| *** clayton has joined #openstack-nova | 00:00 | |
| *** zz_dimtruck is now known as dimtruck | 00:02 | |
| *** yamahata has quit IRC | 00:05 | |
| *** shaohe_feng has quit IRC | 00:07 | |
| *** shaohe_feng has joined #openstack-nova | 00:08 | |
| *** chip_ has joined #openstack-nova | 00:10 | |
| *** ctrath has quit IRC | 00:10 | |
| *** shaohe_feng has quit IRC | 00:17 | |
| *** shaohe_feng has joined #openstack-nova | 00:18 | |
| *** dimtruck is now known as zz_dimtruck | 00:21 | |
| *** takashin has joined #openstack-nova | 00:23 | |
| *** ijw has quit IRC | 00:27 | |
| *** shaohe_feng has quit IRC | 00:28 | |
| *** ijw has joined #openstack-nova | 00:28 | |
| *** shaohe_feng has joined #openstack-nova | 00:29 | |
| *** fragatin_ has joined #openstack-nova | 00:30 | |
| *** hoangcx has joined #openstack-nova | 00:31 | |
| *** fragatina has quit IRC | 00:32 | |
| *** fragatin_ has quit IRC | 00:35 | |
| *** shaohe_feng has quit IRC | 00:38 | |
| *** shaohe_feng has joined #openstack-nova | 00:39 | |
| *** jamesdenton has joined #openstack-nova | 00:43 | |
| *** ircuser-1 has joined #openstack-nova | 00:45 | |
| *** shaohe_feng has quit IRC | 00:48 | |
| *** shaohe_feng has joined #openstack-nova | 00:49 | |
| *** rama_y has joined #openstack-nova | 00:52 | |
| *** salv-orlando has quit IRC | 00:54 | |
| *** salv-orlando has joined #openstack-nova | 00:54 | |
| *** ijw has quit IRC | 00:55 | |
| *** thorst has joined #openstack-nova | 00:55 | |
| *** thorst has quit IRC | 00:56 | |
| *** cheneydc has joined #openstack-nova | 00:57 | |
| *** boomsauce has quit IRC | 00:57 | |
| *** ljxiash has joined #openstack-nova | 00:58 | |
| *** shaohe_feng has quit IRC | 00:58 | |
| *** shaohe_feng has joined #openstack-nova | 00:59 | |
| *** ljxiash has quit IRC | 01:03 | |
| *** raunak has joined #openstack-nova | 01:05 | |
| *** ijw has joined #openstack-nova | 01:06 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove policy checkpoints for SecurityGroupAPI and NetworkAPI https://review.openstack.org/325641 | 01:07 |
|---|---|---|
| openstackgerrit | Alex Xu proposed openstack/nova: Remove policy checkpoints from ComputeAPI https://review.openstack.org/320752 | 01:07 |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove the skip_policy_check flags https://review.openstack.org/325684 | 01:07 |
| *** shaohe_feng has quit IRC | 01:09 | |
| *** shaohe_feng has joined #openstack-nova | 01:10 | |
| *** jamesdenton has quit IRC | 01:16 | |
| *** ijw has quit IRC | 01:16 | |
| *** jamesdenton has joined #openstack-nova | 01:17 | |
| *** shaohe_feng has quit IRC | 01:19 | |
| *** ljxiash has joined #openstack-nova | 01:19 | |
| *** shaohe_feng has joined #openstack-nova | 01:20 | |
| *** chip_ has quit IRC | 01:21 | |
| *** chip_ has joined #openstack-nova | 01:23 | |
| *** EinstCrazy has joined #openstack-nova | 01:25 | |
| *** shaohe_feng has quit IRC | 01:29 | |
| *** ducnc has joined #openstack-nova | 01:29 | |
| *** EinstCrazy has quit IRC | 01:29 | |
| *** shaohe_feng has joined #openstack-nova | 01:30 | |
| *** EinstCrazy has joined #openstack-nova | 01:30 | |
| *** ljxiash has quit IRC | 01:37 | |
| *** mriedem has quit IRC | 01:38 | |
| *** shaohe_feng has quit IRC | 01:39 | |
| *** shaohe_feng has joined #openstack-nova | 01:40 | |
| *** Kevin_Zheng has joined #openstack-nova | 01:43 | |
| *** ljxiash has joined #openstack-nova | 01:46 | |
| *** shaohe_feng has quit IRC | 01:50 | |
| *** shaohe_feng has joined #openstack-nova | 01:50 | |
| *** yamahata has joined #openstack-nova | 01:51 | |
| *** jamesden_ has joined #openstack-nova | 01:52 | |
| *** jamesdenton has quit IRC | 01:55 | |
| *** thorst has joined #openstack-nova | 01:56 | |
| *** jamesden_ has quit IRC | 01:57 | |
| *** thorst has quit IRC | 01:59 | |
| *** sacharya has joined #openstack-nova | 01:59 | |
| *** thorst has joined #openstack-nova | 01:59 | |
| *** zz_dimtruck is now known as dimtruck | 01:59 | |
| *** jamesdenton has joined #openstack-nova | 01:59 | |
| *** thorst has quit IRC | 02:00 | |
| *** shaohe_feng has quit IRC | 02:00 | |
| *** thorst has joined #openstack-nova | 02:00 | |
| *** shaohe_feng has joined #openstack-nova | 02:01 | |
| *** jamesdenton has quit IRC | 02:04 | |
| *** Daisy has joined #openstack-nova | 02:04 | |
| *** thorst has quit IRC | 02:05 | |
| *** shaohe_feng has quit IRC | 02:10 | |
| *** shaohe_feng has joined #openstack-nova | 02:11 | |
| *** Daisy has quit IRC | 02:14 | |
| *** chip_ has quit IRC | 02:14 | |
| *** Daisy has joined #openstack-nova | 02:14 | |
| *** Daisy_ has joined #openstack-nova | 02:16 | |
| *** Daisy has quit IRC | 02:16 | |
| *** sacharya has quit IRC | 02:19 | |
| *** vishwanathj has joined #openstack-nova | 02:20 | |
| *** shaohe_feng has quit IRC | 02:20 | |
| *** shaohe_feng has joined #openstack-nova | 02:21 | |
| *** Jeffrey4l_ has joined #openstack-nova | 02:24 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add check of capabilities for detach interface https://review.openstack.org/308115 | 02:27 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add support matrix for attach and detach interfaces https://review.openstack.org/312801 | 02:27 |
| openstackgerrit | Chen Fan proposed openstack/nova: trivial: remove no more usage local variables https://review.openstack.org/325702 | 02:30 |
| *** shaohe_feng has quit IRC | 02:31 | |
| *** shaohe_feng has joined #openstack-nova | 02:31 | |
| *** chip_ has joined #openstack-nova | 02:31 | |
| *** raunak has quit IRC | 02:37 | |
| *** ctrath has joined #openstack-nova | 02:38 | |
| *** gmann has quit IRC | 02:38 | |
| *** Daisy_ has quit IRC | 02:41 | |
| *** shaohe_feng has quit IRC | 02:41 | |
| *** Daisy has joined #openstack-nova | 02:41 | |
| *** zhangjl has joined #openstack-nova | 02:41 | |
| *** zhangjl has left #openstack-nova | 02:42 | |
| *** ozialien10 has quit IRC | 02:42 | |
| *** gmann has joined #openstack-nova | 02:42 | |
| *** shaohe_feng has joined #openstack-nova | 02:42 | |
| *** chip_ has quit IRC | 02:42 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add support matrix for attach and detach interfaces https://review.openstack.org/312801 | 02:44 |
| *** chip_ has joined #openstack-nova | 02:44 | |
| *** gcb has joined #openstack-nova | 02:45 | |
| *** Daisy_ has joined #openstack-nova | 02:47 | |
| *** Daisy has quit IRC | 02:48 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: parameter verification for os-aggregates https://review.openstack.org/315517 | 02:50 |
| *** shaohe_feng has quit IRC | 02:51 | |
| *** shaohe_feng has joined #openstack-nova | 02:52 | |
| *** salv-orl_ has joined #openstack-nova | 03:00 | |
| *** shaohe_feng has quit IRC | 03:01 | |
| *** Daisy_ has quit IRC | 03:02 | |
| *** Daisy has joined #openstack-nova | 03:02 | |
| *** shaohe_feng has joined #openstack-nova | 03:02 | |
| *** thorst has joined #openstack-nova | 03:03 | |
| *** salv-orlando has quit IRC | 03:03 | |
| *** gcb has quit IRC | 03:03 | |
| *** gcb has joined #openstack-nova | 03:03 | |
| *** Daisy_ has joined #openstack-nova | 03:04 | |
| *** mmandel has joined #openstack-nova | 03:05 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Fix white spaces in api-ref https://review.openstack.org/325705 | 03:07 |
| *** Daisy has quit IRC | 03:08 | |
| *** sheel has joined #openstack-nova | 03:09 | |
| *** thorst has quit IRC | 03:10 | |
| *** shaohe_feng has quit IRC | 03:12 | |
| *** shaohe_feng has joined #openstack-nova | 03:12 | |
| *** gcb has quit IRC | 03:15 | |
| *** sacharya has joined #openstack-nova | 03:16 | |
| *** mmandel has quit IRC | 03:21 | |
| *** shaohe_feng has quit IRC | 03:22 | |
| *** shaohe_feng has joined #openstack-nova | 03:27 | |
| *** chip_ has quit IRC | 03:28 | |
| *** gcb has joined #openstack-nova | 03:30 | |
| *** ljxiash has quit IRC | 03:32 | |
| *** shaohe_feng has quit IRC | 03:32 | |
| *** shaohe_feng has joined #openstack-nova | 03:33 | |
| *** raunak has joined #openstack-nova | 03:35 | |
| *** Daisy_ has quit IRC | 03:36 | |
| openstackgerrit | Yang Shengming proposed openstack/nova: prevent resize down with ceph backend in libvirt driver https://review.openstack.org/325708 | 03:37 |
| *** Daisy has joined #openstack-nova | 03:37 | |
| *** dave-mccowan has quit IRC | 03:41 | |
| *** Daisy has quit IRC | 03:41 | |
| *** shaohe_feng has quit IRC | 03:42 | |
| *** shaohe_feng has joined #openstack-nova | 03:43 | |
| *** raunak has quit IRC | 03:44 | |
| *** zerda2 has joined #openstack-nova | 03:46 | |
| *** amotoki has joined #openstack-nova | 03:47 | |
| *** mmandel has joined #openstack-nova | 03:50 | |
| *** ljxiash has joined #openstack-nova | 03:52 | |
| *** shaohe_feng has quit IRC | 03:53 | |
| *** shaohe_feng has joined #openstack-nova | 03:54 | |
| *** mmandel has quit IRC | 03:55 | |
| *** ljxiash has quit IRC | 03:57 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Add swap volume notifications (start, end) https://review.openstack.org/250283 | 03:57 |
| *** shaohe_feng has quit IRC | 04:03 | |
| *** shaohe_feng has joined #openstack-nova | 04:04 | |
| *** amotoki has quit IRC | 04:08 | |
| *** fragatina has joined #openstack-nova | 04:08 | |
| *** fragatina has quit IRC | 04:09 | |
| *** thorst has joined #openstack-nova | 04:09 | |
| *** fragatina has joined #openstack-nova | 04:09 | |
| *** shaohe_feng has quit IRC | 04:13 | |
| *** shaohe_feng has joined #openstack-nova | 04:14 | |
| *** raunak has joined #openstack-nova | 04:14 | |
| *** thorst has quit IRC | 04:16 | |
| *** sree has joined #openstack-nova | 04:16 | |
| *** dimtruck is now known as zz_dimtruck | 04:19 | |
| *** amotoki has joined #openstack-nova | 04:22 | |
| *** shaohe_feng has quit IRC | 04:23 | |
| *** shaohe_feng has joined #openstack-nova | 04:25 | |
| *** vishwanathj has quit IRC | 04:28 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: parameter verification for os-aggregates https://review.openstack.org/315517 | 04:29 |
| *** vishwanathj has joined #openstack-nova | 04:30 | |
| *** amotoki has quit IRC | 04:30 | |
| *** shaohe_feng has quit IRC | 04:34 | |
| *** shaohe_feng has joined #openstack-nova | 04:35 | |
| *** penick has joined #openstack-nova | 04:37 | |
| *** amotoki has joined #openstack-nova | 04:37 | |
| *** jaosorior has joined #openstack-nova | 04:41 | |
| *** rama_y has quit IRC | 04:43 | |
| *** shaohe_feng has quit IRC | 04:44 | |
| *** shaohe_feng has joined #openstack-nova | 04:45 | |
| *** jwcroppe has joined #openstack-nova | 04:46 | |
| *** jwcroppe has quit IRC | 04:46 | |
| *** irenab has joined #openstack-nova | 04:46 | |
| *** jwcroppe has joined #openstack-nova | 04:46 | |
| *** Daisy has joined #openstack-nova | 04:48 | |
| *** penick has quit IRC | 04:48 | |
| *** irenab has quit IRC | 04:48 | |
| *** shaohe_feng has quit IRC | 04:54 | |
| *** shaohe_feng has joined #openstack-nova | 04:55 | |
| *** amotoki has quit IRC | 05:02 | |
| *** jwcroppe has quit IRC | 05:04 | |
| *** shaohe_feng has quit IRC | 05:04 | |
| *** shaohe_feng has joined #openstack-nova | 05:05 | |
| *** ctrath has quit IRC | 05:05 | |
| *** haukebruno has joined #openstack-nova | 05:08 | |
| *** jwcroppe has joined #openstack-nova | 05:10 | |
| *** pcaruana has quit IRC | 05:13 | |
| *** thorst has joined #openstack-nova | 05:14 | |
| *** shaohe_feng has quit IRC | 05:15 | |
| *** shaohe_feng has joined #openstack-nova | 05:16 | |
| *** ecelik has joined #openstack-nova | 05:17 | |
| *** thorst has quit IRC | 05:20 | |
| *** sudipto has joined #openstack-nova | 05:22 | |
| *** ljxiash has joined #openstack-nova | 05:23 | |
| *** shaohe_feng has quit IRC | 05:25 | |
| *** ljxiash has quit IRC | 05:27 | |
| *** ljxiash has joined #openstack-nova | 05:27 | |
| *** shaohe_feng has joined #openstack-nova | 05:28 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Add swap volume notifications (start, end) https://review.openstack.org/250283 | 05:29 |
| *** sudipto has quit IRC | 05:29 | |
| *** haukebruno has quit IRC | 05:31 | |
| *** haukebruno has joined #openstack-nova | 05:33 | |
| *** shaohe_feng has quit IRC | 05:35 | |
| *** shaohe_feng has joined #openstack-nova | 05:36 | |
| *** ratailor has joined #openstack-nova | 05:36 | |
| *** prateek has joined #openstack-nova | 05:38 | |
| openstackgerrit | Chen Fan proposed openstack/nova: fix a typo in comment https://review.openstack.org/325728 | 05:41 |
| *** shaohe_feng has quit IRC | 05:45 | |
| *** shaohe_feng has joined #openstack-nova | 05:46 | |
| *** shaohe_feng has quit IRC | 05:56 | |
| *** shaohe_feng has joined #openstack-nova | 05:57 | |
| *** josecastroleon has joined #openstack-nova | 05:57 | |
| *** itamarl has joined #openstack-nova | 05:58 | |
| *** openstackgerrit has quit IRC | 06:02 | |
| *** openstackgerrit has joined #openstack-nova | 06:03 | |
| *** shaohe_feng has quit IRC | 06:06 | |
| *** rdopiera_pto is now known as rdopiera | 06:06 | |
| *** shaohe_feng has joined #openstack-nova | 06:07 | |
| *** mjura has joined #openstack-nova | 06:11 | |
| *** sudipto has joined #openstack-nova | 06:11 | |
| *** gcb has quit IRC | 06:12 | |
| *** Daisy has quit IRC | 06:14 | |
| *** Daisy has joined #openstack-nova | 06:15 | |
| *** itamarl_ has joined #openstack-nova | 06:15 | |
| openstackgerrit | Li Yingjun proposed openstack/nova-specs: Flavor notifications spec https://review.openstack.org/321336 | 06:15 |
| *** Daisy_ has joined #openstack-nova | 06:15 | |
| *** shaohe_feng has quit IRC | 06:16 | |
| *** shaohe_feng has joined #openstack-nova | 06:17 | |
| *** itamarl has quit IRC | 06:17 | |
| *** itamarl_ is now known as itamarl | 06:17 | |
| *** thorst has joined #openstack-nova | 06:18 | |
| *** Daisy has quit IRC | 06:19 | |
| *** fishcried has joined #openstack-nova | 06:20 | |
| *** ljxiash has quit IRC | 06:24 | |
| *** gcb has joined #openstack-nova | 06:25 | |
| *** Daisy_ has quit IRC | 06:25 | |
| *** thorst has quit IRC | 06:26 | |
| *** Daisy has joined #openstack-nova | 06:26 | |
| *** shaohe_feng has quit IRC | 06:26 | |
| *** shaohe_feng has joined #openstack-nova | 06:27 | |
| *** andreas_s has joined #openstack-nova | 06:29 | |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: Introduce a CONF flag to determine glance client version https://review.openstack.org/315963 | 06:29 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Fix config option scheduler_default_filters' invalid override value https://review.openstack.org/325743 | 06:29 |
| *** sridharg has joined #openstack-nova | 06:30 | |
| *** dedery has joined #openstack-nova | 06:30 | |
| *** Daisy has quit IRC | 06:30 | |
| *** sacharya has quit IRC | 06:33 | |
| *** fishcried has quit IRC | 06:33 | |
| *** fishcried has joined #openstack-nova | 06:34 | |
| *** liverpooler has joined #openstack-nova | 06:34 | |
| *** Daisy has joined #openstack-nova | 06:34 | |
| *** ociuhandu has joined #openstack-nova | 06:35 | |
| *** ljxiash has joined #openstack-nova | 06:36 | |
| *** shaohe_feng has quit IRC | 06:37 | |
| *** shaohe_feng has joined #openstack-nova | 06:37 | |
| *** apoorv has joined #openstack-nova | 06:37 | |
| *** dedery_ has joined #openstack-nova | 06:39 | |
| *** dedery has quit IRC | 06:42 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: test: pass enable_pass as kwarg in test_evacuate https://review.openstack.org/325748 | 06:44 |
| *** raunak has quit IRC | 06:44 | |
| *** shaohe_feng has quit IRC | 06:47 | |
| *** shaohe_feng has joined #openstack-nova | 06:47 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Make sure config option public_interface with correct override value https://review.openstack.org/325749 | 06:51 |
| *** dedery_ has quit IRC | 06:52 | |
| *** dedery has joined #openstack-nova | 06:53 | |
| *** tesseract has joined #openstack-nova | 06:53 | |
| *** jwcroppe has quit IRC | 06:54 | |
| *** yamahata has quit IRC | 06:55 | |
| *** ratailor has quit IRC | 06:55 | |
| *** ratailor has joined #openstack-nova | 06:55 | |
| *** jwcroppe has joined #openstack-nova | 06:56 | |
| *** shaohe_feng has quit IRC | 06:57 | |
| *** shaohe_feng has joined #openstack-nova | 06:58 | |
| *** jwcroppe has quit IRC | 07:00 | |
| *** salv-orl_ has quit IRC | 07:03 | |
| *** tanee is now known as tanee_away | 07:03 | |
| *** salv-orlando has joined #openstack-nova | 07:03 | |
| *** tanee_away is now known as tanee | 07:04 | |
| *** tanee is now known as tanee_away | 07:04 | |
| *** tanee_away is now known as tanee | 07:04 | |
| *** tanee is now known as tanee_away | 07:05 | |
| *** tanee_away is now known as tanee | 07:06 | |
| tdurakov | hi folks, who could help me mergin 2 backports into stable branches? | 07:06 |
| *** shaohe_feng has quit IRC | 07:07 | |
| *** shaohe_feng has joined #openstack-nova | 07:08 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: test: pass enable_pass as kwarg in test_evacuate https://review.openstack.org/325748 | 07:10 |
| *** yamahata has joined #openstack-nova | 07:10 | |
| *** ratailor has quit IRC | 07:11 | |
| *** ratailor has joined #openstack-nova | 07:11 | |
| *** ratailor has quit IRC | 07:11 | |
| *** ratailor has joined #openstack-nova | 07:12 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Make sure config option public_interface with correct override value https://review.openstack.org/325749 | 07:12 |
| *** migi_ is now known as migi | 07:12 | |
| *** pcaruana has joined #openstack-nova | 07:14 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Remove config option config_drive_format's invalid value test https://review.openstack.org/325757 | 07:15 |
| *** ivase has quit IRC | 07:17 | |
| *** matrohon has joined #openstack-nova | 07:17 | |
| *** ivase has joined #openstack-nova | 07:18 | |
| *** shaohe_feng has quit IRC | 07:18 | |
| *** shaohe_feng has joined #openstack-nova | 07:18 | |
| *** fishcried1 has joined #openstack-nova | 07:22 | |
| *** jed56 has joined #openstack-nova | 07:22 | |
| *** mpavone has joined #openstack-nova | 07:22 | |
| *** fishcried has quit IRC | 07:23 | |
| *** fishcried1 is now known as fishcried | 07:23 | |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Stop using mox from api/openstack/fakes.py https://review.openstack.org/325759 | 07:23 |
| *** thorst has joined #openstack-nova | 07:25 | |
| *** jlanoux has joined #openstack-nova | 07:27 | |
| *** shaohe_feng has quit IRC | 07:28 | |
| *** fishcried has quit IRC | 07:28 | |
| *** shaohe_feng has joined #openstack-nova | 07:29 | |
| *** thorst has quit IRC | 07:31 | |
| *** vtech has joined #openstack-nova | 07:31 | |
| *** daemontool has joined #openstack-nova | 07:31 | |
| *** tesseract has quit IRC | 07:32 | |
| *** Daisy has quit IRC | 07:32 | |
| *** Daisy has joined #openstack-nova | 07:32 | |
| *** tesseract has joined #openstack-nova | 07:33 | |
| *** sacharya has joined #openstack-nova | 07:33 | |
| *** ishant|away is now known as ishant | 07:34 | |
| *** klkumar has joined #openstack-nova | 07:34 | |
| *** klkumar has quit IRC | 07:37 | |
| *** sacharya has quit IRC | 07:38 | |
| *** Daisy_ has joined #openstack-nova | 07:38 | |
| *** shaohe_feng has quit IRC | 07:38 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: libvirt: Fix wrong override value for config options virt_type and cpu_mode https://review.openstack.org/325761 | 07:39 |
| *** shaohe_feng has joined #openstack-nova | 07:39 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in nova/tests/unit/cells https://review.openstack.org/323287 | 07:39 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Return 400 when SecurityGroupCannotBeApplied is raised https://review.openstack.org/323332 | 07:40 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in fake_notifier https://review.openstack.org/323730 | 07:40 |
| *** ljxiash has quit IRC | 07:41 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove mox in test_block_device_mapping_v1.py https://review.openstack.org/308746 | 07:41 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in volume/encryptors/test_base.py https://review.openstack.org/319736 | 07:42 |
| *** Daisy has quit IRC | 07:42 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Set 'serial' to new volume ID in swap volumes https://review.openstack.org/257135 | 07:42 |
| *** roeyc has joined #openstack-nova | 07:42 | |
| *** ljxiash has joined #openstack-nova | 07:42 | |
| *** tesseract has quit IRC | 07:46 | |
| *** tesseract has joined #openstack-nova | 07:46 | |
| *** Daisy_ has quit IRC | 07:48 | |
| *** shaohe_feng has quit IRC | 07:48 | |
| *** Daisy has joined #openstack-nova | 07:49 | |
| *** shaohe_feng has joined #openstack-nova | 07:49 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: test: remove invalid test method in libvirt/test_imagebackend https://review.openstack.org/325765 | 07:51 |
| *** Oku_OS-away is now known as Oku_OS | 07:53 | |
| openstackgerrit | Abhijeet Malawade proposed openstack/nova: Return HTTPBadRequest if incorrect params are provided to images api https://review.openstack.org/314153 | 07:57 |
| *** shaohe_feng has quit IRC | 07:59 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: xenapi: Remove invalid values for config option image_compression_level https://review.openstack.org/325771 | 07:59 |
| *** shaohe_feng has joined #openstack-nova | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-nova | 08:00 | |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Transform instance.delete notifications https://review.openstack.org/313654 | 08:02 |
| *** Daisy_ has joined #openstack-nova | 08:04 | |
| *** danpb has joined #openstack-nova | 08:04 | |
| *** chenfan has joined #openstack-nova | 08:05 | |
| *** Oku_OS has quit IRC | 08:05 | |
| *** Oku_OS has joined #openstack-nova | 08:05 | |
| *** Daisy has quit IRC | 08:08 | |
| *** shaohe_feng has quit IRC | 08:09 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: libvirt: Fix wrong override value for config options virt_type and cpu_mode https://review.openstack.org/325761 | 08:09 |
| *** shaohe_feng has joined #openstack-nova | 08:10 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: test: remove invalid test method in libvirt/test_imagebackend https://review.openstack.org/325765 | 08:12 |
| *** jichen has joined #openstack-nova | 08:13 | |
| *** shaohe_feng has quit IRC | 08:19 | |
| *** paul-carlton2 has joined #openstack-nova | 08:20 | |
| *** shaohe_feng has joined #openstack-nova | 08:20 | |
| *** jaosorior has quit IRC | 08:23 | |
| *** jaosorior has joined #openstack-nova | 08:23 | |
| *** thorst has joined #openstack-nova | 08:29 | |
| *** shaohe_feng has quit IRC | 08:29 | |
| openstackgerrit | Rong Han proposed openstack/nova: Fix nova-compute start failed when reserved_huge_pages has value. https://review.openstack.org/324379 | 08:30 |
| *** shaohe_feng has joined #openstack-nova | 08:30 | |
| *** PaulMurray has joined #openstack-nova | 08:32 | |
| *** EinstCra_ has joined #openstack-nova | 08:33 | |
| *** EinstCrazy has quit IRC | 08:33 | |
| *** tinbar has joined #openstack-nova | 08:33 | |
| *** thorst has quit IRC | 08:35 | |
| *** yamahata has quit IRC | 08:37 | |
| *** shaohe_feng has quit IRC | 08:40 | |
| *** aloga has quit IRC | 08:40 | |
| *** shaohe_feng has joined #openstack-nova | 08:41 | |
| *** aloga has joined #openstack-nova | 08:41 | |
| *** divya___ has joined #openstack-nova | 08:42 | |
| *** apoorv has quit IRC | 08:42 | |
| divya___ | Failed to provision instance 4ad755ea-6f07-4554-9231-ebb3f371034d: Failed to deploy. Error: 400 Bad Request: Unknown scheme 'swift+config' found in URI (HTTP 400) | 08:43 |
| divya___ | i have copied logs @http://paste.openstack.org/show/508315/ | 08:43 |
| divya___ | can anyone help plz ? | 08:43 |
| *** lpetrut has joined #openstack-nova | 08:45 | |
| *** yassine has joined #openstack-nova | 08:45 | |
| *** zhs has joined #openstack-nova | 08:47 | |
| *** jaosorior is now known as jaosorior_lunch | 08:48 | |
| openstackgerrit | Ankit Agrawal proposed openstack/nova: Remove redundant migration status update https://review.openstack.org/247519 | 08:50 |
| *** shaohe_feng has quit IRC | 08:50 | |
| *** mgoddard has joined #openstack-nova | 08:50 | |
| *** jogo has quit IRC | 08:51 | |
| *** shaohe_feng has joined #openstack-nova | 08:51 | |
| *** apoorv has joined #openstack-nova | 08:51 | |
| openstackgerrit | Rong Han proposed openstack/nova: Fix Residual resources for deleted instances after migrate/resize error https://review.openstack.org/325635 | 08:52 |
| *** mvk has joined #openstack-nova | 08:54 | |
| sudipto | divya___, seems like the node you trying to provision via ironic did not respond back in due time... | 08:56 |
| divya___ | how can i chk it | 08:58 |
| divya___ | how can we resolve it | 08:58 |
| *** shaohe_feng has quit IRC | 09:00 | |
| *** mnestratov has joined #openstack-nova | 09:00 | |
| *** cdent has joined #openstack-nova | 09:00 | |
| divya___ | sudipto how can we resolve it | 09:01 |
| *** shaohe_feng has joined #openstack-nova | 09:01 | |
| *** mgoddard has quit IRC | 09:01 | |
| *** ihrachys has joined #openstack-nova | 09:02 | |
| *** jogo has joined #openstack-nova | 09:04 | |
| gibi | johnthetubaguy: hi! could you mark the bp approved https://blueprints.launchpad.net/nova/+spec/versioned-notification-transformation-newton as the relate spec was approved? | 09:10 |
| *** shaohe_feng has quit IRC | 09:10 | |
| johnthetubaguy | gibi: yeah, let me sync those up | 09:10 |
| *** shaohe_feng has joined #openstack-nova | 09:11 | |
| gibi | johnthetubaguy: thanks! | 09:11 |
| *** hshiina has joined #openstack-nova | 09:13 | |
| *** kaisers1 has joined #openstack-nova | 09:17 | |
| *** dedery has quit IRC | 09:18 | |
| *** sahid has joined #openstack-nova | 09:19 | |
| *** sacharya has joined #openstack-nova | 09:19 | |
| *** moshele has joined #openstack-nova | 09:19 | |
| *** shaohe_feng has quit IRC | 09:21 | |
| *** shaohe_feng has joined #openstack-nova | 09:22 | |
| *** permalac has joined #openstack-nova | 09:22 | |
| *** sacharya has quit IRC | 09:23 | |
| ratailor | johnthetubaguy, could you please review https://review.openstack.org/#/c/294491/ | 09:31 |
| *** shaohe_feng has quit IRC | 09:31 | |
| *** shaohe_feng has joined #openstack-nova | 09:32 | |
| *** gszasz has joined #openstack-nova | 09:32 | |
| johnthetubaguy | ratailor: I am just not sure thats the best way to fix that issue. We have too many periodic tasks. What about checking those things on a SIG_HUP instead (normally triggers a config reload) | 09:32 |
| *** thorst has joined #openstack-nova | 09:33 | |
| ratailor | johnthetubaguy, How can we handle this using SIG_HUP, could you please give a hint, if time permits ? | 09:35 |
| ratailor | johnthetubaguy, IMO normally SIG_HUP is used when we want to use new values for config parameters without restarting the service. right ? | 09:36 |
| *** sambetts|afk is now known as sambetts | 09:36 | |
| *** thorst has quit IRC | 09:40 | |
| *** shaohe_feng has quit IRC | 09:41 | |
| *** shaohe_feng has joined #openstack-nova | 09:42 | |
| *** Daisy_ has quit IRC | 09:47 | |
| johnthetubaguy | ratailor: today thats all we do, I believe the compute manager could do extra things on SIG HUP, its just we would need to wire that up | 09:47 |
| *** Daisy has joined #openstack-nova | 09:48 | |
| *** sree has quit IRC | 09:49 | |
| *** shaohe_feng has quit IRC | 09:51 | |
| *** shaohe_feng has joined #openstack-nova | 09:52 | |
| *** jaosorior_lunch is now known as jaosorior | 09:52 | |
| *** Daisy has quit IRC | 09:52 | |
| *** ljxiash has quit IRC | 09:57 | |
| *** ljxiash has joined #openstack-nova | 09:57 | |
| *** ljxiash has quit IRC | 09:58 | |
| *** ljxiash has joined #openstack-nova | 09:58 | |
| *** ljxiash has quit IRC | 09:58 | |
| *** ljxiash has joined #openstack-nova | 09:59 | |
| *** daemontool has quit IRC | 09:59 | |
| johnthetubaguy | ratailor: my mistake, I think you can do this here: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L555 | 10:01 |
| *** shaohe_feng has quit IRC | 10:02 | |
| *** cheneydc has quit IRC | 10:02 | |
| *** shaohe_feng has joined #openstack-nova | 10:03 | |
| *** ljxiash has quit IRC | 10:03 | |
| *** zenoway has joined #openstack-nova | 10:03 | |
| openstackgerrit | Dinesh Bhor proposed openstack/nova: Return HTTP 200 on list for invalid status https://review.openstack.org/315964 | 10:07 |
| *** roeyc has quit IRC | 10:12 | |
| *** shaohe_feng has quit IRC | 10:12 | |
| *** shaohe_feng has joined #openstack-nova | 10:13 | |
| *** zenoway has quit IRC | 10:17 | |
| *** zenoway has joined #openstack-nova | 10:17 | |
| *** ivase has quit IRC | 10:17 | |
| *** ivase has joined #openstack-nova | 10:18 | |
| ratailor | johnthetubaguy, I am not able to understand, How we can use this reset method to delete instances stuck in deleting task state without restarting compute service. | 10:19 |
| ratailor | johnthetubaguy, Is there anything implemented like this before ? If yes, could you pl point to that. | 10:19 |
| johnthetubaguy | ratailor: that reset method is called when you send SIG_HUP signal to the nova-compute process | 10:19 |
| johnthetubaguy | ratailor: for example, see the code in that method, where it re-calculates the RPC pin | 10:19 |
| johnthetubaguy | ratailor: basically, you can just call your periodic task, I believe, from that reset method | 10:20 |
| *** itamarl has quit IRC | 10:20 | |
| ratailor | johnthetubaguy, but in that case, if would not be automatic. | 10:22 |
| *** shaohe_feng has quit IRC | 10:22 | |
| *** salv-orlando has quit IRC | 10:22 | |
| ratailor | johnthetubaguy, I think admin has to send sighup. Am I right ? | 10:23 |
| johnthetubaguy | ratailor: yes, its not automatic | 10:23 |
| johnthetubaguy | thats why I like it | 10:23 |
| *** salv-orlando has joined #openstack-nova | 10:23 | |
| *** shaohe_feng has joined #openstack-nova | 10:23 | |
| johnthetubaguy | now the admin can add automation to call a SIG_HUP if they ever have to kill/restore their rabbit server | 10:23 |
| ratailor | johnthetubaguy, So you mean to say, whatever we want to do on sighup, we should call that functionality from reset method. | 10:24 |
| johnthetubaguy | maybe, yes | 10:24 |
| johnthetubaguy | although, we need to make sure we have not still about to process a delete, thats quite hard to spot | 10:24 |
| ratailor | johnthetubaguy, yes, On service restart, it is handled but not without restart. | 10:25 |
| johnthetubaguy | what I mean, is restarting the service means we can no longer be processing the delete | 10:26 |
| johnthetubaguy | if there are no rabbit messages for the compute node | 10:26 |
| johnthetubaguy | but its harder if the service is still running | 10:26 |
| johnthetubaguy | how do you know we are not currently processing the delete, etc | 10:27 |
| tdurakov | johnthetubaguy: hey, have you heard any of issues about live-migration of volume-backed instances with config-drive | 10:30 |
| tdurakov | type for config drive is vfat | 10:30 |
| tdurakov | PaulMurray: ^ | 10:30 |
| johnthetubaguy | tdurakov: i remember talk of issues before, but maybe not with vfat | 10:31 |
| johnthetubaguy | its possible recent "fixes" broke other things? | 10:31 |
| tdurakov | johnthetubaguy: it doesn't work for iso9600 | 10:31 |
| *** mfedosin has joined #openstack-nova | 10:31 | |
| tdurakov | but I've just faced with the same for vfat | 10:32 |
| tdurakov | filing bug, | 10:32 |
| tdurakov | there is a problem with block-migration flag calculation | 10:32 |
| *** shaohe_feng has quit IRC | 10:32 | |
| *** shaohe_feng has joined #openstack-nova | 10:33 | |
| *** thorst has joined #openstack-nova | 10:38 | |
| tdurakov | johnthetubaguy: https://bugs.launchpad.net/nova/+bug/1589457 | 10:39 |
| openstack | Launchpad bug 1589457 in OpenStack Compute (nova) "volume-backed instances with config-drive type vfat live-migration fails " [Undecided,New] | 10:39 |
| *** zenoway has quit IRC | 10:39 | |
| ratailor | johnthetubaguy, But on restart, delete is handled in init_instance | 10:39 |
| *** zenoway has joined #openstack-nova | 10:40 | |
| ratailor | johnthetubaguy, we can only take instance with 'deleting' task_state into consideration while cleaning up without restarting the service. | 10:40 |
| johnthetubaguy | ratailor: agreed thats what happens. But it can make lots of assumptions because the service has been restarted, it knows there is nothing happening in nova-compute when it comes back up | 10:40 |
| tdurakov | I feel this is done in wrong place https://github.com/openstack/nova/blob/545d8d8666389f33601b0b003dec844004694919/nova/virt/libvirt/driver.py#L6340 | 10:41 |
| tdurakov | it should be done during pre_live_migration checks not in pre_live_migration itself | 10:41 |
| johnthetubaguy | ratailor: I mean we need to know we are not still trying to delete that instance, and its just taking a long time (the instance lock might be enough to fix that, but that causes other issues) | 10:41 |
| *** ivase has quit IRC | 10:42 | |
| johnthetubaguy | tdurakov: yeah, that should be in check_can_live_migrate, or whatever it is called | 10:42 |
| *** shaohe_feng has quit IRC | 10:43 | |
| openstackgerrit | Anh Tran proposed openstack/nova: Stop using mox from compute/test_multiple_create.py https://review.openstack.org/325819 | 10:43 |
| *** shaohe_feng has joined #openstack-nova | 10:44 | |
| *** hoangcx has quit IRC | 10:44 | |
| *** zenoway has quit IRC | 10:44 | |
| *** thorst has quit IRC | 10:46 | |
| *** ljxiash has joined #openstack-nova | 10:46 | |
| *** tangchen has joined #openstack-nova | 10:47 | |
| *** daemontool has joined #openstack-nova | 10:47 | |
| *** zenoway has joined #openstack-nova | 10:48 | |
| *** boomsauce has joined #openstack-nova | 10:48 | |
| *** smatzek_ has joined #openstack-nova | 10:48 | |
| *** ljxiash has quit IRC | 10:51 | |
| *** shaohe_feng has quit IRC | 10:53 | |
| *** shaohe_feng has joined #openstack-nova | 10:54 | |
| *** sdague has joined #openstack-nova | 10:54 | |
| *** ljxiash has joined #openstack-nova | 10:57 | |
| *** sahid has quit IRC | 11:02 | |
| *** zenoway has quit IRC | 11:03 | |
| *** shaohe_feng has quit IRC | 11:03 | |
| *** shaohe_feng has joined #openstack-nova | 11:04 | |
| *** aarefiev_ is now known as aarefiev | 11:06 | |
| *** cdent has quit IRC | 11:08 | |
| *** rtheis has joined #openstack-nova | 11:10 | |
| *** Dinesh_Bhor has joined #openstack-nova | 11:11 | |
| *** shaohe_feng has quit IRC | 11:13 | |
| *** shaohe_feng has joined #openstack-nova | 11:14 | |
| PaulMurray | tdurakov, sorry, was engrossed in code. vfat has always worked AFAIK. The only thing stopping it being migrated before was a conditional that prevented any migration with a config drive - and that was only there because th iso9600 problem | 11:18 |
| PaulMurray | tdurakov, so if you have an issue I think its a bug - and possibly recently introduced | 11:19 |
| sdague | gmann: I'm trying to understand your concerns around microversions here | 11:19 |
| sdague | https://review.openstack.org/#/c/325308/ | 11:19 |
| *** sacharya has joined #openstack-nova | 11:20 | |
| *** ducnc has quit IRC | 11:20 | |
| *** sudipto has quit IRC | 11:22 | |
| *** jaypipes has joined #openstack-nova | 11:23 | |
| sdague | gmann: oh, I see, I left some bits out | 11:24 |
| *** shaohe_feng has quit IRC | 11:24 | |
| *** sacharya has quit IRC | 11:25 | |
| *** shaohe_feng has joined #openstack-nova | 11:27 | |
| *** roeyc has joined #openstack-nova | 11:27 | |
| *** cdent has joined #openstack-nova | 11:27 | |
| openstackgerrit | Sean Dague proposed openstack/nova: api-ref: verify keypairs https://review.openstack.org/325308 | 11:32 |
| openstackgerrit | Sean Dague proposed openstack/nova: api-ref: add version 2.3 parameters to servers https://review.openstack.org/325502 | 11:32 |
| *** mgoddard has joined #openstack-nova | 11:32 | |
| *** shaohe_feng has quit IRC | 11:34 | |
| *** shaohe_feng has joined #openstack-nova | 11:35 | |
| *** salv-orlando has quit IRC | 11:35 | |
| *** salv-orlando has joined #openstack-nova | 11:35 | |
| *** thorst has joined #openstack-nova | 11:36 | |
| *** rmart04 has joined #openstack-nova | 11:37 | |
| *** salv-orl_ has joined #openstack-nova | 11:39 | |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Transfrom wrap_exception notification to versioned format https://review.openstack.org/310700 | 11:41 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Enhance notification sample test base https://review.openstack.org/321610 | 11:41 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Handle multiple samples per versioned notification https://review.openstack.org/321609 | 11:41 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Do not register notification objects https://review.openstack.org/309454 | 11:41 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Transform instance.delete notifications https://review.openstack.org/313654 | 11:41 |
| *** salv-orlando has quit IRC | 11:42 | |
| *** shaohe_feng has quit IRC | 11:44 | |
| *** shaohe_feng has joined #openstack-nova | 11:45 | |
| *** baoli has joined #openstack-nova | 11:45 | |
| *** baoli_ has joined #openstack-nova | 11:46 | |
| *** dave-mccowan has joined #openstack-nova | 11:46 | |
| *** smatzek_ has quit IRC | 11:46 | |
| *** smatzek has joined #openstack-nova | 11:47 | |
| *** gstepanov_ has quit IRC | 11:48 | |
| *** baoli has quit IRC | 11:50 | |
| *** zenoway has joined #openstack-nova | 11:50 | |
| *** Jeffrey4l_ has quit IRC | 11:51 | |
| Dinesh_Bhor | oomichi: Hi, Fixed your review comment. Please take a look at it: https://review.openstack.org/#/c/315964/ | 11:51 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Improve nova.rpc conf options documentation https://review.openstack.org/302678 | 11:51 |
| *** ratailor has quit IRC | 11:51 | |
| *** shaohe_feng has quit IRC | 11:54 | |
| *** ankit has joined #openstack-nova | 11:56 | |
| ankit | Hi all, can someone please help reviewing https://review.openstack.org/#/c/247519/ a trivial change | 11:56 |
| *** shaohe_feng has joined #openstack-nova | 11:57 | |
| *** salv-orl_ has quit IRC | 11:58 | |
| *** salv-orlando has joined #openstack-nova | 11:58 | |
| *** chip_ has joined #openstack-nova | 11:59 | |
| *** JoseMello has joined #openstack-nova | 12:00 | |
| *** paul-carlton2 has quit IRC | 12:01 | |
| *** markvoelker has joined #openstack-nova | 12:01 | |
| *** jsuchome has joined #openstack-nova | 12:01 | |
| *** chip_ has quit IRC | 12:01 | |
| *** ociuhandu has quit IRC | 12:03 | |
| *** shaohe_feng has quit IRC | 12:05 | |
| *** shaohe_feng has joined #openstack-nova | 12:05 | |
| *** aysyd has joined #openstack-nova | 12:05 | |
| tdurakov | PaulMurray: yes, filed bug, working on fix | 12:07 |
| *** zerda2 has quit IRC | 12:08 | |
| openstackgerrit | Sean Dague proposed openstack/nova: api-ref: verify keypairs https://review.openstack.org/325308 | 12:08 |
| openstackgerrit | Sean Dague proposed openstack/nova: api-ref: add version 2.3 parameters to servers https://review.openstack.org/325502 | 12:08 |
| PaulMurray | tdurakov, what's the problem - I guess it doesn't recognise it as a block device ? | 12:09 |
| tdurakov | PaulMurray: https://bugs.launchpad.net/nova/+bug/1589457 - initial RCA | 12:10 |
| openstack | Launchpad bug 1589457 in OpenStack Compute (nova) "live-migration fails for volume-backed instances with config-drive type vfat" [Medium,Confirmed] - Assigned to Timofey Durakov (tdurakov) | 12:10 |
| openstackgerrit | jichenjc proposed openstack/nova: WIP: add tag support in api-ref https://review.openstack.org/325852 | 12:10 |
| *** armax has joined #openstack-nova | 12:11 | |
| *** armax has left #openstack-nova | 12:11 | |
| tdurakov | PaulMurray: it automatically converts migration to be block one, and then doesn't differ vfat from iso9600 | 12:11 |
| *** armax has joined #openstack-nova | 12:12 | |
| *** salv-orl_ has joined #openstack-nova | 12:13 | |
| openstackgerrit | Tang Chen proposed openstack/nova: Use oslo_log in nova https://review.openstack.org/325855 | 12:13 |
| *** Daisy has joined #openstack-nova | 12:14 | |
| *** shaohe_feng has quit IRC | 12:15 | |
| *** shaohe_feng has joined #openstack-nova | 12:16 | |
| *** salv-orlando has quit IRC | 12:16 | |
| sdague | johnthetubaguy / alex_xu either of you able to land https://review.openstack.org/#/c/320048 then I'll add microversions on top of it | 12:17 |
| *** apoorv has quit IRC | 12:19 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Transfrom wrap_exception notification to versioned format https://review.openstack.org/310700 | 12:20 |
| *** GLaupre has joined #openstack-nova | 12:20 | |
| johnthetubaguy | sdague: I can take a look at that in a few mins | 12:20 |
| GLaupre | Hello everyone. I have a question about SR-IOV and the IP commands that are used in the nova code. To whom can I ask question about the libvirt and specialy SR-IOV? Tks :) | 12:21 |
| *** pradk has joined #openstack-nova | 12:21 | |
| *** takashin has left #openstack-nova | 12:23 | |
| tdurakov | PaulMurray, johnthetubaguy according to this one https://bugs.launchpad.net/nova/+bug/1246201 I expect that live-migration of volume-backed instance with vfat config drive should work, am I right? | 12:24 |
| openstack | Launchpad bug 1246201 in OpenStack Compute (nova) "Live migration fails when the instance has a config_drive_format=iso9660" [High,In progress] - Assigned to Michael Still (mikal) | 12:24 |
| johnthetubaguy | tdurakov: that was my assumption too, but I am not sure | 12:24 |
| kashyap | sdague: If you have stable +2 creds, let's put the final nail in the coffin for this?: https://review.openstack.org/#/c/325566/1 | 12:24 |
| *** ivase has joined #openstack-nova | 12:25 | |
| sdague | kashyap: +2 | 12:25 |
| *** shaohe_feng has quit IRC | 12:25 | |
| *** shaohe_feng has joined #openstack-nova | 12:25 | |
| kashyap | Excellent. | 12:26 |
| *** eharney has joined #openstack-nova | 12:28 | |
| *** nortman has quit IRC | 12:29 | |
| *** jsuchome has quit IRC | 12:30 | |
| *** sree has joined #openstack-nova | 12:31 | |
| *** Daisy has quit IRC | 12:34 | |
| PaulMurray | tdurakov, yes, it should work with vfat | 12:34 |
| openstackgerrit | John Garbutt proposed openstack/nova: add "needs:*" tags to the config option modules https://review.openstack.org/322255 | 12:34 |
| *** Daisy has joined #openstack-nova | 12:34 | |
| PaulMurray | tdurakov, iso9600 fails because its a read only device - qemu creates it at the destination and then tries to copy to it, which fails because it can't | 12:35 |
| PaulMurray | tdurakov, it needs to be created with the content | 12:35 |
| PaulMurray | tdurakov, vfat doesn't have the problem so it should just work | 12:35 |
| *** shaohe_feng has quit IRC | 12:35 | |
| PaulMurray | we always did migrate vfat config drives on our cloud - that was icehouse and earlier | 12:36 |
| tdurakov | PaulMurray: in case of vfat and volume-backed instance, should it be block-migration or true live-migration? | 12:36 |
| *** shaohe_feng has joined #openstack-nova | 12:36 | |
| PaulMurray | It should be block migrate - its a local file that qemu has to copy | 12:37 |
| *** IanGovett has joined #openstack-nova | 12:37 | |
| *** jsuchome has joined #openstack-nova | 12:37 | |
| *** akshai has joined #openstack-nova | 12:38 | |
| tdurakov | PaulMurray: ok, thanks | 12:38 |
| *** Daisy has quit IRC | 12:38 | |
| *** rodrigods has quit IRC | 12:38 | |
| *** rodrigods has joined #openstack-nova | 12:38 | |
| frickler | block migrate in icehouse copied attached volumes onto themselves, that was blocked in Juno iirc | 12:39 |
| *** paul-carlton2 has joined #openstack-nova | 12:39 | |
| *** IanGovett has quit IRC | 12:39 | |
| *** IanGovett has joined #openstack-nova | 12:40 | |
| tdurakov | deployed devstack on liberty, same behaviour, so, I doesn't allow to block-migrate such instance, as frickler mentioned | 12:42 |
| *** haukebruno has quit IRC | 12:42 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: parameter verification for os-hypervisors https://review.openstack.org/316681 | 12:42 |
| tdurakov | in case of true live-migration it fails too, complaining to no shared storage | 12:43 |
| tdurakov | InvalidSharedStorage: node1 is not on shared storage: Live migration can not be used without shared storage except a booted from volume VM which does not have a local disk. | 12:43 |
| * tdurakov start thinking is not a bug | 12:43 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_instance_faults https://review.openstack.org/307126 | 12:44 |
| tdurakov | PaulMurray, johnthetubaguy ^ | 12:44 |
| *** garyk has joined #openstack-nova | 12:44 | |
| *** zz_dimtruck is now known as dimtruck | 12:45 | |
| johnthetubaguy | tdurakov: hmm, yeah, I am not sure then | 12:45 |
| *** shaohe_feng has quit IRC | 12:46 | |
| *** ociuhandu has joined #openstack-nova | 12:46 | |
| *** shaohe_feng has joined #openstack-nova | 12:46 | |
| alex_xu | johnthetubaguy: are you checking on https://review.openstack.org/#/c/320048, if not, let me help on it | 12:47 |
| *** eharney has quit IRC | 12:48 | |
| openstackgerrit | Sean Dague proposed openstack/nova: api-ref: verify keypairs https://review.openstack.org/325308 | 12:49 |
| openstackgerrit | Sean Dague proposed openstack/nova: api-ref: add version 2.3 parameters to servers https://review.openstack.org/325502 | 12:49 |
| johnthetubaguy | alex_xu: just posted my review, sdague I found quite a few issues, I think | 12:49 |
| alex_xu | johnthetubaguy: cool | 12:50 |
| sdague | johnthetubaguy: yep | 12:50 |
| *** vishwanathj has quit IRC | 12:51 | |
| alex_xu | sdague: i added few also :) | 12:53 |
| moshele | lbeliveau: hi | 12:54 |
| PaulMurray | tdurakov, frickler yes, selective block migration was introduced in mitaka | 12:54 |
| PaulMurray | tdurakov, the logic for live-migration --block-migrate flag changed | 12:55 |
| moshele | sfinucan: hi | 12:55 |
| *** tinbar has quit IRC | 12:55 | |
| PaulMurray | --block-migrate=True => everything has to be block migrated | 12:55 |
| moshele | sean-k-mooney: hi | 12:55 |
| PaulMurray | --block-migrate=False => everything has to be true live migration (i.e. shared storage or volumes) | 12:55 |
| PaulMurray | --block-migration=None (default) => nova works it out - i.e. selective block migration | 12:56 |
| *** shaohe_feng has quit IRC | 12:56 | |
| PaulMurray | All ^^^ assuming pi version 2.24 | 12:56 |
| PaulMurray | s/pi/API/ | 12:56 |
| *** sree has quit IRC | 12:56 | |
| *** shaohe_feng has joined #openstack-nova | 12:56 | |
| *** sree has joined #openstack-nova | 12:57 | |
| *** rcernin has joined #openstack-nova | 12:58 | |
| *** roman is now known as rpodolyaka | 12:58 | |
| lbeliveau | moshele: hi | 12:58 |
| *** ociuhandu has quit IRC | 12:58 | |
| PaulMurray | sdague, I have a question about API versions and functional tests | 12:58 |
| sdague | PaulMurray: sure, fire away | 12:58 |
| PaulMurray | sdague, its for console URLs | 12:58 |
| *** IanGovett has quit IRC | 12:59 | |
| *** rcernin has quit IRC | 12:59 | |
| *** figleaf is now known as edleafe | 12:59 | |
| PaulMurray | I want to add a term to the query string on the console URL | 12:59 |
| PaulMurray | so it has a token and an instance uuid | 12:59 |
| *** prateek has quit IRC | 12:59 | |
| PaulMurray | that is backward compatible | 12:59 |
| PaulMurray | because the additional terms are ignored | 12:59 |
| PaulMurray | So should it be an API bump ? | 13:00 |
| moshele | lbeliveau: I tested your patch the confirm-revert didn't work | 13:00 |
| lbeliveau | moshele: just saw your email, gonna have a look in a few hours, thanks ! | 13:01 |
| moshele | lbeliveau: do you know how is working on the numa issue with shelf/unshelf? I thought is was sfinucan but is not on irc | 13:02 |
| *** EinstCra_ has quit IRC | 13:02 | |
| lbeliveau | moshele: I taught he was working on it as well | 13:02 |
| sdague | yes, if we are adding parameters it's an API bump | 13:02 |
| PaulMurray | sdague, ok, so the next part.... | 13:03 |
| PaulMurray | sdague, what should I do with the templates in the functional tests? | 13:03 |
| lbeliveau | moshele: send him an email | 13:04 |
| sdague | PaulMurray: which tests cover this? | 13:04 |
| PaulMurray | I noticed there are a few classes in nova/tests/functional/api_sample_tests/test_remote_consoles.py | 13:04 |
| openstackgerrit | jichenjc proposed openstack/nova: api-ref: body verification of ips.inc https://review.openstack.org/325896 | 13:05 |
| PaulMurray | that use templates and api samples to run tests | 13:05 |
| moshele | lbeliveau: do you know if they open a bug for this? | 13:05 |
| PaulMurray | the classes that deal with new features seem to be versioned | 13:05 |
| PaulMurray | I don't know what I should do here | 13:05 |
| moshele | wznoinsk: hi | 13:05 |
| sdague | PaulMurray: right, the versioned ones only run tests at that new microversion | 13:06 |
| *** lindycoder has joined #openstack-nova | 13:06 | |
| *** shaohe_feng has quit IRC | 13:06 | |
| sdague | which is probably what you want here | 13:06 |
| *** shaohe_feng has joined #openstack-nova | 13:06 | |
| *** rlrossit has joined #openstack-nova | 13:07 | |
| *** pwnall has joined #openstack-nova | 13:07 | |
| PaulMurray | Does that mean I would have to create new tmplates and samples to go with new tests at a new version ? | 13:07 |
| *** ayoung has joined #openstack-nova | 13:07 | |
| lbeliveau | moshele: I think this is this one https://bugs.launchpad.net/nova/+bug/1587386 | 13:07 |
| openstack | Launchpad bug 1587386 in OpenStack Compute (nova) "Unshelve results in duplicated resource deallocated" [Undecided,In progress] - Assigned to Stephen Finucane (sfinucan) | 13:07 |
| moshele | lbeliveau: thanks | 13:08 |
| PaulMurray | sdague, FYI this is the change: https://review.openstack.org/#/c/319964/ | 13:09 |
| openstackgerrit | jichenjc proposed openstack/nova: api-ref: verfiy limits body https://review.openstack.org/325899 | 13:10 |
| PaulMurray | sdague, oops, wrong change | 13:10 |
| PaulMurray | sdague, this one https://review.openstack.org/#/c/325382 | 13:10 |
| *** mdrabe has joined #openstack-nova | 13:11 | |
| PaulMurray | sdague, So I think I would need the addition of the new term to the URL to be conditional on the API version (only add if new version) | 13:11 |
| *** boomsauce has quit IRC | 13:11 | |
| PaulMurray | sdague, but can't see any examples of what to do with the test templates and examples | 13:11 |
| sdague | it's a parameter change, so you don't change the templates, but you put the logic mainline | 13:13 |
| sdague | why are you doing the uuid inside of compute manager instead of at the API layer? | 13:15 |
| *** mtanino has joined #openstack-nova | 13:15 | |
| *** xyang1 has joined #openstack-nova | 13:15 | |
| *** IanGovett has joined #openstack-nova | 13:15 | |
| *** mriedem has joined #openstack-nova | 13:16 | |
| PaulMurray | That's where it is at the moment - the compute managers are configured with a base url, generate a token and make up the URL | 13:16 |
| *** setuid has joined #openstack-nova | 13:16 | |
| *** shaohe_feng has quit IRC | 13:16 | |
| PaulMurray | I think that may have to change | 13:17 |
| PaulMurray | if I need to vary the URL based on the API version | 13:17 |
| *** ankit has quit IRC | 13:18 | |
| sdague | yeh, so I'm a bit under the weather, and might not be thinking this through all the way. But my instinct is we need to get that into the API layer proper | 13:19 |
| sdague | this is just there as a routing hint? | 13:19 |
| *** shaohe_feng has joined #openstack-nova | 13:20 | |
| *** liverpooler has quit IRC | 13:20 | |
| PaulMurray | yes | 13:20 |
| PaulMurray | for cells | 13:20 |
| PaulMurray | the token and connection info will be in the child cell db | 13:20 |
| *** sacharya has joined #openstack-nova | 13:20 | |
| PaulMurray | and the token alone is not enough for the console proxies to find the right cell db | 13:21 |
| *** vishwanathj has joined #openstack-nova | 13:22 | |
| sdague | right... | 13:23 |
| sdague | I guess... I'm trying to understand why the tokens are going to be in the cell dbs | 13:23 |
| *** jichen has quit IRC | 13:23 | |
| tdurakov | PaulMurray, thank you for details | 13:23 |
| *** prateek has joined #openstack-nova | 13:23 | |
| sdague | because, the console manager is going to need talk to the api db, get the cell, talk to the cell db. | 13:24 |
| PaulMurray | sdague, yes, I know, could be in api db - probably not that many of them either. | 13:25 |
| *** sacharya has quit IRC | 13:25 | |
| dansmith | talking about console tokens? | 13:25 |
| sdague | yeh | 13:25 |
| dansmith | why not put them in the cell db? | 13:25 |
| PaulMurray | sdague, I talked to alaski about it - he thought they should go in cell because they are specific to instances | 13:25 |
| dansmith | they're specifically related to instances and compute nodes, | 13:26 |
| dansmith | which means cell db, IMHO | 13:26 |
| tdurakov | PaulMurray: actually I used older libvirt without selective feature for block-migration | 13:26 |
| PaulMurray | dansmith, right | 13:26 |
| PaulMurray | dansmith, also compute managers can look after cleaning up their own expired tokens | 13:26 |
| *** edmondsw has joined #openstack-nova | 13:26 | |
| dansmith | right | 13:26 |
| PaulMurray | tdurakov, that would do it :) | 13:27 |
| *** shaohe_feng has quit IRC | 13:27 | |
| sdague | dansmith: ok, it seems that given the routing issue you basically are always talking to both layers directly | 13:27 |
| PaulMurray | tdurakov, as it says in the error message..... | 13:27 |
| *** shaohe_feng has joined #openstack-nova | 13:27 | |
| dansmith | sdague: but that's the same as, say, instance show right? | 13:27 |
| sdague | and it does seem weird to basically have to put routing hints into the urls for the token to figure out how to find your instance | 13:27 |
| dansmith | routing hints? | 13:27 |
| sdague | https://review.openstack.org/#/c/325382/2/nova/compute/manager.py | 13:28 |
| dansmith | because the uuid? | 13:28 |
| sdague | yeh | 13:28 |
| dansmith | seems fine to me | 13:28 |
| dansmith | I guess I'm pretty surprised it's not already in there | 13:29 |
| johnthetubaguy | thing is, as you add compute nodes, you shouldn't need to massively increase the size of the API DB, so this seems like it has to be in the cell DB | 13:29 |
| johnthetubaguy | dansmith: +1 | 13:29 |
| dansmith | johnthetubaguy: right | 13:29 |
| dansmith | shards perfectly, | 13:29 |
| *** vishwana_ has joined #openstack-nova | 13:29 | |
| dansmith | we just have (apparently) something that kinda assumed it could always look up everything by token, | 13:30 |
| dansmith | but no need to keep doing that | 13:30 |
| sdague | you know, maybe I'd feel better if the instance was in the path | 13:30 |
| dansmith | that's the thing I'm surprised isn't already there | 13:30 |
| mriedem | BobBall: anthonyper: is this a known xen project ci failure or new? http://logs.openstack.xenproject.org/53/318953/2/check/dsvm-tempest-xen/d6f01ff/logs/screen-n-cpu.txt.gz - look for "NovaException: Error from libvirt while getting domain info for instance-00000029: [Error Code 1] internal error: libxl_domain_info failed for domain '59'" | 13:31 |
| rlrossit | sdague: now that legacy_v2 is gone, can't half of policy.json go away now too? | 13:31 |
| sdague | rlrossit: it's in the patch stack somewhere | 13:31 |
| rlrossit | saweeeeet | 13:31 |
| * rlrossit waves hands around happily | 13:31 | |
| *** vishwanathj has quit IRC | 13:31 | |
| *** jwcroppe has joined #openstack-nova | 13:32 | |
| *** moshele has quit IRC | 13:32 | |
| PaulMurray | sdague, dansmith does that mean you would rather this became a proper resource path like everything else ? | 13:32 |
| sdague | PaulMurray: how much pain would it be to make the console urls be - CONF.vnc.novncproxy_base_url/{instance_uuid} | 13:32 |
| sdague | or is that going to screw up something with novnc? | 13:33 |
| PaulMurray | sdague, changing the parameter in the qurey string is backward compatible - existing proxies will understand it and do the right thing | 13:33 |
| dansmith | yeah | 13:33 |
| dansmith | that's what I was wondering | 13:33 |
| PaulMurray | sdague, I don't know what cahnging the path will do to each proxy | 13:33 |
| *** sudipto has joined #openstack-nova | 13:33 | |
| PaulMurray | its not a regular nova API that is going to consume the URL | 13:33 |
| dansmith | this doesn't bother me at all, but whichever way is okay with me | 13:33 |
| sdague | I'm not entirely sure what backwards compatible means here | 13:33 |
| sdague | because... these are basically opaque to a running cloud right? | 13:34 |
| PaulMurray | I mean if you add &uuid=<uuid> to any existing console URL it still works | 13:34 |
| *** vishwanathj has joined #openstack-nova | 13:34 | |
| sdague | sure, but it's not like if you have a console url bookmarked, and your cloud upgrades to cells v2, it still works | 13:35 |
| PaulMurray | right - just a list of parameters and they pick out the one they are interested in | 13:35 |
| sdague | these are not being constructed by users | 13:35 |
| *** ishant is now known as ishant|away | 13:35 | |
| *** smatzek has quit IRC | 13:35 | |
| PaulMurray | no, but if you upgrade some compute managers before proxies they will work | 13:35 |
| *** mtanino has quit IRC | 13:36 | |
| PaulMurray | but that's not really a good argument | 13:36 |
| PaulMurray | we want to move the url build to the api probably | 13:36 |
| anthonyper | mriedem: This error looks new, I don't recognize it, and can not find anything usefull in libvirt logs | 13:36 |
| sdague | getting a cell db is an atomic point, right? | 13:36 |
| mriedem | anthonyper: ok i'll open a bug | 13:36 |
| PaulMurray | atomic? | 13:36 |
| sdague | I'm trying to understand what upgrade story we think is better served by a parameter than a path | 13:37 |
| *** shaohe_feng has quit IRC | 13:37 | |
| *** vishwanathj has quit IRC | 13:37 | |
| dansmith | sdague: what's the problem with a parameter? | 13:37 |
| sdague | also, I reverse my previous statement about API version, this is basically opaque internals | 13:37 |
| *** vishwanathj has joined #openstack-nova | 13:37 | |
| *** armax_ has joined #openstack-nova | 13:37 | |
| sdague | parameters are really supposed to be optional things | 13:37 |
| *** anteaya has joined #openstack-nova | 13:38 | |
| *** shaohe_feng has joined #openstack-nova | 13:38 | |
| sdague | this is definitely not optional in cells | 13:38 |
| dansmith | um, really? | 13:38 |
| dansmith | token isn't optional | 13:38 |
| PaulMurray | sdague, to be honest I think its more important to see how they proxies handle the path | 13:38 |
| dansmith | it's exactly as important as token | 13:38 |
| *** vishwana_ has quit IRC | 13:38 | |
| PaulMurray | sdague, and I haven't looked at that | 13:38 |
| *** armax has quit IRC | 13:39 | |
| *** armax_ has quit IRC | 13:39 | |
| *** armax has joined #openstack-nova | 13:40 | |
| sdague | well, especially as it's being used as a sharing hint, getting that in the path seems more appropriate | 13:40 |
| *** xyang1 has quit IRC | 13:40 | |
| sdague | s/sharing/sharding/ | 13:40 |
| mriedem | anthonyper: fyi https://bugs.launchpad.net/nova/+bug/1589507 | 13:40 |
| openstack | Launchpad bug 1589507 in OpenStack Compute (nova) "XenProject CI fails auto->manual disk config tests with "[Error Code 1] internal error: libxl_domain_info failed for domain"" [Undecided,New] | 13:40 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Enhance notification sample test base https://review.openstack.org/321610 | 13:40 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Handle multiple samples per versioned notification https://review.openstack.org/321609 | 13:40 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Transform instance.delete notifications https://review.openstack.org/313654 | 13:40 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Transform instance.update notification https://review.openstack.org/325923 | 13:40 |
| sdague | anyway, it would be nice to know if path was any different on the upgrade story for real. | 13:40 |
| *** sridharg has quit IRC | 13:41 | |
| *** xyang1 has joined #openstack-nova | 13:41 | |
| PaulMurray | sdague, I'll see what the impact on the proxies is and come back on that | 13:41 |
| PaulMurray | sdague, in the meantime, I didn't understand what to do with the functional tests (and presumably tempest as well) | 13:42 |
| sdague | PaulMurray: and, on testing, this should have tests. This is not an MV bump, because this is really an opaque internal url that the user should not be generating. | 13:42 |
| *** sree has quit IRC | 13:43 | |
| *** roeyc has quit IRC | 13:43 | |
| PaulMurray | ok, so no microversion bump | 13:43 |
| *** armax_ has joined #openstack-nova | 13:43 | |
| jsuchome | mriedem: hi, would you consider looking at https://review.openstack.org/#/c/225119 again at some point? | 13:44 |
| sdague | PaulMurray: do we have any code that can figure out if vnc seemed to attach? | 13:44 |
| PaulMurray | I don't think so | 13:44 |
| *** mgoddard_ has joined #openstack-nova | 13:44 | |
| PaulMurray | I haven't looked at tempest through | 13:44 |
| *** armax has quit IRC | 13:44 | |
| *** armax_ is now known as armax | 13:44 | |
| sdague | because... this really seems like it would be covered by a Tempest test that vnc gets past auth | 13:44 |
| *** dmk0202 has joined #openstack-nova | 13:44 | |
| PaulMurray | the functional tests look for the form of the url | 13:44 |
| *** vishwana_ has joined #openstack-nova | 13:44 | |
| sdague | right, the form of the url is really more of a unit test here | 13:45 |
| sdague | because that doesn't mean you actually found a working console | 13:45 |
| mriedem | jsuchome: yeah | 13:45 |
| sdague | if we had a tempest test that got the vnc console url, and at least got enough of the vnc stream to know we had authed correct and were getting data, that would work in all cases | 13:45 |
| sdague | no cells, cells v1, cells v2 | 13:46 |
| PaulMurray | yes, agreed | 13:46 |
| sdague | that would be ideal here | 13:46 |
| *** vishwanathj has quit IRC | 13:47 | |
| *** mgoddard has quit IRC | 13:47 | |
| *** shaohe_feng has quit IRC | 13:47 | |
| PaulMurray | So for the functional part I should just change the template for the URL so it works | 13:47 |
| PaulMurray | ? | 13:47 |
| *** eharney has joined #openstack-nova | 13:47 | |
| *** shaohe_feng has joined #openstack-nova | 13:48 | |
| sdague | sure | 13:48 |
| *** tbachman has joined #openstack-nova | 13:48 | |
| *** prateek has quit IRC | 13:48 | |
| PaulMurray | I'll look at what I can do to refactor the get_console_X calls so more is dealt with in the api | 13:49 |
| *** hshiina has quit IRC | 13:49 | |
| *** tanee is now known as tanee_away | 13:49 | |
| PaulMurray | I'll have a go and push some patches up | 13:50 |
| *** cleong has joined #openstack-nova | 13:50 | |
| *** roeyc has joined #openstack-nova | 13:50 | |
| PaulMurray | thanks | 13:50 |
| *** armax has quit IRC | 13:51 | |
| *** sdake has joined #openstack-nova | 13:51 | |
| *** ecelik has quit IRC | 13:51 | |
| *** ctrath has joined #openstack-nova | 13:53 | |
| *** permalac has quit IRC | 13:54 | |
| *** mrkz has joined #openstack-nova | 13:55 | |
| edleafe | Nova scheduler meeting in 5 minutes in #openstack-meeting-alt | 13:56 |
| *** sdake_ has joined #openstack-nova | 13:56 | |
| *** GLaupre has quit IRC | 13:57 | |
| *** ametts has joined #openstack-nova | 13:57 | |
| *** shaohe_feng has quit IRC | 13:57 | |
| *** shaohe_feng has joined #openstack-nova | 13:58 | |
| *** mgoddard_ has quit IRC | 13:58 | |
| *** sdake has quit IRC | 13:58 | |
| *** mgoddard has joined #openstack-nova | 13:58 | |
| *** vishwanathj has joined #openstack-nova | 14:00 | |
| *** BjoernT has joined #openstack-nova | 14:01 | |
| *** smatzek has joined #openstack-nova | 14:02 | |
| *** vishwana_ has quit IRC | 14:03 | |
| mriedem | jsuchome: comments inline | 14:03 |
| mriedem | jsuchome: i think that's one that's worth bringing up in the operators list | 14:03 |
| mriedem | since it changes the behavior there | 14:03 |
| *** paul-carlton2 has quit IRC | 14:03 | |
| mriedem | jsuchome: but the cinder.cross_az_attach=False restriction is really just kind of terrible | 14:03 |
| *** fawadkhaliq has joined #openstack-nova | 14:05 | |
| jsuchome | mriedem: thanks | 14:06 |
| *** dimtruck is now known as zz_dimtruck | 14:08 | |
| *** shaohe_feng has quit IRC | 14:08 | |
| *** shaohe_feng has joined #openstack-nova | 14:09 | |
| mriedem | mfedosin: sdague: oh wow, got the nova + glance v2 stack passing in tempest w/ glance v1 disabled | 14:09 |
| mriedem | https://review.openstack.org/#/c/325322/ | 14:09 |
| mriedem | we're going to need https://review.openstack.org/#/c/325569/ | 14:09 |
| *** mmandel has joined #openstack-nova | 14:09 | |
| sdague | mriedem: yeh, the stack looked quite promissing, so I'm not hugely surprised | 14:10 |
| mfedosin | cool! | 14:10 |
| mriedem | sdague: well, tempest assumes glance v1 client everywhere :( | 14:10 |
| mriedem | i fixed that over the weekend | 14:10 |
| sdague | mriedem: ah cool | 14:10 |
| mriedem | it could be cleaner, but it was the weekend | 14:10 |
| mfedosin | I'll update patches soon and fix merge conflicts | 14:10 |
| sdague | mfedosin: johnthetubaguy and I just approved the conf flag | 14:10 |
| sdague | once you get the stack updated, I'll do another review today | 14:11 |
| *** amotoki has joined #openstack-nova | 14:11 | |
| *** mgoddard_ has joined #openstack-nova | 14:12 | |
| *** ljxiash has quit IRC | 14:12 | |
| *** MikeG451 has joined #openstack-nova | 14:12 | |
| *** ljxiash has joined #openstack-nova | 14:13 | |
| mfedosin | sdague: okay, got it | 14:13 |
| *** zz_dimtruck is now known as dimtruck | 14:15 | |
| *** mgoddard has quit IRC | 14:15 | |
| *** vishwana_ has joined #openstack-nova | 14:16 | |
| *** paul-carlton2 has joined #openstack-nova | 14:17 | |
| *** shaohe_feng has quit IRC | 14:18 | |
| *** vishwanathj has quit IRC | 14:18 | |
| *** ljxiash has quit IRC | 14:18 | |
| *** shaohe_feng has joined #openstack-nova | 14:19 | |
| *** armax has joined #openstack-nova | 14:21 | |
| *** tanee_away is now known as tanee | 14:22 | |
| *** ljxiash has joined #openstack-nova | 14:23 | |
| *** shaohe_feng has quit IRC | 14:28 | |
| *** shaohe_feng has joined #openstack-nova | 14:29 | |
| *** armax has quit IRC | 14:29 | |
| *** armax has joined #openstack-nova | 14:30 | |
| *** vishwanathj has joined #openstack-nova | 14:31 | |
| tdurakov | mriedem: hi, could you please review this one https://review.openstack.org/#/c/252850/ | 14:32 |
| *** permalac has joined #openstack-nova | 14:33 | |
| *** vishwana_ has quit IRC | 14:34 | |
| *** mtanino has joined #openstack-nova | 14:34 | |
| *** rbradf_not_found is now known as rbradfor | 14:35 | |
| *** rk4n has joined #openstack-nova | 14:35 | |
| *** sajeesh has joined #openstack-nova | 14:37 | |
| *** salv-orlando has joined #openstack-nova | 14:38 | |
| *** amotoki has quit IRC | 14:38 | |
| *** shaohe_feng has quit IRC | 14:38 | |
| *** armax has quit IRC | 14:39 | |
| *** shaohe_feng has joined #openstack-nova | 14:39 | |
| *** cfarquhar has quit IRC | 14:39 | |
| *** armax has joined #openstack-nova | 14:40 | |
| *** salv-orl_ has quit IRC | 14:42 | |
| *** cfarquhar has joined #openstack-nova | 14:42 | |
| *** cfarquhar has quit IRC | 14:42 | |
| *** cfarquhar has joined #openstack-nova | 14:42 | |
| *** rama_y has joined #openstack-nova | 14:43 | |
| *** amotoki has joined #openstack-nova | 14:44 | |
| *** josecastroleon has quit IRC | 14:44 | |
| *** armax has quit IRC | 14:44 | |
| *** armax has joined #openstack-nova | 14:44 | |
| *** sdake_ has quit IRC | 14:45 | |
| *** vishwana_ has joined #openstack-nova | 14:46 | |
| *** amotoki has quit IRC | 14:47 | |
| *** amotoki has joined #openstack-nova | 14:48 | |
| *** vishwanathj has quit IRC | 14:49 | |
| *** shaohe_feng has quit IRC | 14:49 | |
| *** armax has quit IRC | 14:49 | |
| *** shaohe_feng has joined #openstack-nova | 14:49 | |
| *** fawadkhaliq has quit IRC | 14:49 | |
| mdbooth | diana_clarke: Pushed the WIP stuff here: https://github.com/mdbooth/nova/tree/matt_working | 14:51 |
| diana_clarke | mdbooth: Thanks! | 14:51 |
| *** rama_y has quit IRC | 14:51 | |
| mdbooth | It's a bit messier than I intended, but I figured you were just going to rip it apart anyway :) | 14:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: rest api version bumped for async pre live migration checks https://review.openstack.org/314932 | 14:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Make live migration checks async https://review.openstack.org/311116 | 14:52 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Added instance actions for conductor https://review.openstack.org/311115 | 14:52 |
| *** vishwana_ is now known as vishwanathj | 14:52 | |
| *** permalac has quit IRC | 14:52 | |
| *** jwcroppe has quit IRC | 14:55 | |
| *** jwcroppe has joined #openstack-nova | 14:56 | |
| *** mdrabe has quit IRC | 14:56 | |
| *** amotoki has quit IRC | 14:57 | |
| *** mgoddard_ has quit IRC | 14:58 | |
| *** gagehugo has joined #openstack-nova | 14:58 | |
| *** wanghao has quit IRC | 14:59 | |
| * cdent coffees up | 14:59 | |
| *** wanghao has joined #openstack-nova | 14:59 | |
| *** shaohe_feng has quit IRC | 14:59 | |
| *** mvk has quit IRC | 14:59 | |
| *** shaohe_feng has joined #openstack-nova | 15:00 | |
| *** jwcroppe has quit IRC | 15:01 | |
| *** mjura has quit IRC | 15:01 | |
| *** mjura has joined #openstack-nova | 15:01 | |
| *** vishwana_ has joined #openstack-nova | 15:01 | |
| *** mpavone has quit IRC | 15:02 | |
| dansmith | jaypipes: there is another meeting starting in that channel, can we continue here/ | 15:02 |
| jaypipes | cdent, dansmith, bauzas: yeah, sorry. see my last comment from the meeting channel pls. | 15:02 |
| *** penick has joined #openstack-nova | 15:03 | |
| dansmith | jaypipes: if you want.. I don't really know what refactor you plan to do there, so if you'd rather discuss it after you've done it then I guess that's okay, but.. | 15:03 |
| bauzas | jaypipes: okay, sorry, I was afk for a coffee | 15:04 |
| dansmith | maybe I'm silly for still thinking we can get ourselves on the same page ahead of time | 15:04 |
| *** josecastroleon has joined #openstack-nova | 15:04 | |
| *** vishwanathj has quit IRC | 15:04 | |
| bauzas | what I really want here is to help, not block you, so again sorry if you feel the contrary :/ | 15:04 |
| *** dmk0202 has quit IRC | 15:05 | |
| *** ijw has joined #openstack-nova | 15:05 | |
| jaypipes | dansmith: the refactoring I would like to do to cdent's objects patches I chatted with cdent about this morning. Basically, making InventoryList and AllocationList plain-old-data objects and putting all CRUD operations for aggregate association, atomic inventory updates, and atomic allocation changes on the ResourceProvider object itself. | 15:05 |
| cdent | dansmith: the reasoning behind that was because all of those things need the generation of the resource provider | 15:07 |
| jaypipes | bauzas, dansmith: in the meantime, cdent's patch to add a name to the resource provider should be relatively painless and conflict-free: https://review.openstack.org/#/c/281945/ | 15:08 |
| dansmith | jaypipes: hmm, okay | 15:08 |
| bauzas | jaypipes: you mean updating an inventory would cause a new RP generation set ? | 15:08 |
| jaypipes | bauzas: yes. | 15:08 |
| *** garyk has quit IRC | 15:09 | |
| bauzas | I see | 15:09 |
| cdent | jaypipes: I'm going to rebase that one on master and detach it from the rest. make sense? | 15:09 |
| jaypipes | cdent: yes | 15:09 |
| cdent | ✔ | 15:09 |
| mriedem | tdurakov: sure, will take a look | 15:09 |
| *** shaohe_feng has quit IRC | 15:09 | |
| dansmith | cdent: jaypipes: we haven't sent the RP object over rpc ever yet right? | 15:10 |
| mriedem | fudge | 15:10 |
| dansmith | or are we already querying it from the compute node side? | 15:10 |
| mriedem | tdurakov: PaulMurray: looks like we have a bug in the live migration path http://logs.openstack.org/03/315703/1/experimental/gate-tempest-dsvm-multinode-live-migration/97e6c2a/logs/screen-n-cpu.txt.gz?level=TRACE | 15:10 |
| *** shaohe_feng has joined #openstack-nova | 15:10 | |
| mriedem | fairly obvious one | 15:10 |
| cdent | dansmith: not in any reality yet | 15:10 |
| mriedem | Live Migration failure: migrateToURI3() got an unexpected keyword argument 'bandwidth' | 15:10 |
| *** ctrath has quit IRC | 15:10 | |
| *** mdrabe has joined #openstack-nova | 15:11 | |
| tdurakov | mriedem: there is a fix on review | 15:11 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1589548 | 15:11 |
| openstack | Launchpad bug 1589548 in OpenStack Compute (nova) "Live Migration failure: migrateToURI3() got an unexpected keyword argument 'bandwidth'" [High,Confirmed] | 15:11 |
| mriedem | tdurakov: link? | 15:12 |
| bauzas | dansmith: cdent: yeah I think we can modify the object straigh away wirthout requiring a version bump, if that's what you implied | 15:12 |
| dansmith | cdent: okay I was going to say you could just forego the version bump to try to hide more of our wishy-washy false starts from the criticism of history | 15:12 |
| cdent | roger that, thanks | 15:13 |
| tdurakov | mriedem: https://review.openstack.org/#/c/325204/ | 15:13 |
| *** mjura has quit IRC | 15:13 | |
| *** MikeG451 has quit IRC | 15:15 | |
| *** lindycoder has quit IRC | 15:15 | |
| *** amotoki has joined #openstack-nova | 15:15 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Remove image_meta https://review.openstack.org/325975 | 15:16 |
| *** lindycoder has joined #openstack-nova | 15:16 | |
| openstackgerrit | Jay Pipes proposed openstack/nova-specs: Amend generic-resource-pools https://review.openstack.org/325976 | 15:16 |
| jaypipes | bauzas, dansmith, cdent, edleafe, johnthetubaguy: please see above ^^ | 15:16 |
| *** vishwanathj has joined #openstack-nova | 15:16 | |
| cdent | jaypipes: noted | 15:17 |
| *** dgonzalez has quit IRC | 15:17 | |
| edleafe | jaypipes: roger that | 15:18 |
| *** penick has quit IRC | 15:19 | |
| *** dgonzalez has joined #openstack-nova | 15:19 | |
| *** tonytan4ever has joined #openstack-nova | 15:19 | |
| *** shaohe_feng has quit IRC | 15:19 | |
| *** shaohe_feng has joined #openstack-nova | 15:20 | |
| *** vishwana_ has quit IRC | 15:20 | |
| *** jlanoux has quit IRC | 15:20 | |
| *** Daisy has joined #openstack-nova | 15:21 | |
| *** Daisy has quit IRC | 15:21 | |
| *** andreas_s has quit IRC | 15:21 | |
| *** sacharya has joined #openstack-nova | 15:21 | |
| mriedem | danpb: is stuffing bandwidth into params for migrateToURI3 correct here? https://review.openstack.org/#/c/325204/ | 15:23 |
| mriedem | danpb: i'm not familiar enough with the internals of the libvirt c tree to see where that bandwidth parameter is used | 15:23 |
| danpb | mriedem: yes, what sahid said is correct | 15:24 |
| *** Daisy has joined #openstack-nova | 15:24 | |
| danpb | the 'params' dict must have bandwidth added to it | 15:24 |
| mriedem | danpb: ok, i'll make the change | 15:24 |
| mriedem | since live migration is busted w/o it in the ci job | 15:24 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Create Instance from BuildRequest if not in a cell https://review.openstack.org/263927 | 15:24 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Raise exception if BuildRequest deleted twice https://review.openstack.org/322324 | 15:24 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Cells: Handle delete with BuildRequest https://review.openstack.org/325985 | 15:24 |
| mriedem | pkoniszewski: or are you going to update https://review.openstack.org/#/c/325204/ ? | 15:25 |
| danpb | mriedem: the "right" thing is probably for this Guest.migrate method to *not* accept a params parameter at all | 15:25 |
| danpb | mriedem: rather it ought to just take explicit python parameters for everything it needs to support | 15:25 |
| danpb | mriedem: and then construct the dict to pass to libvirt itself | 15:25 |
| danpb | that way the caller is better insulated from the differences in libvirt apis | 15:25 |
| mriedem | hmm, yeah https://github.com/openstack/nova/blob/bfc5285b4684f6e459e6232ede58365fd9d50e34/nova/virt/libvirt/driver.py#L5733 | 15:26 |
| danpb | i'm fine if you don't wnat todo that full re-factoring now though and just focus on the quick fix | 15:26 |
| *** sacharya has quit IRC | 15:26 | |
| danpb | mriedem: oh lol, the params dict already has bandwidth set in it | 15:26 |
| mriedem | yeah, which is why pkoniszewski didn't add it | 15:26 |
| danpb | that just re-inforces my point i guess :-) | 15:26 |
| danpb | silly api design leaking libvirt differences out to the caller | 15:26 |
| mriedem | so expanding the 3 entries in that params dict in guest.migrate is what we want right? | 15:27 |
| *** gcb has quit IRC | 15:27 | |
| mriedem | not sure if there are other things | 15:27 |
| danpb | yeah, there will be other things over time, but we don't need them yet | 15:27 |
| mriedem | that's the only caller so must be | 15:27 |
| *** gcb has joined #openstack-nova | 15:27 | |
| mriedem | ok, so i think we should fix this quick and then i'll do the refactor on top | 15:28 |
| mriedem | let me add the bug ref quick and then i'll +2 | 15:28 |
| *** renato_r has joined #openstack-nova | 15:28 | |
| *** rk4n_ has joined #openstack-nova | 15:30 | |
| *** shaohe_feng has quit IRC | 15:30 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: [libvirt] Incorrect parameters passed to migrateToURI3 https://review.openstack.org/325204 | 15:30 |
| mriedem | danpb: if you want to +2 ^ | 15:30 |
| *** shaohe_feng has joined #openstack-nova | 15:31 | |
| *** ljxiash has quit IRC | 15:31 | |
| openstackgerrit | Jay Pipes proposed openstack/nova-specs: Amend generic-resource-pools https://review.openstack.org/325976 | 15:31 |
| *** roeyc has quit IRC | 15:32 | |
| jaypipes | bauzas, dansmith: ^^ fixed typos and answered ?s. | 15:32 |
| *** vishwana_ has joined #openstack-nova | 15:32 | |
| bauzas | jaypipes: okay, thanks for answering | 15:33 |
| *** rk4n has quit IRC | 15:33 | |
| *** josecastroleon has quit IRC | 15:34 | |
| bauzas | jaypipes: okay, so that clarifies one thing, it's the "caller service" responsibility to manage its resources | 15:34 |
| *** Swami has joined #openstack-nova | 15:34 | |
| openstackgerrit | Chris Dent proposed openstack/nova: Add name to ResourceProvider object https://review.openstack.org/281945 | 15:34 |
| bauzas | jaypipes: and not trust Nova for getting it the correct state | 15:34 |
| *** vishwanathj has quit IRC | 15:35 | |
| bauzas | jaypipes: ie. the "caller service" knows its resources and what it sent to Nova, and just compares whether it needs to update it thanks to the generation bit | 15:35 |
| bauzas | okay, gotcha | 15:35 |
| *** mgoddard has joined #openstack-nova | 15:35 | |
| sdague | super easy +A to close a bug on config file generation - https://review.openstack.org/#/c/324576/ | 15:37 |
| *** salv-orlando has quit IRC | 15:38 | |
| *** jsuchome has left #openstack-nova | 15:39 | |
| *** salv-orlando has joined #openstack-nova | 15:39 | |
| *** jdillaman has joined #openstack-nova | 15:39 | |
| *** yassine has quit IRC | 15:39 | |
| *** Daisy_ has joined #openstack-nova | 15:40 | |
| jaypipes | bauzas: yep, xactly. | 15:40 |
| *** yamahata has joined #openstack-nova | 15:40 | |
| *** shaohe_feng has quit IRC | 15:40 | |
| *** Daisy has quit IRC | 15:40 | |
| jaypipes | sdague: done | 15:40 |
| *** Daisy has joined #openstack-nova | 15:41 | |
| *** sacharya has joined #openstack-nova | 15:41 | |
| *** shaohe_feng has joined #openstack-nova | 15:41 | |
| bauzas | ergh, I Wallibied twice | 15:41 |
| * bauzas runs off now | 15:41 | |
| *** ijw has quit IRC | 15:41 | |
| *** nkrinner_afk has quit IRC | 15:42 | |
| mriedem | mikal: you can remove the -2 on https://review.openstack.org/#/c/317739/ now that the bp is approved | 15:42 |
| openstackgerrit | Alexis Lee proposed openstack/nova: Break up _live_migration_monitor https://review.openstack.org/325997 | 15:42 |
| *** Daisy_ has quit IRC | 15:45 | |
| *** kfarr has joined #openstack-nova | 15:45 | |
| *** tpepper has joined #openstack-nova | 15:45 | |
| *** READ10 has joined #openstack-nova | 15:45 | |
| openstackgerrit | Chris Dent proposed openstack/nova: Add Allocation and AllocationList objects https://review.openstack.org/282442 | 15:45 |
| openstackgerrit | Chris Dent proposed openstack/nova: WIP Extend ResourceProvider to have resources and aggregates https://review.openstack.org/284963 | 15:45 |
| openstackgerrit | Chris Dent proposed openstack/nova: Add a get_by_uuid for aggregates https://review.openstack.org/284895 | 15:45 |
| openstackgerrit | Chris Dent proposed openstack/nova: Correct the Inventory object https://review.openstack.org/315288 | 15:45 |
| openstackgerrit | Chris Dent proposed openstack/nova: WIP resource-pools/placement api https://review.openstack.org/293104 | 15:45 |
| *** salv-orl_ has joined #openstack-nova | 15:47 | |
| *** vishwanathj has joined #openstack-nova | 15:48 | |
| *** jlanoux has joined #openstack-nova | 15:48 | |
| *** josecastroleon has joined #openstack-nova | 15:49 | |
| bauzas | jaypipes: just one thing before I leave, but we can discuss that later, what if service updates a RP with generation=4 but the RP is already at 5 (because another service updated it between the last time the service updated it and now) ? | 15:49 |
| *** salv-orlando has quit IRC | 15:50 | |
| dansmith | bauzas: 409 | 15:50 |
| bauzas | jaypipes: I think we should return a 40x right? | 15:50 |
| mriedem | tdurakov: comments inline https://review.openstack.org/#/c/252850/ but that's going to return a 500 from the REST API | 15:50 |
| dansmith | bauzas: it says that in the spec | 15:50 |
| *** shaohe_feng has quit IRC | 15:50 | |
| bauzas | dansmith: sorry, I was reading it as "if the generation bit was the same, but yeah" | 15:50 |
| *** shaohe_feng has joined #openstack-nova | 15:51 | |
| *** vishwana_ has quit IRC | 15:51 | |
| *** vishwanathj has quit IRC | 15:52 | |
| *** Sukhdev has joined #openstack-nova | 15:53 | |
| *** raunak has joined #openstack-nova | 15:54 | |
| *** hshiina has joined #openstack-nova | 15:56 | |
| *** sdake has joined #openstack-nova | 15:56 | |
| *** ijw has joined #openstack-nova | 15:57 | |
| *** jwcroppe has joined #openstack-nova | 15:57 | |
| *** dmk0202 has joined #openstack-nova | 15:58 | |
| auggy | good morning! | 15:59 |
| PaulMurray | morning auggy | 15:59 |
| dansmith | what's so good about it? | 15:59 |
| *** shaohe_feng has quit IRC | 16:00 | |
| *** shaohe_feng has joined #openstack-nova | 16:01 | |
| mdbooth | dansmith: diana_clarke and I spoke about the incremental thing from Friday earlier. I've been having a slightly closer look, and I *think* we might be able to do it if we switch it round. | 16:01 |
| dansmith | mdbooth: okay cool | 16:02 |
| mdbooth | i.e. Update driver.py to call the new interfaces first, and implement the new interfaces using the old interface. | 16:02 |
| mdbooth | Then replace the old interface incrementally. | 16:02 |
| dansmith | ah, okay | 16:02 |
| mdbooth | However, I haven't started writing the actual code yet, so I may yet stumble. | 16:02 |
| mdbooth | Reason is that by updating driver.py first you've already got the extra required context from the caller. | 16:03 |
| mriedem | danpb: that refactor isn't as clean as i initially though - the method args to guest.migrate are used to determine which of the 3 libvirt migrateToURL methods to use :( | 16:03 |
| *** jwcroppe has quit IRC | 16:03 | |
| mriedem | s/though/thought/ | 16:03 |
| mdbooth | Anyway, will give it a go tomorrow, and will proceed if it's not going to be a huge amount of extra work. | 16:03 |
| dansmith | mdbooth: makes sense | 16:03 |
| *** vtech has quit IRC | 16:04 | |
| *** Daisy has quit IRC | 16:04 | |
| mriedem | i guess i can pass a flag into guest.migrate to tell it which to use, or what's available | 16:05 |
| *** mgoddard_ has joined #openstack-nova | 16:06 | |
| *** dmk0202 has quit IRC | 16:06 | |
| *** mgoddard has quit IRC | 16:06 | |
| *** bnemec has joined #openstack-nova | 16:08 | |
| *** shaohe_feng has quit IRC | 16:11 | |
| dansmith | doffm: around? | 16:12 |
| openstackgerrit | Paul Carlton proposed openstack/nova: Permit Live Migration of Rescued Instances https://review.openstack.org/308198 | 16:12 |
| *** shaohe_feng has joined #openstack-nova | 16:14 | |
| *** josecastroleon has quit IRC | 16:20 | |
| *** shaohe_feng has quit IRC | 16:21 | |
| *** ssurana has joined #openstack-nova | 16:22 | |
| *** shaohe_feng has joined #openstack-nova | 16:22 | |
| *** ssurana has quit IRC | 16:22 | |
| *** josecastroleon has joined #openstack-nova | 16:23 | |
| *** jaosorior has quit IRC | 16:23 | |
| *** jaosorior has joined #openstack-nova | 16:24 | |
| *** nic has joined #openstack-nova | 16:24 | |
| *** Daisy has joined #openstack-nova | 16:25 | |
| *** ssurana has joined #openstack-nova | 16:25 | |
| openstackgerrit | Paul Carlton proposed openstack/python-novaclient: [microversion] Bump to 2.31 https://review.openstack.org/316026 | 16:25 |
| mdbooth | Has anybody else seen weird errors related to gettext and cinderclient recently? | 16:25 |
| *** sudipto has quit IRC | 16:25 | |
| *** rk4n has joined #openstack-nova | 16:26 | |
| mriedem | mdbooth: you probably need to rebase | 16:26 |
| mriedem | that was fixed last week i think | 16:26 |
| mdbooth | mriedem: I'm all up to date :/ | 16:26 |
| *** rama_y has joined #openstack-nova | 16:26 | |
| mdbooth | Got a link? | 16:26 |
| diana_clarke | mdbooth: and -r when you tox | 16:26 |
| mdbooth | diana_clarke: -r ? Does that blow away the venv? | 16:26 |
| *** woodster_ has joined #openstack-nova | 16:26 | |
| *** ssurana has quit IRC | 16:26 | |
| mdbooth | If so, I did that manually. | 16:27 |
| *** mat128 is now known as mat128|lunch | 16:27 | |
| mdbooth | A bunch of times, just to be sure :) | 16:27 |
| mriedem | mdbooth: http://lists.openstack.org/pipermail/openstack-dev/2016-May/096190.html i think | 16:27 |
| mdbooth | mriedem: Thanks, looks related | 16:28 |
| * mdbooth will dig | 16:28 | |
| *** ssurana has joined #openstack-nova | 16:28 | |
| mriedem | yay more / new live migration failures now http://logs.openstack.org/04/325204/2/experimental/gate-tempest-dsvm-multinode-live-migration/2b050a0/logs/screen-n-cpu.txt.gz?level=TRACE#_2016-06-06_16_07_04_867 | 16:28 |
| mriedem | Live Migration failure: Operation not supported: Selecting disks to migrate is not implemented for tunnelled migration | 16:28 |
| *** rk4n_ has quit IRC | 16:28 | |
| *** diga has joined #openstack-nova | 16:29 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Permit Live Migration of Rescued Instances https://review.openstack.org/308198 | 16:29 |
| *** vishwanathj has joined #openstack-nova | 16:30 | |
| *** ijw has quit IRC | 16:30 | |
| *** Daisy has quit IRC | 16:30 | |
| *** stewie925 has joined #openstack-nova | 16:30 | |
| doffm | dansmith: Yep. | 16:31 |
| *** shaohe_feng has quit IRC | 16:31 | |
| *** shaohe_feng has joined #openstack-nova | 16:32 | |
| *** fragatina has quit IRC | 16:33 | |
| diga | auggy: Hi | 16:33 |
| *** artom has quit IRC | 16:33 | |
| dansmith | doffm: maybe this is a gap in my (weak) understanding, but are you saying that if you are inside a transaction and go to insert a row that will conflict with something in another transaction, you'll fail? | 16:33 |
| *** tesseract has quit IRC | 16:33 | |
| dansmith | doffm: I thought you'd only fail when you commit | 16:33 |
| *** pcaruana has quit IRC | 16:34 | |
| *** Apoorva has joined #openstack-nova | 16:35 | |
| doffm | dansmith: No, I guess you are right. Nothing is actually going to the db in these functions. | 16:35 |
| *** matrohon has quit IRC | 16:35 | |
| doffm | Well, we are building a transaction that gets sent to the db. | 16:35 |
| stewie925 | hi auggy good morning | 16:35 |
| dansmith | doffm: right, but your save() won't raise the duplicateentry, right? the commit will? | 16:35 |
| doffm | Which means the exceptions have to go outside of the function (and transaction wrapper). | 16:36 |
| doffm | Yeah the commit will. | 16:36 |
| dansmith | right, exactly | 16:36 |
| mriedem | kashyap: looks like you have some experience with this https://bugs.launchpad.net/nova/+bug/1589591 | 16:36 |
| openstack | Launchpad bug 1589591 in OpenStack Compute (nova) "gate-tempest-dsvm-multinode-live-migration fails with "Live Migration failure: Operation not supported: Selecting disks to migrate is not implemented for tunnelled migration"" [High,Confirmed] | 16:36 |
| mriedem | kashyap: there is an older related bug | 16:36 |
| *** ihrachys has quit IRC | 16:36 | |
| doffm | dansmith: Isn't that a bug that could have been introduced in a ton of places by the db changes? | 16:36 |
| *** ihrachys has joined #openstack-nova | 16:37 | |
| *** vishwana_ has joined #openstack-nova | 16:37 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Permit Live Migration of Rescued Instances https://review.openstack.org/308198 | 16:37 |
| dansmith | doffm: you mean because of the switch to enginefacade? | 16:37 |
| dansmith | doffm: I think I said that in my review right? | 16:37 |
| doffm | dansmith: Yeah. | 16:37 |
| dansmith | if not, I thought it :) | 16:37 |
| doffm | You did. | 16:37 |
| doffm | Let me go off and write a test to check this. | 16:38 |
| dansmith | cool | 16:38 |
| rlrossit | mriedem: what's the kosher way of getting a traceback in nova? | 16:38 |
| *** ijw has joined #openstack-nova | 16:38 | |
| mriedem | rlrossit: LOG.exception? | 16:38 |
| rlrossit | mriedem: nah like hunting down a trace | 16:39 |
| rlrossit | using inspect or traceback? | 16:39 |
| rlrossit | (looking at https://review.openstack.org/#/c/310700/15) | 16:39 |
| mriedem | rlrossit: https://review.openstack.org/#/c/274714/11/nova/objects/instance.py ? | 16:39 |
| *** lpetrut has quit IRC | 16:39 | |
| *** vishwanathj has quit IRC | 16:40 | |
| rlrossit | now I remember where I was doing this stuff... it was in the cinder exception wrapping | 16:40 |
| rlrossit | that uses sys.exc_info() | 16:41 |
| *** mtanino_ has joined #openstack-nova | 16:41 | |
| *** sudipto has joined #openstack-nova | 16:41 | |
| *** shaohe_feng has quit IRC | 16:41 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Permit Live Migration of Rescued Instances https://review.openstack.org/308198 | 16:42 |
| *** mtanino has quit IRC | 16:42 | |
| *** shaohe_feng has joined #openstack-nova | 16:42 | |
| *** vtech has joined #openstack-nova | 16:44 | |
| *** dimtruck is now known as zz_dimtruck | 16:45 | |
| *** nic has quit IRC | 16:46 | |
| mriedem | paul-carlton: do you know anything about this libvirt.live_migration_tunnelled flag? https://github.com/openstack/nova/blob/master/nova/conf/libvirt.py#L98 | 16:47 |
| mriedem | the live and block migrate options are deprecated in favor of that option | 16:48 |
| mriedem | and the option help says, "If not set, Nova will choose a sensible default" | 16:48 |
| mriedem | but looking at the code, it looks like if it's not set, it doesn't do anything | 16:48 |
| mriedem | https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L559 | 16:49 |
| *** tonytan4ever has quit IRC | 16:49 | |
| *** artom has joined #openstack-nova | 16:49 | |
| paul-carlton2 | mriedem, not a lot, I think we don't set it | 16:50 |
| mriedem | markmc: ^ looks like you added that? https://review.openstack.org/#/c/263434/ | 16:50 |
| dansmith | mriedem: yeah, the removal of caring about that flag happened recently I think | 16:51 |
| dansmith | trying to remember who that was | 16:51 |
| doffm | dansmith: Turns out I already had a test. https://review.openstack.org/#/c/296597/12/nova/tests/unit/objects/test_aggregate.py "test_aggregate_host_add_to_db_duplicate" | 16:51 |
| doffm | Still looking in to the issue. | 16:51 |
| *** shaohe_feng has quit IRC | 16:52 | |
| *** vishwanathj has joined #openstack-nova | 16:52 | |
| mriedem | dansmith: ok, hrm. i just know our live migration job is busted right now because we're specifying that tunneled flag and passing device names, which makes libvirt puke | 16:52 |
| *** josecastroleon has quit IRC | 16:52 | |
| *** shaohe_feng has joined #openstack-nova | 16:52 | |
| dansmith | hmm | 16:52 |
| mriedem | and reading the help text on that option makes me think it's a magical fairy unicorn that will solve all of my problems | 16:53 |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: split the glance API path based on config https://review.openstack.org/316007 | 16:54 |
| mriedem | i can easily work around this by popping out the tunneled flag when we're passing device_names to migrateToURI3, but that seems wrong if the config told us to do tunneled migration | 16:55 |
| *** vishwana_ has quit IRC | 16:55 | |
| mriedem | i guess alternatively, we don't pass the device names if we're configured to use tunnelled mode | 16:55 |
| *** tbachman has quit IRC | 16:55 | |
| *** stewie925 has quit IRC | 16:55 | |
| *** ihrachys has quit IRC | 16:55 | |
| *** fragatina has joined #openstack-nova | 16:55 | |
| *** vishwanathj has quit IRC | 16:58 | |
| *** mat128|lunch is now known as mat128 | 16:58 | |
| *** tbachman has joined #openstack-nova | 16:59 | |
| *** jwcroppe has joined #openstack-nova | 16:59 | |
| *** xenogear_ has joined #openstack-nova | 17:00 | |
| *** xenogear_ is now known as xenogear | 17:01 | |
| *** stewie925 has joined #openstack-nova | 17:01 | |
| *** shaohe_feng has quit IRC | 17:02 | |
| *** nic has joined #openstack-nova | 17:03 | |
| *** jwcroppe has quit IRC | 17:04 | |
| *** shaohe_feng has joined #openstack-nova | 17:05 | |
| *** amotoki has quit IRC | 17:05 | |
| *** penick has joined #openstack-nova | 17:06 | |
| jaypipes | alaski: https://review.openstack.org/#/c/325436/ | 17:07 |
| *** nic has quit IRC | 17:07 | |
| *** nic has joined #openstack-nova | 17:08 | |
| kashyap | mriedem: Hey, was on a train, looking now... | 17:09 |
| kashyap | mriedem: Oh, this one. | 17:10 |
| *** rmart04 has quit IRC | 17:10 | |
| * kashyap reads the context fully before commenting... | 17:11 | |
| *** josecastroleon has joined #openstack-nova | 17:12 | |
| *** shaohe_feng has quit IRC | 17:12 | |
| *** mtanino has joined #openstack-nova | 17:12 | |
| *** shaohe_feng has joined #openstack-nova | 17:13 | |
| *** MikeG451 has joined #openstack-nova | 17:13 | |
| *** mtanino_ has quit IRC | 17:14 | |
| kashyap | mriedem: That error message is expected, and the related bug you linked is appropriate. | 17:14 |
| kashyap | I'll add a note there | 17:14 |
| *** BobBall is now known as BobBall_AWOL | 17:15 | |
| mriedem | kashyap: ok, i also just replied to the ML thread that was related to mark's changes for the config options, since some of that seems half baked but i'm not sure | 17:16 |
| kashyap | mriedem: The secure tunnelled migration case will be handled when Gate uses QEMU 2.7 (?) -- which has danpb's TLS migration work built into QMEU. | 17:17 |
| *** jwcroppe has joined #openstack-nova | 17:18 | |
| mriedem | we're at 2.5 right now in the ci job | 17:18 |
| mriedem | with xenial | 17:18 |
| *** Apoorva_ has joined #openstack-nova | 17:18 | |
| kashyap | mriedem: Setting this 'live_migration_tunnelled=false' should solve the Gate failure. | 17:20 |
| *** jaosorior has quit IRC | 17:20 | |
| mriedem | kashyap: yeah because it will pop out the tunnelled flag | 17:20 |
| kashyap | mriedem: Yep. | 17:20 |
| mriedem | but it would be nice if we could do something better in the code | 17:20 |
| kashyap | We should unset it, until we use the right QEMU version. | 17:21 |
| mriedem | especially since the live_migrate_tunnelled help text says it will make the right decision for you | 17:21 |
| *** setuid has left #openstack-nova | 17:21 | |
| mriedem | if you don't set it i mean | 17:21 |
| kashyap | mriedem: It will make the right decision when we use the QEMU version which has native encryption | 17:21 |
| danpb | mriedem: whats the problem we're seeing ? | 17:22 |
| *** Apoorva has quit IRC | 17:22 | |
| kashyap | danpb: https://bugs.launchpad.net/nova/+bug/1589591 | 17:22 |
| openstack | Launchpad bug 1589591 in OpenStack Compute (nova) "gate-tempest-dsvm-multinode-live-migration fails with "Live Migration failure: Operation not supported: Selecting disks to migrate is not implemented for tunnelled migration"" [High,Confirmed] | 17:22 |
| *** shaohe_feng has quit IRC | 17:22 | |
| danpb | oh that sucks | 17:23 |
| mriedem | ha | 17:23 |
| kashyap | danpb: It is coming from the "selective block device migration" work (93a19e28 in libvirt) | 17:23 |
| kashyap | ("It" == the error) | 17:24 |
| danpb | mriedem: oh i see, this problem is complex set of moving parts | 17:24 |
| danpb | there are 2 ways in which qemu can do block migration | 17:24 |
| danpb | the old deprecated way was to include it with the migration data stream | 17:24 |
| danpb | the new preferred way is to use an NBD server | 17:25 |
| *** catintheroof has joined #openstack-nova | 17:25 | |
| danpb | the old way worked with tunneled and non-tunelled migration | 17:25 |
| danpb | the new way only works with non-tunnelled migration | 17:25 |
| mriedem | new way is also == migrateToURI3? | 17:25 |
| danpb | now, we added ability to selectively migrate disk, but this only works with the new NBD way | 17:25 |
| *** shaohe_feng has joined #openstack-nova | 17:25 | |
| danpb | no, this is nothing todo with choice of libvirt APIs | 17:26 |
| danpb | its all about internal impl | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Add 'download' method to GlanceImageServiceV2 https://review.openstack.org/321167 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Add 'show' method to GlanceImageServiceV2 https://review.openstack.org/321166 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: [DNM] Test glance v2 in Nova https://review.openstack.org/321551 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Add 'create' method to GlanceImageServiceV2 https://review.openstack.org/321170 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Split the glance API path based on config https://review.openstack.org/316007 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Add 'update' method to GlanceImageServiceV2 https://review.openstack.org/321171 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Make Hyper-V to work with glance v2 api https://review.openstack.org/324670 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Add 'delete' method to GlanceImageServiceV2 https://review.openstack.org/321168 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Make Xenplugin to work with glance v2 api https://review.openstack.org/324536 | 17:26 |
| openstackgerrit | Mike Fedosin proposed openstack/nova: Add 'detail' method to GlanceImageServiceV2 https://review.openstack.org/321169 | 17:26 |
| danpb | so basically nova config has asked for tunnelled migration, which forces libvirt to use the old block migration approach | 17:26 |
| danpb | but nova has also started using select disk migration, which doesn;'t work with old block migration | 17:26 |
| mriedem | i'm with you so far | 17:27 |
| mriedem | we're broken out of the box by default, i like it | 17:27 |
| danpb | yeah, fun | 17:27 |
| danpb | i'm just going to look at the logic nova is using | 17:27 |
| danpb | as it was not supposed to use the selective disk migration unless it has checked it will actually wokr | 17:28 |
| *** sambetts is now known as sambetts|afk | 17:28 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Revert cell DB inventory/allocations migrations https://review.openstack.org/325436 | 17:28 |
| *** diogogmt has joined #openstack-nova | 17:29 | |
| *** vtech has quit IRC | 17:29 | |
| kashyap | mriedem: For later bed-time reading, here's a tiny wiki page that tries to describe the 'new' way -- http://wiki.libvirt.org/page/NBD_storage_migration | 17:30 |
| danpb | mriedem: looks to me like _live_migration_copy_disk_paths() is only checking the libvirt version when deciding whether it cna do selective disk migration | 17:30 |
| danpb | mriedem: if we augment that method to also check if tunnelled migration is requested, then we can skip the selective disk migration | 17:30 |
| kashyap | Ah-ha | 17:33 |
| *** shaohe_feng has quit IRC | 17:33 | |
| mriedem | danpb: ok, i was looking at that one too | 17:33 |
| *** shaohe_feng has joined #openstack-nova | 17:33 | |
| mriedem | danpb: if device_names is empty, should we omit 'migrate_disks' in params to migrateToURI3? | 17:34 |
| mriedem | or will libvirt ignore an empty list? | 17:34 |
| mriedem | probably best to omit i'm guessing | 17:34 |
| danpb | probably safest to omit it, as you never know if libvirt will get stricter in future | 17:34 |
| mriedem | ok, i'll work on a patch this afternoon then | 17:34 |
| * mriedem didn't realize today was going to turn into an all things live migration day | 17:34 | |
| danpb | any day which involves live migration turns into an "all things live migration day" no matter how small it starts out :-) | 17:35 |
| *** nic has quit IRC | 17:35 | |
| mriedem | true. thanks for the help. | 17:35 |
| *** sudipto has quit IRC | 17:36 | |
| *** sudipto has joined #openstack-nova | 17:37 | |
| kashyap | mriedem: In the spirit of 'live migration day', you're also going to add a comment with some analysis there, for the benefit of those who later come to study it and might get befuddled... | 17:38 |
| *** raunak has quit IRC | 17:40 | |
| *** raunak has joined #openstack-nova | 17:41 | |
| *** rahuls has joined #openstack-nova | 17:42 | |
| *** josecastroleon has quit IRC | 17:42 | |
| artom | Did someone say live migration day? Means I can totally grovel for reviews on https://review.openstack.org/#/c/325361/ | 17:42 |
| *** diga has quit IRC | 17:43 | |
| tdurakov | mriedem, want to discuss you comment https://review.openstack.org/#/c/252850/10/nova/compute/manager.py | 17:43 |
| *** raunak has quit IRC | 17:43 | |
| tdurakov | before submitting new ps | 17:43 |
| *** shaohe_feng has quit IRC | 17:43 | |
| mriedem | kashyap: for sure | 17:43 |
| mriedem | tdurakov: sure | 17:43 |
| *** browne has joined #openstack-nova | 17:43 | |
| *** MikeG451 has quit IRC | 17:44 | |
| *** yamahata has quit IRC | 17:44 | |
| tdurakov | log.debug is proposed by sahid, as it's up to user to properly prepare quest agent, so It's ok to use debug here, to not to spam into logs, what do you think? | 17:44 |
| *** daemontool has quit IRC | 17:44 | |
| *** shaohe_feng has joined #openstack-nova | 17:44 | |
| tdurakov | mriedem: ^ | 17:44 |
| tdurakov | s/quest/guest | 17:44 |
| *** stewie925 has quit IRC | 17:45 | |
| *** lucas-afk has quit IRC | 17:45 | |
| *** raunak has joined #openstack-nova | 17:46 | |
| *** sajeesh has quit IRC | 17:46 | |
| *** vishwanathj has joined #openstack-nova | 17:47 | |
| *** zenoway has quit IRC | 17:48 | |
| *** harlowja has joined #openstack-nova | 17:48 | |
| *** daemontool has joined #openstack-nova | 17:48 | |
| mriedem | tdurakov: i guess that's ok | 17:48 |
| tdurakov | mriedem: cool, will resubmit change soon | 17:49 |
| mriedem | the user will get the error back from the api anyway, or should' | 17:49 |
| mriedem | with the right detail | 17:49 |
| *** sdake_ has joined #openstack-nova | 17:49 | |
| *** sdake has quit IRC | 17:49 | |
| tdurakov | mriedem: agree on second comment, just wanted to make sure about log levels | 17:49 |
| mriedem | ok | 17:50 |
| *** lucasagomes has joined #openstack-nova | 17:52 | |
| *** rmart04 has joined #openstack-nova | 17:53 | |
| *** shaohe_feng has quit IRC | 17:53 | |
| *** josecastroleon has joined #openstack-nova | 17:53 | |
| *** nic has joined #openstack-nova | 17:53 | |
| *** shaohe_feng has joined #openstack-nova | 17:54 | |
| *** daemontool has quit IRC | 17:54 | |
| *** daemontool has joined #openstack-nova | 17:55 | |
| *** jlanoux has quit IRC | 17:57 | |
| *** danpb has quit IRC | 17:58 | |
| *** lucasagomes has quit IRC | 17:59 | |
| *** zz_dimtruck is now known as dimtruck | 18:00 | |
| *** sdake_ has quit IRC | 18:01 | |
| tdurakov | mriedem: what `safe` field in NovaException is used for? | 18:03 |
| *** shaohe_feng has quit IRC | 18:03 | |
| *** Apoorva has joined #openstack-nova | 18:04 | |
| *** shaohe_feng has joined #openstack-nova | 18:04 | |
| *** lucasagomes has joined #openstack-nova | 18:04 | |
| openstackgerrit | Matthew Booth proposed openstack/nova: Don't overwrite config disk when using Rbd https://review.openstack.org/326068 | 18:05 |
| openstackgerrit | Matthew Booth proposed openstack/nova: Don't chown a config disk which already exists https://review.openstack.org/326069 | 18:05 |
| openstackgerrit | Matthew Booth proposed openstack/nova: Only chown console log in rescue https://review.openstack.org/326070 | 18:05 |
| openstackgerrit | Karen Bradshaw proposed openstack/nova: api-ref: os-volumes.inc https://review.openstack.org/320048 | 18:05 |
| *** vishwanathj has quit IRC | 18:06 | |
| tdurakov | it should mean it's safe to expose exception to client, right? | 18:06 |
| mdbooth | dansmith: FYI https://review.openstack.org/326068 Just in case you were thinking of writing it yourself. If not don't worry because Jenkins hasn't looked yet. | 18:06 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Revert inventory/allocation child DB linkage https://review.openstack.org/326071 | 18:06 |
| jaypipes | dansmith: ^^ | 18:06 |
| *** vishwanathj has joined #openstack-nova | 18:06 | |
| dansmith | mdbooth: ah, dang, I was going to do that and forgot | 18:07 |
| dansmith | mdbooth: thanks | 18:07 |
| *** Apoorva_ has quit IRC | 18:08 | |
| openstackgerrit | Clif Houck proposed openstack/nova-specs: Spec for supporting authentication methods based on image metadata https://review.openstack.org/326073 | 18:08 |
| clif_h | if anyone can take a look at that spec, and rip it to shreds for me I'd appreciate it, it's pretty short, as I'm mainly looking for feedback at this point | 18:09 |
| *** fragatina has quit IRC | 18:09 | |
| EmilienM | when I read http://docs.openstack.org/developer/nova/man/nova-cert.html#description - it says, it's only used when using EC2 Api, but I've seen in nova/api/openstack/compute/certificates.py that nova-cert is still used. So my question is, do we still need nova-cert even if EC2 API is not deployed? | 18:10 |
| mriedem | EmilienM: good question, i wondered the same a few weeks ago. as far as i can tell, the ec2 api was the only user of nova-cert, and actually has code to go directly to the nova-cert rpc api | 18:11 |
| mriedem | EmilienM: i think the answer is if you're not using ec2 you don't care about nova-cert | 18:11 |
| mriedem | because nothing else in nova relies on nova-cert | 18:11 |
| *** fragatina has joined #openstack-nova | 18:11 | |
| EmilienM | mriedem: ok good. I thought it might be used by nova to manage keypairs | 18:11 |
| *** fragatina has quit IRC | 18:11 | |
| *** rmart04 has quit IRC | 18:12 | |
| *** paul-carlton2 has quit IRC | 18:12 | |
| *** Sukhdev has quit IRC | 18:12 | |
| *** cdent has quit IRC | 18:12 | |
| *** rmart04 has joined #openstack-nova | 18:12 | |
| EmilienM | mriedem: I'm going to submit a patch in devstack to check that dropping it does not break anything | 18:13 |
| mriedem | EmilienM: nope, nova.compute.api.KeypairAPI does that | 18:13 |
| mriedem | EmilienM: ok, good idea | 18:13 |
| EmilienM | cool! thanks for quick reply | 18:13 |
| *** armax has joined #openstack-nova | 18:13 | |
| *** shaohe_feng has quit IRC | 18:14 | |
| *** shaohe_feng has joined #openstack-nova | 18:15 | |
| mriedem | tdurakov: looks like it's used in the FaultWrapper class | 18:15 |
| mriedem | to add an explanation to the fault if it's safe | 18:15 |
| tdurakov | mriedem: yup, so | 18:15 |
| tdurakov | i think it's safe to expose both new exceptions | 18:16 |
| mriedem | tdurakov: honestly i'd never seen that before you asked | 18:16 |
| mriedem | it seems we should default safe=True and then mark the few that aren't safe as False | 18:16 |
| mriedem | but that's less secure by default | 18:16 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/nova: Implement setup_networks_on_host for Neutron networks https://review.openstack.org/275073 | 18:17 |
| mriedem | hence the note | 18:17 |
| mriedem | # NOTE(eglynn): However, it would be over-conservative and | 18:17 |
| mriedem | # inconsistent with the EC2 API to hide every exception, | 18:17 |
| mriedem | # including those that are safe to expose, see bug 1021373 | 18:17 |
| openstack | bug 1021373 in OpenStack Compute (nova) essex "[nova][volumes] Exceeding volumes, gigabytes and floating_ips quotas returns general uninformative HTTP 500 error" [Medium,Fix released] https://launchpad.net/bugs/1021373 - Assigned to Eoghan Glynn (eglynn) | 18:17 |
| mriedem | so i guess we just whitelist whatever is safe | 18:18 |
| *** jichen has joined #openstack-nova | 18:18 | |
| mriedem | tdurakov: in the case of the admin password stuff, those will come back on the rest api as a 409 | 18:18 |
| mriedem | so the fault is less important | 18:18 |
| *** daemontool has quit IRC | 18:19 | |
| *** yamahata has joined #openstack-nova | 18:19 | |
| jaypipes | bauzas, alaski: https://review.openstack.org/326071 | 18:19 |
| tdurakov | mriedem: I've set safe = True for InstanceAgentNotEnabled, SetAdminPasswdNotSupported | 18:19 |
| *** sdake has joined #openstack-nova | 18:20 | |
| tdurakov | mriedem: are you ok with that? | 18:20 |
| openstackgerrit | jichenjc proposed openstack/nova: Prevent boot if ephemeral disk size > flavor value https://review.openstack.org/213762 | 18:20 |
| alaski | jaypipes: added to the queue | 18:21 |
| mriedem | tdurakov: yeah i think that's safe | 18:21 |
| mriedem | get it?! | 18:21 |
| tdurakov | mriedem, local tox -efunctional still in progress | 18:22 |
| *** cdent has joined #openstack-nova | 18:22 | |
| tdurakov | jaypipes: is it revert you've mentioned in ml thread? | 18:23 |
| *** sdake_ has joined #openstack-nova | 18:23 | |
| *** josecastroleon has quit IRC | 18:23 | |
| *** tinbar has joined #openstack-nova | 18:23 | |
| *** shaohe_feng has quit IRC | 18:24 | |
| cdent | jaypipes: a) did you make any headway on your attack on the inventory object? b) I'm happy if you want to change s/generation/etag/ on g-r-p but did you see the questions about whether a view marker is needed on POST of a new inventory or allocation? | 18:24 |
| *** sdake has quit IRC | 18:24 | |
| *** READ10 has quit IRC | 18:24 | |
| cdent | (I ask because I'm going away for a little while and will probably come back with a couple hours of code mind.) | 18:25 |
| *** shaohe_feng has joined #openstack-nova | 18:25 | |
| *** PaulMurray has quit IRC | 18:25 | |
| EmilienM | mriedem: I have another (interesting) question: do nova-consoleauth & nova-vncproxy need to be collocated on same nodes? | 18:26 |
| *** daemontool has joined #openstack-nova | 18:27 | |
| mriedem | EmilienM: i don't know about that one | 18:28 |
| *** pcaruana has joined #openstack-nova | 18:29 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: [libvirt] Incorrect parameters passed to migrateToURI3 https://review.openstack.org/325204 | 18:30 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: cleanup Guest.migrate signature https://review.openstack.org/326082 | 18:30 |
| jaypipes | cdent: a) yes, running tests now. b) ETag instead of generation cool with me, though what should I do for the generation field for the GET /resource_providers response? i.e. each provider in the response has its own generation... | 18:30 |
| jaypipes | cdent: and yes, I did not include the generation in the POST /.../inventories call because there's no need; the server would just raise an exception if the same resource class already existed in the provider's inventory. | 18:31 |
| EmilienM | mriedem: I think no, they communicate via rpc (oslo messaging) but I'm not sure | 18:32 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Create Instance from BuildRequest if not in a cell https://review.openstack.org/263927 | 18:33 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Raise exception if BuildRequest deleted twice https://review.openstack.org/322324 | 18:33 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Cells: Handle delete with BuildRequest https://review.openstack.org/325985 | 18:33 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Do not use instance.id in the name property https://review.openstack.org/326083 | 18:33 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Handle SetAdminPasswdNotSupported raised by libvirt driver https://review.openstack.org/252850 | 18:34 |
| dansmith | jaypipes: cdent ETags are intended to be opaque to the client right/ | 18:34 |
| cdent | jaypipes: a) hawt, b) a common thing to do in that case is to etag the entire collection with an etag that is solely for the caching, and then include an etag attribute in each individual item in the collection list, which is almost going right back to generation, but life is weird | 18:34 |
| *** shaohe_feng has quit IRC | 18:34 | |
| cdent | dansmith: if by opaque you mean: something you don't know until you get it, then yes, but if you mean something else, what do you mean? | 18:34 |
| dansmith | cdent: right, so don't you need an ordering or to maintain a history of tags on the server side? | 18:35 |
| *** shaohe_feng has joined #openstack-nova | 18:35 | |
| cdent | yeah, no history required | 18:35 |
| dansmith | cdent: i.e. if I have etag foo and someone else updates it to bar, then my foo looks different and I regress when I should get a 409 so I check the latest right? | 18:35 |
| cdent | s | 18:36 |
| cdent | yes | 18:36 |
| *** sudipto has quit IRC | 18:36 | |
| dansmith | I guess the server can order it the client just can't know it's ordered, is that your point? | 18:36 |
| cdent | yes | 18:36 |
| dansmith | and do this instead of generation in the object why? | 18:36 |
| dansmith | the REST object I eman | 18:37 |
| cdent | basically so you can 409 sooner if you want to, and maybe even elsewhere in the stack | 18:37 |
| cdent | as well as use the etag as a cache management tool in clients | 18:38 |
| dansmith | so wait, | 18:38 |
| *** josecastroleon has joined #openstack-nova | 18:38 | |
| dansmith | the generation was going to be that the client bumped the generation when it updated, | 18:38 |
| dansmith | but this would be "here's the last value I knew it to be" | 18:38 |
| *** rmart04 has quit IRC | 18:39 | |
| cdent | when I send a new resource to the server, it needs to have the same generation as what's currently there, if it is, at the end of the transaction, the generation is raised | 18:39 |
| cdent | if it is not, 409 | 18:39 |
| cdent | the client should never change the etag/generation | 18:39 |
| dansmith | I thought the point was to bump it from the client? | 18:39 |
| dansmith | hmm | 18:39 |
| jaypipes | dansmith: no, the client does not bump the generation. it *sends* the generation to the server as an indicator of what its view of the state is. the server increments the generation when updating the database. | 18:39 |
| cdent | not as I understand it | 18:39 |
| cdent | so in my mind, if the client sends the etag, at the very beginning of the api call it has to load the resource provider id'd by the given URI. If the etag and the generation don't match, we immediately 409, before ever decoding the body, or doing anything else | 18:41 |
| *** lpetrut has joined #openstack-nova | 18:41 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: cleanup Guest.migrate signature https://review.openstack.org/326082 | 18:41 |
| dansmith | cdent: yeah, okay | 18:41 |
| tdurakov | mriedem: just read thread about migration flags | 18:41 |
| cdent | s/api call it/api call the server/ | 18:41 |
| tdurakov | actually it's libvirt 1.3.1: http://logs.openstack.org/04/325204/2/experimental/gate-tempest-dsvm-multinode-live-migration/2b050a0/logs/dpkg-l.txt.gz | 18:41 |
| Swami | mriedem: https://review.openstack.org/#/c/286855/ https://review.openstack.org/#/c/275073/ the tempest patch has been fixed to address the live migration. Can you bless the nova patch associated with it. | 18:42 |
| cdent | so we have 403, then 404, then 409, and then only if none of those do we start doing any "work" | 18:42 |
| *** krotscheck has quit IRC | 18:42 | |
| mriedem | tdurakov: oh yeah, i knew that too since i looked this morning | 18:42 |
| *** krotscheck_ has joined #openstack-nova | 18:42 | |
| jaypipes | cdent: mostly correct yes. the server doesn't technically need to re-read from the database at the start of the server-side processing of a request to update the inventory. it adds the Etag/generation to the WHERE clause of the UPDATE statement it does at the very end of the transaction which prevents inconsistent writes. | 18:42 |
| * tdurakov reading the rest | 18:42 | |
| dansmith | jaypipes: right, so we've decoded already anyway | 18:42 |
| jaypipes | ya | 18:43 |
| dansmith | so, IMHO, | 18:43 |
| cdent | jaypipes: I'm not saying it has to, I'm saying it should, to avoid all that hoopa joop, because we want to 404 earlier anyway | 18:43 |
| dansmith | cdent: it *has* to put it in the sql query if you want that fine-grained update locking right? | 18:43 |
| *** krotscheck_ is now known as krotscheck | 18:44 | |
| cdent | dansmith: two ambiguious "it"s there | 18:44 |
| cdent | (sorry, I've been doing the same thing) | 18:44 |
| *** shaohe_feng has quit IRC | 18:44 | |
| mfedosin | mriedem: sdague folks, I updated all ps, take a look when you have time https://review.openstack.org/#/c/321551/ | 18:45 |
| *** shaohe_feng has joined #openstack-nova | 18:45 | |
| *** jed56 has quit IRC | 18:45 | |
| dansmith | cdent: really? the generation has to be in the update sql query if you want that fine-grained update locking that jaypipes is going for right/ | 18:45 |
| jaypipes | dansmith: correct. | 18:45 |
| mriedem | mfedosin: ok, will run my test change against it again too | 18:45 |
| dansmith | so we decode anyway, always | 18:45 |
| bauzas | dansmith: cdent: jaypipes: it becomes hard to follow your convo given my TZ, but why are you saying etag ? | 18:45 |
| dansmith | so TBH, it seems more explicit to put it in the body of the PUT to me | 18:45 |
| dansmith | seems more restful to me to make it a data element of the object | 18:46 |
| jaypipes | bauzas: cdent's proposal was to use HTTP ETag instead of returning a generation field in the response to things like GET /resource_providers/{uuid} | 18:46 |
| * dansmith notes he's no expert on restfulness | 18:46 | |
| cdent | dansmith: I'm not sure I'm following on why we decode always? | 18:46 |
| cdent | s/we/you're saying we/ | 18:46 |
| dansmith | cdent: because we don't know if the etag is right or wrong until we are doing the UPDATE call, at which point we've already decoded the body to construct the query | 18:47 |
| bauzas | cdent: dansmith: then, I'm -1 for an ETag | 18:47 |
| bauzas | because it's not a cache related thing | 18:47 |
| dansmith | right | 18:47 |
| dansmith | it's a property of the thing | 18:47 |
| jaypipes | dansmith: I could go either way. But the fact that ETag doesn't work smoothly for collections of resources (i.e. GET /resource_providers) means we need to indicate the generation for each resource provider in the returned list of providers, and a single ETag won't work there. | 18:47 |
| dansmith | (IMHO) | 18:47 |
| dansmith | jaypipes: good point | 18:47 |
| cdent | bauzas: if you get a chance read: https://www.w3.org/1999/04/Editing/ | 18:47 |
| bauzas | agreed with jaypipes here | 18:48 |
| cdent | like I said earlier on, I'm okay with generation | 18:48 |
| bauzas | cdent: well, I was working on Internet since 10 yrs... | 18:48 |
| dansmith | actually7, | 18:48 |
| dansmith | it's the _wrong_ thing to use, | 18:48 |
| dansmith | because if you say the etag of current is foo, and you're PUTing a different thing, with the same etag, that's wrong right? | 18:48 |
| dansmith | you don't know the cache id of the new thing you're updating | 18:49 |
| dansmith | because it's new to the server | 18:49 |
| cdent | dansmith: not if it is "weak etag", which these would be | 18:49 |
| dansmith | weak etag means "semantically equivalent" according to my googling | 18:49 |
| cdent | but really, it's clear this is controversial, so we should just use generation, it's cool. | 18:49 |
| dansmith | which means they are "interchangeable" which would not be the case right? | 18:49 |
| *** zenoway has joined #openstack-nova | 18:51 | |
| cdent | dansmith: the generation or etag is being used to say "I have seen resource provider Z recently and wish to update it" and the request is a representation which is mean to modify the state of the resource provider (even though our proposed uris are not quite aligned with that) | 18:51 |
| dansmith | right | 18:51 |
| *** ducttape_ has joined #openstack-nova | 18:51 | |
| dansmith | providing the etag of the thing you want to replace just seems weird to me, | 18:52 |
| dansmith | which is, FWIW, why I was expecting the client would be providing the updated generation field | 18:52 |
| dansmith | ah, I see, If-Known | 18:53 |
| dansmith | er, If-None-Match or whatever | 18:53 |
| cdent | yeah | 18:53 |
| cdent | the headers are all kinds of make-you-squint | 18:53 |
| cdent | I'm perfectly happy if we want to say the generatio attribute is an externally visible attribute of the r-p representation and it is the operator for doing view marking. | 18:54 |
| cdent | that makes it consistent across the collection | 18:54 |
| cdent | as well as the individual items | 18:55 |
| *** shaohe_feng has quit IRC | 18:55 | |
| bauzas | tbc, I prefer to have one explicit field rather than using some implicit one | 18:55 |
| dansmith | I dunno, this is new information to me.. I've only ever known etags in the world of caching, but I still prefer it be a data element | 18:55 |
| sneti | Hi all, I'm facing issues logging into https://wiki.openstack.org/wiki/Main_Page. I'm getting HTTP Error 500. Can anyone help me with this. | 18:56 |
| *** shaohe_feng has joined #openstack-nova | 18:56 | |
| dansmith | it's more explicit, at the least | 18:56 |
| *** zenoway has quit IRC | 18:56 | |
| cdent | dansmith: I think it just seems that way because you're not used to doing this kind of thing with etags. It was only once I realized that says description of generation was pretty much exactly what I think of etags being for in a read-write API that the generation concept landed in any successful fashion in my brain. | 18:57 |
| cdent | s/says/jay's/ | 18:57 |
| *** zenoway has joined #openstack-nova | 18:57 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Handle SetAdminPasswdNotSupported raised by libvirt driver https://review.openstack.org/252850 | 18:57 |
| tdurakov | mriedem: ^ | 18:58 |
| oomichi | mriedem: sdague: can you take a look at https://review.openstack.org/#/c/315964/ ? | 18:58 |
| *** sdake has joined #openstack-nova | 18:58 | |
| oomichi | mriedem: sdague: this seems necessary for stable branches also | 18:58 |
| oomichi | the corresponding tempest patch is https://review.openstack.org/#/c/325510/ | 18:59 |
| bauzas | cdent: AFAIR, ETags are more for static content or things that can't be changed right? | 18:59 |
| cdent | bauzas: wrong | 18:59 |
| tdurakov | folks, who could help me with merging backports to stable liberty/mitaka? | 19:00 |
| *** sdake_ has quit IRC | 19:00 | |
| bauzas | cdent: I'm more used to see that in the context of non-REST resources | 19:00 |
| bauzas | cdent: and the following makes me think the above http://greenbytes.de/tech/webdav/draft-reschke-http-etag-on-write-latest.html#rfc.section.1.1 | 19:00 |
| *** vtech has joined #openstack-nova | 19:00 | |
| *** fragatina has joined #openstack-nova | 19:00 | |
| cdent | webdav is an alien land that has rather little to do with REST | 19:01 |
| dansmith | I think jaypipes' point about the collection result is the strongest point | 19:02 |
| cdent | dansmith: yes, I agree | 19:02 |
| dansmith | I just don't like it because it doesn't seem natural to me, but he has an actual point | 19:02 |
| bauzas | dansmith: I agree, that's another issue with etags | 19:03 |
| bauzas | anyway, /me shrugs | 19:04 |
| *** gszasz has quit IRC | 19:04 | |
| *** shaohe_feng has quit IRC | 19:05 | |
| cdent | bauzas: the document I linked before, https://www.w3.org/1999/04/Editing/ , has a good deal to say about how to use etags with dynamic resources, and I can find some other stuff if you're curious. | 19:05 |
| *** shaohe_feng has joined #openstack-nova | 19:06 | |
| cdent | bauzas: there's also this too http://specs.openstack.org/openstack/api-wg/guidelines/etags.html , but I assumed you'd prefer something that wasn't written by me :) | 19:08 |
| *** hshiina has quit IRC | 19:08 | |
| cdent | the problem described there is all over the place in OpenStack, and the generation thing (whether etags or generation attributes) is trying to fix it. | 19:08 |
| *** kfarr has quit IRC | 19:08 | |
| bauzas | sure, I know the If-Match header | 19:09 |
| bauzas | now, the thing that we have in place is already persisted | 19:09 |
| *** raunak has quit IRC | 19:09 | |
| *** ctrath has joined #openstack-nova | 19:09 | |
| cdent | and we want to update it | 19:09 |
| bauzas | we could say, "okay, we could return that field in an header hashed some way to be an etag" | 19:09 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove code referencing inventory table in cell DB https://review.openstack.org/326100 | 19:09 |
| bauzas | I just wonder what the benefit of that is | 19:10 |
| jaypipes | dansmith: follow-on patch to the revert of the inventory linkage ^^ | 19:10 |
| *** josecastroleon has quit IRC | 19:11 | |
| cdent | bauzas: I can think of two reasons | 19:11 |
| dansmith | jaypipes: okay, FYI I wasn't arguing about that stuff being combined with the first one, just the removal of the C/U_inventory() stuff | 19:11 |
| cdent | one is that it is a standard way of doing such things in REST-like APIs | 19:11 |
| sdague | oomichi: hmmm... why is get_db_instances() not returning an empty list/. | 19:12 |
| jaypipes | dansmith: ya, understood. As I was teasing apart the c/u inventory stuff I ran into that and figured it was big enough for a separate patch (since it wasn't in the DB API or anything) | 19:12 |
| dansmith | jaypipes: cdent: I'm trying to think of whether the collections thing is actually a probem | 19:12 |
| dansmith | jaypipes: okay | 19:12 |
| bauzas | jaypipes: before +Wing https://review.openstack.org/#/c/326071/ I just need to make sure that you didn't reverted https://review.openstack.org/#/c/286931/15 because it could need 2 reverts ? | 19:12 |
| cdent | two is what I was trying to figure out with dansmith above about returning a 409 as early as possible in a layered web services. In the context of the api we're talking about now, it probably doesn't matter, but in some apis you can do etag comparisons very early on in a multi-layered stack | 19:12 |
| cdent | dansmith: could you re-state that again for my benefit please | 19:13 |
| dansmith | cdent: yeah, that's the other thing.. not sure it really buys us anything other than standards compliance | 19:13 |
| dansmith | cdent: I'm just thinking out loud.. that's all there was | 19:13 |
| cdent | dansmith: yeah, the lack of buying us anything is why I capitulated | 19:13 |
| bauzas | cdent: and you suggest to return an hashed value of the generation field as the etag? | 19:13 |
| jaypipes | bauzas: 286931 was only partially reverted... | 19:14 |
| dansmith | bauzas: it doesn't matter | 19:14 |
| sdague | oomichi: yeh, I'm not thrilled about that patch. | 19:14 |
| bauzas | jaypipes: okay, just wanted to make sure | 19:14 |
| cdent | bauzas: we hadn't gotten around to discussing the actual form of the etag, but it no longer matters | 19:14 |
| oomichi | sdague: ah, that is the same as my question. please see my comment of PS1. KeyError: 'instances' happens | 19:14 |
| jaypipes | bauzas: np | 19:15 |
| *** ijw has quit IRC | 19:15 | |
| *** shaohe_feng has quit IRC | 19:15 | |
| oomichi | sdague: that is empty as you expected, but the get_db_items tris to get it with the key"instances" | 19:15 |
| sdague | oomichi: because https://review.openstack.org/#/c/315964/1/nova/api/openstack/compute/servers.py | 19:16 |
| sdague | which is definitely wrong | 19:16 |
| *** shaohe_feng has joined #openstack-nova | 19:16 | |
| sdague | the whole point of that request cache is that it is the list of instances the request is operating on | 19:16 |
| sdague | empty is an important value there | 19:17 |
| *** kfarr has joined #openstack-nova | 19:18 | |
| *** stewie925 has joined #openstack-nova | 19:19 | |
| *** cdent has quit IRC | 19:20 | |
| jlvillal | jaypipes: Regarding https://review.openstack.org/#/c/325436/ Can you post an update with the new patches? I'm trying to track this for our Ironic Grenade work. | 19:22 |
| openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/nova: Remove context object in oslo.log method https://review.openstack.org/317109 | 19:22 |
| *** baoli_ has quit IRC | 19:22 | |
| *** salv-orl_ has quit IRC | 19:24 | |
| *** salv-orlando has joined #openstack-nova | 19:24 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Qcow2: add create_from_image & create_from_func https://review.openstack.org/320610 | 19:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Lvm: add create_from_image & create_from_func https://review.openstack.org/320611 | 19:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Image: add create_from_image & create_from_func https://review.openstack.org/318948 | 19:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Rbd: add create_from_image & create_from_func https://review.openstack.org/317694 | 19:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Add check_backing_from_func & check_backing_from_image https://review.openstack.org/322285 | 19:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Flat: add create_from_image & create_from_func https://review.openstack.org/317701 | 19:25 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Ploop: add create_from_image & create_from_func https://review.openstack.org/317689 | 19:25 |
| *** vishwanathj has quit IRC | 19:25 | |
| *** sheel has quit IRC | 19:25 | |
| *** shaohe_feng has quit IRC | 19:25 | |
| *** shaohe_feng has joined #openstack-nova | 19:26 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Cells: Simple setup/migration command https://review.openstack.org/301273 | 19:29 |
| *** matrohon has joined #openstack-nova | 19:32 | |
| sdague | mfedosin: ok, I went through the whole stack. It looks quite good. The only -1 I've got is on the xenserver patch, because I think we can delete the old upload_vhd/download_vhd as they only work with 1.2 requests, which won't get sent in code newer than Feb. | 19:32 |
| sdague | that will just simplify the xenserver patch if we do that delete as the prep patch for it | 19:33 |
| sdague | it's all also left in review comments, but +2 on the rest of the stack | 19:33 |
| mfedosin | sdague: okay, thanks! I'll update it tomorrow then | 19:34 |
| sdague | mfedosin: great | 19:34 |
| openstackgerrit | Timofey Durakov proposed openstack/nova: [DNM] Check migrateToURI2 https://review.openstack.org/326111 | 19:35 |
| *** shaohe_feng has quit IRC | 19:36 | |
| *** Sukhdev has joined #openstack-nova | 19:37 | |
| *** shaohe_feng has joined #openstack-nova | 19:37 | |
| tdurakov | mriedem: i've submitted patch, to check how it works with migrateToURI2 ^ need to go now, feel free to update it | 19:37 |
| mriedem | thanks | 19:37 |
| *** lpetrut has quit IRC | 19:41 | |
| doffm | zzzeek: Could you help me out by reviewing a change. https://review.openstack.org/#/c/296597/12/nova/objects/aggregate.py I / We have some questions about transactions and the new enginefacade api. | 19:43 |
| *** lpetrut has joined #openstack-nova | 19:43 | |
| doffm | dansmith: ^^ Totally reached my db knowledge limits. I share your concern that the try block should encompass the whole transaction, but the unit test to check this passes. | 19:44 |
| zzzeek | doffm: sure | 19:44 |
| doffm | I think that is because of the nested transactions. | 19:44 |
| doffm | But I'm not sure that this is safe with multiple concurrent transactions. | 19:44 |
| doffm | zzzeek: Thanks. | 19:44 |
| *** baoli has joined #openstack-nova | 19:46 | |
| *** zenoway has quit IRC | 19:46 | |
| *** shaohe_feng has quit IRC | 19:46 | |
| *** egarbade| has quit IRC | 19:46 | |
| *** ekhugen has quit IRC | 19:46 | |
| *** baoli has quit IRC | 19:46 | |
| *** shaohe_feng has joined #openstack-nova | 19:47 | |
| openstackgerrit | Ronald Bradford proposed openstack/nova: api-ref: parameter verification os-agents https://review.openstack.org/326116 | 19:47 |
| openstackgerrit | Ronald Bradford proposed openstack/nova: api-ref: parameter verification os-security-groups https://review.openstack.org/326117 | 19:47 |
| *** zenoway has joined #openstack-nova | 19:47 | |
| *** eharney has quit IRC | 19:49 | |
| *** egarbade- has joined #openstack-nova | 19:50 | |
| zzzeek | doffm: from what I can see here the _aggreagte_get_from_db call only deals with locating an object of type Aggregate. it has no bearing on whether or not an object of type AggregateHost is found | 19:50 |
| jaypipes | jlvillal: done! | 19:50 |
| jlvillal | jaypipes: Thank you! | 19:50 |
| *** ekhugen has joined #openstack-nova | 19:50 | |
| openstackgerrit | Ronald Bradford proposed openstack/nova: api-ref: parameter verification os-fixed-ips https://review.openstack.org/326120 | 19:50 |
| jaypipes | dansmith: so did we settle on anything w.r.t. the ETag thing? | 19:50 |
| *** baoli has joined #openstack-nova | 19:51 | |
| *** dimtruck is now known as zz_dimtruck | 19:51 | |
| *** mtanino has quit IRC | 19:52 | |
| stewie925 | heu auggy | 19:52 |
| stewie925 | hey oops | 19:52 |
| *** zz_dimtruck is now known as dimtruck | 19:52 | |
| *** kylek3h has joined #openstack-nova | 19:52 | |
| doffm | zzzeek: I guess I'm concerned that the 'try: host_ref.save(session) except: db_exc.DBDuplicateEntry' will be bypassed when the transaction is commited. | 19:54 |
| zzzeek | doffm: the duplicate entry exception would be raised before the commit occurs | 19:55 |
| doffm | zzzeek: You are right the _aggregate_get_from_db call only locates and Aggregate, has nothing to do with the AggregateHost. | 19:55 |
| *** openstackstatus has quit IRC | 19:55 | |
| *** raunak has joined #openstack-nova | 19:55 | |
| sdague | dansmith: https://review.openstack.org/#/c/326068 you've got a +1 on that patch waiting for jenkins, which passed. You might want to upgrade that | 19:56 |
| doffm | zzzeek: Ok, thanks. | 19:56 |
| *** mtanino has joined #openstack-nova | 19:56 | |
| *** shaohe_feng has quit IRC | 19:56 | |
| *** openstackstatus has joined #openstack-nova | 19:57 | |
| *** ChanServ sets mode: +v openstackstatus | 19:57 | |
| auggy | stewie925: hello! | 19:57 |
| *** shaohe_feng has joined #openstack-nova | 19:57 | |
| *** josecastroleon has joined #openstack-nova | 19:57 | |
| *** browne1 has joined #openstack-nova | 20:00 | |
| *** browne has quit IRC | 20:00 | |
| *** cdent has joined #openstack-nova | 20:02 | |
| stewie925 | heya how are you? | 20:03 |
| stewie925 | hi auggy :) | 20:03 |
| cdent | jaypipes: re [t 3f7C], I think we decided to use generation not etag | 20:05 |
| purplerbot | <jaypipes> dansmith: so did we settle on anything w.r.t. the ETag thing? [2016-06-06 19:50:55] [n 3f7C] | 20:05 |
| dansmith | aye | 20:05 |
| *** shaohe_feng has quit IRC | 20:06 | |
| cdent | alaski: I don't understand what you mean by "I would've expected this to be tags" on https://review.openstack.org/#/c/281945/ | 20:07 |
| *** shaohe_feng has joined #openstack-nova | 20:07 | |
| *** ducttape_ has quit IRC | 20:07 | |
| alaski | cdent: based on jaypipes comment "I'd be just as happy leaving it nullable as basically a simple tag the user can apply." | 20:08 |
| *** baoli has quit IRC | 20:08 | |
| *** baoli has joined #openstack-nova | 20:08 | |
| alaski | just thinking that calling it 'tag(s)' might resolve the dissonance | 20:08 |
| cdent | the expected names are things like "my ultra special gold plated nfs server" | 20:08 |
| stewie925 | hi auggy - am wondering if there are any nova documentation stuff that I can help out with ? | 20:09 |
| auggy | stewie925: yes! we currently have the api-ref project going on, i'll paste a link | 20:09 |
| *** baoli has quit IRC | 20:09 | |
| auggy | stewie925: https://wiki.openstack.org/wiki/Nova/Mentoring#Fix_API_Documentation | 20:09 |
| *** lpetrut has quit IRC | 20:10 | |
| alaski | cdent: yeah, calling it name makes sense. But calling it name makes me think they should all have a name, if only some do then it feels like there should be a better term. perhaps "description"? | 20:11 |
| alaski | anyways, I'm fine with it as is. I just wanted to leave room for dansmith to reexpress his concern | 20:12 |
| stewie925 | auggy: I see the link | 20:12 |
| cdent | roger that alaski | 20:12 |
| *** mfedosin has quit IRC | 20:12 | |
| stewie925 | going through it now.... | 20:12 |
| *** baoli has joined #openstack-nova | 20:13 | |
| dansmith | alaski: yeah, it should not be nullable, and we probably should have fixed that in the schema when we moved it | 20:14 |
| dansmith | that was my bad, | 20:14 |
| dansmith | but we should at least make the object right | 20:14 |
| *** armax has quit IRC | 20:15 | |
| alaski | is anything in the db yet? can the db schema just be fixed? | 20:15 |
| dansmith | alaski: it could | 20:16 |
| dansmith | alaski: jaypipes says he thinks it's okay to have them be nullable | 20:16 |
| dansmith | one thing I raised in bristol, | 20:16 |
| dansmith | is that I'm really concerned that we're going to lose track of what these things are | 20:16 |
| dansmith | just a list of ten thousand uuids in a db, not knowing whether they were created by neutron, compute nodes, or the operators really makes me worry | 20:16 |
| dansmith | people will be afraid or unable to prune them | 20:16 |
| dansmith | name came later, presumably because of the UX aspect, but my concern remains | 20:17 |
| cdent | dansmith: Does the can_host thing play in this area at all? | 20:17 |
| dansmith | no | 20:17 |
| *** shaohe_feng has quit IRC | 20:17 | |
| dansmith | because neutron is going to be auto-creating these things like crazy | 20:17 |
| dansmith | can_host narrows it down to "is or used to be a compute node" | 20:17 |
| dansmith | but all the other ones are nameless | 20:17 |
| dansmith | I guess you could look at inventories to see whether the thing was from neutron, i.e. only network resources | 20:18 |
| *** shaohe_feng has joined #openstack-nova | 20:18 | |
| alaski | so that raises a question | 20:18 |
| alaski | what's to keep ironic from updating a network resource, accidentally or maliciously? | 20:19 |
| cdent | that's what I was wondering when we talked about can_host: surely resource_class is relevant for these things? | 20:19 |
| dansmith | alaski: hmm | 20:19 |
| EmilienM | mriedem: the nova-cert deletion fails devstack CI because there are still some tests for it, http://logs.openstack.org/77/326077/1/check/gate-tempest-dsvm-full/61d60d1/console.html#_2016-06-06_19_08_33_692031 | 20:19 |
| alaski | it's starting to feel like a concept of ownership is needed | 20:19 |
| * cdent shivers | 20:19 | |
| dansmith | cdent: yeah, but it means to cull orphans, you list all resource providers with a network resource and then hit neutron's api for each to see if it's still a thing? | 20:20 |
| dansmith | cdent: that sucks | 20:20 |
| *** baoli has quit IRC | 20:20 | |
| dansmith | name doesn't fix that, | 20:20 |
| dansmith | but if name is set to something useful it likely will help | 20:20 |
| *** IanGovett has quit IRC | 20:20 | |
| EmilienM | mriedem: I wonder if tempest.api.compute.certificates.test_certificates.CertificatesV2TestJSON.test_create_root_certificate and tempest.api.compute.certificates.test_certificates.CertificatesV2TestJSON.test_get_root_certificate are really something we want to test | 20:20 |
| dansmith | alaski: yeah, that's a really good point.. one compromised service account and you can blow away all the resource information and accounting for the cloud I guess, eh/ | 20:20 |
| cdent | dansmith: that's kind of my point: name isn't really useful for much except ux, so I'm not sure why it shouldn't be nullable, since there are going to lots of resource providers | 20:20 |
| cdent | that's a generic issue with cross-service interaction, n'est pas? | 20:21 |
| dansmith | cdent: if neutron has to provide something, they can use something in there.. or cinder the hostname of the shared storage thing | 20:21 |
| EmilienM | cdent: n'est-ce pas | 20:21 |
| dansmith | cdent: it would be for human correlation for sure though yeah | 20:21 |
| cdent | EmilienM: thanks | 20:22 |
| dansmith | EmilienM: I enjoyed that more than you could possibly know | 20:22 |
| * cdent growls at dansmith | 20:22 | |
| dansmith | heh | 20:22 |
| mriedem | EmilienM: ah, they are until we have a microversion that makes calling os-certificates a 404 | 20:22 |
| * cdent barely passed french | 20:22 | |
| jaypipes | alaski: these are all privileged operations. I think the fact that Ironic could potentially update a neutron ip subnet pool is not something we need to worry too much about right this sec. | 20:22 |
| EmilienM | mriedem: I guess I can abandon my patch until a microversion removes this feature | 20:23 |
| alaski | jaypipes: sure, I don't want to stop this until that's sorted. but it's something we'll want to get to | 20:23 |
| mriedem | EmilienM: i know we approved a spec from sdague for deprecating proxy APIs, but os-certificates wasn't in that list | 20:24 |
| dansmith | alaski: I guess the thing is, allocations can't be edited en-masse from the api with one of those accounts, I think | 20:24 |
| mriedem | we were going to do a separate spec for the random APIs we don't want anymore, like os-certificates, os-agents, os-cloudpipe, etc but i'm not sure if that happened | 20:24 |
| EmilienM | mriedem: the question is, what does this API now we don't have EC2 things anymore | 20:24 |
| EmilienM | cdent: your french is good, you can also fix my english anytime :-P | 20:25 |
| alaski | cdent: dansmith so my take on this is that it would be nice to have an easy way to associate a db provider with the actual provider (neutron, cinder, compute, etc) and name seems like a way to do it. if that's tackled another way then a nullable name is less concerning I think | 20:25 |
| dansmith | alaski: allocations are where you'd do damage.. | 20:25 |
| dansmith | alaski: right exactly | 20:26 |
| cdent | EmilienM: I have more fun fixing your gabbi problems. | 20:26 |
| dansmith | alaski: not for permissions-checking, I hope | 20:26 |
| dansmith | alaski: but setting the name to cinder-netapp04 would be super useful | 20:26 |
| alaski | right | 20:26 |
| dansmith | alaski: or neutron-vlan123 | 20:26 |
| dansmith | alaski: or compute-c123.provider.com | 20:26 |
| cdent | if we do something like that can we at least make it service-type? | 20:26 |
| cdent | not name | 20:26 |
| alaski | to ease the burden of the ops guy digging into the db | 20:26 |
| *** shaohe_feng has quit IRC | 20:27 | |
| dansmith | cdent: it's more than service-type to me | 20:27 |
| *** josecastroleon has quit IRC | 20:27 | |
| *** amotoki has joined #openstack-nova | 20:27 | |
| *** amotoki has quit IRC | 20:27 | |
| cdent | dansmith: I mean instead of neutron-vlan123 -> network-vlan123 | 20:28 |
| *** ducttape_ has joined #openstack-nova | 20:28 | |
| dansmith | cdent: well, it's provided by the caller, so sure, but it's their decision right? | 20:28 |
| *** BjoernT has quit IRC | 20:28 | |
| cdent | ohhhh. I thought we were talking about automation in those cases when a user-agent hasn't provided a name | 20:29 |
| cdent | (when n-thousand something are being created) | 20:29 |
| alaski | more like make them provide a name | 20:29 |
| dansmith | no | 20:30 |
| dansmith | just make them provide a name | 20:30 |
| *** shaohe_feng has joined #openstack-nova | 20:30 | |
| dansmith | no uniqueness check | 20:30 |
| *** ayoung has quit IRC | 20:30 | |
| dansmith | just provide something.. better than nothing | 20:30 |
| * cdent shrugs | 20:31 | |
| cdent | I'M JUST HAPPY TO BE HERE | 20:32 |
| *** dave-mccowan has quit IRC | 20:32 | |
| *** dave-mccowan has joined #openstack-nova | 20:33 | |
| dansmith | alaski: I threw my -1 back on there... feel free to override, but I think we should have a name | 20:33 |
| alaski | that's fine, that's why I didn't +W earlier | 20:34 |
| *** Sukhdev has quit IRC | 20:36 | |
| alaski | dansmith: unrelated but you may enjoy https://review.openstack.org/#/c/326083/ | 20:36 |
| dansmith | alaski: you can't do that can you? | 20:37 |
| dansmith | alaski: you'll break all existing clouds on libvirt? | 20:37 |
| alaski | will it? | 20:37 |
| *** shaohe_feng has quit IRC | 20:37 | |
| alaski | libvirt relies on a property that can change based on a config option? | 20:38 |
| dansmith | alaski: unfortunately, I think so | 20:38 |
| *** shaohe_feng has joined #openstack-nova | 20:38 | |
| dansmith | alaski: it names lvms by instance.name I thnk | 20:38 |
| dansmith | alaski: I think we use uuid to talk to libvirt about the actual instance now, so I think we're okay there, but... | 20:38 |
| alaski | I'm going to have to borrow a gdi from mriedem here | 20:39 |
| dansmith | 4 instance-00000061 running | 20:39 |
| cdent | jaypipes: if you got a counter on the rp name-not-nullable stuff, shout out, otherwise I'll change it tomorrow in my-morning | 20:39 |
| dansmith | we name the libvirt instance by instance.name though, AFAIK | 20:39 |
| alaski | okay | 20:39 |
| openstackgerrit | jichenjc proposed openstack/nova: api-ref: verify os-server-groups.inc https://review.openstack.org/325269 | 20:39 |
| alaski | so what about persisting the name based on what's generated at boot time? | 20:39 |
| *** IanGovett has joined #openstack-nova | 20:39 | |
| dansmith | alaski: well, it could be configured differently on each compute node or something and then you'll break people if you generate it on another node right? | 20:40 |
| dansmith | alaski: basically instance.name formatting is one of the worst things ever done :) | 20:40 |
| alaski | yes, yes it is | 20:41 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Qcow2: add create_from_image & create_from_func https://review.openstack.org/320610 | 20:41 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Lvm: add create_from_image & create_from_func https://review.openstack.org/320611 | 20:41 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Image: add create_from_image & create_from_func https://review.openstack.org/318948 | 20:41 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Rbd: add create_from_image & create_from_func https://review.openstack.org/317694 | 20:41 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Add check_backing_from_func & check_backing_from_image https://review.openstack.org/322285 | 20:41 |
| *** yamahata has quit IRC | 20:41 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Flat: add create_from_image & create_from_func https://review.openstack.org/317701 | 20:41 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Ploop: add create_from_image & create_from_func https://review.openstack.org/317689 | 20:41 |
| *** yamahata has joined #openstack-nova | 20:41 | |
| alaski | if we can ignore the API side of it this is less of an issue because my issue is that I can't return the right name at boot time | 20:41 |
| alaski | I mean for the boot request | 20:41 |
| dansmith | alaski: yeah | 20:42 |
| alaski | so if fudging that is acceptable I can move forward | 20:42 |
| alaski | and keep the same craziness on the compute side | 20:43 |
| dansmith | I think you'll have to | 20:43 |
| dansmith | it's pretty crazy | 20:43 |
| *** ctrath has quit IRC | 20:43 | |
| alaski | yeah, the idea of different values for that on different hosts... | 20:44 |
| dansmith | alaski: if you're doing shared block and you want their lvms named by the compute host they're on .... | 20:45 |
| dansmith | alaski: we could try to evolve people away from that with some sort of migration process, but it's a whole standalone can of worms | 20:46 |
| alaski | yep. I was thinking about a periodic to run on each compute and persist the local name for migrating. but not something I want to dive into now | 20:46 |
| dansmith | yeah | 20:47 |
| dansmith | sounds unfun to me | 20:47 |
| *** shaohe_feng has quit IRC | 20:47 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: cleanup Guest.migrate signature https://review.openstack.org/326082 | 20:48 |
| *** shaohe_feng has joined #openstack-nova | 20:48 | |
| *** josecastroleon has joined #openstack-nova | 20:51 | |
| *** dimtruck is now known as zz_dimtruck | 20:51 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Create Instance from BuildRequest if not in a cell https://review.openstack.org/263927 | 20:52 |
| *** raunak has quit IRC | 20:57 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Delete old imagebackend methods & tests https://review.openstack.org/322974 | 20:57 |
| *** shaohe_feng has quit IRC | 20:58 | |
| *** gjayavelu has joined #openstack-nova | 20:58 | |
| openstackgerrit | Andy Hsiang proposed openstack/nova: Remove DictCompat from SecurityGroup https://review.openstack.org/321109 | 20:59 |
| *** Sukhdev has joined #openstack-nova | 20:59 | |
| *** shaohe_feng has joined #openstack-nova | 20:59 | |
| *** gagehugo has quit IRC | 20:59 | |
| *** daemontool_ has joined #openstack-nova | 21:00 | |
| *** daemontool has quit IRC | 21:03 | |
| *** ijw has joined #openstack-nova | 21:03 | |
| *** daemontool_ has quit IRC | 21:03 | |
| *** akshai has quit IRC | 21:05 | |
| *** mvk has joined #openstack-nova | 21:05 | |
| mriedem | thorst: seeing some powervm ci comments coming in on changes that aren't showing up as ci | 21:05 |
| *** zz_dimtruck is now known as dimtruck | 21:05 | |
| *** ivase has quit IRC | 21:06 | |
| mriedem | thorst: https://review.openstack.org/#/c/326082/ from https://review.openstack.org/#/c/326082/ | 21:06 |
| mriedem | oops | 21:06 |
| mriedem | Thawngzapum Lian | 21:06 |
| *** ivase has joined #openstack-nova | 21:06 | |
| mriedem | oh nice it takes me directly into their jenkins | 21:07 |
| mikal | Morning | 21:07 |
| mriedem | ooo look at all those jenkins jobs http://9.114.248.53/ | 21:07 |
| mriedem | rlrossit: ^ :) | 21:07 |
| rlrossit | mriedem: only if you're within the great wall | 21:07 |
| mriedem | oh | 21:07 |
| mriedem | let me see if w3 creds work on this... | 21:07 |
| rlrossit | oh jeez... | 21:07 |
| thorst | mriedem: Let me go yell at Lian. | 21:08 |
| *** pcaruana has quit IRC | 21:08 | |
| *** shaohe_feng has quit IRC | 21:08 | |
| *** shaohe_feng has joined #openstack-nova | 21:09 | |
| *** cleong has quit IRC | 21:09 | |
| *** ijw has quit IRC | 21:09 | |
| *** ivase_ has joined #openstack-nova | 21:10 | |
| *** raunak has joined #openstack-nova | 21:10 | |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Improve the help text for compute timeout_opts https://review.openstack.org/317693 | 21:11 |
| *** ivase has quit IRC | 21:11 | |
| thorst | mriedem: I've got a note out to Lian to ask that to stop. I'm honestly not sure what it is...but I'm certain whatever it is it shouldn't be doing that... | 21:12 |
| *** cdent has quit IRC | 21:12 | |
| mriedem | heh, yeah. thanks. | 21:12 |
| mriedem | i do have access to totally mess with some guy on this jenkins server. | 21:12 |
| thorst | o boy. | 21:13 |
| rlrossit | mriedem: total hax | 21:13 |
| *** eharney has joined #openstack-nova | 21:13 | |
| *** ivase_ has quit IRC | 21:14 | |
| rlrossit | thorst: it looks like it only started a half hour ago | 21:14 |
| thorst | rlrossit: then hopefully its super easy for them to turn off. | 21:15 |
| thorst | he just responded to me, so I'll make sure we get it shut off | 21:15 |
| thorst | heading out, but calling him to make sure its off. | 21:16 |
| *** thorst has quit IRC | 21:17 | |
| *** Daisy has joined #openstack-nova | 21:17 | |
| *** FL1SK has quit IRC | 21:18 | |
| openstackgerrit | Anusha Unnam proposed openstack/python-novaclient: Print unicode display_name of an instance https://review.openstack.org/323550 | 21:18 |
| *** shaohe_feng has quit IRC | 21:18 | |
| *** shaohe_feng has joined #openstack-nova | 21:19 | |
| *** josecastroleon has quit IRC | 21:20 | |
| *** Daisy has quit IRC | 21:22 | |
| *** lindycoder has quit IRC | 21:25 | |
| *** kfarr has quit IRC | 21:26 | |
| *** rlrossit has quit IRC | 21:27 | |
| *** shaohe_feng has quit IRC | 21:28 | |
| *** rtheis has quit IRC | 21:29 | |
| *** shaohe_feng has joined #openstack-nova | 21:29 | |
| *** jwcroppe has quit IRC | 21:30 | |
| *** josecastroleon has joined #openstack-nova | 21:32 | |
| *** rk4n has quit IRC | 21:34 | |
| *** thorst has joined #openstack-nova | 21:36 | |
| *** shaohe_feng has quit IRC | 21:39 | |
| *** jwcroppe has joined #openstack-nova | 21:39 | |
| *** thorst_ has joined #openstack-nova | 21:39 | |
| *** shaohe_feng has joined #openstack-nova | 21:39 | |
| *** smatzek has quit IRC | 21:42 | |
| *** thorst has quit IRC | 21:42 | |
| *** jwcroppe_ has joined #openstack-nova | 21:43 | |
| *** jwcroppe has quit IRC | 21:44 | |
| *** thorst_ has quit IRC | 21:45 | |
| *** jichen has quit IRC | 21:45 | |
| *** thorst has joined #openstack-nova | 21:45 | |
| *** catintheroof has quit IRC | 21:46 | |
| *** salv-orl_ has joined #openstack-nova | 21:47 | |
| *** tinbar has quit IRC | 21:47 | |
| *** smatzek has joined #openstack-nova | 21:48 | |
| *** shaohe_feng has quit IRC | 21:49 | |
| *** JoseMello has quit IRC | 21:49 | |
| *** salv-orlando has quit IRC | 21:49 | |
| *** shaohe_feng has joined #openstack-nova | 21:50 | |
| *** xyang1 has quit IRC | 21:51 | |
| *** sdake_ has joined #openstack-nova | 21:52 | |
| *** sdake has quit IRC | 21:54 | |
| *** jwcroppe_ has quit IRC | 21:54 | |
| *** thorst has quit IRC | 21:55 | |
| *** jwcroppe has joined #openstack-nova | 21:55 | |
| *** ijw has joined #openstack-nova | 21:58 | |
| *** edmondsw has quit IRC | 21:58 | |
| *** rk4n has joined #openstack-nova | 21:59 | |
| *** shaohe_feng has quit IRC | 21:59 | |
| *** jwcroppe has quit IRC | 21:59 | |
| *** markvoelker has quit IRC | 22:00 | |
| *** mriedem is now known as mriedem_away | 22:00 | |
| *** shaohe_feng has joined #openstack-nova | 22:00 | |
| *** markvoelker has joined #openstack-nova | 22:00 | |
| *** Sukhdev has quit IRC | 22:01 | |
| *** josecastroleon has quit IRC | 22:02 | |
| *** ijw has quit IRC | 22:03 | |
| *** ayoung has joined #openstack-nova | 22:04 | |
| *** smatzek has quit IRC | 22:06 | |
| *** shaohe_feng has quit IRC | 22:09 | |
| *** Sukhdev has joined #openstack-nova | 22:10 | |
| *** dimtruck is now known as zz_dimtruck | 22:10 | |
| *** zz_dimtruck is now known as dimtruck | 22:10 | |
| *** jaypipes has quit IRC | 22:10 | |
| *** shaohe_feng has joined #openstack-nova | 22:10 | |
| *** r-daneel has joined #openstack-nova | 22:11 | |
| *** FL1SK has joined #openstack-nova | 22:11 | |
| *** nic has quit IRC | 22:11 | |
| *** aysyd has quit IRC | 22:15 | |
| *** sdake has joined #openstack-nova | 22:15 | |
| *** mdrabe has quit IRC | 22:16 | |
| *** sdague has quit IRC | 22:18 | |
| *** zenoway has quit IRC | 22:18 | |
| *** sdake_ has quit IRC | 22:18 | |
| *** mmandel has quit IRC | 22:19 | |
| *** dimtruck is now known as zz_dimtruck | 22:20 | |
| *** shaohe_feng has quit IRC | 22:20 | |
| *** shaohe_feng has joined #openstack-nova | 22:20 | |
| *** ducttape_ has quit IRC | 22:23 | |
| *** salv-orlando has joined #openstack-nova | 22:24 | |
| *** salv-orl_ has quit IRC | 22:25 | |
| *** ametts has quit IRC | 22:25 | |
| *** kylek3h has quit IRC | 22:28 | |
| *** nic has joined #openstack-nova | 22:29 | |
| *** shaohe_feng has quit IRC | 22:30 | |
| *** shaohe_feng has joined #openstack-nova | 22:31 | |
| *** nic has quit IRC | 22:33 | |
| *** nic has joined #openstack-nova | 22:33 | |
| *** sdake has quit IRC | 22:36 | |
| *** timello has quit IRC | 22:38 | |
| *** IanGovett has quit IRC | 22:39 | |
| *** shaohe_feng has quit IRC | 22:40 | |
| *** shaohe_feng has joined #openstack-nova | 22:41 | |
| *** smatzek has joined #openstack-nova | 22:42 | |
| *** smatzek has quit IRC | 22:43 | |
| *** timello has joined #openstack-nova | 22:44 | |
| melwitt | nic: hey, I was wondering if you have heard before of a situation where ceph available disk shows more than what actually is, that is, disk * number of replicas | 22:45 |
| *** shaohe_feng has quit IRC | 22:50 | |
| *** tpepper has quit IRC | 22:51 | |
| *** shaohe_feng has joined #openstack-nova | 22:52 | |
| *** thorst has joined #openstack-nova | 22:52 | |
| *** diogogmt has quit IRC | 22:55 | |
| *** stewie925 has quit IRC | 22:55 | |
| *** ducttape_ has joined #openstack-nova | 22:57 | |
| *** FL1SK has quit IRC | 22:58 | |
| *** penick has quit IRC | 22:58 | |
| *** ijw has joined #openstack-nova | 22:59 | |
| *** thorst has quit IRC | 23:00 | |
| *** sdake has joined #openstack-nova | 23:00 | |
| *** shaohe_feng has quit IRC | 23:01 | |
| *** harlowja has quit IRC | 23:01 | |
| *** penick has joined #openstack-nova | 23:01 | |
| *** shaohe_feng has joined #openstack-nova | 23:02 | |
| *** sdake_ has joined #openstack-nova | 23:03 | |
| *** Apoorva has quit IRC | 23:04 | |
| *** Apoorva has joined #openstack-nova | 23:04 | |
| *** ijw has quit IRC | 23:05 | |
| *** sdake has quit IRC | 23:05 | |
| *** shaohe_feng has quit IRC | 23:11 | |
| *** ijw has joined #openstack-nova | 23:12 | |
| *** jwcroppe has joined #openstack-nova | 23:13 | |
| *** shaohe_feng has joined #openstack-nova | 23:14 | |
| *** bnemec has quit IRC | 23:14 | |
| *** pwnall has quit IRC | 23:15 | |
| *** pwnall has joined #openstack-nova | 23:15 | |
| *** penick has quit IRC | 23:15 | |
| *** FL1SK has joined #openstack-nova | 23:15 | |
| *** jwcroppe has quit IRC | 23:16 | |
| *** markvoelker has quit IRC | 23:19 | |
| *** bnemec has joined #openstack-nova | 23:21 | |
| *** shaohe_feng has quit IRC | 23:21 | |
| *** shaohe_feng has joined #openstack-nova | 23:22 | |
| *** salv-orlando has quit IRC | 23:23 | |
| *** salv-orlando has joined #openstack-nova | 23:24 | |
| *** ducttape_ has quit IRC | 23:29 | |
| *** pwnall has quit IRC | 23:30 | |
| *** yamahata has quit IRC | 23:31 | |
| *** shaohe_feng has quit IRC | 23:31 | |
| *** shaohe_feng has joined #openstack-nova | 23:32 | |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: Update binding:profile for SR-IOV ports https://review.openstack.org/242573 | 23:33 |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: Fix revert on migration with SR-IOV https://review.openstack.org/326174 | 23:33 |
| *** rk4n has quit IRC | 23:38 | |
| *** shaohe_feng has quit IRC | 23:42 | |
| *** shaohe_feng has joined #openstack-nova | 23:43 | |
| *** yamahata has joined #openstack-nova | 23:43 | |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: Fix revert on migration with SR-IOV https://review.openstack.org/326174 | 23:43 |
| *** diogogmt has joined #openstack-nova | 23:44 | |
| *** FL1SK has quit IRC | 23:44 | |
| *** takashin has joined #openstack-nova | 23:48 | |
| *** matrohon has quit IRC | 23:50 | |
| *** shaohe_feng has quit IRC | 23:52 | |
| *** stewie925 has joined #openstack-nova | 23:54 | |
| *** dank_ has quit IRC | 23:54 | |
| *** renato_r has quit IRC | 23:54 | |
| *** zenoway has joined #openstack-nova | 23:55 | |
| *** jwcroppe has joined #openstack-nova | 23:55 | |
| *** shaohe_feng has joined #openstack-nova | 23:55 | |
| *** ducttape_ has joined #openstack-nova | 23:56 | |
| *** thorst has joined #openstack-nova | 23:58 | |
| *** fragatin_ has joined #openstack-nova | 23:58 | |
| *** Swami has quit IRC | 23:58 | |
| *** sacharya has quit IRC | 23:58 | |
| *** zenoway has quit IRC | 23:59 | |
| *** ssurana has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!