also have a nice dashboard | 10:50 |
johnthetubaguy | belmoreira: did you get chance to review that neutron routed network spec, given the discussion on the ops lists about cells and networking | 10:50 |
kashyap | pkoniszewski: Hi, when you get a moment, might want to rebase this? -- https://review.openstack.org/#/c/168916/ | 10:50 |
bauzas | PaulMurray: yeah I remember now http://status.openstack.org/openstack-health/#/ | 10:51 |
*** ivase_ has joined #openstack-nova | 10:51 | |
*** ivase has quit IRC | 10:51 | |
bauzas | interesting, the last data is from one week... | 10:52 |
bauzas | so back to logstash :D | 10:52 |
*** claudiub has quit IRC | 10:52 | |
pkoniszewski | kashyap: sure, thanks for reminder! :) | 10:53 |
kashyap | pkoniszewski: Thanks for working on it! :-) | 10:53 |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Use constant_time_compare from oslo.utils https://review.openstack.org/284635 | 10:54 |
PaulMurray | bauzas, don't worry - I'm juggling too | 10:56 |
lyarwood | ndipanov: any chance you could take another look at https://review.openstack.org/#/c/266095/ if you have time today? | 10:56 |
bauzas | PaulMurray: so, hard to tell | 10:56 |
bauzas | PaulMurray: I'd say, do a recheck | 10:56 |
PaulMurray | :) | 10:56 |
PaulMurray | if in doubt..... | 10:56 |
*** drankis has joined #openstack-nova | 10:56 | |
PaulMurray | switch it off and on again | 10:56 |
*** ivase has joined #openstack-nova | 10:57 | |
*** ivase_ has quit IRC | 10:57 | |
*** gszasz has joined #openstack-nova | 10:58 | |
*** shaohe_feng_ has joined #openstack-nova | 10:58 | |
*** salv-orlando has joined #openstack-nova | 10:59 | |
*** salv-orlando has quit IRC | 10:59 | |
*** salv-orlando has joined #openstack-nova | 10:59 | |
*** haomaiwang has quit IRC | 11:01 | |
*** haomaiwang has joined #openstack-nova | 11:01 | |
PaulMurray | alex_xu, johnthetubaguy bauzas looking for a +W on https://review.openstack.org/#/c/278454 to get a chain moving - simple novaclient change | 11:01 |
kaisers | Hi Novas! Can somebody pls. point to an explanation what exactly is the difference between nova migrate vs nova live-migration? I'm testing a Cinder impl and just realized i can use nova migrate on a running instance. So what is the difference to live-migration. Only the ability to select the target host? | 11:01 |
johnthetubaguy | PaulMurray: I am just struggling to really understand that stuff | 11:02 |
johnthetubaguy | kaisers: we have a doc on that, let me find the link | 11:02 |
*** efoley__ has joined #openstack-nova | 11:02 | |
kaisers | johnthetubaguy: that would be great. Google found me tons of pages on how to configure things but nothing about the semantics... :) | 11:02 |
johnthetubaguy | kaisers: http://developer.openstack.org/api-guide/compute/server_concepts.html#moving-servers | 11:03 |
kaisers | johnthetubaguy: thanks! | 11:03 |
alex_xu | PaulMurray: added to my review list, will try to review asap | 11:03 |
*** zenoway has joined #openstack-nova | 11:03 | |
johnthetubaguy | kaisers: also this one is a bit shorter and specific about the API: http://developer.openstack.org/api-ref-compute-v2.1.html#os-admin-actions-v2.1 | 11:03 |
PaulMurray | alex_xu, thanks | 11:03 |
pkoniszewski | kaisers: also you can check this http://docs.openstack.org/admin-guide-cloud/compute-configuring-migrations.html to see what are the limitations of both types of migrations | 11:03 |
kashyap | pkoniszewski: BTW, is there a simple way to make an RPC request hang? | 11:03 |
alex_xu | PaulMurray: np | 11:04 |
*** efoley_ has quit IRC | 11:04 | |
kaisers | thanks | 11:04 |
kaisers | <--- digs into things | 11:04 |
shaohe_feng_ | johnthetubaguy: ping. | 11:04 |
johnthetubaguy | kaisers: basically we have cold migrate and live-migrate, one turns the VM off when you move it, the other tries to keep it running | 11:04 |
johnthetubaguy | shaohe_feng_: how can i help? | 11:04 |
bauzas | johnthetubaguy: the nova internals are so lovely between resize/cold-migrate and rebuild/evacuate... | 11:06 |
*** sdague has joined #openstack-nova | 11:06 | |
shaohe_feng_ | johnthetubaguy: sdague : do we need to add migrate_type? | 11:07 |
* alex_xu nod to bauzas | 11:07 | |
bauzas | johnthetubaguy: conceptually, I wonder if we shouldn't big-bang the definitions and provide a clean split between the ops | 11:07 |
*** omnipresent has joined #openstack-nova | 11:07 | |
*** zenoway has quit IRC | 11:07 | |
pkoniszewski | kashyap: you mean simple way to reproduce it? you can slow down conductor before it starts async, it will timeout on the API side | 11:07 |
shaohe_feng_ | johnthetubaguy: how we define in progress list. still negative logic ? | 11:07 |
johnthetubaguy | bauzas: yeah, the history here is xenapi implemented migrate/resize, libvirt did resize, then they each added the other one, at least that what I think happened | 11:07 |
johnthetubaguy | shaohe_feng_: I hate the negative logic myself | 11:08 |
bauzas | johnthetubaguy: so, followin the RPC calls just limbo and there is high error rate | 11:08 |
bauzas | but anyway | 11:08 |
bauzas | that's a long story | 11:08 |
kaisers | johnthetubaguy: So at the core, 'nova migrate' does a cold migration where the VM is in fact shut down for a short time, which 'nova live-migration' does not. But both commands can be run on an active instance which pops up/transfers on/to the new host running and active. Is that right? | 11:08 |
kashyap | pkoniszewski: Yeah, a reproducer. Slowing down Conductor - never done that, how do you go about it? | 11:08 |
bauzas | that's just because I'm right on that :-) | 11:08 |
*** ivase has quit IRC | 11:08 | |
shaohe_feng_ | johnthetubaguy: so let me update it in the coming version. | 11:09 |
johnthetubaguy | bauzas: yeah, its stuff we are bad at testing really, and glad to see us fixing that now | 11:09 |
*** ivase has joined #openstack-nova | 11:09 | |
johnthetubaguy | kaisers: yes, its cold vs live, basically | 11:09 |
bauzas | hopefully, things are improving | 11:09 |
kaisers | johnthetubaguy: great, got that now. :) | 11:09 |
kaisers | thanks again! | 11:09 |
johnthetubaguy | kaisers: np | 11:10 |
shaohe_feng_ | johnthetubaguy: and filter [queued, preparing, running, post-migrating] for in progress. | 11:10 |
johnthetubaguy | shaohe_feng_: I am not sure we need the migration_type, if we only return live-migration things, not sure what sdague and alex_xu think though, I am curious | 11:11 |
PaulMurray | bauzas, johnthetubaguy we pushed out refactoring the code for migrations from the last cycle to get some more immediate stuff done | 11:11 |
PaulMurray | bauzas, johnthetubaguy there is plenty of interest in cleaning this stuff up now | 11:11 |
alex_xu | johnthetubaguy: we return all types migration in shaohe_feng_'s patch, shaohe_feng_ right? | 11:11 |
sdague | johnthetubaguy: I don't think we need it | 11:11 |
shaohe_feng_ | johnthetubaguy: yes. | 11:11 |
johnthetubaguy | PaulMurray: agreed, its something I wrote up in a blueprint 2.5 years ago, and still haven't have time, so happy to see it happening | 11:11 |
kashyap | pkoniszewski: I'll try to find out, don't bother here. | 11:12 |
sdague | alex_xu: I thought it was only returning live migrations | 11:12 |
bauzas | PaulMurray: honestly, I'm not talking of refactoring, just cleaning out the definitions and have clear actions | 11:12 |
johnthetubaguy | sdague: I am not sure it does that yet... | 11:12 |
sdague | johnthetubaguy: oh, so we should do that | 11:12 |
johnthetubaguy | sdague: +1 | 11:12 |
pkoniszewski | kashyap: so the easiest way is to make it sleep (conductor, source compute or destination compute during prechecks), another way is to reduce number of file descriptors for rabbitmq on, e.g., source compute node | 11:12 |
PaulMurray | bauzas, oh, right, something we can document you mean? | 11:12 |
sdague | because that was the plan | 11:12 |
sdague | and why we dropped all the flavor stuff | 11:12 |
johnthetubaguy | sdague: the spec was a little vague and contradictory when I checked it | 11:12 |
johnthetubaguy | sdague: yeah | 11:12 |
alex_xu | sdague: ok, if only return live-migration...agree to we needn't that | 11:12 |
kashyap | pkoniszewski: Ah, reducing FDs sound interesting | 11:12 |
bauzas | PaulMurray: no, touching the REST API | 11:12 |
pkoniszewski | kashyap: it will timeout on different layers due to chain of sync calls | 11:12 |
alex_xu | sdague: johnthetubaguy one more question, we didn't return all types, just for first step? | 11:13 |
PaulMurray | bauzas, restructuing the API then? | 11:13 |
alex_xu | s/we/why/ | 11:13 |
bauzas | PaulMurray: and the RPC bits so that one REST action corresponds straight to one RPC call | 11:13 |
kashyap | pkoniszewski: Yep, noted. Thank you | 11:13 |
sdague | alex_xu: because we're only implementing DELETE for one type | 11:13 |
sdague | if we expose all types here, the DELETE patch gets a ton more complicated | 11:13 |
johnthetubaguy | sdague: yeah, I just -1ed the patch for it not checking its live-migrate | 11:13 |
sdague | this was always supposed to be just live-migration to start | 11:14 |
alex_xu | sdague: ok, got it | 11:14 |
shaohe_feng_ | sdague: johnthetubaguy: OK, only return live migrations. | 11:14 |
bauzas | and not for example cold_migrate on the REST side being a compute.api resize which is a migrate_server conductor method which ends up to tasks.migrate | 11:14 |
johnthetubaguy | shaohe_feng_: so I think we need to make sure we only return live-migrations, so that should help with the inprogress, we just need that states for live-migrations | 11:14 |
*** ivase_ has joined #openstack-nova | 11:14 | |
sdague | I'm not even sure what progress you could return on a resize, for instance | 11:14 |
PaulMurray | bauzas, oh, well, I don't get that exactly | 11:14 |
bauzas | PaulMurray: because we factored some code | 11:14 |
johnthetubaguy | sdague: you could do disk copied, but lets not try get into that | 11:14 |
PaulMurray | bauzas, but it does come into the proposals that have been around | 11:14 |
alex_xu | shaohe_feng_: whether /os-migration return all type migrations? | 11:15 |
alex_xu | if yes, what about ref-link in /os-migration? | 11:15 |
pkoniszewski | johnthetubaguy: isn't decorator enough? Im talking about checking migration type in 'abort live migration' | 11:15 |
bauzas | I mean that if that's a cold move, we should provide only one operation for the REST API | 11:15 |
bauzas | that can be asking for resize or not | 11:15 |
johnthetubaguy | pkoniszewski: no, migrating state is not just live-migrate | 11:15 |
*** ivase has quit IRC | 11:15 | |
shaohe_feng_ | alex_xu: it should return all type. let me check it. | 11:15 |
alex_xu | looks like same as non active migration, just 404, sounds ok | 11:15 |
bauzas | if that's a live move, then there is only one thing to do | 11:15 |
sdague | pkoniszewski: no, the representation we have doesn't make any sense for non live migration | 11:16 |
*** cdent has quit IRC | 11:16 | |
pkoniszewski | johnthetubaguy: thought that cold-migrate/resize has other task state | 11:16 |
bauzas | so, evacuate, resize, rebuild and cold-migrate are pretty identical to me | 11:16 |
sdague | really, explain memory_remaining_bytes field for resize | 11:16 |
bauzas | they're just a cold move operation with a different set of constraints | 11:16 |
bauzas | PaulMurray: ^ | 11:16 |
*** derekh has quit IRC | 11:16 | |
johnthetubaguy | bauzas: right, thats what I wrote up in my blueprint 2.5 years ago, totally agree, although admittedly thats before we added evacuate, possibly, not sure | 11:17 |
pkoniszewski | johnthetubaguy: sdague: but okay, we already have the data so we can check it, so we will avoid any troubles, probably i will need to add it to force_complete too | 11:17 |
bauzas | evacuate is just a cold-migrate where the added constraint is that we can't check the source host | 11:17 |
PaulMurray | pkoniszewski, johnthetubaguy sdague the user issues are different for non-live-migration | 11:17 |
PaulMurray | onlly live-migration tends to hang | 11:18 |
johnthetubaguy | bauzas: actually I made live=true/false an option in there, or something like that, so a single admin "move server", "resize"/"rebuild" is a bit different though, anyways, different convo | 11:18 |
PaulMurray | or take time at least | 11:18 |
johnthetubaguy | pkoniszewski: yeah, we should add that to force_complete | 11:18 |
bauzas | johnthetubaguy: conceptually, I was seeing from an UX PoV : | 11:18 |
pkoniszewski | johnthetubaguy: got it, will work on it | 11:19 |
*** smatzek__ has joined #openstack-nova | 11:19 | |
sdague | pkoniszewski: the entire issue that I understood we were trying to solve was the live migation black hole | 11:19 |
bauzas | #1 as an user, I accept downtime and I want to recreate my instance either locally or remotely | 11:19 |
bauzas | well | 11:19 |
bauzas | it's rather | 11:19 |
bauzas | depending on whether I'm a user, I want to recreate my instance | 11:19 |
bauzas | locally or remotely is not something I should know unless I'm an admin | 11:20 |
*** ivase has joined #openstack-nova | 11:20 | |
bauzas | johnthetubaguy: but anyway, we can think a lot of that, but we should maybe resurrect your spec and discuss it as a backlog one | 11:20 |
PaulMurray | bauzas, I think the four examples in the concepts guide are reasonable | 11:20 |
*** erlon has joined #openstack-nova | 11:20 | |
*** ivase_ has quit IRC | 11:20 | |
bauzas | PaulMurray: why not, I'm not opiniated | 11:21 |
johnthetubaguy | bauzas: so I think there are a list of use cases, probably a separate API for each concept, but probably a single set of "move" primitives underneath that implement those API, I think this was pre-specs, I should go find that... | 11:21 |
bauzas | ++ | 11:21 |
bauzas | exactly *that* | 11:21 |
johnthetubaguy | yeah, the docs covers this quite well now | 11:21 |
johnthetubaguy | the use cases that is | 11:21 |
PaulMurray | i agree | 11:21 |
johnthetubaguy | pkoniszewski: shaohe_feng_: do you think you are both unblocked on these changes now? happy about the next steps? | 11:22 |
PaulMurray | most admin use cases are about moving vms | 11:22 |
pkoniszewski | johnthetubaguy: I have one more question actually before I start work | 11:22 |
johnthetubaguy | pkoniszewski: fire away | 11:22 |
PaulMurray | most user use cases are about doing something to a vm that may require it to move behind the scenes | 11:22 |
johnthetubaguy | PaulMurray: +1 | 11:22 |
PaulMurray | bauzas, e.g. anythig a user can do to a vm that would affect the way it was scheduled implies a potential move | 11:23 |
pkoniszewski | so we mark migration as running before we actually start migration: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L5212 | 11:24 |
pkoniszewski | there is a lot of code after this line executed in driver | 11:24 |
pkoniszewski | wonder if I should ask libvirt if is there any job undergoing so we won't end up with a paused VM, instead of forced live migration | 11:24 |
pkoniszewski | im talking about another check at driver level: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L6509 | 11:25 |
*** sudipto has quit IRC | 11:25 | |
pkoniszewski | johnthetubaguy: ^^ | 11:25 |
johnthetubaguy | pkoniszewski: I think we do need that | 11:25 |
*** ivase_ has joined #openstack-nova | 11:26 | |
johnthetubaguy | pkoniszewski: well, I think lets just check we are aborting a live-migrate job? | 11:26 |
shaohe_feng_ | johnthetubaguy: another question. the instance db API level or cell level in cell v2? | 11:26 |
*** ivase has quit IRC | 11:26 | |
johnthetubaguy | pkoniszewski: oh, sorry, I see, you mean for the pause | 11:26 |
pkoniszewski | johnthetubaguy: exactly, we won't pause during abort because of check at compute-manager level i believe | 11:26 |
johnthetubaguy | pkoniszewski: yeah, checking there is an active live-migrate job, before calling pause might do the trick here? and probably double checking there is still one going after we call pause | 11:27 |
*** fawadkhaliq has joined #openstack-nova | 11:27 | |
johnthetubaguy | pkoniszewski: not sure we change the task state, thinking about it... hmm | 11:28 |
*** gongysh has quit IRC | 11:28 | |
johnthetubaguy | pkoniszewski: lets leave that issue for now, and just fix these other ones first | 11:28 |
pkoniszewski | johnthetubaguy: depends what is more important, that VM is paused or that it is being migrated | 11:28 |
pkoniszewski | johnthetubaguy: asking libvirt should avoid potential race condition, yes | 11:28 |
johnthetubaguy | pkoniszewski: right now I think we need to get the new features merged | 11:29 |
pkoniszewski | johnthetubaguy: sure, I just want to clarify it at least a bit before I start work, let's focus on important stuff before freeze | 11:29 |
pkoniszewski | johnthetubaguy: thanks! | 11:29 |
johnthetubaguy | pkoniszewski: np | 11:30 |
johnthetubaguy | pkoniszewski: these are new tools for admins, so I think its better we add the tool, rather than make it perfect, in this particular case | 11:30 |
shaohe_feng_ | johnthetubaguy: I read cell doc. the migration is API level, and the instance is cell level in V2. | 11:30 |
bauzas | shaohe_feng: not sure I understood your questions, but in cells V2 world, instances will be kept on the cell db, not in the api db | 11:30 |
*** houming has quit IRC | 11:31 | |
johnthetubaguy | shaohe_feng_: oh, so passing down the migration will be quite important, eventually... | 11:31 |
bauzas | shaohe_feng: so we have an InstanceMapping table in the API DB for knowing which instance is in which cell | 11:31 |
johnthetubaguy | shaohe_feng_: yeah, where is this question? | 11:31 |
*** ivase has joined #openstack-nova | 11:31 | |
shaohe_feng_ | bauzas: johnthetubaguy: so we can get migration by object directly. but we still get instance by compute_api . right? | 11:31 |
bauzas | johnthetubaguy: that's quite identical as my RequestSpec object for check-dests, if we get any object in the API DB, we need to pass it as payload to the conductor | 11:32 |
*** ivase_ has quit IRC | 11:32 | |
bauzas | johnthetubaguy: shaohe_feng: although it would still be possible to get some object from the api db within a cell, but we'd like to avoid that as much as possible (because of races) | 11:32 |
bauzas | but I don't have much context, shaohe_feng could you please clarify your question ? | 11:33 |
*** wanghao has quit IRC | 11:33 | |
johnthetubaguy | so don't worry about it, the object layer will be hiding most of that detail, I suspect | 11:34 |
johnthetubaguy | lets just do the simple thing now, thats consistent, and worry about those fix ups later | 11:34 |
shaohe_feng_ | bauzas: in nova API, can we access the migration object directly? pass_through computer_api or cell_api | 11:35 |
shaohe_feng_ | johnthetubaguy: ^ | 11:35 |
*** wanghao has joined #openstack-nova | 11:35 | |
bauzas | which migration object ? | 11:35 |
shaohe_feng_ | that's my question. | 11:35 |
bauzas | I'm missing context | 11:35 |
alex_xu | johnthetubaguy: so that is a problem for https://review.openstack.org/#/c/277971/ ? we need pass migration object down | 11:36 |
PaulMurray | johnthetubaguy, bauzas, sdague meanwhile, back in the check queue.... looks like tempest jobs are failing for everyone | 11:36 |
PaulMurray | sdague, who should we ask about this? | 11:36 |
shaohe_feng_ | the migration object which accesses migration db table. | 11:36 |
bauzas | shaohe_feng: you mean the Migration nova object ? | 11:36 |
sdague | PaulMurray: yeh... I think I found the issue there | 11:36 |
johnthetubaguy | so I think we should pass the migration object down, really | 11:36 |
sdague | https://review.openstack.org/284652 | 11:36 |
bauzas | sdague: which was ? I was looking at the issue but hard to tell | 11:36 |
bauzas | it's not super explicit | 11:36 |
sdague | ansible release | 11:36 |
bauzas | okay | 11:37 |
*** ankit_ag has quit IRC | 11:37 | |
*** ivase_ has joined #openstack-nova | 11:37 | |
PaulMurray | sdague, good job | 11:37 |
shaohe_feng_ | bauzas: yes. | 11:37 |
*** ankit_ag has joined #openstack-nova | 11:37 | |
alex_xu | johnthetubaguy: ok, got it | 11:37 |
*** ivase has quit IRC | 11:37 | |
johnthetubaguy | alex_xu: hmm, I am still on the fence, actually maybe its odd we put the migration objects in the API | 11:39 |
alex_xu | johnthetubaguy: why? | 11:39 |
alex_xu | johnthetubaguy: if we have complex logic, that should be in the comptue_api | 11:40 |
johnthetubaguy | alex_xu: oh, sorry, I was talking about something different, totally the complex logic should live in compute_api | 11:40 |
alex_xu | johnthetubaguy: sorry, misunderstand that | 11:41 |
*** aagrawal has quit IRC | 11:41 | |
PaulMurray | sdague, :( your patch failed | 11:41 |
*** gcb has quit IRC | 11:42 | |
sdague | PaulMurray: oh, right 2 = | 11:42 |
*** mingyu has joined #openstack-nova | 11:43 | |
*** EinstCrazy has joined #openstack-nova | 11:43 | |
sdague | hey, I'm only 1/2 a cup of coffee in. :) | 11:43 |
sdague | I get to make some mistakes | 11:43 |
*** sridharg has joined #openstack-nova | 11:43 | |
*** gcb has joined #openstack-nova | 11:43 | |
*** ivase_ has quit IRC | 11:43 | |
PaulMurray | sdague, don't worry, on half a coffee you are still two steps ahead of us | 11:44 |
*** josecastroleon has quit IRC | 11:45 | |
PaulMurray | sdague, speaking of coffee, I have to walk to another floor now... | 11:45 |
*** efoley__ has quit IRC | 11:46 | |
*** mingyu has quit IRC | 11:47 | |
shaohe_feng_ | bauzas: Can rest API access Migration nova object directly, without computer_api or cell_api? | 11:47 |
bauzas | shaohe_feng: again, sorry, not sure I get your question | 11:49 |
bauzas | shaohe_feng: you want to know if a WSGI resource module can fetch the object if it's in the API DB ? then yes | 11:50 |
bauzas | although I tend to prefer using the compute.api module for that | 11:50 |
*** mdnadeem has quit IRC | 11:50 | |
*** mdnadeem has joined #openstack-nova | 11:51 | |
bauzas | in cell v2, there will be no "compute_cell" proxy | 11:51 |
bauzas | I mean, the nova.compute.cells_api module, sortu | 11:51 |
bauzas | sorry | 11:51 |
bauzas | because by default, the regular API will be there | 11:52 |
*** rfolco has joined #openstack-nova | 11:53 | |
*** cdent has joined #openstack-nova | 11:53 | |
*** lyanchih has joined #openstack-nova | 11:54 | |
johnthetubaguy | shaohe_feng: everywhere can access both datastores, if they have to, but for scalability reasons, the api accessing the child cell is fine, but the reverse should be avoided where possible | 11:55 |
johnthetubaguy | does that help? | 11:55 |
alex_xu | johnthetubaguy: so...will you -1 this patch https://review.openstack.org/#/c/277971/ for pass migration down, or I help you -1 it? | 11:56 |
johnthetubaguy | alex_xu: so I think its OK for now, mostly because its consistent with the other one, and really we do need to double check the migration state | 11:57 |
*** Guest75 has joined #openstack-nova | 11:57 | |
*** ekuris has quit IRC | 11:57 | |
alex_xu | johnthetubaguy: ok, got it | 11:57 |
johnthetubaguy | alex_xu: thats my way of saying, I don't think the migration object should live in the API database, where did we say that would be the case, btw? | 11:57 |
alex_xu | johnthetubaguy: oops, I guess I misunderstand, I thought migraiton object should in the API database | 11:59 |
johnthetubaguy | I think shaohe_feng mentioned that, I am just wondering where we said that, it doesn't feel quite right to me | 11:59 |
johnthetubaguy | as its so related to the instance, it feels like a child cell thing | 11:59 |
johnthetubaguy | but I could be missing something | 11:59 |
*** haomaiwang has quit IRC | 12:01 | |
alex_xu | johnthetubaguy: ok | 12:01 |
*** lyanchih has quit IRC | 12:01 | |
*** thorst has joined #openstack-nova | 12:01 | |
*** haomaiwang has joined #openstack-nova | 12:01 | |
* johnthetubaguy runs away for a little bit | 12:02 | |
*** mpavone has quit IRC | 12:03 | |
*** MCoLo has quit IRC | 12:04 | |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Console log copy for block live migratiob https://review.openstack.org/284674 | 12:04 |
*** MCoLo has joined #openstack-nova | 12:04 | |
sdague | I think we probably need a write up about what goes into the API database, and is strict about it. Otherwise we're going to walk ourselves back into a single db solution one table at a time. | 12:04 |
*** aysyd has joined #openstack-nova | 12:05 | |
bauzas | sdague: that's on-going | 12:06 |
bauzas | sdague: lemme find the ref | 12:06 |
bauzas | sdague: yeah this http://docs.openstack.org/developer/nova/cells.html#database-split | 12:06 |
*** smatzek__ is now known as smatzek | 12:06 | |
bauzas | sdague: but the formatting is awful | 12:06 |
bauzas | I can fix that | 12:07 |
*** Guest75 has quit IRC | 12:07 | |
shaohe_feng_ | bauzas: johnthetubaguy : got it. thank you. | 12:17 |
shaohe_feng_ | bauzas: WSGI resource module can fetch the object if it's in the child cell DB ? | 12:19 |
bauzas | like johnthetubaguy said | 12:20 |
shaohe_feng_ | bauzas: seen it. the answer is yes. | 12:22 |
*** mjura has joined #openstack-nova | 12:25 | |
*** haomaiwang has quit IRC | 12:31 | |
*** shaohe_feng_ has quit IRC | 12:31 | |
*** achanda has joined #openstack-nova | 12:33 | |
*** raildo-afk is now known as raildo | 12:37 | |
belmoreira | johnthetubaguy https://review.openstack.org/#/c/225384 yes but not https://review.openstack.org/#/c/263898. I will have a look | 12:37 |
belmoreira | johnthetubaguy thanks for ping me | 12:37 |
*** lucasagomes is now known as lucas-hungry | 12:39 | |
sdague | while we are waiting for devstack jobs to work again, might I suggest some test only fixes to review - https://review.openstack.org/#/c/267097/ - https://review.openstack.org/#/c/283261/ | 12:39 |
*** ljxiash has joined #openstack-nova | 12:40 | |
*** achanda has quit IRC | 12:40 | |
*** rtheis has joined #openstack-nova | 12:42 | |
*** fawadkhaliq has quit IRC | 12:42 | |
*** daemontool_ has joined #openstack-nova | 12:43 | |
*** mingyu has joined #openstack-nova | 12:43 | |
*** fawadkhaliq has joined #openstack-nova | 12:43 | |
*** zul has quit IRC | 12:43 | |
*** zul has joined #openstack-nova | 12:44 | |
*** daemontool has quit IRC | 12:46 | |
*** diogogmt has joined #openstack-nova | 12:46 | |
*** fawadkhaliq has quit IRC | 12:47 | |
*** daemontool_ has quit IRC | 12:47 | |
openstackgerrit | Dongcan Ye proposed openstack/nova: VMware: Add warning log when creating folder already exists https://review.openstack.org/284694 | 12:47 |
*** mingyu has quit IRC | 12:47 | |
sdague | alex_xu: https://review.openstack.org/#/c/283261/ is actually a regression test for the regex bug you are working on | 12:53 |
*** pcaruana has quit IRC | 12:55 | |
*** mpavone has joined #openstack-nova | 12:55 | |
*** baoli has joined #openstack-nova | 12:55 | |
*** salv-orlando has quit IRC | 12:56 | |
*** dave-mccowan has joined #openstack-nova | 12:56 | |
*** efoley__ has joined #openstack-nova | 12:57 | |
*** Daisy has joined #openstack-nova | 12:58 | |
*** daemontool has joined #openstack-nova | 13:01 | |
*** takedakn has joined #openstack-nova | 13:02 | |
openstackgerrit | Zhengguang Ou proposed openstack/nova: libvirt: fix race condition in detach volume and create image https://review.openstack.org/166152 | 13:02 |
*** Daisy has quit IRC | 13:02 | |
*** mpavone has quit IRC | 13:03 | |
*** ildikov has quit IRC | 13:03 | |
*** edmondsw has joined #openstack-nova | 13:04 | |
thorst | If someone has a chance, could they take a peak at https://review.openstack.org/#/c/264157/ Its been sitting for a few weeks, but solves a long standing bug for cloud-init & ipv6 only environments | 13:07 |
*** efoley_ has joined #openstack-nova | 13:08 | |
*** pcaruana has joined #openstack-nova | 13:09 | |
*** diogogmt has quit IRC | 13:10 | |
*** pradk has joined #openstack-nova | 13:10 | |
*** pradk has quit IRC | 13:10 | |
*** efoley__ has quit IRC | 13:11 | |
*** mpavone has joined #openstack-nova | 13:11 | |
*** derekh has joined #openstack-nova | 13:13 | |
*** ekuris has joined #openstack-nova | 13:13 | |
*** tyagiprince has quit IRC | 13:15 | |
*** takedakn has quit IRC | 13:15 | |
*** zz_dimtruck is now known as dimtruck | 13:16 | |
kaisers | Hi Folks! Can anybody here give me a 'minute' to find out why i encounter a migration issue with console.log beeing chown'ed from nova:nova to root:root? | 13:16 |
*** ducttape_ has joined #openstack-nova | 13:17 | |
sdague | PaulMurray / bauzas / et all | 13:18 |
sdague | the gate is unwedged | 13:18 |
sdague | so if we've got patches that need a recheck | 13:18 |
sdague | now is the time | 13:18 |
*** lucas-hungry is now known as lucasagomes | 13:19 | |
*** rotbeard has joined #openstack-nova | 13:20 | |
*** claudiub has joined #openstack-nova | 13:23 | |
*** mubirru has quit IRC | 13:23 | |
*** baoli_ has joined #openstack-nova | 13:26 | |
*** links has quit IRC | 13:26 | |
*** kylek3h__away is now known as kylek3h | 13:28 | |
*** baoli has quit IRC | 13:28 | |
*** mpavone has quit IRC | 13:30 | |
*** josecastroleon has joined #openstack-nova | 13:31 | |
*** ivase has joined #openstack-nova | 13:34 | |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Console log copy for block live migratiob https://review.openstack.org/284674 | 13:35 |
*** ivase has quit IRC | 13:36 | |
*** ducttape_ has quit IRC | 13:36 | |
*** ivase has joined #openstack-nova | 13:36 | |
*** rtheis has quit IRC | 13:36 | |
*** ankit_ag has quit IRC | 13:37 | |
*** achanda has joined #openstack-nova | 13:37 | |
*** toure is now known as toure_ | 13:39 | |
*** hamzy has quit IRC | 13:41 | |
*** achanda has quit IRC | 13:41 | |
*** ivase has quit IRC | 13:41 | |
*** rtheis has joined #openstack-nova | 13:42 | |
*** ccarmack has joined #openstack-nova | 13:43 | |
*** shyama has quit IRC | 13:44 | |
*** rtheis has quit IRC | 13:44 | |
*** rtheis has joined #openstack-nova | 13:45 | |
openstackgerrit | Paul Murray proposed openstack/python-novaclient: Support for abort an ongoing live migration https://review.openstack.org/284296 | 13:45 |
*** burgerk has joined #openstack-nova | 13:46 | |
*** armax has joined #openstack-nova | 13:47 | |
*** dims has joined #openstack-nova | 13:47 | |
*** yassine__ has quit IRC | 13:49 | |
*** shaohe_feng_ has joined #openstack-nova | 13:49 | |
*** ildikov has joined #openstack-nova | 13:50 | |
*** jaypipes has joined #openstack-nova | 13:51 | |
kaisers | Can anybody give me a quick hint on a libvirt/migration related question?: what is the file console.log for and when/where is it created? | 13:51 |
kaisers | I'm hitting issues with that file beeing owned by root:root and nova beeing unable to access it because of that. | 13:51 |
*** xyang1 has joined #openstack-nova | 13:52 | |
pkoniszewski | andreykurilin: hi, there? | 13:52 |
andreykurilin | pkoniszewski: hi! yes | 13:52 |
jaypipes | cdent: good point on the context/get_engine() thing. /me needs to ask rpodolyaka how to do that.. | 13:53 |
jaypipes | rpodolyaka: how do I do that? :) See cdent's comment on https://review.openstack.org/#/c/284484/1/nova/db/sqlalchemy/api.py | 13:53 |
*** fawadkhaliq has joined #openstack-nova | 13:53 | |
cdent | jaypipes: I was pretty surprised I could still see past the blood in my eyes by the time I got to that line. | 13:54 |
jaypipes | cdent: :( /me cries | 13:54 |
cdent | Actually that's not quite right: it is very well written for what it is and I did understand it | 13:54 |
cdent | It's just ... a lot | 13:54 |
jaypipes | cdent: luckily, that won't be there for long... | 13:54 |
pkoniszewski | andreykurilin: i wonder if we should chain new test classes when adding support for new microversion in novaclient | 13:55 |
jaypipes | cdent: once the resource-providers schema is the only thing used for inventory and usage stuff, it will be a lot simpler and more efficient. | 13:55 |
pkoniszewski | andreykurilin: i'm talking about this https://review.openstack.org/#/c/284296/2/novaclient/tests/unit/v2/test_server_migrations.py | 13:55 |
cdent | I had to write a similar many-aliases thing in some other context otherwise it would have been a bit unfamliar | 13:55 |
jaypipes | cdent: i.e. less use of outer joins. more ability to winnow results on the DB side. | 13:55 |
cdent | \o/ | 13:55 |
johnthetubaguy | belmoreira: cool, just thinking its a good time to double check it will work for you :) | 13:56 |
*** Guest75 has joined #openstack-nova | 13:56 | |
*** zzzeek has joined #openstack-nova | 13:57 | |
andreykurilin | pkoniszewski: imo, we should not put them in the chain | 13:57 |
*** zzzeek has quit IRC | 13:57 | |
pkoniszewski | andreykurilin: okay, we chain tests in nova api, thought we follow it there, but that's fine too, thanks! :) | 13:58 |
*** BobBall_ has joined #openstack-nova | 13:58 | |
*** salv-orlando has joined #openstack-nova | 13:58 | |
*** zzzeek has joined #openstack-nova | 13:58 | |
*** BobBall_ is now known as BobBall | 13:58 | |
andreykurilin | pkoniszewski: we do the same in novaclient( https://github.com/openstack/python-novaclient/blob/master/novaclient/tests/unit/v2/test_servers.py#L1050 ) lol, but, I want to fix this:) | 13:59 |
*** daemontool_ has joined #openstack-nova | 13:59 | |
openstackgerrit | Paul Murray proposed openstack/python-novaclient: Support for forcing live migration to complete https://review.openstack.org/279647 | 13:59 |
alaski | cdent: jaypipes I think you can just pass the context in to get engine to preserve the context_manager semantics, by kwarg since it's not the first arg | 13:59 |
jaypipes | alaski: ah, I see. thank you sir. | 14:01 |
pkoniszewski | andreykurilin: ok, thanks for clarifying this! :) | 14:01 |
jaypipes | I need to keep up with the new oslo.db enginefacade stuff. | 14:01 |
*** efoley_ has quit IRC | 14:01 | |
kashyap | pkoniszewski: I only noticed your question just now on this bug, this was discussed on #virt on OFTC | 14:01 |
kashyap | https://bugzilla.redhat.com/show_bug.cgi?id=994882#c3 | 14:01 |
openstack | bugzilla.redhat.com bug 994882 in libvirt "Live block migration results in a zero console log" [High,New] - Assigned to libvirt-maint | 14:01 |
*** shyama has joined #openstack-nova | 14:01 | |
*** daemontool has quit IRC | 14:02 | |
kashyap | I added the comment (#3) from that discussion there. | 14:02 |
openstackgerrit | Paul Murray proposed openstack/python-novaclient: Support for abort an ongoing live migration https://review.openstack.org/284296 | 14:02 |
*** edtubill has joined #openstack-nova | 14:03 | |
*** apearson has joined #openstack-nova | 14:03 | |
*** salv-orlando has quit IRC | 14:03 | |
*** edtubill has quit IRC | 14:04 | |
*** edtubill has joined #openstack-nova | 14:04 | |
*** whenry has quit IRC | 14:05 | |
johnthetubaguy | mikal tjones cburgess jgrimm adrian_otto funzo mjturek jcookekhugen irina_pov krtaylor danpb alexpilotti flip214 jaypipes garyk edleafe dims moshele anteaya Nisha sileht claudiub lxsli neiljerram markus_z swamireddy alevine tonyb andreykurilin ndipanov sc68cal akuriata artom jlvillal mnestratov kashyap aloga rgeragnov bauzas xyang tpatil med_ nic scottda nagyz belliott dguitarbite sdague jroll diana_clarke | 14:06 |
johnthetubaguy | _diana_ raildo jichen gjayavelu mdorman klindgren sorrison belmoreira mrda auggy PaulMurray gcb thorst duncant macsz thomasem tpatzig bobball lbelivea | 14:06 |
johnthetubaguy | nova meeting is currently happening, in the usual place | 14:06 |
johnthetubaguy | oops, sorry, for the late ping | 14:06 |
*** rotbeard has quit IRC | 14:06 | |
kashyap | What is the usual place? | 14:06 |
* kashyap forgets the IRC channel, goes to look up | 14:06 | |
pkoniszewski | kashyap: sorry if i send it twice, but just got DC and dunno if it went out | 14:06 |
pkoniszewski | kashyap: sounds fair that higher-layer should handle this, yes | 14:06 |
kashyap | pkoniszewski: Didn't receive your first message, okay, I'll add the relevant comment in the Nova LP bug as well. | 14:07 |
*** links has joined #openstack-nova | 14:07 | |
*** efoley_ has joined #openstack-nova | 14:07 | |
*** ctrath has joined #openstack-nova | 14:08 | |
*** andymaier has joined #openstack-nova | 14:08 | |
*** apearson has quit IRC | 14:12 | |
*** apearson has joined #openstack-nova | 14:13 | |
garyk | BobBall: is the CI broken? I got a citrox -1 for a VMware test update :) | 14:14 |
BobBall | probably just a race... | 14:14 |
*** rlrossit_ has joined #openstack-nova | 14:15 | |
BobBall | nope | 14:15 |
BobBall | definitely broken :D | 14:15 |
BobBall | I'll check it out | 14:15 |
garyk | gracias | 14:15 |
garyk | BobBall: https://review.openstack.org/284286 | 14:15 |
*** jichen has quit IRC | 14:17 | |
BobBall | interesting - not seen this failure from devstack before http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/86/284286/2/7309/logs/devstacklog.txt.gz | 14:17 |
*** irenab has quit IRC | 14:18 | |
BobBall | Looks like a general failure garyk - see http://osdir.com/ml/openstack-dev/2016-02/msg02021.html for others who have hit this today | 14:19 |
*** rlrossit_ has quit IRC | 14:19 | |
*** rotbeard has joined #openstack-nova | 14:19 | |
*** rlrossit has joined #openstack-nova | 14:19 | |
garyk | BobBall: ok, thanks | 14:20 |
BobBall | Looks as though it killed the powerkvm CI and most of jenkins at the same time | 14:20 |
BobBall | So it's a waiting game I imagine | 14:20 |
*** fawadkhaliq has quit IRC | 14:20 | |
johnthetubaguy | BobBall: is that the ansible issue? | 14:21 |
johnthetubaguy | oh, right, I see now | 14:21 |
thomasem | hey | 14:21 |
BobBall | Possibly johnthetubaguy - is it? :) | 14:21 |
johnthetubaguy | BobBall: not sure now, that might be different | 14:21 |
johnthetubaguy | thomasem: howdy | 14:22 |
johnthetubaguy | sdague: I just writing a response to that os-brick thread on the ML | 14:22 |
BobBall | garyk: I'm on vacation this afternoon so I'll check in later - when jenkins is fixed I'll check if the fix worked for the citrix CI | 14:22 |
johnthetubaguy | sdague: I kinda thought we agreed to the sudo like privsep, to save us reverting out os-brick | 14:22 |
garyk | BobBall: no problem. enjoy the vacation. its not urgent | 14:23 |
*** yassine__ has joined #openstack-nova | 14:24 | |
*** dimtruck is now known as zz_dimtruck | 14:25 | |
*** lyanchih has joined #openstack-nova | 14:25 | |
*** sudipto has joined #openstack-nova | 14:26 | |
*** mgoddard_ has joined #openstack-nova | 14:27 | |
*** rotbeard has quit IRC | 14:29 | |
PaulMurray | johnthetubaguy, I'm going to fix up the abort migration patch: https://review.openstack.org/#/c/277971/15 | 14:30 |
*** mgoddard has quit IRC | 14:30 | |
PaulMurray | johnthetubaguy, I've got a couple of questions to clear up | 14:30 |
PaulMurray | johnthetubaguy, comments https://review.openstack.org/#/c/277971/15/nova/compute/instance_actions.py | 14:31 |
*** smatzek has quit IRC | 14:31 | |
*** erno has quit IRC | 14:31 | |
*** lyanchih has quit IRC | 14:31 | |
*** gongysh_ has joined #openstack-nova | 14:31 | |
PaulMurray | johnthetubaguy, first it was changed from past tense to present, the comments seem to say change it back to past tense....? | 14:32 |
PaulMurray | also there are - vs _ so I'm a bit lost now | 14:32 |
PaulMurray | sdague, you commented on the above ^^ | 14:33 |
*** Jeffrey4l has quit IRC | 14:33 | |
*** claudiub|2 has joined #openstack-nova | 14:33 | |
*** sridharg has quit IRC | 14:33 | |
*** mdrabe has joined #openstack-nova | 14:33 | |
*** claudiub has quit IRC | 14:34 | |
johnthetubaguy | PaulMurray: oh good questions | 14:35 |
johnthetubaguy | PaulMurray: oh my, we have revertResize live-migration and changePassword... | 14:35 |
johnthetubaguy | PaulMurray: I think it should really match the API action | 14:35 |
johnthetubaguy | but this isn't one | 14:36 |
*** rk4n has quit IRC | 14:36 | |
*** rk4n has joined #openstack-nova | 14:36 | |
*** amotoki has quit IRC | 14:37 | |
PaulMurray | the compute_api or the external rest api? the compute_api action is live_migration_abort | 14:38 |
*** rk4n has quit IRC | 14:38 | |
*** burgerk has quit IRC | 14:39 | |
*** zz_dimtruck is now known as dimtruck | 14:40 | |
*** mriedem has joined #openstack-nova | 14:40 | |
johnthetubaguy | PaulMurray: I was meaning rest api | 14:41 |
PaulMurray | the reset api is a delete on the migration resource | 14:41 |
*** jwcroppe has joined #openstack-nova | 14:44 | |
johnthetubaguy | yeah, that doesn't help really | 14:44 |
johnthetubaguy | PaulMurray: I would +1 live-migration-cancel if that helps | 14:44 |
johnthetubaguy | not sure if others would agree | 14:44 |
*** armax has quit IRC | 14:44 | |
PaulMurray | I was just typing that - I'll do that | 14:44 |
johnthetubaguy | cool | 14:45 |
*** rotbeard has joined #openstack-nova | 14:45 | |
PaulMurray | Also I think we decided to stick with the 400 HTTPBadRequest for InvalidMigrationState | 14:46 |
PaulMurray | Personally I don't like that, but can go with the flow? | 14:46 |
PaulMurray | https://review.openstack.org/#/c/277971/15/nova/api/openstack/compute/server_migrations.py | 14:46 |
*** cfriesen has joined #openstack-nova | 14:46 | |
*** Jeffrey4l has joined #openstack-nova | 14:47 | |
johnthetubaguy | PaulMurray: I guess we did settle on that, and conflict for the instance in a bad state, hmm, that does seem odd | 14:47 |
johnthetubaguy | PaulMurray: I like the consistency though | 14:48 |
*** apearson has quit IRC | 14:49 | |
PaulMurray | ok - I can wizz through this now. I may also add a patch to check its live-migration that we pause in force-complete as well as fixing this - but that's less urgent | 14:49 |
*** apearson has joined #openstack-nova | 14:49 | |
* dansmith bows to jaypipes' fu | 14:50 | |
*** dimtruck is now known as zz_dimtruck | 14:51 | |
*** hamzy has joined #openstack-nova | 14:52 | |
*** mrkz has joined #openstack-nova | 14:53 | |
dansmith | jaypipes: failed for legitness on postgres: http://logs.openstack.org/84/284484/1/check/gate-tempest-dsvm-postgres-full/585d0bf/logs/screen-n-api.txt.gz?level=TRACE#_2016-02-25_01_07_21_737 | 14:53 |
jaypipes | dansmith: ya, saw that. fixing it up. | 14:53 |
dansmith | okay cool | 14:53 |
*** mingyu has joined #openstack-nova | 14:54 | |
*** sc68cal has joined #openstack-nova | 14:54 | |
*** erno has joined #openstack-nova | 14:58 | |
mriedem | garyk: this is a pretty old bug but someone internally it hitting something that sounds similar, have you seen this? https://bugs.launchpad.net/nova/+bug/1373054 | 14:58 |
openstack | Launchpad bug 1373054 in OpenStack Compute (nova) "VMWare compute driver incorrectly attaches volumes on iSCSI for target with multiple LUNs" [Low,Confirmed] | 14:58 |
mriedem | *is hitting | 14:58 |
*** diogogmt has joined #openstack-nova | 14:59 | |
jaypipes | edleafe: yes, dansmith answered correctly your question about None. | 14:59 |
*** smatzek has joined #openstack-nova | 14:59 | |
*** eharney has joined #openstack-nova | 15:00 | |
*** pushkaru has joined #openstack-nova | 15:02 | |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Fix backing file copying https://review.openstack.org/282275 | 15:02 |
*** ducttape_ has joined #openstack-nova | 15:03 | |
*** PsionTheory has joined #openstack-nova | 15:04 | |
*** Swami has joined #openstack-nova | 15:05 | |
*** davideagle_ has quit IRC | 15:06 | |
*** burgerk has joined #openstack-nova | 15:07 | |
*** salv-orlando has joined #openstack-nova | 15:08 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: Don't lazy-load instance.services if the instance is deleted https://review.openstack.org/283820 | 15:08 |
openstackgerrit | Matt Riedemann proposed openstack/nova: Add functional regression test for list deleted instances on v2.16 https://review.openstack.org/283837 | 15:08 |
*** rotbeard has quit IRC | 15:09 | |
*** achinm has quit IRC | 15:09 | |
claudiub|2 | hello. if anyone can review this tiny bug fix, it would be great. :) https://review.openstack.org/#/c/258614/ | 15:10 |
*** daemontool__ has joined #openstack-nova | 15:10 | |
*** links has quit IRC | 15:12 | |
*** rotbeard has joined #openstack-nova | 15:13 | |
garyk | mriedem: i have not hit that one. to be honest i only work with the cinder dirver that supports vmdks. | 15:13 |
garyk | so when it comes to iscsi i have no time int he trenches. sorry | 15:13 |
*** davideagle_ has joined #openstack-nova | 15:13 | |
openstackgerrit | Jay Pipes proposed openstack/nova: Use new inventory schema in compute_node_get_all() https://review.openstack.org/284484 | 15:13 |
*** e0ne has joined #openstack-nova | 15:14 | |
*** daemontool_ has quit IRC | 15:14 | |
jaypipes | dansmith, edleafe, alaski, cdent: ^^ addressed your concerns (and the PostgreSQL thing) | 15:14 |
mriedem | garyk: yeah, what's worse is the one this team hit internally was using the ibm xiv ds8k cinder driver, which is basically closed source | 15:14 |
tjones | mriedem: w.r.t. https://bugs.launchpad.net/nova/+bug/1373054 if they come back with info i can get someone to look into it | 15:14 |
openstack | Launchpad bug 1373054 in OpenStack Compute (nova) "VMWare compute driver incorrectly attaches volumes on iSCSI for target with multiple LUNs" [Low,Incomplete] | 15:14 |
mriedem | garyk: they were trying to figure out what sets the disk label, wasn't sure if that was cinder or nova's vmware volumeops | 15:15 |
mriedem | tjones: i asked the people here to report their bug upstream with details | 15:15 |
tjones | yes | 15:15 |
*** salv-orlando has quit IRC | 15:15 | |
tjones | saw that | 15:15 |
*** knikolla has joined #openstack-nova | 15:16 | |
*** nkrinner has quit IRC | 15:16 | |
PaulMurray | danpb, I have a quick question about cancelling a migration | 15:16 |
danpb | ok | 15:16 |
garyk | mriedem: i would think that it is the cinder side. | 15:17 |
PaulMurray | danpb, looking at https://review.openstack.org/#/c/277971/15/nova/virt/libvirt/driver.py | 15:17 |
smcginnis | Vipin Balachandran might be able to help there. | 15:17 |
smcginnis | He's not online right now, but he's our Cinder VMware guy. | 15:17 |
smcginnis | Or at least one of them. | 15:17 |
PaulMurray | danpb, what's the best way to check a live migration is running ? | 15:18 |
mriedem | unrelated, we hit this warning A LOT in n-cpu logs: http://logstash.openstack.org/#dashboard/file/logstash.json?query=message%3A%5C%22Trying%20to%20apply%20a%20migration%20context%20that%20does%20not%20seem%20to%20be%20set%20for%20this%20instance%5C%22%20AND%20tags%3A%5C%22screen-n-cpu.txt%5C%22&from=7d | 15:18 |
*** zz_dimtruck is now known as dimtruck | 15:18 | |
danpb | PaulMurray: err, i commented on that yesterday i thought | 15:18 |
danpb | yes, see https://review.openstack.org/#/c/277971/14/nova/virt/libvirt/driver.py | 15:18 |
PaulMurray | danpb, so you did, sorry I missed that | 15:18 |
edleafe | jaypipes: dansmith: got it. Thanks. | 15:19 |
*** rk4n has joined #openstack-nova | 15:19 | |
danpb | basically if its in the migrating state in nova's DB, you can assume the job is related to migration | 15:19 |
openstackgerrit | Dmitry Stepanenko proposed openstack/nova: Deprecate ComputeNode pci_device_pools field https://review.openstack.org/247692 | 15:19 |
danpb | because we don't support other non-nova apps directly messing with guests behind nova's back | 15:19 |
*** kfarr has joined #openstack-nova | 15:19 | |
PaulMurray | danpb, I have changed the API layer to check it actually is doing a live migration | 15:20 |
*** rk4n has quit IRC | 15:20 | |
PaulMurray | danpb, I'll follow through the code to check, but that should make it safe then | 15:20 |
*** rk4n has joined #openstack-nova | 15:20 | |
mriedem | smcginnis: yeah, but this isn't the cinder vmware vmdk backend :( | 15:21 |
mriedem | it's the ibm xiv dsk8 | 15:22 |
danpb | PaulMurray: you have @check_instance_state(task_state=[task_states.MIGRATING]) in the api entry point which is fine imho | 15:22 |
*** gongysh_ has quit IRC | 15:22 | |
*** mdrabe has quit IRC | 15:22 | |
*** su_zhang has joined #openstack-nova | 15:22 | |
dansmith | jaypipes: that seems to work | 15:23 |
*** ekuris has quit IRC | 15:23 | |
*** penick has joined #openstack-nova | 15:23 | |
*** efoley_ has quit IRC | 15:24 | |
*** apearson has quit IRC | 15:25 | |
*** PsionTheory has quit IRC | 15:26 | |
*** mrsoul has quit IRC | 15:26 | |
*** mrsoul has joined #openstack-nova | 15:26 | |
*** mrsoul has quit IRC | 15:26 | |
*** mrsoul has joined #openstack-nova | 15:26 | |
*** ekuris_ has joined #openstack-nova | 15:27 | |
cfriesen | is this a known tempest issue? "ERROR! Specified hosts options do not match any hosts" | 15:27 |
*** davideagle_ has quit IRC | 15:27 | |
mriedem | jbernard: a thing! https://bugs.launchpad.net/nova/+bug/1549854 | 15:28 |
openstack | Launchpad bug 1549854 in OpenStack Compute (nova) "compute host "has not been heard from in a while" failing ceph jobs" [Undecided,New] | 15:28 |
*** rtheis has quit IRC | 15:28 | |
*** rtheis has joined #openstack-nova | 15:29 | |
*** shaohe_feng_ has quit IRC | 15:30 | |
*** Jeffrey4l has quit IRC | 15:31 | |
*** omnipresent has quit IRC | 15:31 | |
*** rtheis has quit IRC | 15:31 | |
*** rtheis has joined #openstack-nova | 15:32 | |
*** sdague has quit IRC | 15:32 | |
*** davideagle_ has joined #openstack-nova | 15:33 | |
*** armax has joined #openstack-nova | 15:33 | |
*** ekuris_ has quit IRC | 15:33 | |
*** mgoddard_ has quit IRC | 15:33 | |
*** rtheis has quit IRC | 15:35 | |
openstackgerrit | Thomas Maddox proposed openstack/nova: [WIP - DO NOT MERGE] Testing LXC Gate https://review.openstack.org/274792 | 15:35 |
*** liverpooler has quit IRC | 15:36 | |
johnthetubaguy | cfriesen: yeah, I think the fix for that happening, unless I got my wires crossed | 15:36 |
*** penick has quit IRC | 15:36 | |
openstackgerrit | Sean M. Collins proposed openstack/os-vif: Add mtu attribute to Network object https://review.openstack.org/284209 | 15:36 |
*** mgoddard has joined #openstack-nova | 15:37 | |
*** rtheis has joined #openstack-nova | 15:37 | |
*** mdrabe has joined #openstack-nova | 15:39 | |
*** achanda has joined #openstack-nova | 15:39 | |
*** cleong has joined #openstack-nova | 15:40 | |
dansmith | jaypipes: so you're going to have to convert the singular compute_node_get calls as well | 15:40 |
dansmith | jaypipes: otherwise we only see the new values when we query them all | 15:40 |
*** e0ne has quit IRC | 15:40 | |
*** e0ne_ has joined #openstack-nova | 15:41 | |
lbeliveau | Is there another core that could review this fix simple please, already got a +2 but need workflow: https://review.openstack.org/#/c/216049 | 15:41 |
*** sudipto has quit IRC | 15:41 | |
jaypipes | dansmith: ah, yes. Can I do that in a followup patch, or will it need to be in this same one? | 15:41 |
mriedem | xenserver ci seems borked | 15:41 |
mriedem | BobBall: ^ | 15:41 |
jaypipes | dansmith: trying to keep the amount of code down | 15:41 |
dansmith | jaypipes: it has to be underneath mine, regardless of how you want to do it | 15:41 |
dansmith | jaypipes: seems like it might be most contextually-relevant in this one, but as long as it's in front of mine, I don't care | 15:42 |
PaulMurray | danpb, I guess strictly speaking asynchrony between two api servers could invalidate that check | 15:42 |
*** zenoway has joined #openstack-nova | 15:43 | |
*** mtanino has joined #openstack-nova | 15:43 | |
*** knikolla has quit IRC | 15:43 | |
edleafe | jaypipes: want me to work on the singular patch? | 15:44 |
*** achanda has quit IRC | 15:45 | |
mriedem | Vek: are you around to review https://review.openstack.org/#/c/278454/ ? | 15:45 |
mriedem | johnthetubaguy: might be worth sending the rax ping to Vek for ^ | 15:45 |
mriedem | since we're approaching the freeze | 15:46 |
dansmith | jaypipes: so I will base my current patch on your current one and put it up so people can start looking at it, even though it fails because it's not getting the latest stuff out of the db | 15:46 |
dansmith | jaypipes: the save/create migrations work nice though | 15:46 |
*** tomoiaga has quit IRC | 15:46 | |
mriedem | johnthetubaguy: btw, when did you plan on doing the final novaclient release for mitaka? i see the schedule says it's 2/29-3/4 for client FF | 15:46 |
mriedem | *client release freeze | 15:46 |
*** zenoway has quit IRC | 15:47 | |
*** rotbeard has quit IRC | 15:48 | |
johnthetubaguy | mriedem: I was really hoping PaulMurray would say when all the live-migrate stuff lands, but that might be a bit tight | 15:48 |
mriedem | johnthetubaguy: that's blocked on ^ | 15:48 |
mriedem | hence the Vek ping | 15:49 |
johnthetubaguy | ah.. | 15:49 |
mriedem | http://zuul.openstack.xenproject.org/scoreboard/ is also down, so i think i'm going to block xenserver ci | 15:49 |
mriedem | BobBall: ^ fyi | 15:49 |
*** y_sawai has joined #openstack-nova | 15:49 | |
*** aorourke-afk is now known as aorourke | 15:49 | |
PaulMurray | mriedem, johnthetubaguy this patch is blocking others - needs approval https://review.openstack.org/#/c/278454/13 | 15:49 |
PaulMurray | mriedem, johnthetubaguy then next was previously approved, just rebased - needs +W again | 15:51 |
PaulMurray | The next two need reviews but can be done quick enough | 15:52 |
*** e0ne_ has quit IRC | 15:52 | |
PaulMurray | by reviews I mean core reviews | 15:52 |
*** mtanino has quit IRC | 15:52 | |
PaulMurray | johnthetubaguy, mriedem when is the freeze exactly? | 15:52 |
mriedem | PaulMurray: release freeze for clients is 2/29-3/4 | 15:53 |
PaulMurray | mriedem, I think we should be able to get these four patches in for that | 15:53 |
johnthetubaguy | yeah, probably thursday morning | 15:53 |
PaulMurray | I panicked for a minute then, I thought you were going to say tomorrow | 15:54 |
openstackgerrit | Andrew Laski proposed openstack/nova: Create BuildRequest object during boot process https://review.openstack.org/278125 | 15:54 |
openstackgerrit | Andrew Laski proposed openstack/nova: Add build_requests database table and model https://review.openstack.org/263926 | 15:54 |
openstackgerrit | Andrew Laski proposed openstack/nova: Create Instance from BuildRequest if not in a cell https://review.openstack.org/263927 | 15:54 |
openstackgerrit | Andrew Laski proposed openstack/nova: Add BuildRequest object https://review.openstack.org/278124 | 15:54 |
openstackgerrit | Andrew Laski proposed openstack/nova: Make InstanceMappings.cell_id nullable https://review.openstack.org/278598 | 15:54 |
openstackgerrit | Andrew Laski proposed openstack/nova: Populate instance_mappings during boot https://review.openstack.org/263925 | 15:54 |
mriedem | PaulMurray: +W on the 2.22 patch | 15:55 |
mriedem | heh, 2.23 and 2.22 seem to be in the wrong order, | 15:56 |
mriedem | 2.22 allows you to cancel a live migration, | 15:56 |
mriedem | but requires a migration id | 15:56 |
mriedem | 2.23 provides the API to get the migrations | 15:56 |
mriedem | to use in 2.22 | 15:56 |
mriedem | oops | 15:56 |
*** angdraug has joined #openstack-nova | 15:58 | |
PaulMurray | mriedem, that's the way they were landed | 15:58 |
PaulMurray | in nova | 15:58 |
PaulMurray | actually you can get the id from the old migrations api | 15:58 |
PaulMurray | so from novaclient point of view it is wrong, but you can do it | 15:59 |
mriedem | PaulMurray: yeah true http://developer.openstack.org/api-ref-compute-v2.1.html#os-migrations-v2.1 | 15:59 |
*** knikolla has joined #openstack-nova | 15:59 | |
*** mgoddard_ has joined #openstack-nova | 15:59 | |
mriedem | so now we have 2 APIs to list migrations? | 15:59 |
johnthetubaguy | mriedem: almost yes | 16:00 |
PaulMurray | the old one is being deprecated | 16:00 |
PaulMurray | with adding the new one | 16:00 |
mriedem | is there a simple way to explain why? | 16:00 |
mriedem | we can't ever remove the old one unless we bump the minimum required nova microversion | 16:00 |
PaulMurray | no | 16:00 |
PaulMurray | the new one is treating a migration as a sub resource of an instance | 16:01 |
PaulMurray | the old one lists all migrations in the system | 16:01 |
*** burt has joined #openstack-nova | 16:01 | |
PaulMurray | it was 'decided' that the new one is inline with the direction we are taking in nova as a whole | 16:01 |
*** apearson has joined #openstack-nova | 16:02 | |
bauzas | mriedem: just a question coming out of mind now that we're close to FF, are you planning to provide a dot release for the liberty branch ? | 16:02 |
*** mgoddard has quit IRC | 16:03 | |
mriedem | bauzas: yeah, it's on my todo list | 16:03 |
*** mjura has quit IRC | 16:03 | |
bauzas | mriedem: because I remember you claimed for review help on the stable branch | 16:03 |
openstackgerrit | Paul Murray proposed openstack/nova: Abort an ongoing live migration https://review.openstack.org/277971 | 16:03 |
*** mtanino has joined #openstack-nova | 16:03 | |
mriedem | bauzas: yeah https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/liberty | 16:04 |
mriedem | there isn't much | 16:04 |
*** garyk has quit IRC | 16:04 | |
bauzas | k | 16:05 |
bauzas | alaski: I saw you having some merge conflicts | 16:05 |
bauzas | alaski: for the cells patches | 16:05 |
bauzas | alaski: fancy rebasing those ? | 16:05 |
openstackgerrit | Dan Smith proposed openstack/nova: WIP: Migrate compute node stats to Inventory objects https://review.openstack.org/279313 | 16:06 |
openstackgerrit | Dan Smith proposed openstack/nova: Add Service.get_minimum_version_multi() for multiple binaries https://review.openstack.org/279312 | 16:06 |
alaski | bauzas: yeah, I just pushed up new ones about 10 minutes ago | 16:06 |
dansmith | jaypipes: cdent ^ | 16:06 |
*** harlowja_at_home has joined #openstack-nova | 16:06 | |
*** dims has quit IRC | 16:06 | |
bauzas | alaski: ooooh | 16:06 |
mriedem | alaski: i'll get on https://review.openstack.org/#/c/263926/ today | 16:06 |
bauzas | alaski: so https://review.openstack.org/#/c/263926/ as a top notch ? | 16:07 |
alaski | mriedem: thanks | 16:07 |
alaski | bauzas: yeah, that's the starting point for that series | 16:07 |
*** moshele has quit IRC | 16:07 | |
bauzas | k | 16:07 |
PaulMurray | melwitt, are you there? I'm looking for someone to review this patch on python-novaclient and I haven't tried you yet: https://review.openstack.org/#/c/278454/ | 16:08 |
* bauzas refueling his review pipe | 16:08 | |
PaulMurray | melwitt, it has a +2 | 16:08 |
*** mylu has joined #openstack-nova | 16:08 | |
*** diogogmt has quit IRC | 16:08 | |
*** drankis has quit IRC | 16:11 | |
*** y_sawai has quit IRC | 16:13 | |
*** sdague has joined #openstack-nova | 16:14 | |
*** y_sawai has joined #openstack-nova | 16:14 | |
openstackgerrit | ShaoHe Feng proposed openstack/nova: Add new APIs and deprecate old API for migrations https://review.openstack.org/258771 | 16:15 |
*** daemontool__ has quit IRC | 16:16 | |
openstackgerrit | Paul Murray proposed openstack/nova: Abort an ongoing live migration https://review.openstack.org/277971 | 16:17 |
*** y_sawai has quit IRC | 16:18 | |
cdent | pe | 16:18 |
mriedem | alaski: at some point we might want to amend http://specs.openstack.org/openstack/nova-specs/specs/mitaka/approved/cells-scheduling-interaction.html to point out there are data model impacts | 16:20 |
sdague | johnthetubaguy: we completely agreed to the sudo like privsep | 16:20 |
sdague | during the midcycle, with all the cinder folks plugged in | 16:20 |
sdague | I have no idea why this got raised again | 16:20 |
*** jlanoux has joined #openstack-nova | 16:21 | |
*** y_sawai has joined #openstack-nova | 16:22 | |
openstackgerrit | Swaminathan Vasudevan proposed openstack/nova: Implement setup_networks_on_host for Neutron networks https://review.openstack.org/275073 | 16:22 |
alaski | mriedem: the data model impact is described in https://review.openstack.org/#/c/169901/4/specs/liberty/approved/add-buildrequest-obj.rst, which apparently didn't get carried over to M | 16:23 |
alaski | mriedem: but since I've been assigning patches to the scheduling-interaction one I can update it | 16:23 |
*** jlanoux_ has joined #openstack-nova | 16:24 | |
mriedem | so http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/add-buildrequest-obj.html#data-model-impact | 16:24 |
*** haomaiwa_ has joined #openstack-nova | 16:24 | |
openstackgerrit | ShaoHe Feng proposed openstack/nova: Abort an ongoing live migration https://review.openstack.org/277971 | 16:24 |
openstackgerrit | ShaoHe Feng proposed openstack/nova: Add new APIs and deprecate old API for migrations https://review.openstack.org/258771 | 16:24 |
alaski | mriedem: yep | 16:24 |
*** jlanoux has quit IRC | 16:24 | |
*** rtheis_ has joined #openstack-nova | 16:25 | |
*** rtheis has quit IRC | 16:25 | |
*** drankis has joined #openstack-nova | 16:26 | |
*** belmoreira has quit IRC | 16:27 | |
bauzas | alaski: mriedem: yeah I had some concerns about the fields | 16:28 |
bauzas | I had no time yet to ask why we needed both an instance_uuid again but also an Enum for some RBAC policy rules | 16:29 |
johnthetubaguy | sdague: I am glad you said that, I was thinking I was imagining things, I have VC links between rooms | 16:29 |
johnthetubaguy | s/have VC/hate VC/ | 16:29 |
PaulMurray | shaohe_feng, I just saw you post a change - are you around? | 16:29 |
sdague | yeh, and I was pretty sure DuncanT spoke up during that conversation | 16:29 |
alaski | bauzas: ultimately everything is in there in order to have the method in https://review.openstack.org/#/c/263927/4/nova/objects/build_request.py | 16:29 |
*** gszasz has quit IRC | 16:30 | |
*** jlanoux_ has quit IRC | 16:30 | |
*** claudiub|2 has quit IRC | 16:30 | |
*** mylu has quit IRC | 16:30 | |
bauzas | alaski: oh the locked_by thing ? | 16:30 |
bauzas | man, I hate that | 16:30 |
*** zenoway has joined #openstack-nova | 16:30 | |
sdague | johnthetubaguy / PaulMurray - was offline for a bit, how do the live migration API patches stand? | 16:31 |
*** jlanoux has joined #openstack-nova | 16:31 | |
alaski | bauzas: it's just copying what's on the instance | 16:31 |
*** mylu has joined #openstack-nova | 16:31 | |
bauzas | alaski: that said, https://review.openstack.org/#/c/263927/4/nova/objects/build_request.py can still get instance_uuid from the spec obj ;) | 16:31 |
bauzas | alaski: yeah, I'm just double-checking that | 16:31 |
bauzas | alaski: if so, I apologize for the Enum | 16:31 |
PaulMurray | sdague, shaohe_feng and I just pushed new versions of https://review.openstack.org/25877 and https://review.openstack.org/277971 | 16:32 |
*** davideagle_ has quit IRC | 16:32 | |
alaski | bauzas: it can. it's there in order to look up the BuildRequest by uuid. I don't love the duplication but it greatly simplifies querying | 16:32 |
PaulMurray | sdague, oops, wrong patch | 16:32 |
bauzas | alaski: right, I see | 16:32 |
bauzas | alaski: you plan to have some BuildRequest.get_by_instance_uuid ? | 16:33 |
*** MCoLo has quit IRC | 16:33 | |
PaulMurray | sdague, https://review.openstack.org/277971 and https://review.openstack.org/258771 | 16:33 |
bauzas | alaski: if that, I was wondering why not joining with the ReqSpec | 16:33 |
alaski | bauzas: https://review.openstack.org/#/c/278124/8/nova/objects/build_request.py | 16:33 |
*** silos has joined #openstack-nova | 16:33 | |
mriedem | bauzas: alaski: i'm still going through it | 16:33 |
johnthetubaguy | sdague: yeah, there might be a few more open questions, but closer I hope | 16:34 |
johnthetubaguy | I need to hit those | 16:34 |
bauzas | since we agreed to use the RequestSpec initially, but provide a BuildRequest for any external items that weren't in the Spec object | 16:34 |
*** MCoLo has joined #openstack-nova | 16:34 | |
*** Bjoern_ has joined #openstack-nova | 16:35 | |
*** Bjoern_ is now known as Bjoern_zZzZzZzZ | 16:35 | |
*** zenoway has quit IRC | 16:35 | |
alaski | bauzas: I can join on request_spec, I just took the easy route first | 16:35 |
*** claudiub has joined #openstack-nova | 16:36 | |
alaski | I mean, I do join on that. But I can lookup by request_spec.instance_uuid rather than on build_request | 16:36 |
*** y_sawai has quit IRC | 16:36 | |
bauzas | alaski: I just want to make sure that RequestSpec is the initial main object, and BuildRequest is just an extending object for Cell V2 | 16:37 |
*** y_sawai has joined #openstack-nova | 16:37 | |
*** Bjoern_zZzZzZzZ is now known as Bjoern_ | 16:37 | |
bauzas | alaski: but I should think more of that | 16:37 |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Fix wrong watermark state during live migration https://review.openstack.org/279616 | 16:37 |
mriedem | alaski: bauzas: where is the request_specs model defined? | 16:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: Adds HostCapabilities object model and updates ComputeNode object https://review.openstack.org/228280 | 16:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: Adds host_capabilities to HostState https://review.openstack.org/254747 | 16:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: scheduler: Adds HostCapabilitiesFilter filter https://review.openstack.org/255073 | 16:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: Adds ResourceTracker host capabilities update https://review.openstack.org/250166 | 16:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: driver: Adds get_hypervisor_capabilities method https://review.openstack.org/254124 | 16:38 |
openstackgerrit | Claudiu Belu proposed openstack/nova: db: Adds host_capabilities column in compute_nodes https://review.openstack.org/228279 | 16:38 |
*** salv-orlando has joined #openstack-nova | 16:38 | |
alaski | mriedem: db/sqlalchemy/api_models.py | 16:38 |
*** sheel has joined #openstack-nova | 16:39 | |
mriedem | oh gdi | 16:39 |
mriedem | yeah, | 16:39 |
mriedem | my github was stuck on an old hash | 16:39 |
bauzas | mriedem: so the Spec object is pretty small in the DB, just a blob and an index :) | 16:39 |
*** y_sawai_ has joined #openstack-nova | 16:39 | |
bauzas | the index being the instance uuid | 16:39 |
*** silos has left #openstack-nova | 16:39 | |
*** y_sawai has quit IRC | 16:41 | |
*** salv-orlando has quit IRC | 16:41 | |
sdague | PaulMurray: these seem good to me | 16:41 |
*** paul-carlton1 has quit IRC | 16:41 | |
*** salv-orlando has joined #openstack-nova | 16:41 | |
PaulMurray | sdague, good - thanks - there is a novaclient patch that needs +W sitting in our way: https://review.openstack.org/#/c/278454/13 | 16:42 |
PaulMurray | sdague, I feel really bad because I've been bleating on about it to anyone on this channel - any chance you could look at that please ? | 16:43 |
sdague | PaulMurray: that patch is approved? | 16:43 |
sdague | PaulMurray: yeh, Vek got you 5 minutes ago | 16:44 |
PaulMurray | sdague, ah - Vek got there - thanks Vek | 16:44 |
PaulMurray | sdague, and I didn't even ask him ! | 16:45 |
PaulMurray | maybe I should take something away from that.... | 16:45 |
sdague | :) | 16:45 |
*** y_sawai_ has quit IRC | 16:45 | |
*** EinstCrazy has quit IRC | 16:46 | |
mriedem | ooo boy gerrit is slow | 16:46 |
mriedem | alaski: so on this build_requests object, with things like info_cache and instance_metadata, we're just avoiding new tables and joins on other tables and just storing serialized results of those in the build_requests entry? | 16:47 |
alaski | mriedem: exactly. this is just short term storage until a cell is picked by the scheduler at which point that will all be written normally into the cell | 16:48 |
mriedem | and if it is banished to cell0... | 16:49 |
mriedem | then we need enough to show details about the instance... | 16:49 |
*** y_sawai has joined #openstack-nova | 16:49 | |
alaski | cell0 is a normal cell db, so it can still write it all out properly | 16:49 |
*** y_sawai has quit IRC | 16:49 | |
*** josecastroleon has quit IRC | 16:49 | |
*** rdopiera has quit IRC | 16:50 | |
mriedem | what about the flavor? | 16:50 |
johnthetubaguy | flavor only lives at the api_db layer, and cached in the instance record I thought? | 16:50 |
alaski | mriedem: that's in the request_spec | 16:51 |
alaski | mriedem: https://review.openstack.org/#/c/263927/4/nova/objects/build_request.py | 16:51 |
mriedem | https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api_migrations/migrate_repo/versions/004_add_request_spec.py#L33 ? | 16:51 |
alaski | that generates the instance for the API | 16:51 |
alaski | johnthetubaguy: right. this is just clarifying where that exists before an instance record exists | 16:52 |
ababich | sdague: Hi! could you please look at https://review.openstack.org/#/c/280147 & | 16:52 |
*** ivase has joined #openstack-nova | 16:53 | |
*** y_sawai has joined #openstack-nova | 16:53 | |
mriedem | alaski: sorry, i'm looking at the request_specs table in the api db and i'm only see the spec blob | 16:53 |
mriedem | https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api_migrations/migrate_repo/versions/004_add_request_spec.py#L33 | 16:53 |
mriedem | so i'm assuming the flavor info is in there, | 16:53 |
mriedem | but where is that pulled out to populate the RequestSpec object to be used in https://review.openstack.org/#/c/263927/4/nova/objects/build_request.py | 16:53 |
*** mylu has quit IRC | 16:53 | |
alaski | mriedem: https://github.com/openstack/nova/blob/master/nova/objects/request_spec.py | 16:54 |
alaski | that's a better model to look at | 16:54 |
mriedem | so https://github.com/openstack/nova/blob/master/nova/objects/request_spec.py#L369 | 16:54 |
*** dims has joined #openstack-nova | 16:54 | |
alaski | yep. the db just stores the primitive object | 16:54 |
*** amotoki has joined #openstack-nova | 16:54 | |
*** mylu has joined #openstack-nova | 16:54 | |
*** y_sawai has quit IRC | 16:55 | |
*** browne has joined #openstack-nova | 16:55 | |
mriedem | ok | 16:55 |
ababich | sdague: Thanks! | 16:55 |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Console log copy for block live migratiob https://review.openstack.org/284674 | 16:56 |
mriedem | some of this code could use some documentation :) | 16:56 |
mriedem | 'wondering wtf this is?! go look here!' | 16:56 |
*** rfolco has quit IRC | 16:56 | |
mriedem | but meh | 16:56 |
mriedem | once it's in, if anyone ever asks i'll just direct them to alaski and bauzas | 16:56 |
alaski | mriedem: heh. I do comment a lot in later patches when this gets used. But if you point out those bits I'm happy to document stuff | 16:57 |
*** yamahata has joined #openstack-nova | 16:58 | |
johnthetubaguy | are we close enough to have a "how your instance gets created in a cells v2 world" sort of document? | 16:58 |
johnthetubaguy | covering the mapping, cell0, etc, would be fairly cool | 16:58 |
*** rfolco has joined #openstack-nova | 16:59 | |
*** rfolco has quit IRC | 16:59 | |
alaski | johnthetubaguy: the specifics are still spread out in specs, but the high level hasn't changed from https://github.com/openstack/nova/blob/master/doc/source/cells.rst | 16:59 |
alaski | but a lower level writeup would be good to have | 16:59 |
*** haomaiwa_ has quit IRC | 17:01 | |
*** haomaiwang has joined #openstack-nova | 17:01 | |
johnthetubaguy | alaski: yeah, I guess the low level ones are the ones that quickly go out of date, maybe its all normal in a month or two | 17:01 |
*** y_sawai has joined #openstack-nova | 17:02 | |
mriedem | alaski: alright, -1 on https://review.openstack.org/#/c/263926/ with comments inline | 17:04 |
mriedem | bauzas: fyi http://docs.sqlalchemy.org/en/rel_1_0/core/metadata.html?highlight=nullable#sqlalchemy.schema.Column.params.nullable | 17:04 |
*** thorst is now known as thorst_afk | 17:04 | |
mriedem | nullable=True is the default | 17:04 |
cdent | dansmith: Is an objects.Aggregate.get_by_uuid already in progress? | 17:04 |
*** sfinucan has quit IRC | 17:06 | |
mriedem | bauzas: you might be good to review this backport https://review.openstack.org/#/c/276191/ | 17:06 |
*** zenoway has joined #openstack-nova | 17:06 | |
alaski | mriedem: bauzas that's why I don't use the default, because it's hard to remember what it is :) | 17:07 |
*** mgoddard_ has quit IRC | 17:07 | |
mriedem | there was an epic bikeshed in a cinder meeting a few weeks ago for a similar discussion | 17:07 |
*** mgoddard has joined #openstack-nova | 17:07 | |
smcginnis | We like bikes. | 17:07 |
openstackgerrit | Kevin Benton proposed openstack/nova: Mark 'network_device_mtu' as deprecated https://review.openstack.org/284818 | 17:08 |
*** jistr has quit IRC | 17:08 | |
*** cdent has quit IRC | 17:09 | |
alaski | if it was clear what the default would be then I would be all for it, but in this case one isn't more clearly right | 17:09 |
alaski | though if I had to pick I would expect the default to be False | 17:10 |
mriedem | i just remember jogo going through the model once and dropping all of the default=True kwargs | 17:10 |
mriedem | probably b/c we had default=False, default=True, and some with neither specified | 17:10 |
mriedem | so i've tended to try and keep that enforced | 17:10 |
*** zenoway has quit IRC | 17:10 | |
mriedem | *RIP jogo* | 17:11 |
mriedem | *sniff* | 17:11 |
alaski | heh | 17:11 |
alaski | I'm fine dropping the nullable=True | 17:11 |
dansmith | I hate the assumption that nullable is true, fwiw | 17:12 |
wznoinsk | alex_xu and/or other cores, could someone have a look at my doc fixing and let me know is it GTG or not? (https://review.openstack.org/#/c/280299/) | 17:12 |
bauzas | mriedem: man, that happens when you review when doing 2 things at the same time... forgetting the default value :) | 17:12 |
*** scheuran has quit IRC | 17:12 | |
* bauzas bbiab | 17:12 | |
*** thorst_afk is now known as thorst | 17:12 | |
*** paul-carlton1 has joined #openstack-nova | 17:16 | |
*** drankis has quit IRC | 17:17 | |
*** RuiChen has quit IRC | 17:19 | |
*** penick has joined #openstack-nova | 17:20 | |
*** RuiChen has joined #openstack-nova | 17:20 | |
*** krtaylor has quit IRC | 17:21 | |
*** READ10 has joined #openstack-nova | 17:22 | |
*** mylu has quit IRC | 17:22 | |
*** Swami has quit IRC | 17:25 | |
*** haplo37 has joined #openstack-nova | 17:25 | |
*** su_zhang has quit IRC | 17:26 | |
*** su_zhang has joined #openstack-nova | 17:26 | |
mriedem | dansmith: you had a bug open for nixing lazy-load of things from the instance object right? | 17:27 |
dansmith | mriedem: things in general | 17:28 |
mriedem | nvm i'll find it | 17:28 |
mriedem | bug 1540526 | 17:28 |
openstack | bug 1540526 in OpenStack Compute (nova) "Too many lazy-loads in predictable situations" [Undecided,In progress] https://launchpad.net/bugs/1540526 - Assigned to Dan Smith (danms) | 17:28 |
dansmith | cool | 17:28 |
sheel | Dear Core members: | 17:28 |
sheel | need some eyes on https://review.openstack.org/#/c/282862/ | 17:28 |
sheel | please find some spare time to review this small patch | 17:29 |
sheel | thank you!! | 17:29 |
*** nic has joined #openstack-nova | 17:29 | |
mriedem | sheel: didn't that come up in the cross-project meeting last week? | 17:30 |
mriedem | damn cdent isn't around | 17:30 |
mriedem | jaypipes: ^ | 17:30 |
dansmith | aren't we -2ing that? | 17:30 |
dansmith | sdague: ? | 17:30 |
sheel | mriedem: actually this is not related to db changes | 17:30 |
*** yassine__ has quit IRC | 17:30 | |
*** su_zhang has quit IRC | 17:30 | |
*** rtheis_ is now known as rtheis | 17:31 | |
mriedem | sheel: yeah, it's changing the 500 to a 400 | 17:31 |
sheel | mriedem: yep | 17:31 |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Remove an unnecessary variable in a unit test https://review.openstack.org/283575 | 17:31 |
*** yassine__ has joined #openstack-nova | 17:31 | |
dansmith | oh is this just checking for and handling the fail gracefully? | 17:32 |
mriedem | i'd like to see that handled more generically | 17:32 |
mriedem | yeah | 17:32 |
mriedem | rather than copying that around every getter in the db api | 17:32 |
*** krtaylor has joined #openstack-nova | 17:33 | |
*** y_sawai has quit IRC | 17:33 | |
jaypipes | sheel: I can try to do a review of that later today. right now I'm working on a patch for dstepanenko. | 17:33 |
sheel | jaypipes: thank you | 17:34 |
mriedem | sheel: more comments inline | 17:34 |
mriedem | handling in the REST API would be cool if possible, barring that i'd think an error handler wrapper in the db api would be at least mroe re-usable | 17:34 |
sheel | dansmith: yes, its regarding handling fail | 17:35 |
*** y_sawai has joined #openstack-nova | 17:35 | |
mriedem | honestly, this seems like something we could push down into oslo.db | 17:35 |
*** cdent has joined #openstack-nova | 17:35 | |
mriedem | but we'd still have to handle the oslo.db error and return a 400 | 17:35 |
sheel | mriedem: yep, we already had pushed this in oslo db | 17:35 |
*** zenoway has joined #openstack-nova | 17:35 | |
*** achanda has joined #openstack-nova | 17:36 | |
mriedem | patch? | 17:36 |
openstackgerrit | Mark Doffman proposed openstack/nova: Add manage command for cell0. https://review.openstack.org/275266 | 17:36 |
openstackgerrit | Mark Doffman proposed openstack/nova: WIP: Map instances to cell0 on failed schedule https://review.openstack.org/267827 | 17:36 |
openstackgerrit | Mark Doffman proposed openstack/nova: Add a cell and host mapping utility to nova-manage https://review.openstack.org/270565 | 17:36 |
*** cdent has quit IRC | 17:36 | |
openstackgerrit | Mark Doffman proposed openstack/nova: Manage db sync command for cell0 https://review.openstack.org/274885 | 17:36 |
*** davideagle has joined #openstack-nova | 17:36 | |
sheel | mriedem: https://review.openstack.org/#/c/265921/4/oslo_db/tests/sqlalchemy/test_exc_filters.py | 17:36 |
sheel | https://review.openstack.org/#/c/265921/4 | 17:36 |
doffm | alaski: Could you take a quick look at https://review.openstack.org/267827 | 17:36 |
*** cdent has joined #openstack-nova | 17:36 | |
*** y_sawai has quit IRC | 17:37 | |
doffm | Just the code is up, i'm fighting with the tests for that patch. The conductor unit tests are messing with my mind. :/ | 17:37 |
openstackgerrit | Ed Leafe proposed openstack/nova: Use new inventory schema in all compute_node gets https://review.openstack.org/284833 | 17:37 |
doffm | Want to make sure the idea is right before I keep fighting with the tests. | 17:37 |
edleafe | jaypipes: dansmith: first crack at genericizing the compute_node query changes | 17:37 |
edleafe | jaypipes: dansmith: ^^ | 17:38 |
*** mrkz has quit IRC | 17:38 | |
*** mrkz has joined #openstack-nova | 17:38 | |
*** jwcroppe has quit IRC | 17:38 | |
mriedem | sheel: so we need that oslo.db change to properly detect this for mysql and raise the DBDataError exception, right? | 17:39 |
*** jwcroppe has joined #openstack-nova | 17:39 | |
sheel | mriedem: right | 17:39 |
mriedem | sheel: that went into oslo.db 4.4.0 | 17:39 |
mriedem | global-requirements requires oslo.db>=4.1.0 | 17:39 |
sheel | mriedem: oops | 17:39 |
mriedem | so you need to raise the minimum required version of oslo.db in global-requirements to be sure we're getting that change from oslo.db | 17:39 |
sheel | mriedem: yes, right | 17:40 |
sheel | mriedem: good point | 17:40 |
johnthetubaguy | we might be very close to the deadline for changing that I guess | 17:40 |
alaski | doffm: commented | 17:41 |
*** Swami has joined #openstack-nova | 17:42 | |
doffm | alaski: Thanks. | 17:42 |
sheel | johnthetubaguy: hmm, deadline is much closer.. | 17:42 |
*** mylu has joined #openstack-nova | 17:43 | |
*** jwcroppe has quit IRC | 17:43 | |
claudiub | hello. if anyone can take a look at tiny bugfix, that already has a +2, i'd be greatful. :) https://review.openstack.org/#/c/258614/ | 17:43 |
openstackgerrit | Matt Riedemann proposed openstack/nova: DNM: see where instance.pci_devices is getting lazy-loaded https://review.openstack.org/284839 | 17:45 |
jaypipes | edleafe: reviewed. right track, can be limited though to a lot fewer lines of code changed. | 17:46 |
jaypipes | ok, back to dstepanenko's patch :) | 17:46 |
*** y_sawai has joined #openstack-nova | 17:46 | |
* jogo waves to mriedem | 17:46 | |
edleafe | jaypipes: ok, will take a look | 17:46 |
mriedem | he's alive! | 17:46 |
cdent | jaypipes: before you go, also dansmith your input desired too | 17:46 |
mriedem | claudiub: hyper-v ci? | 17:46 |
*** _ducttape_ has joined #openstack-nova | 17:46 | |
mriedem | i guess it was ok on 2/5 | 17:46 |
claudiub | mriedem: PS5: Build succeeded. | 17:46 |
DuncanT | sdague: It was me who spoke up. I underestimated how like the current model the new privsep temporary thing was going to be, but having seen a bit more where it is going I'm happy again | 17:47 |
cdent | jaypipes, dansmith: Working on ResourcePool it seems I either need to add get_by_uuid to Aggregate or just bypass using the object and go straight to the db. This is a question I hit often when working on object stuff. When do you chose to skip objects for related pieces. | 17:47 |
dansmith | cdent: you never bypass | 17:47 |
dansmith | cdent: and yes, you should add a get_by_uuid() I just didn't because I didn't need it | 17:48 |
jaypipes | I thought dansmith had a patch up that added uuid to aggregate? | 17:48 |
cdent | ^^ | 17:48 |
dansmith | jaypipes: just no query method for it | 17:48 |
cdent | I'll do that now and then put what I'm working on underneath it | 17:48 |
dansmith | you mean above | 17:48 |
* dansmith has strong feelings on patch-queue-endianess | 17:48 | |
cdent | I hear what you are saying but that just not how my mind works dansmith | 17:48 |
*** dedery has quit IRC | 17:49 | |
*** ljxiash has quit IRC | 17:49 | |
alaski | gerrit display is canonical IMO | 17:49 |
*** ducttape_ has quit IRC | 17:49 | |
dansmith | alaski: which means I'm right, right? | 17:49 |
cdent | gerrit's UI should never be considered canonical for anything. to say otherwise is pure stockhold syndrome | 17:49 |
dansmith | stockholm? | 17:50 |
cdent | stockhold, huh, that's about right | 17:50 |
cdent | I type by syllables apparently | 17:50 |
DuncanT | sdague: It came round again because the patch looked *nothing* like privsep was supposed to, rather than a bit like it but with lax validation, which is what I was expecting | 17:50 |
alaski | dansmith: depends, I'm not sure what you meant. to me below=before, above=after | 17:51 |
sdague | DuncanT: ok. | 17:51 |
johnthetubaguy | sheel: yeah, I was just checking it wasn't yesterday really :) | 17:51 |
dansmith | alaski: correct | 17:51 |
mriedem | claudiub: +W | 17:51 |
*** apoorvad has joined #openstack-nova | 17:52 | |
sheel | johnthetubaguy: so we have time for now for this... | 17:52 |
dansmith | edleafe: are you swizzling that patch now? | 17:52 |
mriedem | sdague: i think you will totally be in love with this now https://review.openstack.org/#/c/283837/ | 17:52 |
*** y_sawai has quit IRC | 17:53 | |
*** paul-carlton1 has quit IRC | 17:53 | |
*** sc68cal has quit IRC | 17:53 | |
sdague | mriedem: ideally you'd do this as an expectFailure | 17:54 |
*** y_sawai has joined #openstack-nova | 17:54 | |
sdague | mriedem: https://review.openstack.org/#/c/283261/3/nova/tests/functional/regressions/test_bug_1541691.py | 17:54 |
*** matrohon has quit IRC | 17:54 | |
*** browne has quit IRC | 17:54 | |
sdague | self.expectFailure("some message", ... ) | 17:55 |
sdague | then the test you expect to work | 17:55 |
*** claudiub has quit IRC | 17:55 | |
*** abitha has joined #openstack-nova | 17:55 | |
*** rmart04 has joined #openstack-nova | 17:55 | |
*** amotoki has quit IRC | 17:56 | |
openstackgerrit | Ed Leafe proposed openstack/nova: Use new inventory schema in all compute_node gets https://review.openstack.org/284833 | 17:56 |
sdague | although, because of the asserts off the client, that might not work here | 17:56 |
sdague | never mind, good with it | 17:57 |
cdent | Is this worth fixing: https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L5494 | 17:57 |
*** jlanoux has quit IRC | 17:57 | |
cdent | (writer being used where reader should be) | 17:57 |
sheel | mriedem: thanks for your comments, quite qualitative comments | 17:57 |
openstackgerrit | Nikola Dipanov proposed openstack/nova: live-mig: Add claims and proper resource tracking https://review.openstack.org/244489 | 17:57 |
* sdague adds a non fatal mode for the client as a thing we should do | 17:57 | |
ndipanov | PaulMurray, bauzas ^^ | 17:57 |
sheel | mriedem: response to your comments, (mentioning here to fasten up the process :) ) | 17:57 |
sdague | cdent: yes | 17:57 |
ndipanov | finally got around to resurrecting this one | 17:57 |
sheel | https://www.irccloud.com/pastebin/oAbNytAl/ | 17:57 |
ndipanov | please take a look guys | 17:57 |
cdent | sdague: should I make a bug or just fix it? | 17:57 |
sdague | cdent: just fix it | 17:58 |
cdent | roger | 17:58 |
sdague | there was a blueprint for all the engine facade work, you could tag it with that | 17:58 |
*** paul-carlton1 has joined #openstack-nova | 17:58 | |
*** jwcroppe has joined #openstack-nova | 17:58 | |
sdague | sheel: shouldn't we catch this at the json schema layer? | 17:58 |
PaulMurray | ndipanov, thanks, will do | 17:59 |
*** y_sawai has quit IRC | 17:59 | |
sheel | sdague: yes, but for now there is some bug in that | 17:59 |
sheel | sdague: extra characters around about of 1000 lines | 17:59 |
sheel | in output if 4 byte unicode is provided | 18:00 |
sheel | I could see same in keystone and nova | 18:00 |
sheel | in json schema validation | 18:00 |
sdague | sheel: this bug - https://bugs.launchpad.net/nova/+bug/1541691 ? | 18:00 |
openstack | Launchpad bug 1541691 in OpenStack Compute (nova) "server boot with leading and trailing white spaces in name, displays weird error message" [High,In progress] - Assigned to Alex Xu (xuhj) | 18:01 |
*** haomaiwang has quit IRC | 18:01 | |
sdague | or something else | 18:01 |
*** hemna has joined #openstack-nova | 18:01 | |
PaulMurray | mriedem, did I see you say that xenserver CI is playing up? | 18:01 |
sheel | yes this as well as one other(I raised that) | 18:01 |
*** haomaiwang has joined #openstack-nova | 18:01 | |
sheel | sdague: but both represent same | 18:01 |
dansmith | edleafe: we need the result of compute_node_update to do it as well :/ | 18:01 |
sdague | the fixes for 1541691 are all in flight and will probably land this week | 18:01 |
*** mylu has quit IRC | 18:02 | |
*** bjornar__ has joined #openstack-nova | 18:02 | |
sdague | once that is fixed... what more is missing? | 18:02 |
sheel | sdague: I think we can use that then | 18:02 |
sheel | sdague: once fixed | 18:02 |
sdague | ok, it seems like security groups don't get jsonschema validated today | 18:03 |
*** derekh has quit IRC | 18:03 | |
sdague | so I guess we'd need schema for that as well? | 18:03 |
sheel | sdague: yes | 18:03 |
sdague | I'd *much* rather catch this as the API then bounce it off the db | 18:03 |
openstackgerrit | Chris Dent proposed openstack/nova: Make db.aggregate_get a reader not a writer https://review.openstack.org/284846 | 18:04 |
sheel | sdague: agree o nit | 18:04 |
cdent | sdague: I reckon the amont of effort that ^^ change required justifies my wages | 18:04 |
sheel | sdague: agree on it | 18:04 |
sdague | cdent: heh | 18:05 |
sheel | sdague: but I need to see wat needs to fix in schema way | 18:05 |
sheel | sdague: may be require more time for that | 18:05 |
sdague | ok, how about you investigate a schema fix for this | 18:06 |
sdague | jsonschema that is | 18:06 |
sdague | to enforce at the API layer | 18:06 |
sdague | and see if we can get a suitable fix there | 18:06 |
sheel | sdague: yes, that seems acceptable way of fixing | 18:06 |
*** ndipanov has quit IRC | 18:07 | |
*** lucasagomes is now known as lucas-afk | 18:08 | |
sdague | BobBall: xenserver-ci needs to update their devstack-gate pull | 18:08 |
*** lpetrut has quit IRC | 18:08 | |
*** mingyu has quit IRC | 18:09 | |
*** mingyu has joined #openstack-nova | 18:09 | |
*** READ10 has quit IRC | 18:10 | |
sheel | sdague: so, you want me to abandon this and work on jsonschema thing.... or go with this one in this version and update with jsonshema in next release? | 18:12 |
*** Bjoern_ is now known as Bjoern_zZzZzZzZ | 18:12 | |
sdague | sheel: I want to see what a json schema fix looks like | 18:12 |
sdague | to see if it can fully address the issue | 18:12 |
sheel | sdague: ok, sure | 18:12 |
sdague | because if it can, I don't think we want the db patch | 18:12 |
*** rmart04 has quit IRC | 18:12 | |
sdague | if it can't, we can discuss the db patch | 18:12 |
sheel | sdague: sounds good | 18:13 |
sdague | sheel: thanks | 18:13 |
sheel | sdague: thank you for discussion.. | 18:13 |
dansmith | edleafe: actually, scratch that, based on how compute_node_update() works | 18:13 |
*** su_zhang has joined #openstack-nova | 18:15 | |
*** mingyu has quit IRC | 18:16 | |
*** mingyu has joined #openstack-nova | 18:16 | |
edleafe | dansmith: what about the compute_node_statistics method? | 18:16 |
dansmith | edleafe: I dunno | 18:16 |
edleafe | dansmith: it didn't break any tests, so I had my fingers crossed :) | 18:17 |
*** wanghao has quit IRC | 18:17 | |
dansmith | yeah, let's ignore for now | 18:17 |
*** Bjoern_zZzZzZzZ is now known as Bjoern_ | 18:17 | |
*** wanghao has joined #openstack-nova | 18:17 | |
*** annasort has quit IRC | 18:19 | |
*** penick has quit IRC | 18:20 | |
*** rk4n has quit IRC | 18:20 | |
* edleafe heads off to lunch | 18:21 | |
dansmith | edleafe: thanks for doing that | 18:22 |
*** ihrachys has quit IRC | 18:24 | |
sdague | so.... why are we doing that _do_test() pattern instead of context manager mock? | 18:25 |
*** penick has joined #openstack-nova | 18:25 | |
sdague | because - https://review.openstack.org/#/c/281915/1/nova/tests/unit/compute/test_compute_mgr.py shows the definite danger of it | 18:25 |
alaski | for one mock it shouldn't be used | 18:26 |
alaski | when you have multiple that pattern is much cleaner, though admittedly dangerous | 18:26 |
sdague | yeh, the danger seems pretty bad | 18:27 |
sdague | it's getting cargo culted all over the place | 18:27 |
dansmith | it's infinitely better, IMHO | 18:27 |
alaski | I seem to recall there was some issue with using multiple context managers and line breaks | 18:27 |
dansmith | we have cases where we assert mock properties that don't exist too | 18:28 |
sdague | dansmith: except for the part where people forget the _dot_test | 18:28 |
dansmith | yeah, the context mock is just terrrrrible | 18:28 |
dansmith | sdague: yeah, I get it.. I'm saying we assert mock properties that don't exist too | 18:28 |
sdague | dansmith: sorry, explain? | 18:28 |
dansmith | self.assertTrue(mymockything.is_called) | 18:28 |
dansmith | passes because is_called is truthy but we didn't actually call the thing | 18:29 |
dansmith | because it's .called not .is_called | 18:29 |
sdague | ok | 18:29 |
sdague | mock protects us from a bunch of that now | 18:29 |
dansmith | it's gotten a little better yeah | 18:29 |
dansmith | but the nature of the thing doesn't really make it possible to protect it all | 18:30 |
dansmith | I'm just saying, please don't make use use the terrible context mock just because you can forget to call your thing | 18:30 |
sdague | sure, ok, but that's sort of a different issue | 18:30 |
dansmith | same class of thing to me | 18:30 |
sdague | dansmith: why is it terrible? | 18:30 |
dansmith | because it's so messy and the with statement doesn't allow for line breaks in certain confusing cases | 18:31 |
dansmith | and it manifests really strangely | 18:31 |
dansmith | especially when you just get the usual no tests ran, subunit explosion, or whatever because it's a parse error | 18:31 |
dansmith | and multi-line with statements even where it's allowed is a lot uglier and harder to read than a nice neat stack of decorators | 18:31 |
dansmith | I'm surprised flake8 doesn't notice that a thing was defined and never called or referenced | 18:32 |
sdague | ok, so why not have that decorator stack on top of the test_ function? | 18:32 |
dansmith | I bet pylint would catch it | 18:32 |
*** PaulMurray has quit IRC | 18:32 | |
dansmith | sdague: because you can't patch objects that way | 18:32 |
dansmith | because they don't exist | 18:32 |
alaski | sdague: you can't when you're patching 'self' | 18:32 |
*** salv-orlando has quit IRC | 18:32 | |
openstackgerrit | Gleb Stepanov proposed openstack/nova: Fix backing file copying https://review.openstack.org/282275 | 18:33 |
dansmith | I mean, you can patch objects, but not things you have to create first | 18:33 |
*** hemna has quit IRC | 18:33 | |
*** maishsk has joined #openstack-nova | 18:33 | |
*** maishsk has left #openstack-nova | 18:34 | |
dansmith | edleafe: jaypipes: I think something is broken in that get patch.. I'm setting a column and then getting the thing back and it's different | 18:35 |
sdague | dansmith: ok, but you also missed it here - https://github.com/openstack/nova/commit/2d7a851199b12355b36a3982ea0fc7bb5a7cd11f | 18:35 |
sdague | it just seems like the current pattern is pretty error prone | 18:35 |
sdague | and hard to catch correctly in review | 18:36 |
dansmith | sdague: well, I suck, but seriously the alternative is not better, IMHO | 18:36 |
dansmith | why is it hard to catch? | 18:36 |
dansmith | defining a function with no call or reference to it seems pretty obvious to me | 18:36 |
sdague | dansmith: ok, if you want to build somethign for it. But we're clearly landing tests that don't actually run right now | 18:37 |
dansmith | sdague: where else did I miss it in that patch? | 18:37 |
sdague | in that patch, that's just the one instance | 18:37 |
dansmith | ah, 4333 O see | 18:37 |
dansmith | oh, no, nevermind | 18:38 |
sdague | but that's actually not fixed by https://review.openstack.org/#/c/281915/1/nova/tests/unit/compute/test_compute_mgr.py | 18:38 |
sdague | I just decided to read the rest of the file to see if there were other instances | 18:38 |
sdague | and there were | 18:38 |
*** salv-orlando has joined #openstack-nova | 18:39 | |
dansmith | yep, and you know why? I remember this now | 18:39 |
*** haomaiwang has quit IRC | 18:39 | |
dansmith | this was such a disaster of a thing to maintain | 18:39 |
dansmith | I consistently had git merge out those lines because it thought they were the same | 18:39 |
*** rodrigods has quit IRC | 18:40 | |
*** rodrigods has joined #openstack-nova | 18:40 | |
dansmith | which can happen to anything where you have some repetition and git thinks it is smart | 18:40 |
dansmith | if you look at that diff you see why it thinks it's right | 18:40 |
*** hemna has joined #openstack-nova | 18:40 | |
sdague | sure, I get it, I'm just questioning a pattern where there are many failure modes, including git trying to delete it out fromus | 18:41 |
dansmith | well, I question who would ever prefer the alternative.. so there :) | 18:41 |
dansmith | are you going to comment on that one patch to fix up the rest or should I do it separately? | 18:42 |
*** paul-carlton1 has quit IRC | 18:42 | |
sdague | I have an outstanding draft, I was just trying to understand more | 18:43 |
dansmith | okay | 18:43 |
sdague | I left a comment on it | 18:43 |
sdague | I'm still +2, because it fixes a real thing | 18:44 |
*** haomaiwang has joined #openstack-nova | 18:44 | |
sdague | but there are clearly other things to be fixed | 18:44 |
dansmith | okay, I will keep hacking on this rpools thing, but if it doesn't get fixed in this patch I'll gladly do it | 18:44 |
*** haomaiwang has quit IRC | 18:49 | |
openstackgerrit | Pavel Kholkin proposed openstack/nova: WIP: keypairs pagination bp https://review.openstack.org/284868 | 18:49 |
*** hemna has quit IRC | 18:52 | |
mriedem | gdi, now xenproject ci is hosed too | 18:53 |
mriedem | i'm going to turn off both i think | 18:53 |
openstackgerrit | Vladik Romanovsky proposed openstack/nova: Clean up networks with SR-IOV binding on reschedule https://review.openstack.org/283198 | 18:53 |
mriedem | the stakes are high and so am i .. | 18:53 |
sdague | mriedem: right, because of the ansible thing | 18:53 |
sdague | apparently they don't take latest devstack-gate | 18:54 |
*** mvk has joined #openstack-nova | 18:54 | |
mriedem | the xenserver ci log i looked at this morning was taking that actually | 18:54 |
mriedem | but still failed | 18:54 |
sdague | mriedem: really? | 18:54 |
sdague | not the one I just looked at | 18:54 |
sdague | https://review.openstack.org/#/c/284846/ | 18:54 |
*** y_sawai has joined #openstack-nova | 18:55 | |
sdague | 2016-02-25 18:11:01.009 | ++ sudo -H /tmp/ansible/bin/pip install ansible | 18:55 |
sdague | they think they have that code, but they aren't using it? | 18:55 |
mriedem | alright, xenproject ci also removed | 18:56 |
sdague | yeh, they have a fork | 18:56 |
sdague | 2016-02-25 18:10:48.407 | + git clone https://github.com/citrix-openstack/devstack-gate -b master | 18:56 |
*** lpetrut has joined #openstack-nova | 18:56 | |
sdague | which, you know, kind of pointless :) | 18:56 |
*** hemna has joined #openstack-nova | 18:58 | |
mriedem | sdague: damn, that missing call to _test() in that one test you pointed out, adding it makes the test fail - so the test is busted or sometihng :( | 18:59 |
mriedem | i'll see if i can figure it out | 18:59 |
*** Marga_ has quit IRC | 19:00 | |
openstackgerrit | Sean Dague proposed openstack/nova: Deprecate conductor manager option https://review.openstack.org/284332 | 19:00 |
*** y_sawai has quit IRC | 19:00 | |
*** davideagle has quit IRC | 19:00 | |
*** yassine__ has quit IRC | 19:01 | |
sdague | mriedem: or the code is busted | 19:01 |
*** mingyu has quit IRC | 19:01 | |
sdague | mriedem: on the fixing tests front - https://review.openstack.org/#/c/282983/1 is another good one | 19:02 |
*** davideagle_ has joined #openstack-nova | 19:02 | |
sdague | if you want to approve that | 19:02 |
*** mingyu has joined #openstack-nova | 19:02 | |
cdent | dansmith: I've done aggregate_get_by_uuid but I'm not sure where to put it. I need it for the resourcepool stuff, which will slot "above" this: https://review.openstack.org/#/c/281945/ but that stack is stalled pending approval of the modification to the resource providers spec. Should I put the aggregate stuff in that stack, or by itself on the hope/assumption that it will get merged sooner? | 19:04 |
*** josecastroleon has joined #openstack-nova | 19:06 | |
*** vilobhmm11 has joined #openstack-nova | 19:06 | |
*** inc0 has joined #openstack-nova | 19:07 | |
openstackgerrit | Thomas Maddox proposed openstack/nova: Try udevadm settle between lvremove attempts https://review.openstack.org/240611 | 19:07 |
SamYaple | inc0: ping | 19:08 |
inc0 | so guys, question about upgrade_levels | 19:08 |
inc0 | in kolla we do upgrades of nova in a way that we upgrade conductors first and rest of it later | 19:08 |
openstackgerrit | Andrew Laski proposed openstack/nova: Create BuildRequest object during boot process https://review.openstack.org/278125 | 19:09 |
openstackgerrit | Andrew Laski proposed openstack/nova: Add build_requests database table and model https://review.openstack.org/263926 | 19:09 |
openstackgerrit | Andrew Laski proposed openstack/nova: Create Instance from BuildRequest if not in a cell https://review.openstack.org/263927 | 19:09 |
openstackgerrit | Andrew Laski proposed openstack/nova: Add BuildRequest object https://review.openstack.org/278124 | 19:09 |
openstackgerrit | Andrew Laski proposed openstack/nova: Make InstanceMappings.cell_id nullable https://review.openstack.org/278598 | 19:09 |
openstackgerrit | Andrew Laski proposed openstack/nova: Populate instance_mappings during boot https://review.openstack.org/263925 | 19:09 |
inc0 | but sighupping nova services aftes whole process is done seems to be breaking (at least inside docker) | 19:09 |
*** penick has quit IRC | 19:09 | |
inc0 | do we really need it with upgrade_levels=auto? | 19:09 |
alaski | mriedem: bauzas I updated the cells patch series | 19:09 |
inc0 | dansmith ^ | 19:10 |
sdague | jaypipes: https://review.openstack.org/#/c/284257 - responded to your question | 19:10 |
sdague | missed it when I was rebasing last night | 19:10 |
dansmith | inc0: =auto if you want to avoid doing manual pins, and I hope to remove manual pins at some point | 19:11 |
jaypipes | sdague: ah, gotcha. cool, thanks for the explanation | 19:11 |
*** baoli_ has quit IRC | 19:11 | |
*** angdraug has quit IRC | 19:12 | |
inc0 | dansmith, and all of nova services should handle sighup for it right? | 19:12 |
dansmith | yes | 19:12 |
inc0 | or do we need to restart | 19:12 |
*** _ducttape_ has quit IRC | 19:12 | |
SamYaple | dansmith: the question is actually do we _need_ to sighup or can we restart (is restarting going to give the same effect)? | 19:16 |
SamYaple | the issue here is with docker and sighup and we dont have alot of time to track it down atm | 19:16 |
dansmith | SamYaple: yeah same effect | 19:17 |
SamYaple | i thought so. thanks dansmith ! | 19:17 |
dansmith | yep | 19:18 |
*** skoude has joined #openstack-nova | 19:18 | |
*** jichen has joined #openstack-nova | 19:19 | |
skoude | hmm.. I have a problem that instances try to access http://169.254.169.254/2009-04-04/meta-data/instance-id when instead they should access https://169.254.169.254/2009-04-04/meta-data/instance-id.. Any idea where to change the address? | 19:20 |
SamYaple | dansmith: to clear up some confusion. we would need to restart/sighup all nova services after the last nova-compute node is upgraded. does that mean _also_ the nova-compute services again? | 19:20 |
skoude | Tried to find solution, but haven't still found one.. | 19:20 |
dansmith | SamYaple: yeah | 19:20 |
SamYaple | ok we can work with that. thanks again dansmith | 19:21 |
skoude | I have defined nova_metadata_protocol = https, but still it does not connect as https | 19:21 |
*** fawadkhaliq has joined #openstack-nova | 19:22 | |
openstackgerrit | Chris Dent proposed openstack/nova: Add name to ResourceProvider object https://review.openstack.org/281945 | 19:24 |
*** markusz has quit IRC | 19:25 | |
*** ducttape_ has joined #openstack-nova | 19:27 | |
sdague | skoude: it's cloud init | 19:28 |
*** Sukhdev has joined #openstack-nova | 19:29 | |
*** salv-orlando has quit IRC | 19:29 | |
*** andymaier has quit IRC | 19:30 | |
*** redhat034 has joined #openstack-nova | 19:30 | |
mriedem | 'at least you can't complain about good american spooning' | 19:30 |
mriedem | nice quote from the neutron midcycle | 19:30 |
skoude | sdague: yeap, I know, but where do I define it globally? It worked before, but suse upgraded some packages and now it tries to connect http instead of https.. And I'm trying to figure out where is that config. | 19:31 |
openstackgerrit | Chris Dent proposed openstack/nova: Add a get_by_uuid for aggregates https://review.openstack.org/284895 | 19:31 |
sdague | mriedem: heh | 19:32 |
mriedem | people complaining about sharing a hotel room | 19:32 |
*** redhat034 has quit IRC | 19:33 | |
*** su_zhang has quit IRC | 19:35 | |
*** josecastroleon has quit IRC | 19:35 | |
*** su_zhang has joined #openstack-nova | 19:36 | |
*** Marga_ has joined #openstack-nova | 19:36 | |
mriedem | sdague: https://review.openstack.org/#/c/282983/ scares me | 19:36 |
*** salv-orlando has joined #openstack-nova | 19:36 | |
sdague | mriedem: I'm definitely partially to blame on that one | 19:37 |
sdague | because I was reviewing that code, but we were doing so many iterations at the end I missed those | 19:37 |
sdague | and was focussed on other issues with those patches | 19:38 |
sdague | at least on the libosinfo bits | 19:38 |
sdague | mriedem: also - because you loved this patch so much the first time - https://review.openstack.org/#/c/284257/ jaypipes +2ed it | 19:39 |
bauzas | alaski: got a wonder here https://review.openstack.org/#/c/263926/13/nova/db/sqlalchemy/api_migrations/migrate_repo/versions/006_build_request.py@65 | 19:39 |
sdague | bauzas: what is your magic to get @ into the urls? | 19:39 |
bauzas | sdague: thanks to gerrit | 19:40 |
mriedem | sdague: i had it open in a tab, | 19:40 |
bauzas | sdague: just see the hyperlink | 19:40 |
mriedem | but then saw you and dan fighting like an old married couple | 19:40 |
*** su_zhang has quit IRC | 19:40 | |
bauzas | when commenting | 19:40 |
mriedem | and now i'm fixing that other test thing | 19:40 |
bauzas | alaski: btw. FWIW will be on PTO tomorrow | 19:41 |
sdague | bauzas: nope, I don't see how to do it. | 19:41 |
*** su_zhang has joined #openstack-nova | 19:42 | |
openstackgerrit | Jay Pipes proposed openstack/nova: Added supports_requests method to PciDeviceList https://review.openstack.org/242087 | 19:42 |
bauzas | sdague: for example, look at a comment | 19:42 |
bauzas | sdague: you'll see every line having like line <nr> : my_comment | 19:42 |
alaski | bauzas: I'm out tomorrow as well | 19:43 |
alaski | bauzas: are you suggesting pulling the request_spec into the build_request? | 19:43 |
*** toure_ is now known as toure | 19:43 | |
bauzas | alaski: no, just removing the FK | 19:43 |
sdague | oh... | 19:43 |
bauzas | we're already constrained by the UC | 19:43 |
sdague | you have to leave a comment, then get it back at the top page | 19:43 |
bauzas | sdague: well, that's an hyperlink, but you can still provide any line number | 19:44 |
sdague | sure | 19:44 |
sdague | I just wanted it in the diff view | 19:44 |
alaski | bauzas: right, but that makes no guarantee that there is a request_spec | 19:44 |
bauzas | alaski: so the FK is useful for 3 points : | 19:44 |
bauzas | #1 the UC it creates | 19:44 |
*** markusz has joined #openstack-nova | 19:44 | |
bauzas | #2 the parallel removal it could have when deleting the parent key | 19:45 |
*** haomaiwang has joined #openstack-nova | 19:45 | |
bauzas | #3 for making sure that we have a correct id | 19:45 |
alaski | yeah, I'm interested in #3 | 19:46 |
bauzas | alaski: like I said, #1 is made by having an UC, and I think we can have #2 and #2 just by coding that | 19:46 |
mriedem | but if you bork up coding that, | 19:46 |
mriedem | the fkey is there to save you | 19:46 |
mriedem | if you delete the request_spec before the build_request | 19:46 |
mriedem | or attempt to | 19:46 |
alaski | I'm not really sure what I gain by not having it | 19:46 |
Swami | https://review.openstack.org/#/c/275073/ could I get some reviews on this patch - this for handling live migration in neutron. | 19:46 |
bauzas | alaski: I need to remember some old convo I saw about that one | 19:47 |
bauzas | mriedem: sure, but since we have a DB Facade, I thought it's just okay | 19:47 |
alaski | I know that some people really don't like them. But I think they add value | 19:47 |
bauzas | like the host/service lazy relationship we have | 19:47 |
bauzas | alaski: honestly, it's not a strong objection to having a FK | 19:48 |
*** lpetrut has quit IRC | 19:48 | |
bauzas | alaski: I'm not a DB guru so I'm taking both mriedem and your's points | 19:48 |
bauzas | alaski: but I just feel it's not really needed for our use, that's it :) | 19:49 |
mriedem | shall we summon zzzeek and morded since we're talking about foreign keys? :) | 19:49 |
mriedem | let's get ready to rumble!!!! | 19:49 |
bauzas | heh | 19:49 |
alaski | oh boy | 19:49 |
doffm | mriedem: Is there some special summoning dance? | 19:49 |
mriedem | no | 19:49 |
bauzas | who is calling the kracken ? | 19:49 |
*** baoli has joined #openstack-nova | 19:49 | |
mriedem | different epic | 19:50 |
doffm | With the DB split there are going to be a number of relationships that arn't enforced by the DB schema. | 19:50 |
bauzas | s/calling/awakening | 19:50 |
*** shyama has quit IRC | 19:50 | |
alaski | bauzas: the likelihood of a foreign key catching an issue is low right now because we're being careful adding all of this. But as we start messing with request_specs later I prefer to have the sanity check | 19:50 |
*** krotscheck is now known as krotscheck_dcm | 19:50 | |
*** baoli has quit IRC | 19:51 | |
bauzas | alaski: what I'm concerned is by keeping a loose relationship between both | 19:51 |
bauzas | alaski: because the Spec object is aimed to go down to the compute ideallyu | 19:51 |
doffm | But if the tables are in the same db... why not have the extra enforcement? | 19:52 |
mriedem | sdague: got that test passing, pushing up a rev soon | 19:52 |
mriedem | i'll just +W myself | 19:52 |
bauzas | that's fine to say that it's just an instance and that the persisted object will still be in the API DB, but I feel that conceptually, we only need to have this partnership for the initial call | 19:52 |
sdague | mriedem: sure, or show me the link and I'll wallaby | 19:53 |
bauzas | doffm: well, the point is that it doesn't scale great | 19:53 |
*** gmurphy has joined #openstack-nova | 19:53 | |
*** haomaiwang has quit IRC | 19:53 | |
bauzas | but since it's the API DB, I can say it's probably a nitpick | 19:53 |
bauzas | but I dunno | 19:53 |
alaski | bauzas: the buildrequest is short lived though. so the partnership goes away quickly | 19:53 |
bauzas | agreed | 19:54 |
*** lpetrut has joined #openstack-nova | 19:54 | |
bauzas | okay, it seems there is a consensus of 3 people having good opinions and me just picky with my +1 | 19:54 |
bauzas | :) | 19:54 |
bauzas | alaski: okay, will turn my comment | 19:55 |
alaski | there are definitely times that they don't make sense. I just don't see a downside for this usage and I like the extra validation, since we absolutely have to have a request_spec or the build_request is useless | 19:55 |
*** hemna has quit IRC | 19:56 | |
*** eil397 has joined #openstack-nova | 19:56 | |
*** sc68cal has joined #openstack-nova | 19:57 | |
bauzas | alaski: tbc, my concerns are pretty well explained in https://www.percona.com/blog/2006/12/12/innodb-locking-and-foreign-keys/ | 19:57 |
*** y_sawai has joined #openstack-nova | 19:57 | |
bauzas | I'm not saying it's the case here, just me trying to avoid FKS whenever it's possible | 19:57 |
bauzas | (that's an old blogpost I read a couple of years before, probably the mysql engine is performing better now) | 19:59 |
alaski | looks like it's more a concern for write heavy tables, which these are not | 20:00 |
*** danpb has quit IRC | 20:00 | |
bauzas | agreed | 20:00 |
bauzas | it just scales per request :) | 20:00 |
*** ljxiash has joined #openstack-nova | 20:01 | |
*** rk4n has joined #openstack-nova | 20:01 | |
openstackgerrit | Jay Pipes proposed openstack/nova: Added supports_requests method to PciDeviceList https://review.openstack.org/242087 | 20:02 |
*** y_sawai has quit IRC | 20:02 | |
bauzas | anyway, that's not big deal | 20:02 |
bauzas | that's something we can modify later on | 20:02 |
openstackgerrit | Chris Dent proposed openstack/nova: Add a get_by_uuid for aggregates https://review.openstack.org/284895 | 20:02 |
alaski | bauzas: I appreciate the thoroughness of the review. I'm always happy to get constructive feedback. | 20:03 |
openstackgerrit | Matt Riedemann proposed openstack/nova: Added missing execution of the test https://review.openstack.org/281915 | 20:03 |
mriedem | sdague: ^ | 20:03 |
*** eil397 has quit IRC | 20:04 | |
*** eil397 has joined #openstack-nova | 20:05 | |
*** ljxiash has quit IRC | 20:05 | |
openstackgerrit | Dan Smith proposed openstack/nova: WIP: Hacking check to detect uncalled closures in tests https://review.openstack.org/284905 | 20:09 |
dansmith | alaski: sdague ^ | 20:09 |
dansmith | I only found those two examples of this pattern failing, | 20:09 |
*** ducttape_ has quit IRC | 20:09 | |
dansmith | but lots of other things we've defined and orphaned in a test that can just go away | 20:09 |
dansmith | if that looks good to you then I'll fix those other issues | 20:09 |
dansmith | but wanted first impressions ... first | 20:10 |
*** su_zhang has quit IRC | 20:12 | |
*** su_zhang has joined #openstack-nova | 20:13 | |
*** fawadkhaliq has quit IRC | 20:13 | |
*** fawadkhaliq has joined #openstack-nova | 20:13 | |
alaski | you went a little more generic than my attempt so far. I have been trying to narrow down to closures decorated with mocks and checking for a specific call | 20:14 |
alaski | but the general check might be more useful | 20:14 |
*** su_zhang has quit IRC | 20:15 | |
*** penick has joined #openstack-nova | 20:15 | |
*** Marga_ has quit IRC | 20:15 | |
*** su_zhang has joined #openstack-nova | 20:15 | |
*** annasort has joined #openstack-nova | 20:15 | |
mriedem | sdague: do you remember anything recently where nova service-list was listing downed API services? | 20:15 |
*** Marga_ has joined #openstack-nova | 20:15 | |
dansmith | alaski: okay it seemed like the safest thing to go general, but it also hits a lot of things | 20:16 |
dansmith | alaski: interested to see what you come up with though for sure | 20:16 |
mriedem | Sam-I-Am was seeing something where nova service-list was showing downed API services but nova-api was running, | 20:16 |
dansmith | maybe both are useful | 20:16 |
alaski | yeah, I'll keep pushing on this | 20:16 |
mriedem | on my newish devstack those API services don't show up | 20:16 |
*** ducttape_ has joined #openstack-nova | 20:16 | |
*** inc0 has quit IRC | 20:16 | |
alaski | mriedem: I recall reviewing something about that not too long ago | 20:17 |
dansmith | mriedem: api services don't update their records | 20:17 |
alaski | mriedem: https://review.openstack.org/#/c/277881/ | 20:17 |
dansmith | the patch was to filter them out of service list Ithink | 20:17 |
dansmith | yeah that :) | 20:17 |
mriedem | nice, that's exactly it | 20:18 |
mriedem | thanks | 20:18 |
mriedem | i wonder if that's an issue in liberty | 20:18 |
*** salv-orlando has quit IRC | 20:18 | |
mriedem | or something regressed that in mitaka | 20:18 |
dansmith | nope just mitaka I think | 20:19 |
mriedem | cool | 20:19 |
dansmith | edleafe: you back? | 20:19 |
edleafe | dansmith: yup - just got back | 20:19 |
dansmith | edleafe: I assert that your patch is broken | 20:20 |
dansmith | but I don't know why | 20:20 |
*** salv-orlando has joined #openstack-nova | 20:20 | |
edleafe | dansmith: just because you know my work? :) | 20:20 |
dansmith | edleafe: I do a compute_node_update({'memory_mb': 2000}), get back a matching thing, then do another compute_node_get(id) and get back the original un-updated value | 20:20 |
dansmith | edleafe: heh, no :) | 20:20 |
edleafe | dansmith: hmmm - I haven't checked out compute_node_update | 20:21 |
dansmith | edleafe: thought I'd point it out in case you had any ideas or could write some quick tests to prove it's me and no you | 20:21 |
dansmith | and got distracted on something else before I could do it myself | 20:22 |
edleafe | dansmith: ok - just started working on the tests, so I'll add that to the list | 20:22 |
dansmith | okay thanks | 20:22 |
mriedem | sdague: shouldn't v2.1 also be in here? step 4 when creating compute endpoints: http://docs.openstack.org/liberty/install-guide-ubuntu/nova-controller-install.html#prerequisites | 20:23 |
*** kfarr has quit IRC | 20:23 | |
mriedem | or use v2.1 instead of v2, period | 20:23 |
*** Marga_ has quit IRC | 20:25 | |
*** Marga_ has joined #openstack-nova | 20:26 | |
*** Marga_ has quit IRC | 20:27 | |
*** sheel has quit IRC | 20:27 | |
*** su_zhang has quit IRC | 20:30 | |
dansmith | edleafe: http://pastebin.com/NDw6NZqL | 20:30 |
dansmith | was hoping I would be pasting "edsstuffisbroken: PASS" but .. :D | 20:30 |
*** su_zhang has joined #openstack-nova | 20:30 | |
dansmith | testtools.matchers._impl.MismatchError: 1024 != 512 | 20:30 |
dansmith | ooooh | 20:31 |
dansmith | I know what it is | 20:31 |
edleafe | dansmith: is update updating the right thing? | 20:31 |
dansmith | you're returning a dict now and compute_node_get is updating that dict, not the model | 20:31 |
dansmith | sorry, compute_node_update is updating the dict from _get | 20:32 |
edleafe | ah | 20:32 |
dansmith | but, | 20:32 |
edleafe | not the object | 20:32 |
dansmith | we need to keep some of that behavior | 20:32 |
dansmith | lemme make a change and paste you a diff | 20:32 |
edleafe | k | 20:32 |
*** su_zhang has quit IRC | 20:35 | |
*** shawnmaten has joined #openstack-nova | 20:37 | |
*** thorst is now known as thorst_afk | 20:38 | |
*** angdraug has joined #openstack-nova | 20:39 | |
*** toure is now known as toure_ | 20:39 | |
dansmith | Hmm, I dunno how to get sqla not to do a create | 20:39 |
*** jwcroppe has quit IRC | 20:40 | |
dansmith | edleafe: this, but working: http://pastebin.com/MW6zv4h2 | 20:40 |
dansmith | that .save() ends up doing an INSERT, but I just want it to UPDATE | 20:40 |
*** jwcroppe has joined #openstack-nova | 20:40 | |
dansmith | i.e. it would be nice to not have to query out the new way and the old way, update the old one and return the new one | 20:41 |
*** huats has quit IRC | 20:41 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: Update the help for deprecated glance host/port/protocol options https://review.openstack.org/284920 | 20:42 |
edleafe | dansmith: instead of creating a new ComputeNode object, can you do a get for the given id? | 20:42 |
*** matrohon has joined #openstack-nova | 20:42 | |
edleafe | or are the objects smart enough to update if the id exists? | 20:42 |
*** huats has joined #openstack-nova | 20:42 | |
dansmith | edleafe: that's a sqla model | 20:43 |
*** haplo37 has quit IRC | 20:43 | |
dansmith | edleafe: objects just call the db_api method (this one in this case) and pass the dict of values and expect it to be handled, however that may be | 20:43 |
edleafe | dansmith: doh! yeah, I was thinking Nova objects | 20:44 |
dansmith | zzzeek: if I want to just create a model, set the id and a few new values and get sqla to do an UPDATE instead of INSERT.. is there some trick? | 20:44 |
*** jwcroppe has quit IRC | 20:45 | |
dansmith | or jaypipes ^ | 20:45 |
jaypipes | dansmith: session.add(model)? | 20:46 |
dansmith | oh? | 20:46 |
dansmith | nope. | 20:47 |
*** porunov has joined #openstack-nova | 20:48 | |
mriedem | Swami: comments in https://review.openstack.org/#/c/275073/ | 20:49 |
mriedem | Swami: is that migration_to binding:host stuff new for neutron in mitaka? | 20:49 |
*** shawnmaten has left #openstack-nova | 20:49 | |
*** browne has joined #openstack-nova | 20:49 | |
*** rk4n has quit IRC | 20:50 | |
*** haomaiwang has joined #openstack-nova | 20:50 | |
sdague | mriedem: yeh | 20:51 |
sdague | the install guide should say v2.1 | 20:51 |
mriedem | sdague: ok, that's what i told Sam-I-Am | 20:52 |
*** ducttape_ has quit IRC | 20:52 | |
Swami | mriedem: yes it is new, it has not been handled so far. | 20:53 |
*** ducttape_ has joined #openstack-nova | 20:53 | |
*** thorst_afk is now known as thorst | 20:53 | |
mriedem | Swami: i see https://review.openstack.org/#/c/275420/ | 20:54 |
Swami | mriedem: since we already had the binding:profile support in ports, we just mad use of it to address the "migrating_to" | 20:54 |
bauzas | dansmith: jaypipes: where are we with resource-providers ? which change shall I review now ? I can see most of changes are dependent on https://review.openstack.org/#/c/281837/18 right? | 20:54 |
mriedem | yeah, binding:* is apparently just turning into a wild west of stuff to describe a port | 20:54 |
Swami | mriedem: yes you got the right patch. | 20:55 |
Swami | mriedem: I agree | 20:55 |
mriedem | is there a ci job that runs live migration with neutron? sdague? | 20:55 |
*** Marga_ has joined #openstack-nova | 20:55 | |
mriedem | even though this is specific to dvr... | 20:55 |
*** rtheis has quit IRC | 20:55 | |
sdague | mriedem: nothing voting yet | 20:55 |
sdague | there might be experimental jobs | 20:56 |
openstackgerrit | Ronald Bradford proposed openstack/nova: Define context.roles using base class https://review.openstack.org/284925 | 20:56 |
jaypipes | bauzas: https://review.openstack.org/#/c/242087/ could use your eyes. same with https://review.openstack.org/#/c/284484/ | 20:56 |
mriedem | was someone driving those? sc68cal? | 20:56 |
sc68cal | hello | 20:56 |
*** mylu has joined #openstack-nova | 20:57 | |
mriedem | sc68cal: neutron live migration + dvr - go! | 20:57 |
bauzas | jaypipes: ack | 20:57 |
mriedem | sc68cal: + testing! | 20:57 |
*** y_sawai has joined #openstack-nova | 20:57 | |
*** haomaiwang has quit IRC | 20:58 | |
openstackgerrit | Mark Doffman proposed openstack/nova: Map instances to cell0 on failed schedule https://review.openstack.org/267827 | 20:58 |
sc68cal | I think we test live migrations in the neutron-dsvm-multinode-dvr job... | 20:58 |
sc68cal | and I think the neutron-dsvm-multinode - but not 100% sure | 20:58 |
sc68cal | problem is - dvr job has like 50/50 failure rate :-\ | 20:59 |
sc68cal | so yeah long way of saying what sdague said succintly - "not voting yet" | 20:59 |
*** Marga_ has quit IRC | 20:59 | |
sc68cal | does that sound accurate Swami or am I giving you the short shrift? | 21:00 |
*** gyee has joined #openstack-nova | 21:00 | |
sc68cal | re live migrate | 21:00 |
Swami | sc68cal: the most recent failure that we have been seeing is the "test_volume_backend_live_migration" | 21:01 |
*** y_sawai has quit IRC | 21:02 | |
_gryf | sdague, remember our conversation about tox running different python interpreter from yesterday? turns out that my setup.py is doing things, which convinced tox to use certain interpreter, even if it doesn't make sense. so the case is closed, and thanks anyway :) | 21:03 |
Swami | sc68cal: Yes our multinode job is still experimental and not voting yet | 21:03 |
mriedem | test_volume_backend_live_migration wouldn't use block migrate though right | 21:04 |
Swami | mriedem: yes the neutron ci multinode jobs are currently testing the live-migration | 21:04 |
mriedem | Swami: sc68cal: basically, it'd be cool if we had a job (probably on experimental queue) that shows live block migration working with neutron + dvr for https://review.openstack.org/#/c/275073/ | 21:04 |
*** vishwanathj has quit IRC | 21:04 | |
mriedem | or the neutron changes dependent on https://review.openstack.org/#/c/275073/ | 21:04 |
*** eil397 has quit IRC | 21:04 | |
Swami | mriedem: we have been seeing some inconsistency in the live migration tests. | 21:04 |
*** raildo is now known as raildo-afk | 21:04 | |
mriedem | Swami: so i guess we should see the multinode dvr job passing live block migration with https://review.openstack.org/#/c/260738/ ? although i don't know the live migration tests have a floating IP attached... | 21:05 |
*** daemontool has joined #openstack-nova | 21:05 | |
*** eil397 has joined #openstack-nova | 21:06 | |
*** haplo37 has joined #openstack-nova | 21:06 | |
mriedem | http://logs.openstack.org/38/260738/18/check/gate-tempest-dsvm-neutron-dvr-multinode-full/8493f58/console.html#_2016-02-25_06_54_23_635 | 21:06 |
mriedem | tempest.api.compute.admin.test_live_migration.LiveBlockMigrationTestJSON.test_live_block_migration [19.204524s] ... ok | 21:06 |
Swami | mriedem: Yes I had a doubt about it, but this patch should work for both non floatingip case and for floatingip case with respect to dvr, since we pre-create the routers in the agent. | 21:06 |
openstackgerrit | Mark Doffman proposed openstack/nova: Move logging outside of LibvirtConfigObject.to_xml https://review.openstack.org/224216 | 21:06 |
*** rtheis has joined #openstack-nova | 21:06 | |
bauzas | alaski: around ? I have a question about cells v1 | 21:08 |
alaski | bauzas: yep | 21:08 |
bauzas | alaski: so I'm modifying resize_instance() for providing the Spec obj | 21:08 |
mriedem | Swami: sc68cal: so i guess i can't trust those job results b/c that test passes on other neutron changes in the same job | 21:09 |
mriedem | http://logs.openstack.org/76/282876/13/check/gate-tempest-dsvm-neutron-dvr-multinode-full/2879370/console.html#_2016-02-24_20_15_35_438 | 21:09 |
mriedem | to see the failure, we'd need to probably update the test to attach a floating ip to the server that gets migrated | 21:09 |
bauzas | alaski: by modifying https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2632 to add an extra request_spec | 21:09 |
*** aysyd has quit IRC | 21:10 | |
mriedem | we could do that as a one-time change to tempest (do not merge it, just use it for testing) | 21:10 |
Swami | mriedem: make sense | 21:10 |
mriedem | oh, well....to really verify it i'd think we'd have to ssh into the guest before and after it's live migrated | 21:11 |
bauzas | alaski: so, tbc, if we're in cells v1, compute.api.ComputeTaskAPI.resize() will be called but the RPC API will go to the cells ? | 21:11 |
mriedem | i see gate-tempest-dsvm-neutron-dvr-multinode-full already uses floating IPs | 21:11 |
mriedem | http://logs.openstack.org/76/282876/13/check/gate-tempest-dsvm-neutron-dvr-multinode-full/2879370/logs/tempest_conf.txt.gz | 21:11 |
mriedem | ssh_connect_method = floating | 21:11 |
Swami | mriedem: we need to ssh and see if the ssh session does not break. | 21:11 |
*** eil397 has quit IRC | 21:11 | |
mriedem | so the test is passing but networking could be broken | 21:11 |
mriedem | right | 21:11 |
bauzas | alaski: speaking of https://github.com/openstack/nova/blob/master/nova/compute/cells_api.py#L72 | 21:12 |
alaski | bauzas: was just about to link that :) because it's set there the call will go through cells/rpcapi to the cell | 21:12 |
bauzas | alaski: so I'm a bit worried noxw | 21:13 |
alaski | bauzas: https://github.com/openstack/nova/blob/master/nova/cells/rpcapi.py#L557 | 21:13 |
bauzas | alaski: because I would need to also modify the cells RPC API to add this new arg, right? | 21:13 |
*** ducttape_ has quit IRC | 21:13 | |
alaski | yes | 21:13 |
bauzas | oh man, it was merged | 21:14 |
bauzas | I mean the ConductorTaskAPI, without adding the cells RPC API too | 21:14 |
bauzas | alaski: ^ | 21:14 |
*** gmurphy has left #openstack-nova | 21:15 | |
*** davideagle_ has quit IRC | 21:15 | |
bauzas | alaski: meaning https://github.com/openstack/nova/blob/master/nova/conductor/rpcapi.py#L270-L272 | 21:15 |
bauzas | sec, I'm puzzled | 21:15 |
mriedem | hmm, during driver.spawn, libvirt eventually tries to get pci_devices from the instance, but i don't think we have them at that point... | 21:15 |
mriedem | not sure where those get created | 21:16 |
alaski | bauzas: I don't see resize_instance in there | 21:16 |
bauzas | alaski: because compute.api.resize() calls conductor.rebuild() | 21:16 |
*** ildikov has quit IRC | 21:16 | |
bauzas | but ooooh | 21:16 |
alaski | bauzas: but you may have broken rebuild and migrate | 21:17 |
alaski | and unshelve | 21:17 |
bauzas | alaski: that's what I wonder | 21:17 |
bauzas | oh man, those levels of indirection are getting me crazy | 21:17 |
alaski | bauzas: the easiest fix might be to update cells/rpcapi to accept those fields and then not pass the request_spec over the wire | 21:18 |
openstackgerrit | jichenjc proposed openstack/nova: Replace fake_utils by using Fixture https://review.openstack.org/230954 | 21:18 |
*** smatzek has quit IRC | 21:19 | |
alaski | bauzas: and within each cell it will need to fall back to treating instances like older ones with no req_spec | 21:19 |
openstackgerrit | Mark Doffman proposed openstack/nova: Cleanup rescue images on unrescue when using RBD. https://review.openstack.org/241777 | 21:19 |
*** fawadkhaliq has quit IRC | 21:20 | |
*** Sukhdev has quit IRC | 21:21 | |
bauzas | alaski: okay, I just double-checked, and I see why I didn't modified resize_instance() for the conductor RPC API, it's because it's named migrate_server | 21:22 |
bauzas | alaski: yeah, was thinking of that | 21:22 |
mriedem | dansmith: man, figuring out where instance.pci_devices gets set during the instance create, that's not going to be fun | 21:22 |
openstackgerrit | Matt Riedemann proposed openstack/nova: DNM: see where instance.pci_devices is getting lazy-loaded https://review.openstack.org/284839 | 21:22 |
*** jichen has quit IRC | 21:22 | |
*** Marga_ has joined #openstack-nova | 21:22 | |
bauzas | for cells v1, it would mean that this feature wouldn't be available but that's a trade-off | 21:22 |
mriedem | dansmith: but i think we're round-tripping to lazy-load that field on instance spawn and it's empty | 21:22 |
mriedem | so basically pointless | 21:23 |
alaski | bauzas: yeah. you could pass it over the wire as well, there just isn't a good way to test that | 21:23 |
dansmith | mriedem: yep, that was on my list, but I have nothing done for it, because it wasn't obvious | 21:23 |
*** achanda has quit IRC | 21:23 | |
mriedem | dansmith: ^ test patch | 21:23 |
mriedem | trying to find out | 21:23 |
dansmith | mriedem: oh, I have a generic one up | 21:23 |
*** su_zhang has joined #openstack-nova | 21:23 | |
bauzas | alaski: given that we're pretty frozen on cells v1 with a minimal coverage (because we missed that with our CI), I'm honestly fine to just add the arg and not pass it | 21:24 |
dansmith | mriedem: https://review.openstack.org/#/c/274714/8/nova/objects/instance.py | 21:24 |
mriedem | https://review.openstack.org/#/c/211288/? | 21:24 |
alaski | bauzas: sounds good | 21:24 |
bauzas | alaski: so, just writing the fix | 21:24 |
mriedem | dansmith: ok, i already did get a trace on pci_devices | 21:24 |
mriedem | http://logs.openstack.org/39/284839/1/check/gate-tempest-dsvm-full/433bb96//logs/screen-n-cpu.txt.gz?level=TRACE | 21:24 |
bauzas | alaski: thanks for your help, really sorry to have borked that :( | 21:24 |
mriedem | definitely from spawn, i just wanted to confirm it's an empty list | 21:24 |
bauzas | bork, bork | 21:24 |
alaski | heh | 21:24 |
dansmith | yep cool | 21:24 |
*** matrohon has quit IRC | 21:25 | |
*** annasort has quit IRC | 21:25 | |
*** rtheis has quit IRC | 21:26 | |
*** rtheis has joined #openstack-nova | 21:26 | |
*** browne has quit IRC | 21:27 | |
*** Marga_ has quit IRC | 21:27 | |
*** su_zhang has quit IRC | 21:28 | |
*** annasort has joined #openstack-nova | 21:28 | |
*** matrohon has joined #openstack-nova | 21:29 | |
*** apearson has quit IRC | 21:30 | |
*** achanda has joined #openstack-nova | 21:31 | |
*** rk4n has joined #openstack-nova | 21:32 | |
*** apearson has joined #openstack-nova | 21:32 | |
*** ijw has joined #openstack-nova | 21:33 | |
*** achanda has quit IRC | 21:33 | |
bauzas | alaski: last point, I don't see unshelve in the RPC proxy for cells, I suppose it's not supported for cells v1 ? | 21:34 |
alaski | bauzas: it's done differently https://github.com/openstack/nova/blob/master/nova/compute/cells_api.py#L341 | 21:35 |
bauzas | BOOOOM | 21:35 |
bauzas | <head exploded> | 21:35 |
*** ducttape_ has joined #openstack-nova | 21:35 | |
alaski | yeah, cells seemed too simple at first so multiple ways of doing things were added | 21:36 |
alaski | really we just didn't migrate everything to the new ways | 21:36 |
melwitt | hah, that's great | 21:37 |
*** achanda has joined #openstack-nova | 21:37 | |
openstackgerrit | Ed Leafe proposed openstack/nova: Use new inventory schema in all compute_node gets https://review.openstack.org/284833 | 21:38 |
*** thorst is now known as thorst_afk | 21:39 | |
bauzas | alaski: okay, so IIRC, it just cuts off compute_task_api but still calls compute_api.unshevel and use instead cast_to_cells to propagate thru cells,right? | 21:39 |
*** kaisers has quit IRC | 21:39 | |
alaski | yep | 21:39 |
bauzas | in that case, it's fine | 21:39 |
bauzas | okay, writing a few tests for covering that hack, and go | 21:40 |
bauzas | alaski: thanks | 21:40 |
*** hamzy has quit IRC | 21:40 | |
*** sc68cal has quit IRC | 21:41 | |
*** ijw has quit IRC | 21:41 | |
*** kaisers has joined #openstack-nova | 21:41 | |
*** ijw has joined #openstack-nova | 21:42 | |
*** thorst_afk has quit IRC | 21:43 | |
*** thorst_afk has joined #openstack-nova | 21:43 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: Add specific method to lazy-load instance.pci_devices https://review.openstack.org/284945 | 21:44 |
*** annasort has quit IRC | 21:44 | |
mriedem | dansmith: ^ easy one for you | 21:44 |
*** yamahata has quit IRC | 21:44 | |
dansmith | mriedem: looks okay but lemme scrape through the intel pci logs if it runs | 21:45 |
mriedem | i almost thought that was a joke | 21:46 |
dansmith | it might be, we'll see :) | 21:47 |
*** daemontool has quit IRC | 21:48 | |
*** eil397 has joined #openstack-nova | 21:49 | |
*** annasort has joined #openstack-nova | 21:50 | |
*** Marga_ has joined #openstack-nova | 21:51 | |
*** thorst_afk has quit IRC | 21:52 | |
*** shawnmaten has joined #openstack-nova | 21:54 | |
*** haomaiwang has joined #openstack-nova | 21:55 | |
*** eil397 has quit IRC | 21:57 | |
*** eil397 has joined #openstack-nova | 21:57 | |
*** ZZelle_ has joined #openstack-nova | 21:58 | |
*** y_sawai has joined #openstack-nova | 21:58 | |
*** penick_ has joined #openstack-nova | 21:59 | |
*** penick has quit IRC | 21:59 | |
*** penick_ is now known as penick | 21:59 | |
*** delatte has quit IRC | 22:00 | |
mriedem | oooo i think intel nfv ci found a bug | 22:00 |
*** haomaiwang has quit IRC | 22:01 | |
sdague | shut the front door | 22:01 |
*** porunov has quit IRC | 22:01 | |
*** asselin has left #openstack-nova | 22:01 | |
*** ivase has quit IRC | 22:01 | |
*** browne has joined #openstack-nova | 22:01 | |
openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix Cells RPC API by accepting a RequestSpec arg https://review.openstack.org/284959 | 22:02 |
bauzas | alaski: melwitt: (anyone who wants to understand cells v1) ^ | 22:02 |
*** apearson has quit IRC | 22:02 | |
mriedem | https://bugs.launchpad.net/nova/+bug/1550014 | 22:02 |
openstack | Launchpad bug 1550014 in OpenStack Compute (nova) "test_shelve_shelved_server fails in Intel NFV CI for api-compute-on-nfv-flavors scenario" [Undecided,New] | 22:02 |
mriedem | sdague: unless it's an env issue | 22:02 |
*** y_sawai has quit IRC | 22:02 | |
sdague | mriedem: log ? | 22:03 |
*** shawnmaten has quit IRC | 22:03 | |
mriedem | http://intel-openstack-ci-logs.ovh/compute-ci/refs/changes/20/283820/2/compute-nfv-flavors/20160225_183949/screen/n-cpu.log.gz | 22:03 |
mriedem | 2016-02-25 19:15:26.874 ERROR nova.compute.manager [req-078b2da5-cb6e-41ed-925c-01f3acb52264 tempest-ServersNegativeTestJSON-975508431 tempest-ServersNegativeTestJSON-1886110650] [instance: e6395df6-5393-4391-aa8e-6f3790268f8f] Instance failed to spawn | 22:03 |
*** shawnmaten has joined #openstack-nova | 22:05 | |
*** ivase has joined #openstack-nova | 22:05 | |
mriedem | cfriesen: don't you care about cpu pinning? ^ | 22:06 |
*** apearson has joined #openstack-nova | 22:09 | |
*** achanda has quit IRC | 22:09 | |
*** ivase has quit IRC | 22:10 | |
*** eharney has quit IRC | 22:10 | |
*** apearson has quit IRC | 22:10 | |
*** mylu has quit IRC | 22:10 | |
*** dedery has joined #openstack-nova | 22:11 | |
bauzas | mriedem: heh, just created https://bugs.launchpad.net/nova/+bug/1550018 | 22:12 |
openstack | Launchpad bug 1550018 in OpenStack Compute (nova) "Cells v1 is borked for live-migrate and evacuate" [High,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | 22:12 |
bauzas | mriedem: so yeah will be attaching the bug to the change | 22:12 |
*** knikolla has quit IRC | 22:13 | |
openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix Cells RPC API by accepting a RequestSpec arg https://review.openstack.org/284959 | 22:13 |
*** cleong has quit IRC | 22:15 | |
*** mylu has joined #openstack-nova | 22:17 | |
*** daemontool has joined #openstack-nova | 22:17 | |
*** fawadkhaliq has joined #openstack-nova | 22:19 | |
openstackgerrit | Ed Leafe proposed openstack/nova: Use new inventory schema in all compute_node gets https://review.openstack.org/284833 | 22:19 |
edleafe | dansmith: ^^ Let me know what you think about this approach | 22:19 |
*** shawnmaten has quit IRC | 22:22 | |
*** su_zhang has joined #openstack-nova | 22:23 | |
*** shawnmaten has joined #openstack-nova | 22:24 | |
*** delatte has joined #openstack-nova | 22:25 | |
openstackgerrit | Chris Dent proposed openstack/nova: WIP Add ResourcePool object https://review.openstack.org/284963 | 22:26 |
openstackgerrit | Chris Dent proposed openstack/nova: Adjust resource-providers models for resource-pools https://review.openstack.org/281837 | 22:28 |
*** Sukhdev has joined #openstack-nova | 22:28 | |
openstackgerrit | Chris Dent proposed openstack/nova: Add name to ResourceProvider object https://review.openstack.org/281945 | 22:28 |
openstackgerrit | Chris Dent proposed openstack/nova: Add a get_by_uuid for aggregates https://review.openstack.org/284895 | 22:29 |
openstackgerrit | Chris Dent proposed openstack/nova: WIP Add ResourcePool object https://review.openstack.org/284963 | 22:29 |
*** apearson has joined #openstack-nova | 22:29 | |
*** hemna has joined #openstack-nova | 22:30 | |
*** burgerk has quit IRC | 22:33 | |
*** delattec has joined #openstack-nova | 22:33 | |
*** john51_ has joined #openstack-nova | 22:34 | |
openstackgerrit | Andrew Laski proposed openstack/nova: WIP Add a hacking check for test method closures https://review.openstack.org/284970 | 22:35 |
*** mylu has quit IRC | 22:35 | |
alaski | dansmith: ^ my initial attempt | 22:35 |
alaski | it fails on self.assertRaises though | 22:35 |
*** delatte has quit IRC | 22:36 | |
alaski | I'm not walking all nodes though and I may need to convert to doing that | 22:36 |
*** john51 has quit IRC | 22:38 | |
*** ccarmack has quit IRC | 22:40 | |
*** mylu has joined #openstack-nova | 22:42 | |
*** armax has quit IRC | 22:43 | |
*** annasort has quit IRC | 22:43 | |
*** armax has joined #openstack-nova | 22:44 | |
*** mylu has quit IRC | 22:44 | |
*** mylu has joined #openstack-nova | 22:44 | |
bauzas | mriedem: see the commit msg for https://review.openstack.org/#/c/284959/2/nova/cells/rpcapi.py | 22:45 |
bauzas | mriedem: as discussed with alaski we agreed to not modify the RPC API | 22:46 |
mriedem | what would adding a version break? | 22:46 |
openstackgerrit | Sylvain Bauza proposed openstack/nova: WIP: Cold migrate using the RequestSpec object https://review.openstack.org/284974 | 22:47 |
bauzas | mriedem: so bumping a version for that would mean that we touch the RPC API for something we don't really want | 22:48 |
bauzas | mriedem: here, we just want to make sure that the cells api accepts the signature, not really needing to pass it down | 22:48 |
bauzas | mriedem: hence the UTs I added https://review.openstack.org/#/c/284959/2/nova/tests/unit/cells/test_cells_rpcapi.py | 22:49 |
*** thorst_afk has joined #openstack-nova | 22:49 | |
mriedem | unrelated | 22:51 |
mriedem | http://logs.openstack.org/89/283689/3/gate/gate-tempest-dsvm-full-ceph-src-glance_store/f23f9eb/logs/screen-n-sch.txt.gz#_2016-02-25_01_47_37_429 | 22:51 |
mriedem | Seems service is down. Last heartbeat was 2016-02-25 01:43:59. Elapsed time is 218.428975 is_up | 22:51 |
mriedem | that elapsed time seems wrong | 22:51 |
dansmith | edleafe: that doesn't fix compute_node_update() | 22:51 |
mriedem | but i guess it's not | 22:51 |
openstackgerrit | Swaminathan Vasudevan proposed openstack/nova: Implement setup_networks_on_host for Neutron networks https://review.openstack.org/275073 | 22:53 |
* bauzas heading off to bed, good week-end folks (me on PTO tomorrow) | 22:53 | |
*** mdrabe has quit IRC | 22:55 | |
*** hemna has quit IRC | 22:55 | |
openstackgerrit | Mark Doffman proposed openstack/nova: Add manage command for cell0. https://review.openstack.org/275266 | 22:56 |
openstackgerrit | Mark Doffman proposed openstack/nova: Map instances to cell0 on failed schedule https://review.openstack.org/267827 | 22:56 |
openstackgerrit | Mark Doffman proposed openstack/nova: Manage db sync command for cell0 https://review.openstack.org/274885 | 22:56 |
*** thorst_afk has quit IRC | 22:57 | |
*** armax has quit IRC | 22:57 | |
*** y_sawai has joined #openstack-nova | 22:59 | |
*** haomaiwang has joined #openstack-nova | 22:59 | |
*** eil397 has quit IRC | 23:00 | |
*** haomaiwang has quit IRC | 23:01 | |
*** delatte has joined #openstack-nova | 23:01 | |
*** vilobhmm11 has quit IRC | 23:01 | |
*** vilobhmm11 has joined #openstack-nova | 23:02 | |
*** rlrossit has quit IRC | 23:02 | |
*** vilobhmm11 has quit IRC | 23:03 | |
*** shawnmaten has quit IRC | 23:03 | |
*** delattec has quit IRC | 23:03 | |
*** y_sawai has quit IRC | 23:04 | |
*** shawnmaten has joined #openstack-nova | 23:04 | |
*** shawnmaten has quit IRC | 23:04 | |
*** achanda has joined #openstack-nova | 23:04 | |
*** zenoway has quit IRC | 23:05 | |
*** vilobhmm11 has joined #openstack-nova | 23:05 | |
*** hemna has joined #openstack-nova | 23:07 | |
*** salv-orl_ has joined #openstack-nova | 23:08 | |
*** cdent has quit IRC | 23:09 | |
*** edmondsw has quit IRC | 23:11 | |
*** lpetrut has quit IRC | 23:11 | |
*** edtubill has quit IRC | 23:11 | |
*** ctrath has quit IRC | 23:12 | |
*** salv-orlando has quit IRC | 23:12 | |
*** xyang1 has quit IRC | 23:14 | |
*** matrohon has quit IRC | 23:15 | |
*** dedery has quit IRC | 23:16 | |
*** zenoway has joined #openstack-nova | 23:17 | |
*** Daisy has joined #openstack-nova | 23:18 | |
*** xek_ has joined #openstack-nova | 23:18 | |
*** xek has quit IRC | 23:19 | |
*** zenoway has quit IRC | 23:21 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: Add service binary/host to service is down log for context https://review.openstack.org/284980 | 23:22 |
*** Daisy has quit IRC | 23:22 | |
*** mriedem has quit IRC | 23:23 | |
*** ducttape_ has quit IRC | 23:24 | |
*** burt has quit IRC | 23:24 | |
*** haplo37 has quit IRC | 23:25 | |
edleafe | :w | 23:26 |
* edleafe sighs | 23:26 | |
*** bjornar__ has quit IRC | 23:26 | |
*** dimtruck is now known as zz_dimtruck | 23:27 | |
*** hemna has quit IRC | 23:28 | |
*** davideagle has joined #openstack-nova | 23:29 | |
*** Sukhdev has quit IRC | 23:29 | |
*** y_sawai has joined #openstack-nova | 23:30 | |
edleafe | dansmith: this seems to work for me: http://pastebin.com/WK1P9Lrx | 23:34 |
*** rtheis has quit IRC | 23:34 | |
edleafe | dansmith: let me know what issues you're still seeing | 23:34 |
* edleafe leaves for a school meeting (yay!) | 23:35 | |
dansmith | edleafe: I think I got hung up scrolling and didn't see that you updated cmopute_node_update | 23:36 |
*** yamahata has joined #openstack-nova | 23:36 | |
*** Bjoern_ has quit IRC | 23:36 | |
*** davideagle has quit IRC | 23:38 | |
*** Swami has quit IRC | 23:38 | |
*** achanda has quit IRC | 23:38 | |
*** salv-orl_ has quit IRC | 23:39 | |
*** RichardRaseley has joined #openstack-nova | 23:39 | |
*** y_sawai has quit IRC | 23:42 | |
*** angdraug has quit IRC | 23:43 | |
*** fawadkhaliq has quit IRC | 23:47 | |
*** achanda has joined #openstack-nova | 23:49 | |
*** edtubill has joined #openstack-nova | 23:49 | |
*** zenoway has joined #openstack-nova | 23:52 | |
*** shawnmaten has joined #openstack-nova | 23:54 | |
*** thorst_afk has joined #openstack-nova | 23:55 | |
*** su_zhang has quit IRC | 23:57 | |
*** zenoway has quit IRC | 23:57 | |
*** su_zhang has joined #openstack-nova | 23:57 | |
*** krtaylor has quit IRC | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!