| *** takashin has joined #openstack-nova | 00:00 | |
| *** gouthamr has joined #openstack-nova | 00:01 | |
| *** shaohe_feng has quit IRC | 00:04 | |
| *** shaohe_feng has joined #openstack-nova | 00:04 | |
| *** thorst has joined #openstack-nova | 00:05 | |
| *** thorst has quit IRC | 00:06 | |
| *** thorst has joined #openstack-nova | 00:10 | |
| *** shaohe_feng has quit IRC | 00:14 | |
| *** shaohe_feng has joined #openstack-nova | 00:15 | |
| *** ducttape_ has joined #openstack-nova | 00:15 | |
| *** thorst has quit IRC | 00:15 | |
| *** Jeffrey4l_ has joined #openstack-nova | 00:20 | |
| *** Jeffrey4l__ has quit IRC | 00:20 | |
| *** baoli has joined #openstack-nova | 00:22 | |
| *** hoangcx2 has quit IRC | 00:22 | |
| *** ducttape_ has quit IRC | 00:22 | |
| *** shaohe_feng has quit IRC | 00:24 | |
| *** shaohe_feng has joined #openstack-nova | 00:25 | |
| *** shaohe_feng has quit IRC | 00:34 | |
| *** shaohe_feng has joined #openstack-nova | 00:35 | |
| *** qeelee has joined #openstack-nova | 00:39 | |
| *** ducttape_ has joined #openstack-nova | 00:40 | |
| *** morgabra has quit IRC | 00:41 | |
| *** morgabra has joined #openstack-nova | 00:41 | |
| *** shaohe_feng has quit IRC | 00:45 | |
| *** shaohe_feng has joined #openstack-nova | 00:46 | |
| *** wanghao_ has joined #openstack-nova | 00:47 | |
| *** fragatina has joined #openstack-nova | 00:48 | |
| *** hoangcx has joined #openstack-nova | 00:48 | |
| *** thorst has joined #openstack-nova | 00:50 | |
| *** thorst has quit IRC | 00:50 | |
| *** thorst has joined #openstack-nova | 00:51 | |
| *** devkulkarni has joined #openstack-nova | 00:54 | |
| *** shaohe_feng has quit IRC | 00:55 | |
| *** shaohe_feng has joined #openstack-nova | 00:55 | |
| *** cheneydc has joined #openstack-nova | 00:57 | |
| *** code-R has joined #openstack-nova | 00:57 | |
| *** thorst has quit IRC | 00:59 | |
| *** ducttape_ has quit IRC | 01:00 | |
| *** shaohe_feng has quit IRC | 01:05 | |
| *** shaohe_feng has joined #openstack-nova | 01:06 | |
| *** cheneydc has quit IRC | 01:08 | |
| *** ducttape_ has joined #openstack-nova | 01:08 | |
| *** vdrok has quit IRC | 01:09 | |
| *** wanghao_ has quit IRC | 01:10 | |
| *** serverascode has quit IRC | 01:10 | |
| *** ameade has quit IRC | 01:10 | |
| *** sballe_ has quit IRC | 01:10 | |
| *** tdurakov has quit IRC | 01:10 | |
| *** xenogear has quit IRC | 01:11 | |
| *** rpodolyaka has quit IRC | 01:11 | |
| *** scottda has quit IRC | 01:11 | |
| *** betherly has quit IRC | 01:11 | |
| *** zhiyan has quit IRC | 01:11 | |
| *** raj_singh_ has joined #openstack-nova | 01:15 | |
| *** raj_singh_ has quit IRC | 01:15 | |
| *** shaohe_feng has quit IRC | 01:15 | |
| *** mtanino has joined #openstack-nova | 01:16 | |
| *** shaohe_feng has joined #openstack-nova | 01:16 | |
| *** salv-orlando has joined #openstack-nova | 01:17 | |
| *** ducttape_ has quit IRC | 01:20 | |
| *** ducttape_ has joined #openstack-nova | 01:21 | |
| *** salv-orlando has quit IRC | 01:22 | |
| *** rpodolyaka has joined #openstack-nova | 01:23 | |
| *** scottda has joined #openstack-nova | 01:23 | |
| *** xiaoyang has joined #openstack-nova | 01:23 | |
| *** lixiaoy1 has joined #openstack-nova | 01:24 | |
| *** magic has joined #openstack-nova | 01:24 | |
| *** wanghao_ has joined #openstack-nova | 01:25 | |
| *** vdrok has joined #openstack-nova | 01:25 | |
| *** sballe_ has joined #openstack-nova | 01:25 | |
| *** magic is now known as Guest47074 | 01:25 | |
| *** ameade has joined #openstack-nova | 01:25 | |
| *** sdake_ has quit IRC | 01:25 | |
| *** shaohe_feng has quit IRC | 01:26 | |
| *** tdurakov has joined #openstack-nova | 01:26 | |
| *** serverascode has joined #openstack-nova | 01:26 | |
| *** shaohe_feng has joined #openstack-nova | 01:27 | |
| *** zhiyan has joined #openstack-nova | 01:27 | |
| *** betherly has joined #openstack-nova | 01:27 | |
| *** xiaoyang has quit IRC | 01:28 | |
| *** ducttape_ has quit IRC | 01:28 | |
| *** Guest47074 has quit IRC | 01:28 | |
| *** ducttape_ has joined #openstack-nova | 01:29 | |
| *** jamielennox is now known as jamielennox|away | 01:30 | |
| *** sdake has joined #openstack-nova | 01:30 | |
| *** thorst has joined #openstack-nova | 01:32 | |
| *** thorst has quit IRC | 01:32 | |
| openstackgerrit | wanghao proposed openstack/nova: Fix invalid Image&block device mapping combination https://review.openstack.org/171984 | 01:35 |
|---|---|---|
| *** shaohe_feng has quit IRC | 01:36 | |
| *** baoli has quit IRC | 01:36 | |
| *** kaisers_ has joined #openstack-nova | 01:36 | |
| *** thorst has joined #openstack-nova | 01:37 | |
| *** shaohe_feng has joined #openstack-nova | 01:37 | |
| *** devkulkarni1 has joined #openstack-nova | 01:40 | |
| *** EinstCrazy has joined #openstack-nova | 01:41 | |
| *** kaisers_ has quit IRC | 01:41 | |
| *** EinstCrazy has quit IRC | 01:42 | |
| *** thorst has quit IRC | 01:42 | |
| *** EinstCrazy has joined #openstack-nova | 01:42 | |
| *** devkulkarni has quit IRC | 01:43 | |
| *** jamielennox|away is now known as jamielennox | 01:43 | |
| *** baoli has joined #openstack-nova | 01:44 | |
| *** shaohe_feng has quit IRC | 01:46 | |
| *** shaohe_feng has joined #openstack-nova | 01:47 | |
| *** wangqun has joined #openstack-nova | 01:47 | |
| *** ducttape_ has quit IRC | 01:47 | |
| *** baoli has quit IRC | 01:49 | |
| *** thorst has joined #openstack-nova | 01:52 | |
| *** devkulkarni has joined #openstack-nova | 01:54 | |
| *** thorst has quit IRC | 01:54 | |
| *** shaohe_feng has quit IRC | 01:56 | |
| *** baoli has joined #openstack-nova | 01:56 | |
| *** devkulkarni1 has quit IRC | 01:58 | |
| *** shaohe_feng has joined #openstack-nova | 01:58 | |
| *** EinstCrazy has quit IRC | 02:00 | |
| *** vishwanathj has joined #openstack-nova | 02:01 | |
| *** EinstCrazy has joined #openstack-nova | 02:01 | |
| *** su_zhang has joined #openstack-nova | 02:02 | |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Fix floating-ip-list to show associated instance info https://review.openstack.org/352612 | 02:05 |
| *** shaohe_feng has quit IRC | 02:07 | |
| *** shaohe_feng has joined #openstack-nova | 02:08 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Fix MonitorMetric obj_make_compatible https://review.openstack.org/361792 | 02:08 |
| *** annp has joined #openstack-nova | 02:10 | |
| *** vishwanathj is now known as vishwanathj_zzz | 02:10 | |
| *** ducttape_ has joined #openstack-nova | 02:11 | |
| *** yamahata has quit IRC | 02:14 | |
| *** devkulkarni has quit IRC | 02:16 | |
| *** ducttape_ has quit IRC | 02:16 | |
| *** shaohe_feng has quit IRC | 02:17 | |
| *** EinstCrazy has quit IRC | 02:17 | |
| *** EinstCrazy has joined #openstack-nova | 02:18 | |
| *** shaohe_feng has joined #openstack-nova | 02:19 | |
| *** amotoki has joined #openstack-nova | 02:24 | |
| *** salv-orlando has joined #openstack-nova | 02:24 | |
| *** shaohe_feng has quit IRC | 02:27 | |
| *** salv-orlando has quit IRC | 02:27 | |
| *** shaohe_feng has joined #openstack-nova | 02:27 | |
| *** devkulkarni has joined #openstack-nova | 02:30 | |
| *** vishwanathj_zzz has quit IRC | 02:33 | |
| *** vishwanathj_zzz has joined #openstack-nova | 02:33 | |
| *** raunak has joined #openstack-nova | 02:33 | |
| *** shaohe_feng has quit IRC | 02:37 | |
| *** vishwanathj_zzz has quit IRC | 02:41 | |
| *** vishwanathj_zzz has joined #openstack-nova | 02:41 | |
| *** shaohe_feng has joined #openstack-nova | 02:42 | |
| *** vishwanathj_zzz has quit IRC | 02:44 | |
| *** vishwanathj_zzz has joined #openstack-nova | 02:44 | |
| *** shaohe_feng has quit IRC | 02:48 | |
| *** shaohe_feng has joined #openstack-nova | 02:48 | |
| *** haplo37__ has quit IRC | 02:49 | |
| *** thorst has joined #openstack-nova | 02:55 | |
| *** vishwanathj_zzz has quit IRC | 02:56 | |
| *** vishwanathj_zzz has joined #openstack-nova | 02:56 | |
| *** shaohe_feng has quit IRC | 02:58 | |
| *** code-R has quit IRC | 02:58 | |
| *** shaohe_feng has joined #openstack-nova | 02:58 | |
| *** tiendc has joined #openstack-nova | 02:59 | |
| *** tuanla_ has joined #openstack-nova | 02:59 | |
| *** ljxiash has joined #openstack-nova | 03:01 | |
| *** jianghuaw has quit IRC | 03:02 | |
| *** thorst has quit IRC | 03:03 | |
| *** kevinz has joined #openstack-nova | 03:04 | |
| *** shaohe_feng has quit IRC | 03:08 | |
| *** devkulkarni1 has joined #openstack-nova | 03:09 | |
| *** shaohe_feng has joined #openstack-nova | 03:10 | |
| *** rkrum has joined #openstack-nova | 03:11 | |
| *** devkulkarni has quit IRC | 03:12 | |
| *** ducttape_ has joined #openstack-nova | 03:17 | |
| *** shaohe_feng has quit IRC | 03:18 | |
| *** ljxiash has quit IRC | 03:19 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/nova: Using assertIsNotNone() is preferred over assertIsNot(None,) https://review.openstack.org/361815 | 03:19 |
| openstackgerrit | wanghao proposed openstack/nova: Fix invalid Image&block device mapping combination https://review.openstack.org/171984 | 03:19 |
| *** su_zhang has quit IRC | 03:20 | |
| *** ljxiash has joined #openstack-nova | 03:20 | |
| *** su_zhang has joined #openstack-nova | 03:20 | |
| *** jamielennox is now known as jamielennox|away | 03:21 | |
| *** amotoki has quit IRC | 03:21 | |
| *** shaohe_feng has joined #openstack-nova | 03:21 | |
| *** ducttape_ has quit IRC | 03:21 | |
| *** ljxiash has quit IRC | 03:24 | |
| *** su_zhang has quit IRC | 03:24 | |
| *** kaisers_ has joined #openstack-nova | 03:26 | |
| *** markvoelker has joined #openstack-nova | 03:27 | |
| *** shaohe_feng has quit IRC | 03:29 | |
| *** markvoelker_ has joined #openstack-nova | 03:29 | |
| *** shaohe_feng has joined #openstack-nova | 03:30 | |
| *** kaisers_ has quit IRC | 03:30 | |
| *** markvoelker has quit IRC | 03:33 | |
| *** code-R has joined #openstack-nova | 03:38 | |
| *** shaohe_feng has quit IRC | 03:39 | |
| *** raunak has quit IRC | 03:39 | |
| *** amotoki has joined #openstack-nova | 03:40 | |
| *** shaohe_feng has joined #openstack-nova | 03:40 | |
| *** ManishD has joined #openstack-nova | 03:40 | |
| *** rmcall has joined #openstack-nova | 03:42 | |
| *** code-R has quit IRC | 03:45 | |
| *** aswadr_ has joined #openstack-nova | 03:45 | |
| *** code-R has joined #openstack-nova | 03:46 | |
| *** shaohe_feng has quit IRC | 03:49 | |
| *** ljxiash has joined #openstack-nova | 03:50 | |
| *** code-R has quit IRC | 03:50 | |
| *** shaohe_feng has joined #openstack-nova | 03:50 | |
| *** ljxiash has quit IRC | 03:50 | |
| *** ljxiash has joined #openstack-nova | 03:51 | |
| *** markvoelker_ has quit IRC | 03:53 | |
| *** markvoelker has joined #openstack-nova | 03:54 | |
| *** devkulkarni1 has quit IRC | 03:56 | |
| *** ljxiash has quit IRC | 03:56 | |
| *** jamielennox|away is now known as jamielennox | 03:56 | |
| *** takashin has left #openstack-nova | 03:57 | |
| *** raunak has joined #openstack-nova | 03:57 | |
| *** shaohe_feng has quit IRC | 03:59 | |
| *** gouthamr has quit IRC | 04:01 | |
| *** shaohe_feng has joined #openstack-nova | 04:01 | |
| *** thorst has joined #openstack-nova | 04:01 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Should not be able to sort instances based on joined tables https://review.openstack.org/359134 | 04:03 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Should not be able to sort instances based on joined tables https://review.openstack.org/359134 | 04:05 |
| openstackgerrit | huan proposed openstack/nova: XenAPI: Fix VM live-migrate with iSCSI SR volume https://review.openstack.org/359548 | 04:09 |
| *** thorst has quit IRC | 04:09 | |
| *** shaohe_feng has quit IRC | 04:10 | |
| *** diga has joined #openstack-nova | 04:11 | |
| *** shaohe_feng has joined #openstack-nova | 04:11 | |
| *** tonytan4ever has quit IRC | 04:13 | |
| *** unicell has joined #openstack-nova | 04:14 | |
| *** salv-orlando has joined #openstack-nova | 04:18 | |
| *** ducttape_ has joined #openstack-nova | 04:18 | |
| *** lixiaoy1 has quit IRC | 04:19 | |
| *** shaohe_feng has quit IRC | 04:20 | |
| *** suresh12 has joined #openstack-nova | 04:20 | |
| *** shaohe_feng has joined #openstack-nova | 04:20 | |
| *** amotoki has quit IRC | 04:21 | |
| *** salv-orlando has quit IRC | 04:23 | |
| *** ducttape_ has quit IRC | 04:23 | |
| *** suresh12 has quit IRC | 04:25 | |
| *** suresh12 has joined #openstack-nova | 04:26 | |
| *** coolsvap_ has joined #openstack-nova | 04:28 | |
| *** unicell has quit IRC | 04:28 | |
| *** coolsvap_ is now known as coolsvap_\ | 04:28 | |
| *** coolsvap_\ is now known as coolsvap | 04:28 | |
| *** unicell has joined #openstack-nova | 04:28 | |
| *** shaohe_feng has quit IRC | 04:30 | |
| *** shaohe_feng has joined #openstack-nova | 04:31 | |
| *** raunak has quit IRC | 04:32 | |
| *** sheel has joined #openstack-nova | 04:33 | |
| *** links has joined #openstack-nova | 04:33 | |
| *** ManishD has quit IRC | 04:37 | |
| *** ManishD has joined #openstack-nova | 04:38 | |
| *** shaohe_feng has quit IRC | 04:40 | |
| *** suresh12 has quit IRC | 04:40 | |
| *** shaohe_feng has joined #openstack-nova | 04:41 | |
| *** amotoki has joined #openstack-nova | 04:42 | |
| *** code-R has joined #openstack-nova | 04:44 | |
| *** fragatin_ has joined #openstack-nova | 04:45 | |
| *** suresh12 has joined #openstack-nova | 04:46 | |
| *** fragati__ has joined #openstack-nova | 04:48 | |
| *** fragatina has quit IRC | 04:48 | |
| *** amotoki has quit IRC | 04:48 | |
| *** fragatin_ has quit IRC | 04:50 | |
| *** shaohe_feng has quit IRC | 04:51 | |
| *** shaohe_feng has joined #openstack-nova | 04:51 | |
| *** jaosorior has joined #openstack-nova | 04:52 | |
| *** xiaoyang has joined #openstack-nova | 04:52 | |
| *** fragati__ has quit IRC | 04:53 | |
| ManishD | HI All, Please can you have a look when you have some free time: https://review.openstack.org/#/c/357131/ | 04:55 |
| *** ecelik has joined #openstack-nova | 04:56 | |
| *** magic has joined #openstack-nova | 04:58 | |
| *** magic is now known as Guest98474 | 04:59 | |
| *** vishwanathj_zzz has quit IRC | 05:01 | |
| *** shaohe_feng has quit IRC | 05:01 | |
| *** vishwanathj_zzz has joined #openstack-nova | 05:01 | |
| *** xiaoyang has quit IRC | 05:02 | |
| *** amotoki has joined #openstack-nova | 05:03 | |
| *** shaohe_feng has joined #openstack-nova | 05:05 | |
| *** ratailor has joined #openstack-nova | 05:06 | |
| *** thorst has joined #openstack-nova | 05:07 | |
| *** sdake_ has joined #openstack-nova | 05:10 | |
| *** sudipto has joined #openstack-nova | 05:10 | |
| *** shaohe_feng has quit IRC | 05:11 | |
| *** sdake_ is now known as dake | 05:13 | |
| *** sdake has quit IRC | 05:13 | |
| *** shaohe_feng has joined #openstack-nova | 05:13 | |
| *** dake is now known as sdake | 05:13 | |
| *** tonytan4ever has joined #openstack-nova | 05:13 | |
| *** thorst has quit IRC | 05:14 | |
| *** kaisers_ has joined #openstack-nova | 05:14 | |
| *** mtanino has quit IRC | 05:15 | |
| *** code-R_ has joined #openstack-nova | 05:17 | |
| *** ekuris has joined #openstack-nova | 05:18 | |
| *** lixiaoy1 has joined #openstack-nova | 05:18 | |
| *** tonytan4ever has quit IRC | 05:18 | |
| *** kaisers_ has quit IRC | 05:19 | |
| *** code-R has quit IRC | 05:20 | |
| *** shaohe_feng has quit IRC | 05:21 | |
| *** shaohe_feng has joined #openstack-nova | 05:22 | |
| *** ManishD has quit IRC | 05:24 | |
| *** jianghuaww has joined #openstack-nova | 05:26 | |
| *** sdake has quit IRC | 05:26 | |
| *** jianghuaww has quit IRC | 05:30 | |
| *** jianghuaw_ has joined #openstack-nova | 05:31 | |
| *** shaohe_feng has quit IRC | 05:32 | |
| *** shaohe_feng has joined #openstack-nova | 05:32 | |
| *** markvoelker has quit IRC | 05:36 | |
| *** salv-orlando has joined #openstack-nova | 05:36 | |
| *** suresh12 has quit IRC | 05:36 | |
| *** code-R_ has quit IRC | 05:39 | |
| *** code-R has joined #openstack-nova | 05:39 | |
| *** shaohe_feng has quit IRC | 05:42 | |
| *** shaohe_feng has joined #openstack-nova | 05:42 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/360894 | 05:43 |
| openstackgerrit | Tuan Luong-Anh proposed openstack/nova: Using assertIsNone() is preferred over assertEqual() https://review.openstack.org/361847 | 05:44 |
| *** pbandark has joined #openstack-nova | 05:45 | |
| *** jianghuaw_ is now known as jianghuaw | 05:45 | |
| *** pbandark has quit IRC | 05:50 | |
| *** liverpooler has joined #openstack-nova | 05:50 | |
| *** pbandark has joined #openstack-nova | 05:50 | |
| *** ssurana has joined #openstack-nova | 05:51 | |
| *** shaohe_feng has quit IRC | 05:52 | |
| *** shaohe_feng has joined #openstack-nova | 05:52 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Emit warning when use 'user_id' in policy rule https://review.openstack.org/356963 | 05:52 |
| *** ljxiash has joined #openstack-nova | 05:53 | |
| *** lpetrut has joined #openstack-nova | 05:58 | |
| *** tojuvone_ is now known as tojuvone | 05:58 | |
| *** sridharg has joined #openstack-nova | 05:59 | |
| *** shaohe_feng has quit IRC | 06:02 | |
| *** salv-orlando has quit IRC | 06:05 | |
| *** ssurana has quit IRC | 06:06 | |
| *** mjura has joined #openstack-nova | 06:06 | |
| *** moshele has joined #openstack-nova | 06:06 | |
| *** markvoelker has joined #openstack-nova | 06:07 | |
| *** shaohe_feng has joined #openstack-nova | 06:07 | |
| *** salv-orlando has joined #openstack-nova | 06:07 | |
| *** prateek has joined #openstack-nova | 06:07 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Emit warning when use 'user_id' in policy rule https://review.openstack.org/356963 | 06:08 |
| *** raunak has joined #openstack-nova | 06:09 | |
| openstackgerrit | gengchc2 proposed openstack/nova: Fix a typo in the telnet_example.rst file https://review.openstack.org/361861 | 06:10 |
| *** markvoelker has quit IRC | 06:11 | |
| *** mjura has quit IRC | 06:12 | |
| *** thorst has joined #openstack-nova | 06:12 | |
| *** mjura has joined #openstack-nova | 06:12 | |
| *** code-R has quit IRC | 06:12 | |
| *** shaohe_feng has quit IRC | 06:13 | |
| *** code-R has joined #openstack-nova | 06:13 | |
| *** shaohe_feng has joined #openstack-nova | 06:13 | |
| *** matrohon has quit IRC | 06:17 | |
| *** code-R has quit IRC | 06:17 | |
| *** ManishD has joined #openstack-nova | 06:18 | |
| *** thorst has quit IRC | 06:19 | |
| *** ducttape_ has joined #openstack-nova | 06:20 | |
| *** nkrinner_afk is now known as nkrinner | 06:22 | |
| *** shaohe_feng has quit IRC | 06:23 | |
| *** shaohe_feng has joined #openstack-nova | 06:24 | |
| *** ducttape_ has quit IRC | 06:24 | |
| *** fragatina has joined #openstack-nova | 06:25 | |
| *** lixiaoy1 has quit IRC | 06:26 | |
| *** fragatina has quit IRC | 06:26 | |
| *** fragatina has joined #openstack-nova | 06:26 | |
| *** amotoki has quit IRC | 06:29 | |
| *** gcb has joined #openstack-nova | 06:30 | |
| *** shaohe_feng has quit IRC | 06:33 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Replace compute_req_id middlware by RequestId oslo middleware https://review.openstack.org/358375 | 06:33 |
| *** lixiaoy1 has joined #openstack-nova | 06:34 | |
| *** shaohe_feng has joined #openstack-nova | 06:35 | |
| *** markvoelker has joined #openstack-nova | 06:36 | |
| *** praneshp has quit IRC | 06:38 | |
| *** pbandark has quit IRC | 06:38 | |
| *** pbandark has joined #openstack-nova | 06:39 | |
| *** markvoelker has quit IRC | 06:41 | |
| *** shaohe_feng has quit IRC | 06:43 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Adds nova-policy-check cmd https://review.openstack.org/322944 | 06:44 |
| *** Dinesh_Bhor has joined #openstack-nova | 06:44 | |
| *** shaohe_feng has joined #openstack-nova | 06:44 | |
| *** kaisers_ has joined #openstack-nova | 06:45 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Adds nova-policy-check cmd https://review.openstack.org/322944 | 06:46 |
| *** rcernin has joined #openstack-nova | 06:47 | |
| *** andreas_s has joined #openstack-nova | 06:49 | |
| *** amotoki has joined #openstack-nova | 06:50 | |
| *** shaohe_feng has quit IRC | 06:54 | |
| *** shaohe_feng has joined #openstack-nova | 06:56 | |
| *** amotoki has quit IRC | 06:58 | |
| *** pcaruana has joined #openstack-nova | 07:00 | |
| *** qeelee has quit IRC | 07:00 | |
| *** tesseract- has joined #openstack-nova | 07:01 | |
| *** aswadr_ has quit IRC | 07:01 | |
| *** qeelee has joined #openstack-nova | 07:02 | |
| *** lpetrut has quit IRC | 07:02 | |
| *** shaohe_feng has quit IRC | 07:04 | |
| *** jdurgin has quit IRC | 07:04 | |
| *** Hosam has joined #openstack-nova | 07:04 | |
| *** shaohe_feng has joined #openstack-nova | 07:04 | |
| *** bwallis has joined #openstack-nova | 07:05 | |
| *** markvoelker has joined #openstack-nova | 07:06 | |
| *** markvoelker has quit IRC | 07:12 | |
| *** jichen has joined #openstack-nova | 07:13 | |
| *** shaohe_feng has quit IRC | 07:14 | |
| *** shaohe_feng has joined #openstack-nova | 07:15 | |
| *** thorst has joined #openstack-nova | 07:17 | |
| *** amotoki has joined #openstack-nova | 07:19 | |
| *** jdurgin has joined #openstack-nova | 07:20 | |
| *** pbandark has quit IRC | 07:20 | |
| *** ekuris has quit IRC | 07:22 | |
| *** markus_z has joined #openstack-nova | 07:24 | |
| *** thorst has quit IRC | 07:24 | |
| *** shaohe_feng has quit IRC | 07:24 | |
| *** shaohe_feng has joined #openstack-nova | 07:25 | |
| *** markus_z has quit IRC | 07:28 | |
| *** acabot has joined #openstack-nova | 07:28 | |
| *** acabot has quit IRC | 07:29 | |
| *** acabot has joined #openstack-nova | 07:29 | |
| *** code-R has joined #openstack-nova | 07:29 | |
| *** markus_z has joined #openstack-nova | 07:29 | |
| *** karthiks_afk is now known as karthikc | 07:31 | |
| *** karthikc is now known as karthiks | 07:31 | |
| *** jdurgin has quit IRC | 07:32 | |
| *** matrohon has joined #openstack-nova | 07:34 | |
| *** ekuris has joined #openstack-nova | 07:34 | |
| *** EinstCrazy has quit IRC | 07:34 | |
| *** shaohe_feng has quit IRC | 07:35 | |
| *** EinstCrazy has joined #openstack-nova | 07:35 | |
| *** rkrum has quit IRC | 07:35 | |
| *** amotoki has quit IRC | 07:36 | |
| *** raunak has quit IRC | 07:36 | |
| *** markvoelker has joined #openstack-nova | 07:36 | |
| *** suresh12 has joined #openstack-nova | 07:37 | |
| *** shaohe_feng has joined #openstack-nova | 07:37 | |
| *** Guest98474 has quit IRC | 07:37 | |
| *** xiaoyang has joined #openstack-nova | 07:38 | |
| *** tiendc has quit IRC | 07:38 | |
| *** pbandark has joined #openstack-nova | 07:39 | |
| *** tiendc has joined #openstack-nova | 07:40 | |
| *** markvoelker has quit IRC | 07:41 | |
| *** suresh12 has quit IRC | 07:41 | |
| *** xiaoyang has quit IRC | 07:42 | |
| *** xiaoyang has joined #openstack-nova | 07:42 | |
| *** shaohe_feng has quit IRC | 07:45 | |
| *** shaohe_feng has joined #openstack-nova | 07:45 | |
| *** amotoki has joined #openstack-nova | 07:45 | |
| *** code-R_ has joined #openstack-nova | 07:45 | |
| *** tanee is now known as tanee_away | 07:46 | |
| *** tanee_away is now known as tanee | 07:46 | |
| *** Sukhdev has quit IRC | 07:46 | |
| *** code-R has quit IRC | 07:48 | |
| *** pilgrimstack has joined #openstack-nova | 07:49 | |
| *** EinstCra_ has joined #openstack-nova | 07:50 | |
| *** qeelee has quit IRC | 07:50 | |
| *** vishwanathj_zzz has quit IRC | 07:53 | |
| *** vishwanathj_zzz has joined #openstack-nova | 07:53 | |
| *** EinstCrazy has quit IRC | 07:53 | |
| *** markus_z has quit IRC | 07:54 | |
| *** shaohe_feng has quit IRC | 07:55 | |
| *** shaohe_feng has joined #openstack-nova | 07:55 | |
| *** yassine has joined #openstack-nova | 07:57 | |
| *** yassine is now known as Guest91620 | 07:58 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** daidv has joined #openstack-nova | 08:00 | |
| *** Guest91620 is now known as yassou | 08:00 | |
| *** markus_z has joined #openstack-nova | 08:01 | |
| *** zzzeek has joined #openstack-nova | 08:01 | |
| *** code-R_ has quit IRC | 08:02 | |
| *** code-R has joined #openstack-nova | 08:02 | |
| *** markvoelker has joined #openstack-nova | 08:05 | |
| *** shaohe_feng has quit IRC | 08:05 | |
| *** links has quit IRC | 08:06 | |
| *** shaohe_feng has joined #openstack-nova | 08:07 | |
| *** markvoelker has quit IRC | 08:10 | |
| *** derekh has joined #openstack-nova | 08:12 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: The server_group_members quota cannot work when bulk boot instances https://review.openstack.org/360187 | 08:13 |
| *** roeyc has joined #openstack-nova | 08:13 | |
| *** greghaynes has quit IRC | 08:13 | |
| *** daidv has quit IRC | 08:13 | |
| *** daidv has joined #openstack-nova | 08:13 | |
| *** shaohe_feng has quit IRC | 08:16 | |
| *** kuzko has quit IRC | 08:16 | |
| *** shaohe_feng has joined #openstack-nova | 08:16 | |
| *** kuzko has joined #openstack-nova | 08:18 | |
| *** daidv has quit IRC | 08:19 | |
| *** mnestratov has joined #openstack-nova | 08:20 | |
| *** lucas-afk is now known as lucasagomes | 08:21 | |
| *** coolsvap is now known as coolsvap_ | 08:21 | |
| *** ducttape_ has joined #openstack-nova | 08:21 | |
| *** thorst has joined #openstack-nova | 08:22 | |
| *** qeelee has joined #openstack-nova | 08:25 | |
| *** ducttape_ has quit IRC | 08:25 | |
| *** shaohe_feng has quit IRC | 08:26 | |
| *** shaohe_feng has joined #openstack-nova | 08:26 | |
| tdurakov | morning nova | 08:29 |
| *** thorst has quit IRC | 08:29 | |
| *** markvoelker has joined #openstack-nova | 08:33 | |
| *** code-R has quit IRC | 08:34 | |
| *** bwallis has quit IRC | 08:35 | |
| *** trisq has joined #openstack-nova | 08:36 | |
| *** shaohe_feng has quit IRC | 08:36 | |
| *** ekuris has quit IRC | 08:36 | |
| *** shaohe_feng has joined #openstack-nova | 08:37 | |
| *** markvoelker has quit IRC | 08:38 | |
| *** rmart04 has joined #openstack-nova | 08:39 | |
| *** greghaynes has joined #openstack-nova | 08:40 | |
| *** tonytan4ever has joined #openstack-nova | 08:44 | |
| *** shaohe_feng has quit IRC | 08:46 | |
| *** shaohe_feng has joined #openstack-nova | 08:47 | |
| *** qeelee has quit IRC | 08:48 | |
| *** tonytan4ever has quit IRC | 08:49 | |
| *** salv-orlando has quit IRC | 08:49 | |
| *** ekuris has joined #openstack-nova | 08:53 | |
| openstackgerrit | Alex Szarka proposed openstack/nova: Reduce code complexity - libvirt/config.py https://review.openstack.org/359879 | 08:54 |
| *** lpetrut has joined #openstack-nova | 08:57 | |
| *** shaohe_feng has quit IRC | 08:57 | |
| *** shaohe_feng has joined #openstack-nova | 08:57 | |
| *** jdurgin has joined #openstack-nova | 09:00 | |
| *** lixiaoy1 has quit IRC | 09:00 | |
| *** rmart04 has quit IRC | 09:01 | |
| *** markvoelker has joined #openstack-nova | 09:03 | |
| *** qeelee has joined #openstack-nova | 09:05 | |
| *** shaohe_feng has quit IRC | 09:07 | |
| *** markvoelker has quit IRC | 09:07 | |
| *** shaohe_feng has joined #openstack-nova | 09:08 | |
| *** wznoinsk_ is now known as wznoinsk | 09:09 | |
| *** shaohe_feng has quit IRC | 09:17 | |
| *** shaohe_feng has joined #openstack-nova | 09:20 | |
| *** roeyc has quit IRC | 09:20 | |
| *** salv-orlando has joined #openstack-nova | 09:20 | |
| *** roeyc has joined #openstack-nova | 09:20 | |
| *** Hosam_ has joined #openstack-nova | 09:21 | |
| openstackgerrit | Alex Szarka proposed openstack/nova: Reduce code complexity - libvirt/config.py https://review.openstack.org/359879 | 09:22 |
| *** ducttape_ has joined #openstack-nova | 09:22 | |
| *** Hosam has quit IRC | 09:25 | |
| *** ducttape_ has quit IRC | 09:26 | |
| *** thorst has joined #openstack-nova | 09:27 | |
| *** shaohe_feng has quit IRC | 09:27 | |
| *** rmart04 has joined #openstack-nova | 09:27 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/nova: Python 3.x compatibility modules urllib https://review.openstack.org/361964 | 09:30 |
| *** markvoelker has joined #openstack-nova | 09:31 | |
| *** shaohe_feng has joined #openstack-nova | 09:31 | |
| *** qeelee has quit IRC | 09:33 | |
| *** gszasz has joined #openstack-nova | 09:34 | |
| *** gabor_antal has joined #openstack-nova | 09:34 | |
| *** thorst has quit IRC | 09:34 | |
| *** flaper87 has quit IRC | 09:35 | |
| *** markvoelker has quit IRC | 09:35 | |
| *** code-R has joined #openstack-nova | 09:37 | |
| *** shaohe_feng has quit IRC | 09:38 | |
| *** rmart04 has quit IRC | 09:38 | |
| bauzas | good (jetlagged) morning, Nova | 09:38 |
| *** shaohe_feng has joined #openstack-nova | 09:38 | |
| *** ljxiash has quit IRC | 09:39 | |
| *** code-R_ has joined #openstack-nova | 09:39 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/nova: Python 3.x compatibility https://review.openstack.org/361969 | 09:40 |
| *** code-R has quit IRC | 09:42 | |
| *** rkrum has joined #openstack-nova | 09:42 | |
| *** mamandle has joined #openstack-nova | 09:42 | |
| mamandle | I'm running into an issue after upgrading from Liberty to Mitaka, it seems that uuid was added in ComputeNode in Mitaka. Do we need any special handling of that as part of upgrade? | 09:45 |
| *** rkrum has quit IRC | 09:47 | |
| bauzas | mamandle: a db sync ? :) | 09:48 |
| *** shaohe_feng has quit IRC | 09:48 | |
| *** qeelee has joined #openstack-nova | 09:49 | |
| *** ljxiash has joined #openstack-nova | 09:49 | |
| *** shaohe_feng has joined #openstack-nova | 09:49 | |
| mamandle | yeap doing that, just wanted to confirm that its not something that needs to be handled at service startup or anything similar. | 09:49 |
| *** Hosam has joined #openstack-nova | 09:53 | |
| *** flaper87 has joined #openstack-nova | 09:53 | |
| *** ljxiash has quit IRC | 09:54 | |
| *** flaper87 has quit IRC | 09:54 | |
| *** flaper87 has joined #openstack-nova | 09:54 | |
| *** roeyc has quit IRC | 09:54 | |
| *** Hosam_ has quit IRC | 09:56 | |
| *** zigo_ is now known as zigo | 09:56 | |
| openstackgerrit | Alex Szarka proposed openstack/nova: Reduce code complexity - instance.py https://review.openstack.org/359878 | 09:57 |
| *** vishwanathj_zzz has quit IRC | 09:58 | |
| *** shaohe_feng has quit IRC | 09:58 | |
| *** vishwanathj_zzz has joined #openstack-nova | 09:58 | |
| *** shaohe_feng has joined #openstack-nova | 10:00 | |
| *** yassou has quit IRC | 10:00 | |
| *** markvoelker has joined #openstack-nova | 10:00 | |
| *** ihrachys has joined #openstack-nova | 10:03 | |
| *** markvoelker has quit IRC | 10:05 | |
| *** coolsvap_ is now known as coolsvap | 10:06 | |
| *** shaohe_feng has quit IRC | 10:08 | |
| *** shaohe_feng has joined #openstack-nova | 10:09 | |
| *** shaohe_feng has quit IRC | 10:19 | |
| *** shaohe_feng has joined #openstack-nova | 10:19 | |
| *** tiendc has quit IRC | 10:27 | |
| *** markvoelker has joined #openstack-nova | 10:28 | |
| *** oomichi has quit IRC | 10:28 | |
| *** roeyc has joined #openstack-nova | 10:28 | |
| *** shaohe_feng has quit IRC | 10:29 | |
| *** hoangcx has quit IRC | 10:30 | |
| *** oomichi has joined #openstack-nova | 10:30 | |
| *** mamandle has quit IRC | 10:31 | |
| *** shaohe_feng has joined #openstack-nova | 10:31 | |
| *** cdent has joined #openstack-nova | 10:31 | |
| *** zenoway has joined #openstack-nova | 10:32 | |
| *** markvoelker has quit IRC | 10:32 | |
| *** Hosam_ has joined #openstack-nova | 10:35 | |
| *** qeelee has quit IRC | 10:37 | |
| *** Hosam has quit IRC | 10:38 | |
| *** shaohe_feng has quit IRC | 10:39 | |
| *** qeelee has joined #openstack-nova | 10:39 | |
| *** garyk has joined #openstack-nova | 10:41 | |
| *** amotoki has quit IRC | 10:41 | |
| *** shaohe_feng has joined #openstack-nova | 10:42 | |
| *** code-R_ has quit IRC | 10:43 | |
| *** code-R has joined #openstack-nova | 10:44 | |
| *** wangqun has quit IRC | 10:44 | |
| *** tonytan4ever has joined #openstack-nova | 10:45 | |
| *** amotoki has joined #openstack-nova | 10:46 | |
| *** amotoki has quit IRC | 10:46 | |
| *** mamandle has joined #openstack-nova | 10:48 | |
| *** tonytan4ever has quit IRC | 10:49 | |
| *** shaohe_feng has quit IRC | 10:49 | |
| *** amotoki has joined #openstack-nova | 10:50 | |
| *** shaohe_feng has joined #openstack-nova | 10:50 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: conf: Make list->dict conversion more specific https://review.openstack.org/359338 | 10:51 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: conf: Add 'deprecated_reason' to osapi opts https://review.openstack.org/357061 | 10:51 |
| *** cdent has quit IRC | 10:53 | |
| openstackgerrit | Peng Li proposed openstack/nova: adsfsadfdfs https://review.openstack.org/362011 | 10:54 |
| *** thorst has joined #openstack-nova | 10:54 | |
| *** kolesovdv has joined #openstack-nova | 10:54 | |
| openstackgerrit | Ha Van Tu proposed openstack/nova: Fix using filter() to meet python2,3 https://review.openstack.org/362014 | 10:57 |
| *** tiendc has joined #openstack-nova | 10:59 | |
| *** markvoelker has joined #openstack-nova | 10:59 | |
| *** shaohe_feng has quit IRC | 11:00 | |
| *** shaohe_feng has joined #openstack-nova | 11:00 | |
| *** bwallis has joined #openstack-nova | 11:01 | |
| *** amotoki has quit IRC | 11:02 | |
| *** markvoelker has quit IRC | 11:03 | |
| *** tuanla_ has quit IRC | 11:05 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: Add deprecated_since parameter https://review.openstack.org/353631 | 11:05 |
| sfinucan | morning, johnthetubaguy | 11:07 |
| sfinucan | I've a question about your first comment on https://review.openstack.org/#/c/289431/12/nova/tests/unit/virt/xenapi/test_xenapi.py Surely it's better to be locally consistent and change everything over at once? | 11:07 |
| sfinucan | I can do a follow up patch that did the mox->mock conversion for that file, if it would help? | 11:08 |
| *** roeyc has quit IRC | 11:08 | |
| *** trisq has quit IRC | 11:09 | |
| *** roeyc has joined #openstack-nova | 11:10 | |
| *** shaohe_feng has quit IRC | 11:10 | |
| *** EinstCra_ has quit IRC | 11:11 | |
| *** cdent has joined #openstack-nova | 11:12 | |
| *** shaohe_feng has joined #openstack-nova | 11:12 | |
| *** rodrigods has quit IRC | 11:13 | |
| *** rodrigods has joined #openstack-nova | 11:13 | |
| *** cdent has quit IRC | 11:13 | |
| *** sdague has joined #openstack-nova | 11:14 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: objects: Move 'arch' to 'fields.Architecture' https://review.openstack.org/356067 | 11:15 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: objects: Move 'vm_mode' to 'fields.VMMode' https://review.openstack.org/356553 | 11:15 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: objects: Move 'hv_type' to 'fields.HVType' https://review.openstack.org/356424 | 11:15 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: objects: Move 'cpumodel' to 'fields.CPU*' https://review.openstack.org/356398 | 11:15 |
| *** bwallis has quit IRC | 11:15 | |
| openstackgerrit | Prateek Arora proposed openstack/nova: Delete traces of in-progress snapshot on VM being deleted https://review.openstack.org/294513 | 11:18 |
| *** shaohe_feng has quit IRC | 11:20 | |
| *** shaohe_feng has joined #openstack-nova | 11:21 | |
| *** bvanhav has joined #openstack-nova | 11:21 | |
| *** ducttape_ has joined #openstack-nova | 11:23 | |
| *** markvoelker has joined #openstack-nova | 11:28 | |
| *** sridharg has quit IRC | 11:28 | |
| *** ducttape_ has quit IRC | 11:28 | |
| *** shaohe_feng has quit IRC | 11:30 | |
| *** shaohe_feng has joined #openstack-nova | 11:32 | |
| *** ratailor has quit IRC | 11:32 | |
| *** markvoelker has quit IRC | 11:32 | |
| *** sudipto has quit IRC | 11:33 | |
| *** yassine has joined #openstack-nova | 11:35 | |
| *** yassine is now known as Guest4347 | 11:35 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: virtlogd: use virtlogd for char devices https://review.openstack.org/323765 | 11:36 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: remove py26 compat code in "get_console_output" https://review.openstack.org/334480 | 11:36 |
| *** tiendc has quit IRC | 11:40 | |
| *** shaohe_feng has quit IRC | 11:41 | |
| *** sridharg has joined #openstack-nova | 11:41 | |
| *** pradk has joined #openstack-nova | 11:42 | |
| *** shaohe_feng has joined #openstack-nova | 11:43 | |
| *** Guest85979 is now known as beagles | 11:44 | |
| *** annp has quit IRC | 11:44 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Emit warning when use 'user_id' in policy rule https://review.openstack.org/356963 | 11:46 |
| *** pradk has quit IRC | 11:46 | |
| *** rtheis has joined #openstack-nova | 11:47 | |
| *** shaohe_feng has quit IRC | 11:51 | |
| *** shaohe_feng has joined #openstack-nova | 11:51 | |
| *** code-R_ has joined #openstack-nova | 11:55 | |
| *** lucasagomes is now known as lucas-hungry | 11:55 | |
| *** tblakes has joined #openstack-nova | 11:56 | |
| *** code-R has quit IRC | 11:56 | |
| *** markvoelker has joined #openstack-nova | 11:57 | |
| *** ducttape_ has joined #openstack-nova | 11:58 | |
| *** code-R has joined #openstack-nova | 11:58 | |
| *** sridharg has quit IRC | 12:00 | |
| *** code-R_ has quit IRC | 12:01 | |
| *** markvoelker has quit IRC | 12:01 | |
| *** shaohe_feng has quit IRC | 12:01 | |
| *** mamandle has quit IRC | 12:02 | |
| *** jaosorior has quit IRC | 12:02 | |
| *** shaohe_feng has joined #openstack-nova | 12:02 | |
| *** jaosorior has joined #openstack-nova | 12:03 | |
| *** rfolco has joined #openstack-nova | 12:03 | |
| *** mamandle has joined #openstack-nova | 12:05 | |
| *** gouthamr has joined #openstack-nova | 12:06 | |
| *** salv-orl_ has joined #openstack-nova | 12:08 | |
| *** vishwanathj_zzz has quit IRC | 12:08 | |
| *** vishwanathj_zzz has joined #openstack-nova | 12:08 | |
| *** aswadr_ has joined #openstack-nova | 12:08 | |
| *** pradk has joined #openstack-nova | 12:09 | |
| *** salv-orlando has quit IRC | 12:10 | |
| *** shaohe_feng has quit IRC | 12:11 | |
| *** ducttape_ has quit IRC | 12:12 | |
| *** Hosam has joined #openstack-nova | 12:13 | |
| *** shaohe_feng has joined #openstack-nova | 12:14 | |
| *** rmart04 has joined #openstack-nova | 12:14 | |
| *** devkulkarni1 has joined #openstack-nova | 12:15 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova-specs: Enable to share PCI devices between numa nodes https://review.openstack.org/361140 | 12:16 |
| *** salv-orl_ has quit IRC | 12:16 | |
| *** amotoki has joined #openstack-nova | 12:17 | |
| *** Hosam_ has quit IRC | 12:17 | |
| *** amotoki has quit IRC | 12:17 | |
| *** baoli_ has joined #openstack-nova | 12:18 | |
| *** baoli has quit IRC | 12:21 | |
| *** _diana_ has joined #openstack-nova | 12:21 | |
| *** shaohe_feng has quit IRC | 12:22 | |
| *** shaohe_feng has joined #openstack-nova | 12:22 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: ensure resource provider records exist from RT https://review.openstack.org/358797 | 12:23 |
| *** markvoelker has joined #openstack-nova | 12:26 | |
| *** magic has joined #openstack-nova | 12:26 | |
| *** magic is now known as Guest94048 | 12:26 | |
| *** rmart04 has quit IRC | 12:26 | |
| *** xiaoyang has quit IRC | 12:28 | |
| *** diana_clarke has joined #openstack-nova | 12:28 | |
| *** xiaoyang has joined #openstack-nova | 12:28 | |
| *** salv-orlando has joined #openstack-nova | 12:29 | |
| *** sudipto has joined #openstack-nova | 12:29 | |
| *** markvoelker has quit IRC | 12:30 | |
| *** Guest94048 has quit IRC | 12:31 | |
| *** prateek has quit IRC | 12:32 | |
| *** shaohe_feng has quit IRC | 12:32 | |
| *** shaohe_feng has joined #openstack-nova | 12:32 | |
| *** mamandle has quit IRC | 12:40 | |
| *** shaohe_feng has quit IRC | 12:42 | |
| *** shaohe_feng has joined #openstack-nova | 12:43 | |
| *** ManishD has quit IRC | 12:45 | |
| *** pradk has quit IRC | 12:46 | |
| *** pilgrimstack1 has joined #openstack-nova | 12:50 | |
| *** amotoki has joined #openstack-nova | 12:51 | |
| *** shaohe_feng has quit IRC | 12:52 | |
| *** trisq has joined #openstack-nova | 12:53 | |
| *** pilgrimstack has quit IRC | 12:53 | |
| *** garyk has quit IRC | 12:53 | |
| *** shaohe_feng has joined #openstack-nova | 12:54 | |
| *** devkulkarni1 has quit IRC | 12:54 | |
| *** diga has quit IRC | 12:54 | |
| *** roeyc has quit IRC | 12:55 | |
| *** rlrossit has joined #openstack-nova | 12:55 | |
| *** markvoelker has joined #openstack-nova | 12:56 | |
| *** lucas-hungry is now known as lucasagomes | 12:56 | |
| *** tbachman has joined #openstack-nova | 12:57 | |
| *** jichen has quit IRC | 12:59 | |
| *** prateek has joined #openstack-nova | 12:59 | |
| *** markvoelker has quit IRC | 13:00 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Modify workflow of rebuild_instance when schedule fails https://review.openstack.org/348111 | 13:01 |
| *** eandersson__ has quit IRC | 13:02 | |
| *** pcaruana has quit IRC | 13:02 | |
| *** shaohe_feng has quit IRC | 13:03 | |
| *** sdake has joined #openstack-nova | 13:03 | |
| *** _diana_ has joined #openstack-nova | 13:04 | |
| *** shaohe_feng has joined #openstack-nova | 13:04 | |
| *** tonytan4ever has joined #openstack-nova | 13:06 | |
| *** daemontool has joined #openstack-nova | 13:06 | |
| *** roeyc has joined #openstack-nova | 13:06 | |
| *** markvoelker has joined #openstack-nova | 13:06 | |
| *** markvoelker_ has joined #openstack-nova | 13:07 | |
| *** ducttape_ has joined #openstack-nova | 13:07 | |
| *** tbachman_ has joined #openstack-nova | 13:07 | |
| *** ducttape_ has quit IRC | 13:08 | |
| *** ljxiash has joined #openstack-nova | 13:10 | |
| *** markvoelker_ has quit IRC | 13:10 | |
| *** markvoelker has quit IRC | 13:10 | |
| *** markvoelker_ has joined #openstack-nova | 13:10 | |
| *** kylek3h has joined #openstack-nova | 13:11 | |
| *** tbachman has quit IRC | 13:11 | |
| *** tbachman_ is now known as tbachman | 13:11 | |
| *** markvoelker has joined #openstack-nova | 13:11 | |
| *** shaohe_feng has quit IRC | 13:13 | |
| *** markvoelker_ has quit IRC | 13:13 | |
| *** shaohe_feng has joined #openstack-nova | 13:13 | |
| *** cleong has joined #openstack-nova | 13:14 | |
| *** bkopilov has quit IRC | 13:14 | |
| *** marst has quit IRC | 13:16 | |
| *** pcaruana has joined #openstack-nova | 13:17 | |
| *** EinstCrazy has joined #openstack-nova | 13:18 | |
| *** xyang1 has joined #openstack-nova | 13:20 | |
| *** jaypipes has joined #openstack-nova | 13:20 | |
| *** erhudy has joined #openstack-nova | 13:20 | |
| *** devkulkarni has joined #openstack-nova | 13:21 | |
| *** nkrinner has quit IRC | 13:21 | |
| *** sudipto has quit IRC | 13:22 | |
| openstackgerrit | Hussain Chachuliya proposed openstack/nova: Use min attribute from oslo_config https://review.openstack.org/362110 | 13:22 |
| *** sdake_ has joined #openstack-nova | 13:23 | |
| *** shaohe_feng has quit IRC | 13:23 | |
| *** jaypipes has quit IRC | 13:24 | |
| *** nkrinner has joined #openstack-nova | 13:24 | |
| *** shaohe_feng has joined #openstack-nova | 13:24 | |
| *** devkulkarni has quit IRC | 13:25 | |
| *** mriedem has joined #openstack-nova | 13:25 | |
| mriedem | ahoy | 13:25 |
| *** sdake has quit IRC | 13:25 | |
| bauzas | yoha | 13:25 |
| mriedem | sdague: can we get this in for the 6.0 novaclient release? https://review.openstack.org/#/c/353018/ | 13:26 |
| mriedem | sdague: neutron put in the fix for get me a network tests failing last week, i have the tempest revert up here https://review.openstack.org/#/c/361671/ | 13:26 |
| mriedem | rechecked it 4 times yesterday, no failures | 13:26 |
| sdague | mriedem: ok | 13:26 |
| sdague | mriedem: where is the fix ? | 13:26 |
| *** BjoernT has joined #openstack-nova | 13:26 | |
| sdague | just so I can educate myself better | 13:26 |
| mriedem | https://review.openstack.org/#/c/358753/ | 13:27 |
| sdague | when is our 6.0 release cutoff? | 13:27 |
| mriedem | thursday | 13:27 |
| sdague | How about we get the revert landed for tempest, get more data today, if all goes well I'll approve tomorrow? | 13:28 |
| mriedem | alright | 13:28 |
| sdague | I just approved the tempest revert | 13:28 |
| sdague | also, thanks for getting that sorted | 13:28 |
| mriedem | np, armax did all of the heavy lifting on that before i got back | 13:29 |
| mriedem | speaking of, it's my first day back and i have 2 hours of meetings this morning and then jury duty this afternoon, | 13:29 |
| mriedem | so are there any things that need my immediate attention? | 13:29 |
| sdague | so... having a really excellent day? | 13:29 |
| mriedem | yeah :) i even had to dress like an adult for work. | 13:29 |
| sdague | the only thing I needed you for was the get-me-a-network issue, which armax solved | 13:29 |
| sdague | so \o/ | 13:30 |
| mriedem | cool | 13:30 |
| *** markvoelker has quit IRC | 13:30 | |
| *** suresh12 has joined #openstack-nova | 13:31 | |
| *** nkrinner has quit IRC | 13:31 | |
| sdague | alex_xu: I don't seem to see warning messages with your policy bit - http://logs.openstack.org/43/360943/2/check/gate-tempest-dsvm-full-ubuntu-xenial/e271ea0/logs/screen-n-api.txt.gz?level=WARNING ? | 13:31 |
| *** pwnall has joined #openstack-nova | 13:33 | |
| * dansmith yawns | 13:33 | |
| *** shaohe_feng has quit IRC | 13:33 | |
| *** hoonetorg has quit IRC | 13:33 | |
| *** suresh12_ has joined #openstack-nova | 13:34 | |
| *** bhagyashris has quit IRC | 13:34 | |
| *** prateek has quit IRC | 13:34 | |
| *** shaohe_feng has joined #openstack-nova | 13:34 | |
| *** bhagyashris has joined #openstack-nova | 13:34 | |
| *** prateek has joined #openstack-nova | 13:35 | |
| *** suresh12 has quit IRC | 13:36 | |
| *** suresh12_ has quit IRC | 13:36 | |
| alaski | dansmith: wanted to run something past you. I proposed https://review.openstack.org/#/c/361436/ to fix the issue of the cell0 db not being setup properly, but this weekend I was considering just having the simple_cell_setup run the migrations. | 13:36 |
| alaski | thoughts on direction here? | 13:36 |
| * dansmith looks | 13:36 | |
| *** eharney has joined #openstack-nova | 13:37 | |
| dansmith | alaski: doing them together may be confusing because the docs will have to say "if this is the first time you're setting up a cell, the migrations are already done and don't have to be run again like normal" | 13:37 |
| dansmith | alaski: so I think just doing that reorder patch is better | 13:37 |
| *** kevinz has quit IRC | 13:38 | |
| *** ducttape_ has joined #openstack-nova | 13:38 | |
| *** efried has joined #openstack-nova | 13:38 | |
| mriedem | alaski: looks like the grenade failure on the nova change which depends on that grenade change is unrelated http://logs.openstack.org/38/356138/13/check/gate-grenade-dsvm-neutron-ubuntu-trusty/5011f3a/logs/grenade.sh.txt.gz#_2016-08-26_20_28_07_019 | 13:38 |
| *** hshiina has joined #openstack-nova | 13:39 | |
| mriedem | hmm, but all three grenade jobs failed | 13:39 |
| alaski | dansmith: okay. it would be a lot to add in the migrations there, but is simpler. I like the explicitness of the reorder though | 13:39 |
| *** tbachman has quit IRC | 13:39 | |
| *** pwnall has quit IRC | 13:39 | |
| dansmith | alaski: yeah, I think it's more explicit and less confusing to just make it separate.. they won't ever run cell_setup again, so.. | 13:40 |
| mriedem | http://logs.openstack.org/38/356138/13/check/gate-grenade-dsvm-neutron-ubuntu-trusty/5011f3a/logs/new/screen-n-api.txt.gz?level=TRACE#_2016-08-26_20_28_09_168 | 13:40 |
| alaski | mriedem: huh | 13:40 |
| mriedem | alaski: ^ same in all 3 job failures | 13:40 |
| alaski | mriedem: yeah, that's the failure the grenade patch should fix | 13:40 |
| alaski | can I depends-on a grenade patch? | 13:40 |
| mriedem | can haz depends-on grenade patch? | 13:41 |
| mriedem | i thought you could | 13:41 |
| alaski | I just know some things don't work, but have no clue which things. | 13:42 |
| alaski | I'll have the grenade patch depends-on the nova patch and see what happens | 13:42 |
| mriedem | hold up | 13:42 |
| mriedem | the depends-on from the nova change to the grenade change worked | 13:42 |
| mriedem | http://logs.openstack.org/38/356138/13/check/gate-grenade-dsvm-neutron-ubuntu-trusty/5011f3a/logs/devstack-gate-setup-workspace-new.txt.gz#_2016-08-26_20_03_16_709 | 13:42 |
| mriedem | it pulled in the grenade change there ^ | 13:42 |
| *** prateek_ has joined #openstack-nova | 13:42 | |
| alaski | hmm | 13:43 |
| mriedem | but what creates the nova_api_cell0 db? | 13:43 |
| mriedem | isn't that in devstack? | 13:43 |
| *** shaohe_feng has quit IRC | 13:44 | |
| alaski | it should happen in http://logs.openstack.org/38/356138/13/check/gate-grenade-dsvm-neutron-ubuntu-trusty/5011f3a/logs/grenade.sh.txt.gz#_2016-08-26_20_25_43_606 | 13:44 |
| alaski | it was reverted from devstack | 13:44 |
| alaski | and hasn't re merged | 13:44 |
| *** nkrinner has joined #openstack-nova | 13:44 | |
| *** sc68cal_ is now known as sc68cal | 13:44 | |
| alaski | https://review.openstack.org/#/c/355602/ | 13:44 |
| mriedem | ok, so make the grenade change depends on the devstack change? or the nova change depend on both? | 13:45 |
| *** su_zhang has joined #openstack-nova | 13:45 | |
| *** shaohe_feng has joined #openstack-nova | 13:45 | |
| alaski | it shouldn't need the devstack change I don't think. devstack does the same thing grenade is doing | 13:45 |
| *** prateek has quit IRC | 13:46 | |
| mriedem | true | 13:46 |
| *** ayoung has joined #openstack-nova | 13:47 | |
| alaski | oh http://logs.openstack.org/38/356138/13/check/gate-grenade-dsvm-neutron-ubuntu-trusty/5011f3a/logs/grenade.sh.txt.gz#_2016-08-26_20_25_49_113 | 13:47 |
| *** burt has joined #openstack-nova | 13:47 | |
| mriedem | yeah, that's why i'm wondering what creates the nova_api_cell0 db | 13:47 |
| alaski | the simple_cell_setup commands creates the connection string, and I thought the migrations would create the db | 13:48 |
| *** prateek has joined #openstack-nova | 13:48 | |
| mriedem | i don't think that's how it works | 13:48 |
| alaski | does the database need to be present before migrations are run? | 13:48 |
| *** Hosam has quit IRC | 13:48 | |
| mriedem | the migrations create the tables in the db | 13:48 |
| mriedem | devstack creates the db | 13:48 |
| alaski | ahh | 13:49 |
| *** Hosam has joined #openstack-nova | 13:49 | |
| mriedem | there is a recreate_database function in devstack | 13:50 |
| mriedem | off to meetings | 13:50 |
| *** prateek_ has quit IRC | 13:50 | |
| *** sdake_ has quit IRC | 13:50 | |
| alaski | mriedem: found it. thanks | 13:50 |
| *** sdake has joined #openstack-nova | 13:51 | |
| *** rlrossit has quit IRC | 13:52 | |
| *** prateek_ has joined #openstack-nova | 13:53 | |
| *** shaohe_feng has quit IRC | 13:54 | |
| *** Hosam has quit IRC | 13:54 | |
| *** woodster_ has joined #openstack-nova | 13:54 | |
| *** mriedem1 has joined #openstack-nova | 13:54 | |
| openstackgerrit | Prateek Arora proposed openstack/nova: Delete traces of in-progress snapshot on VM being deleted https://review.openstack.org/294513 | 13:55 |
| *** mriedem has quit IRC | 13:55 | |
| *** pwnall has joined #openstack-nova | 13:55 | |
| *** pcaruana has quit IRC | 13:55 | |
| *** shaohe_feng has joined #openstack-nova | 13:55 | |
| *** prateek has quit IRC | 13:56 | |
| *** markvoelker has joined #openstack-nova | 13:57 | |
| *** jamesdenton has joined #openstack-nova | 13:57 | |
| *** tbachman has joined #openstack-nova | 13:58 | |
| edleafe- | Scheduler subteam meeting in 2 minutes in #openstack-meeting-alt | 13:58 |
| *** vladikr has joined #openstack-nova | 13:58 | |
| *** edleafe- is now known as edleafe | 13:59 | |
| *** ecelik has quit IRC | 14:01 | |
| *** rmcall has quit IRC | 14:01 | |
| *** markvoelker has quit IRC | 14:01 | |
| *** pradk has joined #openstack-nova | 14:02 | |
| *** rmcall has joined #openstack-nova | 14:02 | |
| *** tbachman_ has joined #openstack-nova | 14:02 | |
| *** tbachman has quit IRC | 14:04 | |
| *** tbachman_ is now known as tbachman | 14:04 | |
| bauzas | alaski: dansmith: so, my question I had was related to cell0 being an hard requirement | 14:04 |
| *** tpatzig_ has joined #openstack-nova | 14:04 | |
| *** shaohe_feng has quit IRC | 14:04 | |
| *** marcusvrn_ has quit IRC | 14:04 | |
| bauzas | alaski: so, you think we should accept for Newton operators not running the simple cell creation ? | 14:04 |
| dansmith | no | 14:04 |
| bauzas | dansmith: context is https://review.openstack.org/#/c/356138/12/nova/compute/api.py | 14:05 |
| *** tpatzig has quit IRC | 14:06 | |
| *** shaohe_feng has joined #openstack-nova | 14:06 | |
| *** mkoderer has quit IRC | 14:06 | |
| alaski | bauzas: two things. I would like to get to a point where cell0 is necessary before forcing it, which I'm trying to get to in N. And even if we force it in N I'm not sure we should force that it must be run before N code can be used | 14:06 |
| alaski | so I think we should still protect against it not being setup | 14:06 |
| *** mtanino has joined #openstack-nova | 14:07 | |
| bauzas | alaski: that's actually a good upgrade question | 14:07 |
| bauzas | when should we ask our operators to run the simple cell call ? | 14:07 |
| dansmith | alaski: everything we have now is still graceful about not having the cells setup right? | 14:07 |
| alaski | dansmith: yes | 14:08 |
| dansmith | I definitely think we need to have a cycle where they have to run it before upgrading to the next cycle, | 14:08 |
| dansmith | which means N can't break if they haven't run it, otherwise we didn't give them a graceful experience | 14:08 |
| bauzas | okay, so we should leave it cool | 14:08 |
| alaski | bauzas: sometime after upgrading to N but before upgrading to O | 14:08 |
| bauzas | gotcha | 14:08 |
| *** mkoderer has joined #openstack-nova | 14:08 | |
| *** pcaruana has joined #openstack-nova | 14:09 | |
| *** sridharg has joined #openstack-nova | 14:12 | |
| *** annegentle has joined #openstack-nova | 14:13 | |
| *** sridharg has quit IRC | 14:13 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: test_keypairs_list_for_different_users for v2.10 https://review.openstack.org/347462 | 14:14 |
| *** shaohe_feng has quit IRC | 14:14 | |
| bauzas | alaski: I was also having another nit in https://review.openstack.org/#/c/357396/6/nova/compute/api.py@1012 | 14:16 |
| bauzas | alaski: not sure you saw it when rebasing | 14:16 |
| pkholkin | mriedem1: alaski: hi folks, please review this patch with test, thanks! https://review.openstack.org/#/c/347462 | 14:16 |
| pkholkin | Andrew, thanks for your comments | 14:16 |
| *** ekuris has quit IRC | 14:16 | |
| *** mlavalle has joined #openstack-nova | 14:17 | |
| alaski | bauzas: yeah, I need to respond to that | 14:17 |
| *** moshele has quit IRC | 14:18 | |
| *** shaohe_feng has joined #openstack-nova | 14:19 | |
| *** pwnall has quit IRC | 14:20 | |
| *** vishwanathj_zzz is now known as vishwanathj | 14:20 | |
| *** annegent_ has joined #openstack-nova | 14:21 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: ensure resource provider records exist from RT https://review.openstack.org/358797 | 14:21 |
| *** zz_dimtruck is now known as dimtruck | 14:22 | |
| bhagyashris | sdague: Hi, could you please review this patch https://review.openstack.org/#/c/360941/ ? | 14:22 |
| *** jaypipes has joined #openstack-nova | 14:23 | |
| *** vladikr has quit IRC | 14:24 | |
| *** annegentle has quit IRC | 14:24 | |
| *** jaosorior is now known as jaosorior_away | 14:25 | |
| *** shaohe_feng has quit IRC | 14:25 | |
| *** rkrum has joined #openstack-nova | 14:25 | |
| *** tbachman has quit IRC | 14:27 | |
| sdague | bhagyashris: +2 | 14:27 |
| *** shaohe_feng has joined #openstack-nova | 14:27 | |
| *** bhagyashris has quit IRC | 14:28 | |
| *** sridharg has joined #openstack-nova | 14:29 | |
| *** pwnall has joined #openstack-nova | 14:31 | |
| *** hemanthm|away is now known as hemanthm | 14:31 | |
| *** tbachman has joined #openstack-nova | 14:31 | |
| *** clenimar has joined #openstack-nova | 14:34 | |
| *** sridharg has quit IRC | 14:34 | |
| *** shaohe_feng has quit IRC | 14:35 | |
| alex_xu | sdague: emm...strange, let me check | 14:35 |
| *** aghadge has joined #openstack-nova | 14:35 | |
| *** browne has joined #openstack-nova | 14:36 | |
| *** shaohe_feng has joined #openstack-nova | 14:36 | |
| *** sheel has quit IRC | 14:36 | |
| alex_xu | mriedem1: good morning, I have question about this patch https://review.openstack.org/328157, I saw you -2 the patch after non-priority features, then remove the -2 again. But you didn't leave any comment, so just want to ask whether we allow to that patch? | 14:37 |
| *** devkulkarni has joined #openstack-nova | 14:38 | |
| mriedem1 | alex_xu: that looks like the -2 was dropped when i removed myself from the reviewers list | 14:38 |
| mriedem1 | i didn't realize it would do that, but it makes sense | 14:39 |
| *** vladikr has joined #openstack-nova | 14:39 | |
| *** mriedem1 is now known as mriedem | 14:39 | |
| mriedem | i just didn't want all of the email since i wasn't actively reviewing that but it kept changing | 14:39 |
| mriedem | alex_xu: i re-applied the -2 | 14:39 |
| *** tbachman has quit IRC | 14:40 | |
| alex_xu | mriedem: thanks, I'm clear now | 14:41 |
| *** tonytan_brb has joined #openstack-nova | 14:42 | |
| *** cfriesen has joined #openstack-nova | 14:44 | |
| *** tonytan4ever has quit IRC | 14:45 | |
| *** shaohe_feng has quit IRC | 14:45 | |
| *** tonytan_brb is now known as tonytan4ever | 14:45 | |
| *** parora has joined #openstack-nova | 14:46 | |
| *** shaohe_feng has joined #openstack-nova | 14:46 | |
| *** kaisers_ has quit IRC | 14:46 | |
| *** code-R has quit IRC | 14:47 | |
| *** kevinz has joined #openstack-nova | 14:47 | |
| *** prateek_ has quit IRC | 14:49 | |
| jroll | oh, removing yourself drops votes? lame | 14:51 |
| jroll | thanks for clearing that one up mriedem | 14:51 |
| *** eandersson has joined #openstack-nova | 14:52 | |
| *** sridharg has joined #openstack-nova | 14:53 | |
| *** su_zhang has quit IRC | 14:53 | |
| *** mamandle has joined #openstack-nova | 14:53 | |
| alaski | bauzas: responded on https://review.openstack.org/#/c/357396/6 | 14:54 |
| *** rlrossit has joined #openstack-nova | 14:54 | |
| *** anush__ has joined #openstack-nova | 14:54 | |
| *** sridharg has quit IRC | 14:54 | |
| *** shaohe_feng has quit IRC | 14:55 | |
| *** _ducttape_ has joined #openstack-nova | 14:56 | |
| *** shaohe_feng has joined #openstack-nova | 14:56 | |
| *** pbandark has quit IRC | 14:56 | |
| bauzas | alaski: so for example, I was thinking on a DB deadlock that would prevent the instance delete | 14:57 |
| alaski | bauzas: wouldn't that raise an exception? | 14:57 |
| *** marst has joined #openstack-nova | 14:58 | |
| bauzas | alaski: indeed | 14:58 |
| *** kfarr has joined #openstack-nova | 14:58 | |
| *** rrecio has joined #openstack-nova | 14:58 | |
| alaski | and that would stop the instance_mapping from being deleted | 14:58 |
| openstackgerrit | Sean Dague proposed openstack/nova: Make api-ref bug link point to nova https://review.openstack.org/362170 | 14:59 |
| *** eharney has quit IRC | 14:59 | |
| bauzas | mmm, you're right, if we just raise an exception, then we would stop the loop | 14:59 |
| *** ducttape_ has quit IRC | 14:59 | |
| bauzas | which is sad but at least we wouldn't delete the mappings without deleting the related instance | 14:59 |
| alaski | yeah | 14:59 |
| *** sridharg has joined #openstack-nova | 15:00 | |
| alaski | this is also short term as the next patch I put up should move instance creation to the conductor | 15:00 |
| *** sridharg has quit IRC | 15:00 | |
| bauzas | right | 15:00 |
| *** kevinz has quit IRC | 15:01 | |
| *** kevinz has joined #openstack-nova | 15:01 | |
| *** roeyc has quit IRC | 15:02 | |
| openstackgerrit | Peter Hamilton proposed openstack/nova-specs: Add support for certificate validation https://review.openstack.org/357151 | 15:04 |
| *** mjura has quit IRC | 15:05 | |
| *** shaohe_feng has quit IRC | 15:06 | |
| *** shaohe_feng has joined #openstack-nova | 15:06 | |
| mriedem | mdbooth: is there a reason why a unit test is not in this change https://review.openstack.org/#/c/355415/ ? | 15:07 |
| mriedem | mdbooth: i read some of the comments and it basically sounded like, 'it's hard so let's just not do it' | 15:07 |
| *** daemontool has quit IRC | 15:07 | |
| *** hemnafk has quit IRC | 15:08 | |
| *** daemontool has joined #openstack-nova | 15:09 | |
| *** jaosorior_away is now known as jaosorior | 15:10 | |
| *** devkulkarni has quit IRC | 15:10 | |
| syjulian | dansmith, jaypipes: hi, jay and dan. would you be able to review this change if you guys are available? it should be the last part of gibi's json-schema bp https://review.openstack.org/#/c/360095/ | 15:11 |
| *** devkulkarni has joined #openstack-nova | 15:13 | |
| *** hshiina has quit IRC | 15:13 | |
| *** gongysh has joined #openstack-nova | 15:14 | |
| *** aghadge has quit IRC | 15:14 | |
| mriedem | syjulian: we're past non-client lib releases for newton | 15:15 |
| *** armax has joined #openstack-nova | 15:15 | |
| mriedem | which means oslo.versionedobjects | 15:15 |
| *** beagles is now known as beagles_brb | 15:15 | |
| *** shaohe_feng has quit IRC | 15:16 | |
| *** praneshp has joined #openstack-nova | 15:16 | |
| *** hoonetorg has joined #openstack-nova | 15:16 | |
| *** shaohe_feng has joined #openstack-nova | 15:16 | |
| *** sridharg has joined #openstack-nova | 15:17 | |
| *** markvoelker has joined #openstack-nova | 15:17 | |
| *** rlrossit has quit IRC | 15:18 | |
| *** sdake_ has joined #openstack-nova | 15:18 | |
| *** sdake has quit IRC | 15:20 | |
| *** rlrossit has joined #openstack-nova | 15:20 | |
| *** andreas_s has quit IRC | 15:20 | |
| *** mriedem1 has joined #openstack-nova | 15:20 | |
| *** prometheanfire has joined #openstack-nova | 15:21 | |
| prometheanfire | I think I found another oslo bump that's gonna cause nova to fail tests | 15:22 |
| prometheanfire | https://review.openstack.org/360476 | 15:22 |
| prometheanfire | nova.tests.unit.test_fixtures.TestStableObjectJsonFixture fails with a versioned-objects bump | 15:22 |
| *** mriedem has quit IRC | 15:22 | |
| *** rkrum has quit IRC | 15:23 | |
| *** sridharg has quit IRC | 15:23 | |
| *** markus_z has quit IRC | 15:24 | |
| *** sdake has joined #openstack-nova | 15:24 | |
| *** sdake_ has quit IRC | 15:26 | |
| *** hemna has joined #openstack-nova | 15:26 | |
| *** shaohe_feng has quit IRC | 15:26 | |
| *** shaohe_feng has joined #openstack-nova | 15:27 | |
| *** browne has quit IRC | 15:29 | |
| *** kfarr has quit IRC | 15:29 | |
| *** eharney has joined #openstack-nova | 15:30 | |
| *** Sukhdev has joined #openstack-nova | 15:31 | |
| *** harlowja_at_home has joined #openstack-nova | 15:32 | |
| *** markvoelker has quit IRC | 15:32 | |
| mriedem1 | hmm http://logs.openstack.org/76/360476/2/check/gate-cross-nova-python27-db-ubuntu-xenial/14ae1cc/console.html#_2016-08-29_04_56_08_994629 | 15:33 |
| *** mriedem1 is now known as mriedem | 15:33 | |
| prometheanfire | mriedem: yarp | 15:34 |
| mriedem | probably due to this https://github.com/openstack/oslo.versionedobjects/commit/39a057becc10d1cfb5a5d5024bfcbbe6db1b56be | 15:34 |
| *** hieulq_ has joined #openstack-nova | 15:35 | |
| mriedem | dansmith: probably need your help handling ^ | 15:35 |
| mriedem | should we even have nova.tests.unit.test_fixtures.TestStableObjectJsonFixture in nova anymore? | 15:36 |
| mriedem | looks like no | 15:36 |
| *** shaohe_feng has quit IRC | 15:36 | |
| *** yamahata has joined #openstack-nova | 15:37 | |
| kashyap | mriedem: or others with more time, this (non-intrusive) patch from Pawel has been sitting there for a while -- https://review.openstack.org/#/c/292826/14 ("Remove block_migration from LM rollback") | 15:37 |
| mriedem | StableObjectJsonFixture was released in 1.9.0, we require >=1.13.0 in g-r, so we can probably safely remove the duplicate fixture from nova and it's now-failing unit tests | 15:37 |
| *** shaohe_feng has joined #openstack-nova | 15:37 | |
| prometheanfire | that'd be nice, removing code to fix a bug is always nice :D | 15:38 |
| mriedem | prometheanfire: i'll try and work up a patch quick | 15:38 |
| *** Guest4347 has quit IRC | 15:38 | |
| prometheanfire | thanks, I can rework the requirements patch to be dependant on yours (or you can, doesn't mater) | 15:39 |
| *** EinstCrazy has quit IRC | 15:40 | |
| *** browne has joined #openstack-nova | 15:40 | |
| dansmith | mriedem: okay yeah | 15:41 |
| dansmith | mriedem: I had to fix the test in o.vo that was testing the wrong thing after that fix, so nova's is similarly broken | 15:41 |
| dansmith | mriedem: I can patch it out | 15:42 |
| mriedem | yup | 15:42 |
| mriedem | i'm working a patch | 15:42 |
| dansmith | then I shall +2 said ptch | 15:42 |
| *** tesseract- has quit IRC | 15:42 | |
| mriedem | yar har | 15:42 |
| *** pcaruana has quit IRC | 15:43 | |
| *** rcernin has quit IRC | 15:44 | |
| *** shaohe_feng has quit IRC | 15:47 | |
| *** tongli has joined #openstack-nova | 15:47 | |
| *** shaohe_feng has joined #openstack-nova | 15:47 | |
| *** READ10 has joined #openstack-nova | 15:48 | |
| *** yamahata has quit IRC | 15:48 | |
| *** matrohon has quit IRC | 15:48 | |
| *** gszasz has quit IRC | 15:51 | |
| mriedem | dansmith: https://review.openstack.org/#/c/362214/ | 15:54 |
| *** vladikr has quit IRC | 15:54 | |
| mriedem | prometheanfire: updated https://review.openstack.org/#/c/360476/ | 15:54 |
| dansmith | mriedem: thanks | 15:54 |
| *** amotoki has quit IRC | 15:55 | |
| prometheanfire | k | 15:55 |
| jaypipes | mriedem, dansmith: in https://review.openstack.org/#/c/358797/, I'm attempting to use the same mechanism that the cinder volume driver uses to get the endpoint for the placement service. I'm having trouble though, since it's not clear that we have any entry for the new placement service in the Keystone catalog... | 15:56 |
| *** shaohe_feng has quit IRC | 15:57 | |
| *** gongysh has quit IRC | 15:58 | |
| mriedem | jaypipes: in here https://review.openstack.org/#/c/358797/5/nova/scheduler/client/report.py@41 | 15:58 |
| mriedem | i thought cdent had a devstack change to add the placement service entry to the service catalog? | 15:58 |
| mriedem | and deploy and start the service, etc | 15:58 |
| *** su_zhang has joined #openstack-nova | 15:59 | |
| *** yamahata has joined #openstack-nova | 16:00 | |
| mriedem | jaypipes: added a comment inline | 16:00 |
| mriedem | can we attempt to guess the URL? | 16:00 |
| mriedem | or provide a backdoor option for configuring the service name/type etc? | 16:01 |
| mriedem | sdague would probably hate that | 16:01 |
| *** david-lyle_ is now known as david-lyle | 16:01 | |
| jaypipes | mriedem: yeah, I specifically did not want to add an endpoint override URL configuration option (see commit message) | 16:01 |
| *** shaohe_feng has joined #openstack-nova | 16:02 | |
| *** kevinz has quit IRC | 16:03 | |
| *** nic has joined #openstack-nova | 16:03 | |
| *** hieulq_ has quit IRC | 16:03 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: plugins/xenserver: Add '.py' extension https://review.openstack.org/289431 | 16:04 |
| *** browne has quit IRC | 16:04 | |
| *** devananda|OPS is now known as devananda | 16:05 | |
| *** nkrinner is now known as nkrinner_afk | 16:05 | |
| *** _ducttape_ has quit IRC | 16:06 | |
| *** ducttape_ has joined #openstack-nova | 16:07 | |
| *** shaohe_feng has quit IRC | 16:07 | |
| *** shaohe_feng has joined #openstack-nova | 16:08 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/nova: Implement setup_networks_on_host for Neutron networks https://review.openstack.org/275073 | 16:08 |
| cmartin | Can somebody tell me where instance_type gets set when used in the creation of a new instance using the nova boot command | 16:08 |
| *** browne has joined #openstack-nova | 16:09 | |
| *** wanghao_ has quit IRC | 16:10 | |
| *** prateek_ has joined #openstack-nova | 16:10 | |
| *** su_zhang has quit IRC | 16:11 | |
| *** su_zhang has joined #openstack-nova | 16:11 | |
| anteaya | cmartin: clarify you are asking in service of fixing a bug is often helpful for your goal | 16:11 |
| bauzas | jaypipes: mriedem: well, I unfortunately think we should ask our operators to provide the placement URL in the catalog, or bypass the whole placement API if not | 16:11 |
| bauzas | but not use any hack to guess it | 16:12 |
| anteaya | s/clarify/clarifing | 16:12 |
| lbeliveau | jaypipes: we fixed this patch, would be great if you could review so that it gets merged for the release (we would then have PCI migration and revert functionnality working) https://review.openstack.org/#/c/349060/ | 16:12 |
| *** beagles_brb is now known as beagles | 16:12 | |
| *** pilgrimstack1 has quit IRC | 16:12 | |
| *** parora has quit IRC | 16:13 | |
| sfinucan | wznoinsk: any idea what's up with intel-nfv-ci? https://review.openstack.org/#/c/356398/ | 16:13 |
| mriedem | bauzas: jaypipes: yeah i'm fine with requiring the service catalog to make this work, that's what we should probably require on any new things since we want to get off hacky deployment-specific config options for endpoint overrides, | 16:14 |
| *** liangy has quit IRC | 16:14 | |
| mriedem | and given the scheduler isn't using the placement API data yet, it's probably not the end of the world if you're running newton w/o all of the data from RT going there right out of the gate right? | 16:14 |
| mriedem | but at some point we're going to have to switch that over | 16:14 |
| mriedem | i'm not sure if there is a way to make that a bit painful | 16:15 |
| wznoinsk | sfinucan: 10min, hiccup, fixed, I left a comment for a recheck | 16:15 |
| mriedem | besides just the error/warning in the logs that it's not setup | 16:15 |
| *** zenoway has quit IRC | 16:15 | |
| cmartin | Ok just to clarify, I am working on this bug https://bugs.launchpad.net/nova/+bug/1587108 and I am trying to find the root cause, from what I gather the root_gb value is what sets min_disk, and root_gb is obtained from instance_type so I am trying to figure out where instance_type is set because it is above the create method. I do know it seems to be taken from the flavor instead of the image which is what the bug is | 16:15 |
| openstack | Launchpad bug 1587108 in OpenStack Compute (nova) "Booting an instance from snapshot (which is an instance booted from Cinder volume) fails with: "Volume is smaller than the minimum size specified in image metadata."" [Medium,In progress] - Assigned to Chris Martin (cm876n) | 16:15 |
| *** su_zhang has quit IRC | 16:15 | |
| *** zenoway has joined #openstack-nova | 16:16 | |
| *** diana_clarke has left #openstack-nova | 16:16 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/nova: Using assertIsNotNone() instead of assertIsNot(None,) https://review.openstack.org/361815 | 16:16 |
| *** raunak has joined #openstack-nova | 16:17 | |
| *** shaohe_feng has quit IRC | 16:17 | |
| *** shaohe_feng has joined #openstack-nova | 16:18 | |
| *** liangy has joined #openstack-nova | 16:18 | |
| bauzas | mriedem: yup, I wasn't thinking that a service catalog entry was needed for Newton | 16:19 |
| *** sdake has quit IRC | 16:19 | |
| bauzas | early Ocata, maybe | 16:19 |
| mriedem | but if your data isn't migrated, how are we going to handle that in ocata? same as always? | 16:19 |
| mriedem | migrated/duplicated | 16:19 |
| *** armax has quit IRC | 16:20 | |
| *** zenoway has quit IRC | 16:20 | |
| *** rama_y has joined #openstack-nova | 16:20 | |
| mriedem | i guess this is already doing a migration of sorts, if the RP doesn't exist from the RT call it creates it | 16:21 |
| sfinucan | wznoinsk: Lovely, thanks :) | 16:21 |
| *** browne has quit IRC | 16:21 | |
| mriedem | jaypipes: i think you need this https://review.openstack.org/#/c/342362/ | 16:22 |
| *** armax has joined #openstack-nova | 16:22 | |
| bauzas | mriedem: mmm, that's actually a good question | 16:23 |
| mriedem | https://review.openstack.org/#/c/342362/8/lib/placement@107 creates the service catalog entries for the placement service | 16:24 |
| mriedem | so needs that as a depends-on for the nova change | 16:24 |
| bauzas | mriedem: if we say computes need to call the REST API for writing inventories, I was thinking we would directly call out the API stack using requests | 16:24 |
| mriedem | i'd think for newton we just log the error but don't fail if the SC isn't configured for the placement api | 16:24 |
| mriedem | for ocata the SC entry is a requirement, but we still fallback to creating RPs on the fly | 16:24 |
| bauzas | mriedem: for Newton, we could assume that computes can read the placement conf entries | 16:25 |
| mriedem | bauzas: i don't think we want to make that assumption, we don't make that assumption today for the nova-compute being able to read the nova-api config | 16:25 |
| mriedem | or that the configs are the same | 16:26 |
| *** salv-orlando has quit IRC | 16:27 | |
| *** salv-orlando has joined #openstack-nova | 16:27 | |
| bauzas | mriedem: well, that's an assumption we didn't enforced | 16:27 |
| jaypipes | mriedem: for the record, I didn't need that patch in the revision before this. I was asked by edleafe, Yingxin and cdent to not use the strategy I was previously using of having a _placement_api_error marker variable to short-circuit error handling and ensure that the report client object didn't bomb. | 16:27 |
| bauzas | mriedem: and by working on the disk/cpu/ram ratios to RT, I saw that we were basically reading those values | 16:28 |
| *** shaohe_feng has quit IRC | 16:28 | |
| openstackgerrit | Tuan Luong-Anh proposed openstack/nova: Using assertIsNone() instead of assertEqual(None) https://review.openstack.org/361847 | 16:28 |
| bauzas | anyway, I need to bail out :/ | 16:28 |
| *** yamahata has quit IRC | 16:28 | |
| openstackgerrit | Brent Eagles proposed openstack/nova: Add bigswitch command to compute rootwrap filters https://review.openstack.org/358024 | 16:29 |
| *** shaohe_feng has joined #openstack-nova | 16:29 | |
| *** sdake has joined #openstack-nova | 16:30 | |
| *** salv-orlando has quit IRC | 16:30 | |
| mriedem | jaypipes: ok, but either way with this change the placement API / service is still optional isn't it? | 16:33 |
| *** _diana_ has left #openstack-nova | 16:33 | |
| mriedem | we want people using it though so the switchover isn't as painful when it comes | 16:34 |
| *** kaisers_ has joined #openstack-nova | 16:35 | |
| mriedem | we need cdent (or someone) to fix the bashate failure in https://review.openstack.org/#/c/342362/ | 16:36 |
| mriedem | and then i think we want a depends-on in the nova series to show this working | 16:36 |
| mriedem | jaypipes: ^ | 16:36 |
| *** lpetrut has quit IRC | 16:38 | |
| *** shaohe_feng has quit IRC | 16:38 | |
| *** shaohe_feng has joined #openstack-nova | 16:38 | |
| Guest35918 | sdague: RE profiling of nova-scheduler with Ironic, I found that this line takes 1s to complete: https://github.com/openstack/nova/blob/kilo-eol/nova/scheduler/host_manager.py#L561 | 16:39 |
| *** Guest35918 is now known as mgagne | 16:39 | |
| mgagne | sdague: was me ^ | 16:39 |
| *** mgagne has quit IRC | 16:39 | |
| *** mgagne has joined #openstack-nova | 16:39 | |
| mriedem | mgagne: is that still in master? | 16:39 |
| *** harlowja_at_home has quit IRC | 16:39 | |
| mgagne | mriedem: checking, just started my journey into profiling that code =) | 16:40 |
| *** kaisers_ has quit IRC | 16:40 | |
| mriedem | mgagne: ok, i'd be interested to see if you're hitting this https://github.com/openstack/nova/blob/kilo-eol/nova/scheduler/host_manager.py#L590 | 16:40 |
| mriedem | i.e. backlevel computes | 16:40 |
| mgagne | mriedem: challenge is that we aren't running master or anything remotely recent | 16:40 |
| *** Apoorva has joined #openstack-nova | 16:40 | |
| mgagne | mriedem: will check right now | 16:40 |
| sdague | mriedem: sorry, was away, what's the SC thing I would hate? | 16:40 |
| *** prometheanfire has left #openstack-nova | 16:40 | |
| mriedem | sdague: endpoint overrides | 16:40 |
| sdague | mgagne: great, lets start stacking that up on a bug so we can figure out if it's still an issue | 16:41 |
| jaypipes | mriedem: yes, the placement API was entirely optional. It's just with edleafe and cdent's requests, the report client continually outputs error messages in the log file over and over again every time update_available_resource() is called. | 16:41 |
| *** Apoorva has quit IRC | 16:41 | |
| mriedem | sdague: looks like we need to get https://review.openstack.org/#/c/342362/ in so that https://review.openstack.org/#/c/358797/5/nova/scheduler/client/report.py@41 works | 16:41 |
| mgagne | mriedem: so I'm hitting objects.InstanceList.get_by_host | 16:41 |
| mriedem | mgagne: yeah, so i'm thinking there are backlevel computes not sending the instances over the wire to the scheduler | 16:42 |
| mriedem | which is why the scheduler is then looking them up in the db every time | 16:42 |
| sdague | mriedem / jaypipes: if it's entirely optional, why do we need it? Optional should mean we don't error if it's not there. | 16:42 |
| *** Apoorva has joined #openstack-nova | 16:42 | |
| mriedem | warning might be more appropriate, | 16:42 |
| mriedem | we want people to use it so that we start creating resource providers in the placement service asap | 16:43 |
| sdague | maybe step back, what is the desired behavior if there is no SC entry? | 16:43 |
| sdague | and when do we want it non optional? | 16:43 |
| mriedem | i thought ocata | 16:43 |
| sdague | and how is that transition? | 16:43 |
| armax | mriedem: how are we doing as far as bug 1615710 goes? | 16:43 |
| openstack | bug 1615710 in neutron "test_server_multi_create_auto_allocate failing with "Failed to allocate the network(s), not rescheduling"" [Critical,Fix released] https://launchpad.net/bugs/1615710 - Assigned to Kevin Benton (kevinbenton) | 16:43 |
| armax | did you notice any bump in the road so far? | 16:43 |
| *** unicell has quit IRC | 16:43 | |
| mriedem | armax: the tempest skip revert merged this morning | 16:43 |
| armax | mriedem: ok, still early to say then? | 16:44 |
| mriedem | armax: so far no bumps | 16:44 |
| mriedem | http://status.openstack.org//elastic-recheck/index.html#1616498 | 16:44 |
| mgagne | mriedem: is it related to the scheduler_tracks_instance_changes config? | 16:44 |
| armax | mriedem: ack | 16:44 |
| sdague | armax: thanks for digging into that | 16:44 |
| mriedem | mgagne: i don't know off the top of my head, would have to dig into that code | 16:44 |
| armax | sdague: np | 16:44 |
| mgagne | ok, will check | 16:44 |
| jaypipes | sdague: the prior revision of that patch did *not* error (just logged a warning). edleafe and others asked me to change things to repeatedly log errors and try connecting to the placement API instead of the short-circuit error marker approach I was using in the previous patch revision. | 16:44 |
| *** lucasagomes is now known as lucas-afk | 16:45 | |
| mriedem | jaypipes: i think a warning in newton is ok | 16:45 |
| mriedem | and then we hard fail in ocata if it's not there right? | 16:45 |
| jaypipes | mriedem: that was indeed my original plan. | 16:45 |
| jaypipes | and my original revision. | 16:45 |
| mriedem | indubitably | 16:45 |
| mriedem | so let's do that | 16:46 |
| mriedem | but also get cdent's devstack change landed so we can see this working | 16:46 |
| jaypipes | le sigh... do you feel like arguing with edleafe and Yingxin on the patch comments? | 16:46 |
| mgagne | mriedem: well... help message is a bit confusing: https://github.com/openstack/nova/blob/282c257aff6b53a1b6bb4b4b034a670c450d19d8/nova/conf/scheduler.py#L174-L177 | 16:46 |
| mriedem | jaypipes: not really, what's their reason? raise the big red flag that you need this thing setup even though the scheduler isn't using it yet? | 16:46 |
| sdague | mriedem: hold on a sec, because the devstack change means that we don't test the optionality part | 16:46 |
| *** Swami has joined #openstack-nova | 16:47 | |
| mriedem | jaypipes: i don't think we need an error in tens of thousands of cern compute nodes when they upgrade to newton :) | 16:47 |
| sdague | because this code is definitely not going to quietly not work if the SC entry isn't there | 16:47 |
| mriedem | for an optional thing | 16:47 |
| sdague | this is going to blast errors left right and center | 16:47 |
| mgagne | mriedem: "You should disable it to gain performance if you don't need it BUT info could be loaded from DB is needed." I'm not sure *what* needs it, looks to be systematic... | 16:47 |
| jaypipes | sdague: the previous revision did just that (quietly not work if the SC entry wasnt there, with a single warning in the logs) | 16:47 |
| *** mamandle has quit IRC | 16:47 | |
| mriedem | mgagne: it being the option | 16:48 |
| mriedem | set scheduler_tracks_instance_changes=False to disable 'it' | 16:48 |
| mgagne | mriedem: yes | 16:48 |
| mriedem | jaypipes: i think that's what we want | 16:48 |
| *** shaohe_feng has quit IRC | 16:48 | |
| sdague | jaypipes: ok, so that seems better | 16:48 |
| mriedem | then we bring the pain in ocata | 16:49 |
| *** shaohe_feng has joined #openstack-nova | 16:49 | |
| *** trisq has quit IRC | 16:49 | |
| edleafe | jaypipes: the issue I had was that once an error hit, there was no reviving things short of restarting the service | 16:49 |
| edleafe | jaypipes: I didn't want the logs flooded with repeated error messages | 16:50 |
| *** READ10 has quit IRC | 16:50 | |
| mriedem | so can we do both? time cache the sc lookup result, but log warning if it's not there (instead of error) | 16:50 |
| sdague | mriedem: why would we want to do that? | 16:51 |
| mriedem | sdague: so it doesn't warn every minute | 16:51 |
| sdague | mriedem: I guess I'm not sure I understand why it warns every minute | 16:51 |
| mriedem | b/c it's in the RT update_available_resource | 16:51 |
| mriedem | which by defaults runs every 60s | 16:51 |
| sdague | so just set a global | 16:51 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/360894 | 16:51 |
| jaypipes | edleafe: sorry, was Yingxin who said "If not, a misconfigured placement service may cause all the compute nodes fall into unrecoverable errors. And deployer must restart all the compute nodes to fix the problem. I don't think there is any unrecoverable errors in compute nodes." | 16:52 |
| mriedem | ^ is the complaint against setting a global right? | 16:53 |
| sdague | HAS_WARNED_ABOUT_PLACEMENT=True | 16:53 |
| jaypipes | mriedem: honestly, I'm not sure any more. :( | 16:53 |
| sdague | don't warn again if it's true | 16:54 |
| sdague | but still try to rebuild the client | 16:54 |
| mriedem | sdague: ok yeah i think that's fine | 16:54 |
| sdague | and unset it if you find a working placement endpoint | 16:54 |
| jaypipes | https://review.openstack.org/#/c/358797/1/nova/scheduler/client/report.py | 16:54 |
| mriedem | was mixing up what the flag was going to block | 16:54 |
| jaypipes | ^^ please see above. | 16:54 |
| mgagne | mriedem: ok, this looks to be the config to enable. It will load all instance info in an async manner, maybe in-memory. Once info are all loaded, _add_instance_info takes no time to complete and db isn't hit anymore. | 16:55 |
| jaypipes | there used to be a ScheduerReportClient._placement_api_error marker variable that was used to short-circuit to prevent constantly trying to connect to the placement service when it didn't exist. | 16:55 |
| mriedem | jaypipes: i think we want to try and keep reconnecting | 16:55 |
| mriedem | because once that's set in the sc then we're good | 16:55 |
| mriedem | w/o having to restart nova-compute | 16:55 |
| jroll | +1 | 16:56 |
| sdague | mriedem: where is this object built? | 16:56 |
| mriedem | but use the global to not dump a warning every 60s | 16:56 |
| jaypipes | I was asked to remove that _placment_api_error marker variable because it was too "strong". From Chris: "The -1 is because I'm not sure the behaviour of the error flag is sufficiently clear and without that clarity is hard to know if it is correct. At this stage it seems like it is too strong and not resilient in transitory failure states." | 16:56 |
| *** tonytan4ever has quit IRC | 16:57 | |
| jaypipes | mriedem: ok then. this patch will be 1K LOCC of more... just warning you. | 16:57 |
| mriedem | so (1) keep trying to reconnect, (2) log a warning on first failure, set a global, (3) don't warn on connection failure if we already have once before, (4) unset the flag once we connect | 16:57 |
| mriedem | i might be missing something, i obviously don't have the review history here | 16:58 |
| mriedem | anyway, jury duty time for me to get to the courthouse | 16:58 |
| *** tonytan4ever has joined #openstack-nova | 16:58 | |
| sdague | jaypipes: right, so a lot of history happened, let's not entirely unwind that now. I mostly want to focus on what we think the correct behavior should be. | 16:58 |
| jaypipes | mriedem: I've been trying to keep the patch size small. doing things like retrying connections adds a significant amount of lines of code. | 16:58 |
| *** vladikr has joined #openstack-nova | 16:58 | |
| *** vladikr has quit IRC | 16:58 | |
| *** shaohe_feng has quit IRC | 16:58 | |
| *** mriedem has quit IRC | 16:58 | |
| *** shaohe_feng has joined #openstack-nova | 16:59 | |
| jaypipes | sdague: the correct behaviour should be that the report client doesn't take a dump if there is no placement service available. | 16:59 |
| mgagne | sdague: so I found the culprit taking to mriedem | 16:59 |
| jroll | sure, that much is clear - the second question is whether we think folks should need to restart n-cpu once they get the placement service running? | 16:59 |
| *** lucas-afk is now known as lucasagomes | 17:00 | |
| *** jaosorior has quit IRC | 17:00 | |
| sdague | mgagne: awesome | 17:00 |
| jaypipes | sdague: whether or not *this particular patch* should add a bunch of code for handling retrying of the placement service connectivity along with logic for when and how often to log some warning/error is what is in question here. The longer the patch is, the less likely anything is to merge, right? | 17:00 |
| mgagne | sdague: scheduler_tracks_instance_changes needs to be True. Default value in Nova is True. BUT looking at Ironic install-guide, they recommend to set it to False: http://docs.openstack.org/developer/ironic/deploy/install-guide.html | 17:00 |
| sdague | mgagne: ah... great | 17:00 |
| mgagne | sdague: yea.... super awesome | 17:00 |
| *** mpaolino has joined #openstack-nova | 17:01 | |
| * jroll jumping into ironic meeting but would be curious what the problem there is | 17:01 | |
| sdague | mgagne: have you been able to test it in an environment to see the perf difference? | 17:01 |
| jaypipes | jroll: It would be great if they don't have to restrt n-cpu. The question is do we have to do all of this in the same patch. | 17:01 |
| *** tonytan4ever has quit IRC | 17:01 | |
| jaypipes | jroll: or can I please add a followup patch that does the reconnecting tries. | 17:01 |
| jroll | jaypipes: ah, fair enough | 17:01 |
| jaypipes | jroll: otherwise this patch is going to be enormous. | 17:01 |
| mgagne | sdague: adding a ton of debug logs in an environment with lot of hosts. I can see clear improvement in timing. 1s -> 0.01s | 17:01 |
| sdague | mgagne: very cool | 17:01 |
| mgagne | sdague: at least it got fixed. I'm a bit frustrated that our team didn't catch it before today :-/ | 17:02 |
| mgagne | we knew it was in the get_all_host_states function but didn't push our investigation, big mistake | 17:03 |
| sdague | mgagne: well, it happens | 17:03 |
| mgagne | sdague: my coworker found this: https://bugs.launchpad.net/nova/+bug/1568188 | 17:03 |
| openstack | Launchpad bug 1568188 in OpenStack Compute (nova) "Scheduler should respect the setting of scheduler_tracks_instance_changes" [Undecided,Opinion] | 17:03 |
| edleafe | jaypipes: what about setting a timeout on failure? This way it isn't constantly trying to re-connect, but will eventually reconnect if the problem is transient. | 17:03 |
| sdague | edleafe: I don't understand how the problem will be transient | 17:05 |
| *** browne has joined #openstack-nova | 17:05 | |
| jroll | mgagne: btw, I really wouldn't recommend scheduler_tracks_instance_changes=True in ironic unless you're at a very small scale | 17:05 |
| mgagne | sdague: so I think that if you really don't care about details of instance located on nodes (like you would with an affinity filter), there is no way to disable the tracking of this info. You enable the option, it builds in an async manner, you disable it, it hits the DB every times | 17:05 |
| mgagne | jroll: well =False saved me 0.99s load time for every node. Please explain the logic here =) | 17:06 |
| jroll | mgagne: select * from instances where host='the-only-ironic-n-cpu-which-has-hundreds-or-thousands-of-instances'; | 17:07 |
| mgagne | jroll: could be the behavior got fixed in latest Nova but so far, you NEED to have scheduler_tracks_instance_changes=True to scale Nova scheduler for Ironic. | 17:07 |
| *** yamahata has joined #openstack-nova | 17:07 | |
| jroll | mgagne: a cycle or two ago that was OOM'ing 8gb scheduler hosts at startup | 17:07 |
| jroll | because it loads all instances at scheduler startup | 17:07 |
| sdague | jroll: this is kilo level | 17:07 |
| mgagne | jroll: adding debug logs shows that DB is hit for every hypervisor/ironic node loaded by scheduler. | 17:07 |
| jroll | sdague: okay, I think this was liberty | 17:08 |
| *** tonytan4ever has joined #openstack-nova | 17:08 | |
| mgagne | jroll: timing we have for one single node loaded by scheduler: http://paste.openstack.org/show/564621/ | 17:08 |
| *** shaohe_feng has quit IRC | 17:09 | |
| *** vladikr has joined #openstack-nova | 17:09 | |
| mgagne | 1s spent in add_instance_info which in turn calls objects.InstanceList.get_by_host(context, host_name) | 17:09 |
| mgagne | here https://github.com/openstack/nova/blob/kilo-eol/nova/scheduler/host_manager.py#L590 | 17:09 |
| *** shaohe_feng has joined #openstack-nova | 17:10 | |
| jroll | mgagne: ouch. I don't have time to dig into code right now, but I do recall that config causing us huge problems in liberty | 17:10 |
| mgagne | so I think there is no way to disable loading of instance info in Ironic context. Either it's done in an async manner (OOM problem?) or it's systematic for each nodes scheduler loads from database. (takes forever) | 17:11 |
| *** jaypipes has quit IRC | 17:12 | |
| jroll | mmm | 17:12 |
| jroll | yeah | 17:12 |
| sdague | jaypipes: how does https://review.openstack.org/#/c/358797/5/nova/scheduler/client/report.py even work if the SC entry is there? It's using a naked requests session, and I don't see where it hands keystone auth to it anywhere? | 17:12 |
| mgagne | that's why I think we should reopen this bug: https://bugs.launchpad.net/nova/+bug/1568188 | 17:13 |
| openstack | Launchpad bug 1568188 in OpenStack Compute (nova) "Scheduler should respect the setting of scheduler_tracks_instance_changes" [Undecided,Opinion] | 17:13 |
| mgagne | it has a huge performance impact, even for huge cloud if you don't need affinity filters | 17:13 |
| *** jaypipes has joined #openstack-nova | 17:13 | |
| mgagne | the reporter is describing the *exact* same situation we are in | 17:15 |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova-specs: Improve usability of booting guest with SR-IOV https://review.openstack.org/362287 | 17:16 |
| jroll | mgagne: why do I feel like I had a similar bug files | 17:16 |
| jroll | filed* | 17:16 |
| *** mdrabe has joined #openstack-nova | 17:16 | |
| *** ManishD has joined #openstack-nova | 17:17 | |
| jroll | mgagne: https://bugs.launchpad.net/nova/+bug/1479124 | 17:17 |
| openstack | Launchpad bug 1479124 in OpenStack Compute (nova) "Scheduler doesn't respect tracks_instance_changes in all cases" [Low,Fix released] - Assigned to Ed Leafe (ed-leafe) | 17:17 |
| jroll | mgagne: doesn't fix it the way you'd think, but does fix it for ironic | 17:18 |
| mgagne | jroll: will read bug and commit | 17:18 |
| jroll | mgagne: enjoy :) | 17:19 |
| *** shaohe_feng has quit IRC | 17:19 | |
| mgagne | jroll: I think you implemented something similar to us but we limited scheduler to empty compute nodes at the DB level instead: https://gist.github.com/mgagne/1fbeca4c0b60af73f019bc2e21eb4a80 | 17:19 |
| jroll | mgagne: actually, maybe it's not what you need, sorry I'm distracted with a meeting | 17:20 |
| jroll | ah, interesting | 17:20 |
| mgagne | jroll: yea, I don't care about configurability, I don't mind having instance tracking disabled forever in Ironic context | 17:21 |
| *** mpaolino has quit IRC | 17:21 | |
| *** su_zhang has joined #openstack-nova | 17:21 | |
| *** shaohe_feng has joined #openstack-nova | 17:22 | |
| mgagne | jroll: I think our little hack should be part of upstream, I see little reasons for nova-scheduler to ever consider hypervisor with at least 1 instance in Ironic context. but you would lose some debug info since filtering would be done beforehand in database, not scheduler filters | 17:24 |
| *** tri2sing has joined #openstack-nova | 17:24 | |
| *** ducttape_ has quit IRC | 17:25 | |
| jroll | mgagne: yeah, that's a good point | 17:26 |
| ManishD | Dansmith,BobBall,Jay Pipes, Nova team: HI, Please can you have a look when you have some free time: https://review.openstack.org/#/c/357131/ | 17:27 |
| *** lpetrut has joined #openstack-nova | 17:28 | |
| *** shaohe_feng has quit IRC | 17:29 | |
| *** shaohe_feng has joined #openstack-nova | 17:30 | |
| *** aghadge has joined #openstack-nova | 17:30 | |
| *** aghadge has quit IRC | 17:31 | |
| *** TravT_ has joined #openstack-nova | 17:32 | |
| *** yamahata__ has quit IRC | 17:33 | |
| *** unicell has joined #openstack-nova | 17:33 | |
| *** TravT has quit IRC | 17:34 | |
| *** ducttape_ has joined #openstack-nova | 17:37 | |
| *** mriedem has joined #openstack-nova | 17:38 | |
| *** maestropandy has joined #openstack-nova | 17:39 | |
| *** zul has quit IRC | 17:39 | |
| mriedem | he's guilty! | 17:39 |
| mriedem | jury duty cancelled suckas | 17:39 |
| *** shaohe_feng has quit IRC | 17:39 | |
| mriedem | the real crime was making me wear dress pants when it's 90 out | 17:40 |
| openstackgerrit | viswesuwara nathan proposed openstack/python-novaclient: The incorrect output of "nova show" when VM has long user data fixed. New argument "--wrap" is added for "nova show" command to wrap the output. The default wrap length is set to 80 now. Example: nova show --wrap 60 <server-id> https://review.openstack.org/362299 | 17:40 |
| *** maestropandy has left #openstack-nova | 17:40 | |
| *** ihrachys has quit IRC | 17:40 | |
| *** praefect has joined #openstack-nova | 17:40 | |
| openstackgerrit | viswesuwara nathan proposed openstack/python-novaclient: The incorrect output of "nova show" when VM has long user data fixed. https://review.openstack.org/362299 | 17:41 |
| *** tonytan4ever has quit IRC | 17:42 | |
| dansmith | mriedem: hah | 17:42 |
| *** zul has joined #openstack-nova | 17:42 | |
| *** shaohe_feng has joined #openstack-nova | 17:42 | |
| openstackgerrit | viswesuwara nathan proposed openstack/python-novaclient: The incorrect output of "nova show" when VM has long user data. https://review.openstack.org/362299 | 17:42 |
| *** jaypipes has quit IRC | 17:43 | |
| *** fragatina has quit IRC | 17:44 | |
| *** ManishD has quit IRC | 17:44 | |
| *** ManishD has joined #openstack-nova | 17:44 | |
| *** Sukhdev has quit IRC | 17:45 | |
| mriedem | sdague: so did jaypipes get his RT thing sorted out? i wasn't catching what the issue was with the retry | 17:45 |
| edleafe | sdague: sorry, was AFK. If there is no placement service, then no, it won't be transient. | 17:48 |
| edleafe | sdague: the concern was *with* the placement service, a network hiccup could turn it off. | 17:48 |
| *** eharney_ has joined #openstack-nova | 17:49 | |
| *** shaohe_feng has quit IRC | 17:50 | |
| *** praefect has quit IRC | 17:50 | |
| sdague | edleafe: ah, yes | 17:50 |
| sdague | right, that seems wrong for sure | 17:50 |
| *** shaohe_feng has joined #openstack-nova | 17:51 | |
| sdague | mriedem: I do not think we are yet sorted | 17:51 |
| *** eharney has quit IRC | 17:51 | |
| mgagne | mriedem: so jroll found this bug: https://bugs.launchpad.net/nova/+bug/1479124 | 17:51 |
| openstack | Launchpad bug 1479124 in OpenStack Compute (nova) "Scheduler doesn't respect tracks_instance_changes in all cases" [Low,Fix released] - Assigned to Ed Leafe (ed-leafe) | 17:51 |
| *** bdeschenes has joined #openstack-nova | 17:52 | |
| openstackgerrit | viswesuwara nathan proposed openstack/python-novaclient: The incorrect output of "nova show" when VM has long user data fixed https://review.openstack.org/362303 | 17:52 |
| mriedem | sdague: ok, i just left a comment in the patch, i'm not sure where the extra 1K LOC comes in | 17:52 |
| *** markvoelker has joined #openstack-nova | 17:53 | |
| mriedem | mgagne: can we back up - i'm assuming something came out of the ops summit last week that is related to scheduler perf profiling or something? | 17:53 |
| mriedem | mgagne: so the fix for that bug doesn't actually fix the problem for what you were looking at, it was really a partial fix, and only for ironic | 17:54 |
| mgagne | mriedem: yea, I complained about performance in our setup. sdague came and asked for more details. I explained and we came up with the idea of profiling the setup. In the end, we found this was the root cause. | 17:54 |
| mgagne | mriedem: I only care about Ironic at this point | 17:54 |
| mgagne | mriedem: because it hasn't affected our cloud yet | 17:55 |
| mriedem | mgagne: ok, well, you're testing on kilo and https://review.openstack.org/#/c/206736/ was in liberty | 17:55 |
| *** eharney_ is now known as eharney | 17:55 | |
| mgagne | mriedem: yes but I'm sure the fix would be similar in Kilo | 17:55 |
| mriedem | ok...this is feature freeze week so perf issues on kilo seem like a distraction right now | 17:56 |
| mgagne | mriedem: yea, my expectations aren't that a backport will be made, I can take care of that on my side. I was only looking for a root cause to this performance issue and hoping it's not something that is still present in master | 17:57 |
| *** dharinic is now known as dharinic|lunch | 17:58 | |
| *** shaohe_feng has quit IRC | 18:00 | |
| *** isq has joined #openstack-nova | 18:01 | |
| sdague | mgagne: yeh, so given the mad rush this week to meet freeze, it might be better to pick it up next week in more detail. Especially if jroll can be involved as he'll understand more of the ironic implications. | 18:01 |
| *** Sukhdev has joined #openstack-nova | 18:01 | |
| *** shaohe_feng has joined #openstack-nova | 18:01 | |
| openstackgerrit | Lucian Petrut proposed openstack/nova: HyperV: use os-brick for volume related operations https://review.openstack.org/273504 | 18:02 |
| *** moshele has joined #openstack-nova | 18:02 | |
| *** fragatina has joined #openstack-nova | 18:02 | |
| mgagne | sdague: I think the bug is "solved" in Liberty+, since we are running Kilo, we are out of luck. We will take care of backporting process ourselves on our sides. I'm sure there is more to do to improve performance but so far, I'm satisfied with the bug jroll found and fixed. | 18:03 |
| sdague | mgagne: ok, great | 18:03 |
| mgagne | sdague: thanks for the help btw, much appreciated =) | 18:03 |
| sdague | mgagne: no prob. Glad you all got it solved. | 18:04 |
| *** mugsie_ is now known as mugsie | 18:05 | |
| *** rmart04 has joined #openstack-nova | 18:06 | |
| openstackgerrit | Vu Tran proposed openstack/nova: Not allow overcommit ratios to be negative https://review.openstack.org/344254 | 18:06 |
| *** tbachman has joined #openstack-nova | 18:06 | |
| prateek_ | mriedem, jaypipes , can you please review https://review.openstack.org/#/c/294513/ whenever you have some free time | 18:06 |
| *** markvoelker has quit IRC | 18:09 | |
| *** shaohe_feng has quit IRC | 18:10 | |
| *** rmart04 has quit IRC | 18:10 | |
| *** vishwanathj has quit IRC | 18:11 | |
| *** vishwanathj has joined #openstack-nova | 18:11 | |
| *** shaohe_feng has joined #openstack-nova | 18:12 | |
| *** kolesovdv has quit IRC | 18:15 | |
| *** lpetrut has quit IRC | 18:15 | |
| oomichi | mriedem: can we push https://review.openstack.org/#/c/328157 ? I and alex_xu are fine on current one, we just need 10 sec to push | 18:17 |
| ManishD | mriedem, sdageue, johnthetubaguy, jaypipes, can you please review https://review.openstack.org/#/c/357131/ whenever you have some free time | 18:18 |
| *** jvgrant__ has quit IRC | 18:18 | |
| *** jvgrant has joined #openstack-nova | 18:18 | |
| mriedem | oomichi: you're the 3rd person to ask me about that one now - why should we get that in so far past non-priority feature freeze? | 18:19 |
| *** isq has quit IRC | 18:19 | |
| *** moshele has quit IRC | 18:19 | |
| *** jvgrant_ has joined #openstack-nova | 18:19 | |
| *** annegent_ has quit IRC | 18:19 | |
| *** tanee is now known as tanee_away | 18:20 | |
| *** isq has joined #openstack-nova | 18:20 | |
| *** shaohe_feng has quit IRC | 18:20 | |
| *** shaohe_feng has joined #openstack-nova | 18:22 | |
| *** ihrachys has joined #openstack-nova | 18:22 | |
| oomichi | mriedem: yeah, I know. we spend much time for that, and the sooner merging, the better quality by using it on many chances. | 18:22 |
| *** su_zhang has quit IRC | 18:23 | |
| *** TravT_ has quit IRC | 18:23 | |
| *** jvgrant has quit IRC | 18:23 | |
| mriedem | it should have been merged 2 months ago then | 18:24 |
| ducttape_ | anyone able to speak to the timeline on https://blueprints.launchpad.net/nova/+spec/enable-volume-types ? wondering when we can launch instances with other volume types, in a single call | 18:24 |
| *** kaisers_ has joined #openstack-nova | 18:24 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Pull from cell0 and build_requests for instance list https://review.openstack.org/356138 | 18:25 |
| *** su_zhang has joined #openstack-nova | 18:25 | |
| *** isq has quit IRC | 18:26 | |
| mriedem | ducttape_: see the ML thread on that - no one ever pushed a spec for it | 18:27 |
| *** moshele has joined #openstack-nova | 18:27 | |
| ducttape_ | k thanks mriedem | 18:28 |
| oomichi | mriedem: yeah, and you have a good point on that now. I lose | 18:29 |
| *** kaisers_ has quit IRC | 18:29 | |
| *** moshele has quit IRC | 18:30 | |
| *** shaohe_feng has quit IRC | 18:31 | |
| mriedem | this also seems like a thing we'd block on capabilities, like for live resize | 18:32 |
| *** shaohe_feng has joined #openstack-nova | 18:32 | |
| mriedem | because it's totally dependent on how ironic is configured it looks like | 18:32 |
| *** lucasagomes is now known as lucas-dinner | 18:33 | |
| *** isq has joined #openstack-nova | 18:34 | |
| *** dharinic|lunch is now known as dharinic | 18:34 | |
| *** salv-orlando has joined #openstack-nova | 18:34 | |
| mriedem | oomichi: another thing is, shouldn't ironic have a microversion bump for this? | 18:36 |
| mriedem | it's a new console type / capability in their API isn't it? so how would nova, as a client, know if it's talking to an ironic that's new enough to support this type of console? | 18:37 |
| *** stewie925_ has joined #openstack-nova | 18:37 | |
| jroll | it's a new driver, the API already existed | 18:37 |
| *** mnestratov has quit IRC | 18:37 | |
| *** tonytan4ever has joined #openstack-nova | 18:38 | |
| *** stewie925 has quit IRC | 18:38 | |
| *** stewie925_ is now known as stewie925 | 18:38 | |
| *** jvgrant_ has quit IRC | 18:38 | |
| mriedem | so i guess we just have https://review.openstack.org/#/c/328157/23/nova/virt/ironic/driver.py@1335 | 18:39 |
| *** jvgrant_ has joined #openstack-nova | 18:39 | |
| jroll | yeah | 18:39 |
| mriedem | which if it's not socat translates to a 400 in the nova API | 18:39 |
| * jroll has not reviewed this yet | 18:39 | |
| *** ihrachys has quit IRC | 18:39 | |
| *** shaohe_feng has quit IRC | 18:41 | |
| *** salv-orlando has quit IRC | 18:42 | |
| mriedem | i'm not sure we'd want to block this on host capabilities, since this has less to do with placement/scheduling, although in the far out future i could make a scheduling request that i'd prefer a host that supports serial console | 18:42 |
| *** shaohe_feng has joined #openstack-nova | 18:42 | |
| mriedem | as part of jay's required/preferred build request stuff | 18:42 |
| *** cdent has joined #openstack-nova | 18:42 | |
| mriedem | also, that patch is a landmine of LOG.error | 18:42 |
| *** ManishD has quit IRC | 18:44 | |
| stewie925 | hello guys - I investigated Nova bug #1117923 and found that it looks like a non-issue at all. Please review my comments. Inputs appreciated. | 18:44 |
| openstack | bug 1117923 in OpenStack Compute (nova) "instance_system_metadata column type truncates on snapshot" [Medium,In progress] https://launchpad.net/bugs/1117923 - Assigned to STEW TY (stewie925) | 18:44 |
| *** Sukhdev has quit IRC | 18:46 | |
| *** su_zhang has quit IRC | 18:46 | |
| *** liverpooler has quit IRC | 18:47 | |
| *** shaohe_feng has quit IRC | 18:51 | |
| *** shaohe_feng has joined #openstack-nova | 18:53 | |
| cdent | mriedem, sdague I just tried to catch up in the log about the resource tracker sending inventories stuff, and I'm not sure I was able to extract a good summary. I can say, howerver, that my concerns about hard versus soft failure were not about the service catalog not being up to date, but with the placement service suffering a failure because of network partition | 18:53 |
| cdent | edleafe: you too on ^, I guess jaypipes is no longer with us today | 18:54 |
| mriedem | cdent: looking at patch set 1 on that change, the error flag was set one of two ways i think, (1) sc lookup fails and (2) some placement API query fails | 18:56 |
| mriedem | and then the flag wasn't unset | 18:56 |
| sdague | cdent: right, that makes a ton of sense | 18:56 |
| cdent | mriedem: yeah, and 2 was my arear of concern | 18:56 |
| sdague | I agree that placement API falling over shouldn't wedge things | 18:56 |
| *** tongli has quit IRC | 18:56 | |
| mriedem | cdent: agree on that | 18:56 |
| mriedem | i commented in the latest patch set on the change | 18:56 |
| sdague | I'm fine if a change in #1 requires nova-compute restart | 18:56 |
| mriedem | set a flag for warning on (1) | 18:56 |
| *** praneshp has quit IRC | 18:56 | |
| * cdent is fixing the bashate thing on the devstack-for-placement review | 18:56 | |
| sdague | I still don't quite get how the auth is passed though | 18:57 |
| cdent | sdague: I have that concern too | 18:57 |
| *** annegentle has joined #openstack-nova | 18:57 | |
| mriedem | cdent: ah ok, i was just going to fix that bashate failure, figured you were done for the day | 18:57 |
| cdent | mriedem: I'm not really here today, is a holiday | 18:57 |
| *** annegentle has quit IRC | 18:57 | |
| cdent | but I needed to exercise my typing fingers | 18:57 |
| *** TravT has joined #openstack-nova | 18:58 | |
| sdague | so... I still don't think we want to land that devstack change this cycle, because for newton we should default test the "it's optional" state | 18:59 |
| sdague | and add it in as soon as we branch devstack | 18:59 |
| cdent | I'm agnostic/ambivalent: I've needed it around for my own testing, so it is handy to have it exist, but I'm not certain it needs to be merged any time soon | 19:00 |
| melwitt | mriedem: do you know if there's something different than needs to be done with glance v2 when doing ceph snapshots, to allow the rbd driver to update the image location in glance? I thought I remembered there being a difference between v1 and v2 glance for how the location update works | 19:00 |
| melwitt | like a config setting | 19:01 |
| mriedem | melwitt: https://github.com/openstack/devstack-plugin-ceph/blob/master/devstack/lib/ceph#L559 | 19:01 |
| *** shaohe_feng has quit IRC | 19:01 | |
| mriedem | melwitt: i don't think we had to change anything for the ceph job to keep working with glance v2 | 19:02 |
| *** ihrachys has joined #openstack-nova | 19:02 | |
| *** annegentle has joined #openstack-nova | 19:02 | |
| *** ljxiash has quit IRC | 19:02 | |
| *** shaohe_feng has joined #openstack-nova | 19:02 | |
| bauzas | mriedem: sdague: cdent: so I agree with sdague, we should make the placement call optionak | 19:02 |
| mriedem | sdague: cdent: if not merged for newton, we should probably have something at the top of the stack running with it enabled to make sure the nova bits underneath don't totally explode when it's running | 19:02 |
| bauzas | I thought about that during dinner | 19:02 |
| sdague | cdent: and you've seen it all work? I'm still a little confused on how a naked requests client that doesn't seem to add keystone headers work | 19:02 |
| melwitt | mriedem: thanks! yeah, I thought if anything it was just a config thing, nothing else different | 19:03 |
| sdague | mriedem: yeh, sure | 19:03 |
| mriedem | sdague: cdent: because i thought the point was, make it available but optional, but to start migrating/duplicating data now before it's all required | 19:03 |
| sdague | cdent: but you can also get back to vacationing today | 19:03 |
| sdague | mriedem: sure | 19:03 |
| bauzas | and I think we should just make sure that if there is some problem with the scheduler client, then it won't update the resource provider API, but it will still update the current compute nodes | 19:03 |
| cdent | sdague: no, I _haven't_ seen it work, thus the concern. By "my testing" I mean testing (with gabbi-run) of the placement api, not the resource tracker. | 19:03 |
| sdague | cdent: ok, yeh, I expect that | 19:03 |
| mriedem | cdent: ok, so that's where we need a top-level DNM patch that runs with the latest nova RT code plus the devstack change to enable the placement API | 19:04 |
| bauzas | cdent: mriedem: sdague: that way, operators needing to use the placement API would probably add a new service type, but for the others, then it will still work | 19:04 |
| mriedem | because w/o that the RT code could be calling a non-existing placement API or something totally wonky | 19:04 |
| dansmith | alaski: melwitt: sanity check me here: https://review.openstack.org/#/c/326906/5 | 19:04 |
| mriedem | given the number of changes and specs that wouldn't surprise me | 19:04 |
| cdent | mriedem: I think we probably need to have an email thread or a chat in one of the meetings to get everyone on the same page because I'm not sure anyone really knows what the expected bits are | 19:06 |
| *** TravT has quit IRC | 19:06 | |
| cdent | It needs to be written down, somewhere other than gerrit or specs | 19:06 |
| mriedem | how about a tattoo on jay's butt? | 19:06 |
| cdent | oh, yeah, the devstack stuff needs to -W until the placement db stuff that got reverted is redone (which I'm going to do tomorrow morning) | 19:07 |
| *** vladikr has quit IRC | 19:07 | |
| mriedem | from the midcycle etherpad: "Jay's personal goal for Newton is for the resource tracker to be writing inventory and allocation data via the placement API. Get the data pumping into the placement API in Newton so we can start using it in Ocata." | 19:08 |
| *** vladikr has joined #openstack-nova | 19:08 | |
| mriedem | jay's working on the inventory/allocation stuff on top of https://review.openstack.org/#/c/358797/ | 19:08 |
| mriedem | while on PTO i guess | 19:09 |
| cdent | family reunion plans don't align with openstack release cycles | 19:10 |
| mriedem | so today SchedulerReportClient.update_resource_stats just calls compute_node.save() | 19:10 |
| mriedem | https://review.openstack.org/#/c/358797/ updates it to start creating the resource provider in the placement service if it doesn't exist | 19:10 |
| *** parora has joined #openstack-nova | 19:11 | |
| alaski | dansmith: makes sense to me, commented my agreement | 19:11 |
| mriedem | "These resource provider objects will be used in followup patches that add creation of inventory and allocation records to the scheduler report client." | 19:11 |
| dansmith | alaski: okay | 19:11 |
| *** shaohe_feng has quit IRC | 19:12 | |
| mriedem | so i think we just need at least a nova DNM change that depends on the top of the generic-resource-pools series and depends on the devstack change to enable the placement api | 19:12 |
| mriedem | unless we had an experimental job that runs with the placement API | 19:12 |
| *** tbachman has quit IRC | 19:13 | |
| *** shaohe_feng has joined #openstack-nova | 19:13 | |
| *** praneshp has joined #openstack-nova | 19:13 | |
| mriedem | sdague: couldn't we do that ^ ? devstack defaults to not enable the placement API in newton, but we have a job that adds placement to ENABLED_SERVICES and run that in the experimental queue until ocata? | 19:13 |
| cdent | Creating an experimental job against the placement API is on my todo list for "very soon", but my todo list is way bigger than realistic at the moment :( | 19:13 |
| *** prateek_ has quit IRC | 19:14 | |
| cdent | (I was just gonna do a simple `gabbi-run < some-simple.yaml` (borrowed from integration tests in telemetry and a few other places)) | 19:14 |
| sdague | mriedem: yes, we can do that. | 19:14 |
| mriedem | cdent: we can have other people work on that job definition | 19:14 |
| cdent | that would be very useful | 19:15 |
| melwitt | dansmith, alaski: makes sense to me too, thanks for catching it | 19:15 |
| mriedem | cdent: so this is the -W right? https://review.openstack.org/#/c/342362/8/lib/placement@123 | 19:15 |
| mriedem | since there isn't a placement DB | 19:15 |
| cdent | right | 19:15 |
| cdent | well | 19:16 |
| cdent | the placement DB is optional | 19:16 |
| cdent | but the way in which it is optional has changed since a weeka go | 19:16 |
| *** prateek_ has joined #openstack-nova | 19:16 | |
| mriedem | right | 19:16 |
| mriedem | so default is api db | 19:16 |
| mriedem | but could be a separate optional placement DB with the API DB schema | 19:17 |
| mriedem | cdent: which are you going to do in your devstack change? | 19:17 |
| bauzas | cdent: don't reopen the wound :) | 19:17 |
| cdent | the new way is triggered by how config is set, so it would just be an initset, I don't really have an opinion | 19:17 |
| mriedem | i'd think we'd want to go with the default in newton of just use the api db | 19:17 |
| cdent | that's fine | 19:18 |
| mriedem | dansmith or jaypipes or alaski might have different ideas there | 19:18 |
| mriedem | my thinking is test with the default that people are going to be using most likely in newton since the placement api is still optional | 19:18 |
| bauzas | mriedem: so the idea we discussed at midcycle was to make optional a connection string allowing operators who would to run a separate api db sync so they could have a 3rd engine for the placement stuff | 19:19 |
| bauzas | (and later, delete the non-related tables) | 19:19 |
| *** parora has quit IRC | 19:19 | |
| alaski | because the new db is very optional in N I would have devstack not use it for now, then switch to it in O if we're ready to urge people to move over | 19:19 |
| bauzas | ++ | 19:20 |
| bauzas | the schemas will be the same, it's just for helping operators long-term | 19:20 |
| bauzas | there is no need to run it in devstack for Newton | 19:20 |
| * cdent alaski I feel in a rut | 19:22 | |
| cdent | a big rut | 19:22 |
| cdent | oh wel | 19:22 |
| cdent | l | 19:22 |
| *** shaohe_feng has quit IRC | 19:22 | |
| cdent | So what's the immediate plan: | 19:22 |
| cdent | fix the devstack thing to not use the placement db and fix the bashate | 19:22 |
| mriedem | cdent: yes | 19:23 |
| *** whenry has joined #openstack-nova | 19:23 | |
| mriedem | cdent: i'll work on a job for using the placement api | 19:23 |
| cdent | do the stuff described in: https://etherpad.openstack.org/p/placement-db | 19:23 |
| *** shaohe_feng has joined #openstack-nova | 19:23 | |
| bauzas | cdent: the above in the etherpad is totally necessary but unrelated to the RT | 19:24 |
| cdent | and jay continues working on the resource tracker stuff, but it needs to be resilient in the face of both the service catalog having no placement entry _and_ placement experiencing a partition | 19:24 |
| mriedem | i haven't read that, assuming it was a post-mortem from the placement db revert last week? | 19:24 |
| *** raunak has quit IRC | 19:24 | |
| cdent | bauzas: I'm trying to make an overview | 19:24 |
| mriedem | cdent: yes i think that's correct | 19:24 |
| cdent | because nobody seems to have a clear view | 19:24 |
| mriedem | jay still needs to get the inventory/allocation changes up on top of that RT series | 19:24 |
| mriedem | cdent: you work on the devstack change | 19:24 |
| bauzas | cdent: here, we need to allow #1 operators to be able to set their own placement DB if they wish, #2 map the liaison between aggregates and RPs | 19:24 |
| mriedem | i'll work on the job to use the placement api | 19:24 |
| mriedem | s/api/service/ | 19:25 |
| mriedem | then we can get the devstack change merged and run the job in the experimental queue on jay's changes (or any changes that use the placement service) | 19:25 |
| cdent | mriedem: there's still work to be done on the objects and and the api for allocations | 19:25 |
| *** su_zhang has joined #openstack-nova | 19:25 | |
| mriedem | cdent: ok, well those can be done in parallel right? | 19:25 |
| cdent | yes | 19:25 |
| mriedem | cool | 19:25 |
| cdent | just making sure there's awareness | 19:25 |
| mriedem | i can summarize next steps in the ML if that helps since we're all on disjointed schedules | 19:26 |
| cdent | that would be awesome, at least from my perspective | 19:27 |
| bauzas | so, just to make it clear, we're at 3 days from FF about to add a new devstack change and an experimental job in order to test our RT change that is still WIP ? :) | 19:27 |
| alaski | mriedem: a summary would be great. even when we all seem to be on the same page reality proves different | 19:27 |
| mriedem | bauzas: yeah | 19:27 |
| bauzas | awesome goal | 19:27 |
| bauzas | I'm on an EU timezone, so I'll try to be present as much as I could | 19:28 |
| bauzas | mriedem: cdent: just to clarify, the devstack change is focusing on exposing the new placement service type in Keystone, you agree ? | 19:29 |
| bauzas | nothing related to an optional placement DB | 19:29 |
| mriedem | correct, the devstack change will just use the API DB | 19:30 |
| bauzas | k, that's doable then | 19:30 |
| mriedem | in ocata we can make devstack start using a separate placement DB when we're ready to start migrating data | 19:30 |
| *** su_zhang has quit IRC | 19:31 | |
| bauzas | mriedem: cdent: so it would be a devstack flag that a specific job would turn on, right? | 19:31 |
| *** _ducttape_ has joined #openstack-nova | 19:31 | |
| bauzas | or even a new n-pl service... | 19:31 |
| cdent | bauzas: enable_service placement-api or something like that | 19:31 |
| mriedem | correct | 19:32 |
| mriedem | that's what i'll be working on | 19:32 |
| *** bvanhav has quit IRC | 19:32 | |
| *** shaohe_feng has quit IRC | 19:32 | |
| bauzas | and again, comes my question whether we should prefix it placement- :) | 19:32 |
| bauzas | I guess placement-api is the best idea, but I just think it signals it as not being nova :) | 19:33 |
| *** shaohe_feng has joined #openstack-nova | 19:33 | |
| cdent | well, honestly, bauzas, we should split all the pieces of nova out to something , not just placement, so from that perspective just splitting placement is limited, not a huge step. | 19:34 |
| *** ducttape_ has quit IRC | 19:34 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: New discover command to add new hosts to a cell https://review.openstack.org/328491 | 19:36 |
| mriedem | i don't want to rathole on if the devstack service is placement-api or nova-placement-api or whatever | 19:38 |
| mriedem | we're writing it to be split out, so it should be placement-api | 19:38 |
| *** fragatina has quit IRC | 19:39 | |
| *** fragatina has joined #openstack-nova | 19:39 | |
| cdent | cool, I'm going to go back to being on holiday, go to bed relatively early and be fresh faced and enthusiastic and get shti done in the morning. | 19:40 |
| * cdent waves | 19:40 | |
| *** cdent has quit IRC | 19:41 | |
| *** salv-orlando has joined #openstack-nova | 19:42 | |
| *** shaohe_feng has quit IRC | 19:42 | |
| *** shaohe_feng has joined #openstack-nova | 19:43 | |
| *** kfarr has joined #openstack-nova | 19:43 | |
| *** salv-orlando has quit IRC | 19:44 | |
| *** pradk has quit IRC | 19:46 | |
| *** thorst_ has joined #openstack-nova | 19:46 | |
| *** ijw has joined #openstack-nova | 19:47 | |
| *** sdake has quit IRC | 19:48 | |
| *** thorst has quit IRC | 19:48 | |
| *** annegentle has quit IRC | 19:50 | |
| *** zul has quit IRC | 19:51 | |
| *** annegentle has joined #openstack-nova | 19:51 | |
| *** aswadr_ has quit IRC | 19:51 | |
| *** markvoelker has joined #openstack-nova | 19:52 | |
| *** shaohe_feng has quit IRC | 19:53 | |
| *** shaohe_feng has joined #openstack-nova | 19:53 | |
| *** pradk has joined #openstack-nova | 19:59 | |
| *** shaohe_feng has quit IRC | 20:03 | |
| *** kaisers_ has joined #openstack-nova | 20:03 | |
| *** derekh has quit IRC | 20:03 | |
| *** tbachman has joined #openstack-nova | 20:03 | |
| *** raunak has joined #openstack-nova | 20:03 | |
| *** devkulkarni has quit IRC | 20:03 | |
| *** shaohe_feng has joined #openstack-nova | 20:04 | |
| *** gus has quit IRC | 20:05 | |
| *** darrenc has quit IRC | 20:05 | |
| *** jhesketh has quit IRC | 20:05 | |
| *** TravT has joined #openstack-nova | 20:05 | |
| *** TravT has quit IRC | 20:05 | |
| *** TravT has joined #openstack-nova | 20:06 | |
| *** darrenc has joined #openstack-nova | 20:07 | |
| *** markvoelker has quit IRC | 20:08 | |
| cmartin | Hey if anybody is available to help I have some questions about a bug I am working on and how to proceed | 20:09 |
| *** tbachman has quit IRC | 20:11 | |
| *** tbachman has joined #openstack-nova | 20:11 | |
| openstackgerrit | Breno Leitão proposed openstack/nova: libvirt: add hugepages support for Power https://review.openstack.org/303564 | 20:11 |
| *** lpetrut has joined #openstack-nova | 20:11 | |
| *** _ducttape_ has quit IRC | 20:12 | |
| *** jhesketh has joined #openstack-nova | 20:12 | |
| *** shaohe_feng has quit IRC | 20:13 | |
| *** ducttape_ has joined #openstack-nova | 20:13 | |
| *** shaohe_feng has joined #openstack-nova | 20:15 | |
| *** gus has joined #openstack-nova | 20:15 | |
| *** markvoelker has joined #openstack-nova | 20:17 | |
| *** tonytan_brb has joined #openstack-nova | 20:18 | |
| *** su_zhang has joined #openstack-nova | 20:19 | |
| *** Sukhdev has joined #openstack-nova | 20:20 | |
| *** tonytan4ever has quit IRC | 20:21 | |
| mriedem | sdague: almost seems like this devstack change for the placement service should live as a devstack plugin within nova, but throwing that into the mix would probably kill it from landing this week... | 20:22 |
| *** devkulkarni has joined #openstack-nova | 20:23 | |
| *** shaohe_feng has quit IRC | 20:23 | |
| *** shaohe_feng has joined #openstack-nova | 20:24 | |
| *** RuiChen has quit IRC | 20:25 | |
| *** devkulkarni has quit IRC | 20:26 | |
| *** TravT has quit IRC | 20:27 | |
| sdague | mriedem: yeh, so, here is what I propose, in the hopes that cdent might see this in the morning | 20:28 |
| sdague | cdent gets devstack change up for review | 20:29 |
| sdague | we use that to test the world with Depends-On, but we don't really land it | 20:29 |
| *** TravT has joined #openstack-nova | 20:29 | |
| sdague | that will let us local verify the new thing works, test the fail back works in the gate, and we can even do Depends-On to test the new thing in the gate | 20:30 |
| sdague | we can get ourselves an experimental job next week so we can touch the code going forward | 20:30 |
| sdague | and I think I can retool this patch to actually work with that devstack bit out there | 20:31 |
| mriedem | stating that in the ML thread will probably get more attention than irc scrollback if you can | 20:31 |
| mriedem | i went through his change, there are quite a few updates needed | 20:31 |
| sdague | yeh | 20:31 |
| sdague | mriedem: so... after we land this change, we are at the finish line? | 20:32 |
| mriedem | for now i'm going to just do a d-g change that depends on the devstack and nova changes | 20:32 |
| mriedem | which change? | 20:32 |
| mriedem | jay's RT change? | 20:32 |
| mriedem | honestly we should have landed the devstack thing and an experimental job months ago | 20:32 |
| mriedem | but nothing was using it until recently so i guess there wasn't pressure | 20:33 |
| openstackgerrit | Vu Tran proposed openstack/nova: Improve error log when snapshot fails https://review.openstack.org/356020 | 20:33 |
| sdague | 358797 | 20:34 |
| *** shaohe_feng has quit IRC | 20:34 | |
| sdague | mriedem: I also agree that the devstack infrastructure should have existed much earlier than freeze week | 20:34 |
| *** shaohe_feng has joined #openstack-nova | 20:34 | |
| sdague | mriedem: as I sort out my next couple of days, I'm trying to figure out where the finish like is with the placement API so that we can depend on it in ocata | 20:35 |
| *** Jeffrey4l_ has quit IRC | 20:35 | |
| mriedem | https://review.openstack.org/#/c/358797/ is only the beginning muwahaha | 20:36 |
| *** Jeffrey4l_ has joined #openstack-nova | 20:36 | |
| mriedem | seriously though, https://review.openstack.org/#/c/358797/ is the groundwork for the inventory/allocation records being written to the placement API from the RT | 20:36 |
| mriedem | then that's the end for newton | 20:36 |
| mriedem | but those inventory/allocatoin patches don't exist yet, and cdent said he still has to work on the placement API code for the allocation objects | 20:36 |
| sdague | right, I'm trying to understand the newton finish line | 20:36 |
| mriedem | it's the inventory/allocation records in the placement DB from the RT | 20:37 |
| sdague | and that if we accomplish it, we actually accomplished something useful for the release | 20:37 |
| sdague | because if the answer is even with 358797 it's all pointless, there are other things to spend time on this week | 20:37 |
| mriedem | agree | 20:37 |
| *** zzelle has joined #openstack-nova | 20:38 | |
| sdague | so, that's the part I don't have an answer to | 20:38 |
| mriedem | i could be missing something since jaypipes isn't around, that's why i put it in the ML to hopefully make sure that's correct | 20:38 |
| sdague | mriedem: ok, so you are handling the devstack patch? | 20:38 |
| sdague | (just getting to scrollback) | 20:38 |
| mriedem | i'm working the d-g/test/job change | 20:38 |
| dansmith | so, | 20:38 |
| dansmith | I think jay's rt patch is the finish line, | 20:39 |
| mriedem | sdague: cdent was going to cleanup the devstack change | 20:39 |
| dansmith | assuming that the devstack job shows that it's running and not throwing the warning about the placement service not being reachable | 20:39 |
| mriedem | dansmith: to just create the resource providers? | 20:39 |
| sdague | ok, so if I have a working devstack patch in the morning, I can rework 358797 to a landing state tomorrow I think | 20:39 |
| openstackgerrit | Andrew Laski proposed openstack/nova: New discover command to add new hosts to a cell https://review.openstack.org/328491 | 20:39 |
| dansmith | I didn't follow all the api stuff to know what state the server side is in, but I think it's supposed to be ready and I think that we'll confirm that with jay's patch running on top of the devstack support | 20:39 |
| *** Sukhdev_ has joined #openstack-nova | 20:40 | |
| dansmith | mriedem: no it creates them and updates the inventory and allocations right? | 20:40 |
| *** Sukhdev has quit IRC | 20:40 | |
| dansmith | oh yeah 797 does not do all that, just create them | 20:40 |
| dansmith | guh | 20:40 |
| mriedem | right | 20:40 |
| dansmith | I thought he had that work up, so maybe there's another patch somewhere? | 20:41 |
| sdague | where are these other patches? | 20:41 |
| mriedem | nope | 20:41 |
| mriedem | he hasn't written them yet | 20:41 |
| sdague | because I don't see any related changes | 20:41 |
| sdague | ..... o_O | 20:41 |
| *** rama_y has quit IRC | 20:41 | |
| * dansmith looks around | 20:41 | |
| mriedem | "Directly following that patch (coming today) is a patch that adds the inventory record setting to the report client. After the the allocation records are added. " | 20:41 |
| mriedem | is what jay told me | 20:41 |
| mriedem | that patch being 797 | 20:42 |
| mriedem | so he might have them local, i'm not sure | 20:42 |
| dansmith | okay, yeah, talking to him I also thought they were ready | 20:42 |
| sdague | even if not ready, at least something written | 20:42 |
| mriedem | he said he had it all in a giant patch and was working on splitting them up | 20:42 |
| mriedem | but is working from a starbucks in the UK while on vacation | 20:43 |
| mriedem | so...you know | 20:43 |
| dansmith | hmm. | 20:43 |
| sdague | I wish he would have pushed that then so others could hunt / peck here | 20:43 |
| dansmith | yeah | 20:43 |
| dansmith | maybe we can get him to do that next time he pops up | 20:43 |
| mriedem | i'll reply to his email with me saying the same | 20:43 |
| *** shaohe_feng has quit IRC | 20:44 | |
| *** vladikr has quit IRC | 20:44 | |
| jroll | jay's a bfd now, email is probably lower latency to reach him than irc anyway | 20:46 |
| JayF | Thanks for that, jroll, I feel extra special now <.< >.> | 20:46 |
| dansmith | jroll: he was recently promoted out of BFD to GFD as I understand it | 20:46 |
| jroll | ha! | 20:46 |
| jroll | dansmith: orly, what's the G there, great? grand? gross? | 20:47 |
| dansmith | jroll: you choose :P | 20:47 |
| melwitt | I was guessing Giant | 20:47 |
| jroll | heh | 20:47 |
| dansmith | I meant giant yeah | 20:47 |
| jroll | yay for him | 20:47 |
| *** markvoelker has quit IRC | 20:48 | |
| jroll | I think that means you need to email his assistant | 20:48 |
| *** shaohe_feng has joined #openstack-nova | 20:49 | |
| *** Sukhdev_ has quit IRC | 20:49 | |
| mriedem | i emailed him yesterday about some of this and he got back to me so i think he'll pick it up | 20:49 |
| *** rama_y has joined #openstack-nova | 20:49 | |
| sdague | mriedem: ok, I piled on | 20:50 |
| *** salv-orlando has joined #openstack-nova | 20:51 | |
| mriedem | http://s2.quickmeme.com/img/0c/0c0fcf30234ad83af446aaa5f1484afa60a6dd9dac943371fb4c026c7802fb04.jpg | 20:52 |
| mriedem | great | 20:52 |
| *** sdake has joined #openstack-nova | 20:52 | |
| *** shaohe_feng has quit IRC | 20:54 | |
| mriedem | here is the d-g test change https://review.openstack.org/#/c/362441/ | 20:56 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Pull from cell0 and build_requests for instance list https://review.openstack.org/356138 | 20:57 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Clean up instance mappings, build requests on quota failure https://review.openstack.org/357396 | 20:57 |
| openstackgerrit | Andrew Laski proposed openstack/nova: DNM logs for debugging https://review.openstack.org/362442 | 20:57 |
| *** mnestratov has joined #openstack-nova | 20:58 | |
| *** su_zhang has quit IRC | 20:59 | |
| *** salv-orlando has quit IRC | 20:59 | |
| *** su_zhang has joined #openstack-nova | 20:59 | |
| *** shaohe_feng has joined #openstack-nova | 20:59 | |
| *** gouthamr has quit IRC | 21:02 | |
| *** cleong has quit IRC | 21:03 | |
| *** shaohe_feng has quit IRC | 21:04 | |
| dansmith | alaski: thinking about this: https://review.openstack.org/#/c/328491/6 | 21:05 |
| dansmith | alaski: are we sure that the scheduler won't pick a host with no hostmapping? If it did, it will schedule to that host, but fail to contact it via rpc | 21:05 |
| *** shaohe_feng has joined #openstack-nova | 21:06 | |
| alaski | dansmith: at this point the scheduler could pick a host with no mapping. but the conductor would fail the build because it wouldn't know where to send the request | 21:06 |
| alaski | or reschedule | 21:06 |
| dansmith | right | 21:06 |
| dansmith | so that kindof sucks | 21:06 |
| dansmith | might be worth a "known issue" reno for that in general | 21:06 |
| dansmith | because now people will have to do that any time they add a compute node, which is new behavior right? | 21:07 |
| alaski | yeah | 21:07 |
| alaski | I can flesh out the reno | 21:07 |
| dansmith | I think it would probably be a good idea to do that | 21:08 |
| bauzas | mriedem: in https://review.openstack.org/#/c/358797/5/releasenotes/notes/placement-config-section-59891ba38e0749e7.yaml you're saying "find the keysyone entry or bust", but do you agree it's fine for operators that they could upgrade to Newton without first adding the service entry and just add it later ? | 21:10 |
| *** su_zhang has quit IRC | 21:10 | |
| bauzas | at least before upgrading to Ocata of course | 21:10 |
| *** su_zhang has joined #openstack-nova | 21:10 | |
| bauzas | mriedem: the "or bust" thing made me think you want the scheduler client to just complain | 21:11 |
| *** vishwanathj has quit IRC | 21:11 | |
| dansmith | bauzas: I think he means "or not report and warn" | 21:11 |
| bauzas | okay | 21:11 |
| dansmith | I don't think we should add another config variable just to bridge us to the keystone catalog, if at all possible | 21:11 |
| *** vishwanathj has joined #openstack-nova | 21:11 | |
| dansmith | the more of those we have, the more we have to clean up | 21:11 |
| *** daemontool has quit IRC | 21:11 | |
| mriedem | bauzas: correct | 21:12 |
| mriedem | bauzas: 'or bust' meaning we use the service catalog, no override config options | 21:12 |
| mriedem | not, break all of nova :) | 21:12 |
| mriedem | errr what dan said above | 21:13 |
| bauzas | coolness then, clarifying your point | 21:13 |
| *** thorst_ has quit IRC | 21:13 | |
| *** daemontool has joined #openstack-nova | 21:13 | |
| *** thorst has joined #openstack-nova | 21:14 | |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Add a new release note https://review.openstack.org/362448 | 21:15 |
| *** shaohe_feng has quit IRC | 21:15 | |
| *** shaohe_feng has joined #openstack-nova | 21:15 | |
| *** salv-orlando has joined #openstack-nova | 21:16 | |
| *** thorst has quit IRC | 21:18 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: WIP conf opt: added check for help msg line lenght https://review.openstack.org/362454 | 21:21 |
| pumaranikar | mriedem, checking your comment on https://review.openstack.org/#/c/247176/ Not sure if any updates/changes required in current patch set | 21:22 |
| mriedem | pumaranikar: i think i was mostly talking to myself there | 21:23 |
| mriedem | pumaranikar: i can't think of reasons not to make that change, but i feel like someone else might | 21:24 |
| mriedem | someone like alaski | 21:24 |
| mriedem | he's always causing trouble | 21:24 |
| alaski | it's what I do | 21:25 |
| pumaranikar | mriedem, :) can you +2 then | 21:25 |
| *** shaohe_feng has quit IRC | 21:25 | |
| alaski | yeah, I didn't actually like that patch but I'm not going to try to stop it | 21:25 |
| openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/nova: Expand feature classification matrix with gate checks https://review.openstack.org/362456 | 21:26 |
| *** shaohe_feng has joined #openstack-nova | 21:26 | |
| mriedem | pumaranikar: generally i don't +2 something i'm not comfortable with | 21:26 |
| alaski | to me the updated_at is a db field and should be updated when the db model changes. it being exposed in the api seems poorly thought out for exactly the reasons the patch is trying to address | 21:26 |
| mriedem | yeah we had this same discussion with the instance actions / events stuff | 21:27 |
| mriedem | for paging | 21:27 |
| mriedem | and sorting | 21:27 |
| mriedem | it was a little different though | 21:27 |
| alaski | a little bit, but same sort of thing | 21:28 |
| alaski | pumaranikar: the use case for this is that a volume operation should mean the instance shows up in a changes-since query? | 21:29 |
| mriedem | right attach/detach volume would change the instance updated_at | 21:30 |
| *** rlrossit has quit IRC | 21:30 | |
| alaski | I'm curious why that's important | 21:30 |
| alaski | unless it's for API sorting | 21:31 |
| alaski | or filtering | 21:31 |
| mriedem | looking at the default sort keys for listing instances, | 21:31 |
| mriedem | https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L2389 | 21:31 |
| mriedem | updated_at isn't in there anyway | 21:31 |
| alaski | oh right, it's created_at | 21:32 |
| pumaranikar | alaski, bug description says updated_at value of each instance need to be changed for any operation performed on the vm | 21:32 |
| alaski | pumaranikar: right, but why should updated_at be updated? | 21:32 |
| *** sdake has quit IRC | 21:33 | |
| mriedem | because the bug says it should duh :) | 21:33 |
| *** swatson has joined #openstack-nova | 21:33 | |
| alaski | oh, now I see :) | 21:33 |
| swatson | Hey all, I heard Nova recently had a few config option changes. I was wondering if anyone here had experience with those. I have some questions regarding deprecation | 21:33 |
| *** eharney has quit IRC | 21:34 | |
| mriedem | alaski: there is also some discussion with ndipanov and russellb in the original version of this https://review.openstack.org/#/c/55824 | 21:34 |
| pumaranikar | alaski, about accessing db fields directly, am not sure why we have _instance_update() in compute manager. I am using the same | 21:34 |
| *** cdent has joined #openstack-nova | 21:34 | |
| mriedem | really, this bug has been around since 2013 | 21:35 |
| mriedem | i don't think anyone is dying by not having it in the api | 21:35 |
| alaski | pumaranikar: I think this proposal needs some more discussion about the scope of when updated_at should be changed. Should it be changed for network attach/detach events as well? | 21:35 |
| *** shaohe_feng has quit IRC | 21:35 | |
| mriedem | i asked about that also | 21:35 |
| mriedem | we should be consistent | 21:35 |
| mriedem | and this isn't | 21:35 |
| alaski | yeah | 21:35 |
| pumaranikar | alaski, I can work on that as well. | 21:36 |
| *** shaohe_feng has joined #openstack-nova | 21:37 | |
| alaski | pumaranikar: in my opinion I don't see this as a bug, it's a misunderstanding of what updated_at means. However I could see changing updated_at to work from an API perspective, e.g. attach/detach events update the instance, but I see that as a feature change | 21:38 |
| cdent | sdague, mriedem: did another log run through, only things I can comment on with any clarity is that the devstack code for placement has been in existence since early july and I've had wip code for all the endpoints of the placement api since quite a while before that, but the aggregates and allocations apis have changes a few times. I had hoped to do the placement api stuff as a devstack plugin but I didn't because | 21:38 |
| cdent | optional wasn't an option...until today | 21:38 |
| *** rtheis_ has joined #openstack-nova | 21:38 | |
| alaski | pumaranikar: which means this would need a blueprint/spec and would need to wait until O. If we're going to change this let's do it properly | 21:39 |
| cdent | but yeah, I'll make it go in its rudimentary form in the morning | 21:39 |
| mriedem | cdent: yeah, like i said, things weren't using it until recently either | 21:39 |
| mriedem | so it wasn't on my radar | 21:39 |
| cdent | I think we should use this entire process as a reference point for a "how to do things better next time" retrospective or something. | 21:40 |
| *** rtheis has quit IRC | 21:40 | |
| cdent | Because I think there's been a lot of misspent energy | 21:41 |
| pumaranikar | alaski, ok. Can you please add comment under review as well. https://review.openstack.org/#/c/247176/ | 21:41 |
| alaski | pumaranikar: done | 21:41 |
| pumaranikar | alaski, thanks. | 21:42 |
| *** ijw has quit IRC | 21:43 | |
| mriedem | cdent: i don't disagree. more regular f2f calls or something might have been better for this so more people were on the same page, but we had a lot going on this cycle and there's only so much time and attention for everything. | 21:43 |
| pumaranikar | alaski, I will connect with you regarding how this change request can be addressed. | 21:44 |
| alaski | okay | 21:44 |
| mriedem | having said that, there has been quite a bit of progress made if you consider the placement api is a thing in just a single cycle | 21:44 |
| pumaranikar | mriedem, thanks for your comment | 21:44 |
| *** mriedem is now known as mriedem_afk | 21:44 | |
| cdent | mriedem: yeah (on both your statements). I'm only bringing it up because I think it is a general conversation we have to have sooner than later and having something concrete on which to base observations, rather than speculation, will probably help keep things from floating off into uselessness. | 21:45 |
| *** shaohe_feng has quit IRC | 21:45 | |
| cdent | A first thing we need to do is figure out a sane way to make "there's only so much time and attention for everything" less of an issue. | 21:45 |
| * cdent hasn't got any ideas, yet | 21:46 | |
| *** tblakes has quit IRC | 21:46 | |
| *** shaohe_feng has joined #openstack-nova | 21:46 | |
| * cdent goes to bed | 21:46 | |
| cdent | good night | 21:46 |
| *** cdent has quit IRC | 21:46 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: New discover command to add new hosts to a cell https://review.openstack.org/328491 | 21:49 |
| *** BjoernT has quit IRC | 21:52 | |
| *** zzelle has quit IRC | 21:53 | |
| *** shaohe_feng has quit IRC | 21:56 | |
| *** lucas-dinner has quit IRC | 21:56 | |
| *** burt has quit IRC | 21:56 | |
| *** shaohe_feng has joined #openstack-nova | 21:57 | |
| *** sai has quit IRC | 21:58 | |
| *** kevinz has joined #openstack-nova | 21:59 | |
| *** devananda has quit IRC | 21:59 | |
| *** sai has joined #openstack-nova | 21:59 | |
| *** lucasagomes has joined #openstack-nova | 21:59 | |
| *** devananda has joined #openstack-nova | 22:00 | |
| *** kfarr has quit IRC | 22:00 | |
| *** devananda has quit IRC | 22:01 | |
| dansmith | alaski: cool, looks nicer :) | 22:02 |
| *** devananda has joined #openstack-nova | 22:02 | |
| *** kevinz has quit IRC | 22:04 | |
| *** TravT_ has joined #openstack-nova | 22:04 | |
| *** annegentle has quit IRC | 22:05 | |
| *** shaohe_feng has quit IRC | 22:06 | |
| *** shaohe_feng has joined #openstack-nova | 22:06 | |
| *** ayoung has quit IRC | 22:07 | |
| *** TravT has quit IRC | 22:07 | |
| *** marst has quit IRC | 22:07 | |
| *** jamesdenton has quit IRC | 22:08 | |
| *** mnestratov has quit IRC | 22:13 | |
| *** efried has quit IRC | 22:16 | |
| *** shaohe_feng has quit IRC | 22:16 | |
| *** mdrabe has quit IRC | 22:16 | |
| *** chrichip has joined #openstack-nova | 22:18 | |
| *** marst has joined #openstack-nova | 22:18 | |
| *** ihrachys has quit IRC | 22:20 | |
| *** shaohe_feng has joined #openstack-nova | 22:20 | |
| *** ijw has joined #openstack-nova | 22:22 | |
| *** bdeschenes has quit IRC | 22:23 | |
| *** ijw has quit IRC | 22:23 | |
| *** chrichip has quit IRC | 22:25 | |
| *** shaohe_feng has quit IRC | 22:26 | |
| *** devkulkarni has joined #openstack-nova | 22:26 | |
| *** rtheis_ has quit IRC | 22:28 | |
| *** devkulkarni has quit IRC | 22:28 | |
| *** shaohe_feng has joined #openstack-nova | 22:29 | |
| *** ducttape_ has quit IRC | 22:33 | |
| *** devkulkarni has joined #openstack-nova | 22:34 | |
| *** thorst has joined #openstack-nova | 22:34 | |
| *** fragatin_ has joined #openstack-nova | 22:35 | |
| *** yamahata has quit IRC | 22:36 | |
| *** shaohe_feng has quit IRC | 22:37 | |
| *** shaohe_feng has joined #openstack-nova | 22:37 | |
| *** fragatina has quit IRC | 22:38 | |
| *** devkulkarni1 has joined #openstack-nova | 22:41 | |
| *** xyang1 has quit IRC | 22:41 | |
| *** mlavalle has quit IRC | 22:42 | |
| *** browne has quit IRC | 22:43 | |
| *** devkulkarni has quit IRC | 22:44 | |
| *** shaohe_feng has quit IRC | 22:47 | |
| *** shaohe_feng has joined #openstack-nova | 22:47 | |
| *** daemontool has quit IRC | 22:49 | |
| *** daemontool has joined #openstack-nova | 22:50 | |
| *** pwnall has quit IRC | 22:51 | |
| *** sdague has quit IRC | 22:55 | |
| *** yamahata has joined #openstack-nova | 22:55 | |
| *** shaohe_feng has quit IRC | 22:57 | |
| *** shaohe_feng has joined #openstack-nova | 22:59 | |
| *** trisq has joined #openstack-nova | 23:00 | |
| *** markvoelker has joined #openstack-nova | 23:02 | |
| *** dimtruck is now known as zz_dimtruck | 23:02 | |
| *** mmedvede has quit IRC | 23:02 | |
| *** annegentle has joined #openstack-nova | 23:04 | |
| *** lpetrut has quit IRC | 23:04 | |
| alaski | dansmith: yep. thanks for pointing that out | 23:06 |
| dansmith | np | 23:06 |
| *** mmedvede has joined #openstack-nova | 23:06 | |
| *** gouthamr has joined #openstack-nova | 23:07 | |
| *** annegentle has quit IRC | 23:07 | |
| *** shaohe_feng has quit IRC | 23:07 | |
| *** bdeschenes has joined #openstack-nova | 23:10 | |
| *** shaohe_feng has joined #openstack-nova | 23:10 | |
| *** ducttape_ has joined #openstack-nova | 23:11 | |
| *** tri2sing has quit IRC | 23:12 | |
| *** praneshp has quit IRC | 23:14 | |
| *** pradk has quit IRC | 23:17 | |
| *** shaohe_feng has quit IRC | 23:18 | |
| *** vishwanathj has quit IRC | 23:18 | |
| *** vishwanathj has joined #openstack-nova | 23:18 | |
| *** ayoung has joined #openstack-nova | 23:20 | |
| *** shaohe_feng has joined #openstack-nova | 23:21 | |
| *** devkulkarni1 has quit IRC | 23:23 | |
| *** rkrum has joined #openstack-nova | 23:24 | |
| *** salv-orlando has quit IRC | 23:25 | |
| *** trisq has quit IRC | 23:26 | |
| *** shaohe_feng has quit IRC | 23:28 | |
| *** rama_y has quit IRC | 23:28 | |
| *** vishwanathj has quit IRC | 23:29 | |
| *** shaohe_feng has joined #openstack-nova | 23:29 | |
| *** zul has joined #openstack-nova | 23:29 | |
| *** vishwanathj has joined #openstack-nova | 23:29 | |
| openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/nova: Expand feature matrix with server admin actions https://review.openstack.org/362408 | 23:30 |
| *** vishwanathj has quit IRC | 23:31 | |
| *** browne has joined #openstack-nova | 23:31 | |
| *** vishwanathj has joined #openstack-nova | 23:31 | |
| *** vishwanathj has quit IRC | 23:37 | |
| *** vishwanathj has joined #openstack-nova | 23:37 | |
| *** praneshp has joined #openstack-nova | 23:38 | |
| *** shaohe_feng has quit IRC | 23:38 | |
| *** rrecio has quit IRC | 23:39 | |
| *** markstur has joined #openstack-nova | 23:40 | |
| *** shaohe_feng has joined #openstack-nova | 23:42 | |
| markstur | Anyone out there happen to know what happens if you mix 2 different cpu_allocation_ratio settings on one hypervisor? | 23:45 |
| markstur | I think you can do this with overlapping host aggregates. | 23:45 |
| markstur | And then I think the math would not work well. | 23:45 |
| markstur | You could nearly fill up with 1:1 ratio vCPUs and then add a whole bunch of 16:1 vCPUs because I think the multiplier is only used at placement and not reflected in the vcpus_used. | 23:45 |
| markstur | Please holler if I read something wrong. | 23:45 |
| *** TravT_ has quit IRC | 23:46 | |
| *** vishwanathj has quit IRC | 23:47 | |
| *** vishwanathj has joined #openstack-nova | 23:47 | |
| *** shaohe_feng has quit IRC | 23:48 | |
| *** shaohe_feng has joined #openstack-nova | 23:50 | |
| *** ducttape_ has quit IRC | 23:54 | |
| *** markvoelker has quit IRC | 23:58 | |
| *** anush__ has quit IRC | 23:58 | |
| *** shaohe_feng has quit IRC | 23:59 | |
| *** shaohe_feng has joined #openstack-nova | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
