| *** ducttape_ has quit IRC | 00:00 | |
| *** thorst_ has joined #openstack-nova | 00:00 | |
| *** hoangcx2 has joined #openstack-nova | 00:01 | |
| *** raunak has quit IRC | 00:03 | |
| *** baoli has quit IRC | 00:04 | |
| *** raunak has joined #openstack-nova | 00:05 | |
| *** hoangcx has quit IRC | 00:05 | |
| *** shaohe_feng has quit IRC | 00:06 | |
| *** shaohe_feng has joined #openstack-nova | 00:07 | |
| *** thorst_ has quit IRC | 00:09 | |
| *** thorst_ has joined #openstack-nova | 00:10 | |
| *** baoli has joined #openstack-nova | 00:12 | |
| *** shaohe_feng has quit IRC | 00:16 | |
| *** shaohe_feng has joined #openstack-nova | 00:17 | |
| *** thorst_ has quit IRC | 00:19 | |
| *** su_zhang has joined #openstack-nova | 00:22 | |
| *** jwcroppe has joined #openstack-nova | 00:23 | |
| *** shaohe_feng has quit IRC | 00:27 | |
| *** shaohe_feng has joined #openstack-nova | 00:28 | |
| *** jwcroppe has quit IRC | 00:28 | |
| openstackgerrit | Li Yingjun proposed openstack/nova: Adds view builders for hypervisors controller https://review.openstack.org/335282 | 00:31 |
|---|---|---|
| *** thorst_ has joined #openstack-nova | 00:34 | |
| *** raunak has quit IRC | 00:37 | |
| *** shaohe_feng has quit IRC | 00:37 | |
| *** shaohe_feng has joined #openstack-nova | 00:38 | |
| *** thorst_ has quit IRC | 00:39 | |
| *** mingyu has joined #openstack-nova | 00:42 | |
| *** mingyu has quit IRC | 00:46 | |
| *** shaohe_feng has quit IRC | 00:47 | |
| *** shaohe_feng has joined #openstack-nova | 00:48 | |
| *** greghaynes has quit IRC | 00:48 | |
| *** greghaynes has joined #openstack-nova | 00:49 | |
| *** thorst_ has joined #openstack-nova | 00:53 | |
| *** thorst_ has quit IRC | 00:53 | |
| *** hoangcx has joined #openstack-nova | 00:54 | |
| *** shaohe_feng has quit IRC | 00:57 | |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: WIP: Skip nova-network-only tests if using Neutron https://review.openstack.org/355215 | 00:58 |
| *** mriedem has quit IRC | 00:58 | |
| *** shaohe_feng has joined #openstack-nova | 00:58 | |
| *** cheneydc has joined #openstack-nova | 00:59 | |
| *** ducttape_ has joined #openstack-nova | 01:01 | |
| *** bswartz has joined #openstack-nova | 01:02 | |
| *** cheneydc has quit IRC | 01:05 | |
| *** ducttape_ has quit IRC | 01:05 | |
| *** shaohe_feng has quit IRC | 01:08 | |
| *** shaohe_feng has joined #openstack-nova | 01:08 | |
| *** smatzek has joined #openstack-nova | 01:09 | |
| *** lixiaoy1 has quit IRC | 01:10 | |
| *** bswartz has quit IRC | 01:13 | |
| *** zhangjl has joined #openstack-nova | 01:14 | |
| *** zhangjl has left #openstack-nova | 01:14 | |
| *** su_zhang has quit IRC | 01:17 | |
| *** su_zhang has joined #openstack-nova | 01:17 | |
| *** shaohe_feng has quit IRC | 01:18 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Fix Newton approved specs list https://review.openstack.org/335306 | 01:18 |
| *** shaohe_feng has joined #openstack-nova | 01:18 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Fix blueprints and specs link in ocata template https://review.openstack.org/335312 | 01:18 |
| *** lixiaoy1 has joined #openstack-nova | 01:19 | |
| *** takashin has joined #openstack-nova | 01:19 | |
| *** mingyu has joined #openstack-nova | 01:20 | |
| *** EinstCrazy has joined #openstack-nova | 01:23 | |
| *** kevinz has joined #openstack-nova | 01:27 | |
| *** shaohe_feng has quit IRC | 01:28 | |
| *** shaohe_feng has joined #openstack-nova | 01:31 | |
| *** ljxiash has joined #openstack-nova | 01:32 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Add a hacking rule for string interpolation at logging https://review.openstack.org/334844 | 01:32 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Abort Cold Migration https://review.openstack.org/334732 | 01:33 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Add swap volume notifications https://review.openstack.org/340715 | 01:33 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Enable cold migration with target host https://review.openstack.org/334286 | 01:33 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova-specs: Check the destination host in cold migration https://review.openstack.org/334725 | 01:34 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Fix an error in archiving 'migrations' table https://review.openstack.org/323684 | 01:34 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Fix wrong parameters in os-volumes.inc https://review.openstack.org/347112 | 01:34 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Fix 'id' (attachment_id) parameters https://review.openstack.org/349863 | 01:34 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Set 'serial' to new volume ID in swap volumes https://review.openstack.org/257135 | 01:34 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in nova/tests/unit/cells https://review.openstack.org/323287 | 01:35 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Stop using mox stubs in nova/tests/unit/virt/disk https://review.openstack.org/322778 | 01:35 |
| *** kevinz has quit IRC | 01:35 | |
| *** shaohe_feng has quit IRC | 01:38 | |
| *** shaohe_feng has joined #openstack-nova | 01:39 | |
| *** yamahata has joined #openstack-nova | 01:41 | |
| *** smatzek has quit IRC | 01:41 | |
| *** tonytan4ever has joined #openstack-nova | 01:42 | |
| *** hoangcx2 has quit IRC | 01:42 | |
| *** vishwanathj has joined #openstack-nova | 01:43 | |
| *** vishwanathj has quit IRC | 01:43 | |
| *** hoangcx2 has joined #openstack-nova | 01:45 | |
| *** ducttape_ has joined #openstack-nova | 01:46 | |
| *** wangqun has joined #openstack-nova | 01:46 | |
| *** tonytan4ever has quit IRC | 01:47 | |
| *** shaohe_feng has quit IRC | 01:49 | |
| *** shaohe_feng has joined #openstack-nova | 01:49 | |
| *** thorst_ has joined #openstack-nova | 01:52 | |
| *** ducttape_ has quit IRC | 01:56 | |
| *** baoli has quit IRC | 01:56 | |
| *** su_zhang has quit IRC | 01:59 | |
| *** shaohe_feng has quit IRC | 01:59 | |
| *** shaohe_feng has joined #openstack-nova | 02:00 | |
| *** ljxiash has quit IRC | 02:00 | |
| *** tonytan4ever has joined #openstack-nova | 02:01 | |
| *** thorst_ has quit IRC | 02:07 | |
| *** thorst_ has joined #openstack-nova | 02:08 | |
| *** shaohe_feng has quit IRC | 02:09 | |
| *** jamielennox is now known as jamielennox|away | 02:10 | |
| *** shaohe_feng has joined #openstack-nova | 02:10 | |
| *** thorst_ has quit IRC | 02:16 | |
| *** ljxiash has joined #openstack-nova | 02:17 | |
| *** kevinz has joined #openstack-nova | 02:18 | |
| *** shaohe_feng has quit IRC | 02:19 | |
| *** shaohe_feng has joined #openstack-nova | 02:20 | |
| *** jwcroppe has joined #openstack-nova | 02:26 | |
| *** shaohe_feng has quit IRC | 02:30 | |
| *** jamielennox|away is now known as jamielennox | 02:30 | |
| *** shaohe_feng has joined #openstack-nova | 02:30 | |
| *** gongysh has joined #openstack-nova | 02:30 | |
| *** jwcroppe has quit IRC | 02:31 | |
| *** ducttape_ has joined #openstack-nova | 02:32 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: libvirt: cleanup never used migratable flag checking https://review.openstack.org/310352 | 02:33 |
| *** gongysh has quit IRC | 02:37 | |
| *** shaohe_feng has quit IRC | 02:40 | |
| *** gothicmindfood has quit IRC | 02:40 | |
| *** lixiaoy1 has quit IRC | 02:40 | |
| *** shaohe_feng has joined #openstack-nova | 02:41 | |
| openstackgerrit | Hieu LE proposed openstack/nova: Config options: improve libvirt help text (2) https://review.openstack.org/316550 | 02:45 |
| *** shaohe_feng has quit IRC | 02:50 | |
| *** zerda2 has joined #openstack-nova | 02:51 | |
| *** shaohe_feng has joined #openstack-nova | 02:51 | |
| *** amotoki has quit IRC | 02:54 | |
| *** amotoki has joined #openstack-nova | 02:55 | |
| *** amotoki has quit IRC | 02:56 | |
| *** shaohe_feng has quit IRC | 03:00 | |
| *** shaohe_feng has joined #openstack-nova | 03:02 | |
| *** ducttape_ has quit IRC | 03:02 | |
| *** gcb has joined #openstack-nova | 03:02 | |
| *** hoangcx2 has quit IRC | 03:03 | |
| *** shaohe_feng has quit IRC | 03:11 | |
| *** shaohe_feng has joined #openstack-nova | 03:11 | |
| *** thorst_ has joined #openstack-nova | 03:15 | |
| *** shaohe_feng has quit IRC | 03:21 | |
| *** thorst_ has quit IRC | 03:21 | |
| *** baoli has joined #openstack-nova | 03:24 | |
| *** shaohe_feng has joined #openstack-nova | 03:24 | |
| openstackgerrit | Clenimar Filemon proposed openstack/nova: Support Identity v3 when connecting to Ironic https://review.openstack.org/300154 | 03:28 |
| *** shaohe_feng has quit IRC | 03:31 | |
| *** shaohe_feng has joined #openstack-nova | 03:32 | |
| *** baoli has quit IRC | 03:39 | |
| *** amotoki has joined #openstack-nova | 03:39 | |
| *** shaohe_feng has quit IRC | 03:41 | |
| openstackgerrit | LisaLi proposed openstack/nova: Attach/detach encrypted volume during swap_volume https://review.openstack.org/252809 | 03:42 |
| *** shaohe_feng has joined #openstack-nova | 03:44 | |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Remove mox from api/openstack/compute/test_deferred_delete.py https://review.openstack.org/320346 | 03:45 |
| *** tonytan4ever has quit IRC | 03:46 | |
| *** su_zhang has joined #openstack-nova | 03:52 | |
| *** shaohe_feng has quit IRC | 03:52 | |
| *** shaohe_feng has joined #openstack-nova | 03:53 | |
| *** EinstCrazy has quit IRC | 03:53 | |
| *** _diana_ has joined #openstack-nova | 03:54 | |
| *** ljxiash has quit IRC | 04:00 | |
| *** amotoki has quit IRC | 04:01 | |
| *** shaohe_feng has quit IRC | 04:02 | |
| *** ducttape_ has joined #openstack-nova | 04:02 | |
| *** shaohe_feng has joined #openstack-nova | 04:03 | |
| *** ekuris has joined #openstack-nova | 04:03 | |
| *** hshiina has joined #openstack-nova | 04:07 | |
| *** ducttape_ has quit IRC | 04:07 | |
| *** amotoki has joined #openstack-nova | 04:09 | |
| *** shaohe_feng has quit IRC | 04:12 | |
| *** mingyu has quit IRC | 04:12 | |
| *** shaohe_feng has joined #openstack-nova | 04:13 | |
| *** mingyu has joined #openstack-nova | 04:13 | |
| *** amotoki has quit IRC | 04:13 | |
| *** amotoki has joined #openstack-nova | 04:17 | |
| *** thorst_ has joined #openstack-nova | 04:18 | |
| *** mingyu has quit IRC | 04:18 | |
| *** shaohe_feng has quit IRC | 04:22 | |
| *** shaohe_feng has joined #openstack-nova | 04:23 | |
| *** armax has joined #openstack-nova | 04:24 | |
| *** vishwanathj has joined #openstack-nova | 04:25 | |
| *** thorst_ has quit IRC | 04:26 | |
| *** armax has quit IRC | 04:29 | |
| *** jwcroppe has joined #openstack-nova | 04:29 | |
| *** shaohe_feng has quit IRC | 04:33 | |
| *** gouthamr has quit IRC | 04:33 | |
| *** EinstCrazy has joined #openstack-nova | 04:33 | |
| *** shaohe_feng has joined #openstack-nova | 04:33 | |
| *** jwcroppe has quit IRC | 04:33 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Remove user_id policy enforcement from server start https://review.openstack.org/354662 | 04:39 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Log warning when user set improper config option value https://review.openstack.org/354532 | 04:42 |
| *** ljxiash has joined #openstack-nova | 04:42 | |
| *** shaohe_feng has quit IRC | 04:43 | |
| *** shaohe_feng has joined #openstack-nova | 04:44 | |
| *** jaosorior has joined #openstack-nova | 04:47 | |
| *** ljxiash has quit IRC | 04:47 | |
| openstackgerrit | Rikimaru Honjo proposed openstack/nova: Add checks for preventing unexpected detaching volume https://review.openstack.org/354617 | 04:50 |
| *** shaohe_feng has quit IRC | 04:53 | |
| *** shaohe_feng has joined #openstack-nova | 04:54 | |
| *** EinstCrazy has quit IRC | 04:54 | |
| *** EinstCrazy has joined #openstack-nova | 04:55 | |
| *** vishwanathj has quit IRC | 04:55 | |
| *** _diana_ has quit IRC | 04:57 | |
| *** EinstCrazy has quit IRC | 04:59 | |
| *** EinstCrazy has joined #openstack-nova | 05:00 | |
| *** shaohe_feng has quit IRC | 05:04 | |
| *** ducttape_ has joined #openstack-nova | 05:04 | |
| *** shaohe_feng has joined #openstack-nova | 05:04 | |
| *** bswartz has joined #openstack-nova | 05:07 | |
| *** ducttape_ has quit IRC | 05:08 | |
| *** pbandark has joined #openstack-nova | 05:14 | |
| *** shaohe_feng has quit IRC | 05:14 | |
| *** markmc has quit IRC | 05:15 | |
| *** shaohe_feng has joined #openstack-nova | 05:15 | |
| *** markmc has joined #openstack-nova | 05:15 | |
| *** vishwanathj has joined #openstack-nova | 05:15 | |
| *** lucas-dinner has quit IRC | 05:15 | |
| *** vishwanathj is now known as vishwanathj_zzz | 05:16 | |
| *** mingyu has joined #openstack-nova | 05:19 | |
| *** lucasagomes has joined #openstack-nova | 05:19 | |
| *** GB21 has joined #openstack-nova | 05:21 | |
| *** mingyu has quit IRC | 05:23 | |
| *** shaohe_feng has quit IRC | 05:24 | |
| *** shaohe_feng has joined #openstack-nova | 05:24 | |
| *** dmsimard has quit IRC | 05:25 | |
| *** rcernin has joined #openstack-nova | 05:25 | |
| *** thorst_ has joined #openstack-nova | 05:25 | |
| *** EinstCrazy has quit IRC | 05:26 | |
| *** dmsimard has joined #openstack-nova | 05:27 | |
| *** EinstCrazy has joined #openstack-nova | 05:27 | |
| *** EinstCra_ has joined #openstack-nova | 05:30 | |
| *** thorst_ has quit IRC | 05:31 | |
| *** EinstCrazy has quit IRC | 05:31 | |
| *** shaohe_feng has quit IRC | 05:34 | |
| *** shaohe_feng has joined #openstack-nova | 05:35 | |
| *** ljxiash has joined #openstack-nova | 05:39 | |
| *** esracelik has joined #openstack-nova | 05:41 | |
| *** esracelik is now known as ecelik | 05:41 | |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Remove mox from unit/api/openstack/compute/test_serversV21.py https://review.openstack.org/320821 | 05:42 |
| *** shaohe_feng has quit IRC | 05:45 | |
| *** shaohe_feng has joined #openstack-nova | 05:45 | |
| *** rkrum has joined #openstack-nova | 05:47 | |
| openstackgerrit | Hieu LE proposed openstack/nova: Config options: improve libvirt help text (2) https://review.openstack.org/316550 | 05:50 |
| *** baoli has joined #openstack-nova | 05:51 | |
| *** shaohe_feng has quit IRC | 05:55 | |
| *** ljxiash has quit IRC | 05:55 | |
| *** shaohe_feng has joined #openstack-nova | 05:55 | |
| *** baoli has quit IRC | 05:56 | |
| *** ljxiash has joined #openstack-nova | 05:57 | |
| *** ljxiash_ has joined #openstack-nova | 05:58 | |
| openstackgerrit | Li Yingjun proposed openstack/nova: Adds view builders for hypervisors controller https://review.openstack.org/335282 | 05:58 |
| *** liverpooler has joined #openstack-nova | 05:59 | |
| *** ljxiash_ has quit IRC | 06:00 | |
| *** ljxiash_ has joined #openstack-nova | 06:00 | |
| *** hieulq has quit IRC | 06:00 | |
| *** agireud has quit IRC | 06:01 | |
| *** ljxiash has quit IRC | 06:01 | |
| *** ducttape_ has joined #openstack-nova | 06:04 | |
| *** shaohe_feng has quit IRC | 06:05 | |
| *** shaohe_feng has joined #openstack-nova | 06:06 | |
| *** hieulq has joined #openstack-nova | 06:07 | |
| *** ducttape_ has quit IRC | 06:09 | |
| *** agireud has joined #openstack-nova | 06:10 | |
| *** shaohe_feng has quit IRC | 06:15 | |
| *** shaohe_feng has joined #openstack-nova | 06:15 | |
| *** ljxiash_ has quit IRC | 06:16 | |
| *** moshele has joined #openstack-nova | 06:18 | |
| *** jwcroppe has joined #openstack-nova | 06:23 | |
| *** shaohe_feng has quit IRC | 06:26 | |
| *** armax has joined #openstack-nova | 06:26 | |
| *** shaohe_feng has joined #openstack-nova | 06:27 | |
| *** jwcroppe has quit IRC | 06:28 | |
| *** thorst_ has joined #openstack-nova | 06:29 | |
| *** armax has quit IRC | 06:31 | |
| *** david-lyle has joined #openstack-nova | 06:36 | |
| *** shaohe_feng has quit IRC | 06:36 | |
| *** pcaruana has joined #openstack-nova | 06:36 | |
| *** shaohe_feng has joined #openstack-nova | 06:36 | |
| *** thorst_ has quit IRC | 06:36 | |
| *** houming has joined #openstack-nova | 06:38 | |
| *** david-lyle_ has quit IRC | 06:39 | |
| *** shaohe_feng has quit IRC | 06:46 | |
| *** shaohe_feng has joined #openstack-nova | 06:47 | |
| *** tonytan4ever has joined #openstack-nova | 06:47 | |
| *** belmoreira has joined #openstack-nova | 06:48 | |
| *** phuongnh has joined #openstack-nova | 06:49 | |
| *** tonytan4ever has quit IRC | 06:52 | |
| *** ljxiash has joined #openstack-nova | 06:54 | |
| *** wangqun has quit IRC | 06:54 | |
| *** gcb has quit IRC | 06:56 | |
| *** shaohe_feng has quit IRC | 06:56 | |
| *** shaohe_feng has joined #openstack-nova | 06:57 | |
| *** ljxiash has quit IRC | 06:58 | |
| *** Thelo has joined #openstack-nova | 07:02 | |
| *** ducttape_ has joined #openstack-nova | 07:05 | |
| *** shaohe_feng has quit IRC | 07:07 | |
| *** rkrum has quit IRC | 07:07 | |
| *** shaohe_feng has joined #openstack-nova | 07:07 | |
| *** ducttape_ has quit IRC | 07:09 | |
| *** Thelo has quit IRC | 07:10 | |
| *** ljxiash has joined #openstack-nova | 07:10 | |
| *** Thelo has joined #openstack-nova | 07:12 | |
| *** shaohe_feng has quit IRC | 07:17 | |
| *** shaohe_feng has joined #openstack-nova | 07:17 | |
| *** Thelo has left #openstack-nova | 07:17 | |
| *** su_zhang has quit IRC | 07:25 | |
| *** markus_z has joined #openstack-nova | 07:26 | |
| *** shaohe_feng has quit IRC | 07:27 | |
| *** shaohe_feng has joined #openstack-nova | 07:28 | |
| *** matrohon has joined #openstack-nova | 07:28 | |
| *** ihrachys has joined #openstack-nova | 07:31 | |
| *** amotoki_ has joined #openstack-nova | 07:32 | |
| *** thorst_ has joined #openstack-nova | 07:34 | |
| *** lpetrut has joined #openstack-nova | 07:34 | |
| *** amotoki has quit IRC | 07:35 | |
| *** yamahata has quit IRC | 07:36 | |
| *** shaohe_feng has quit IRC | 07:37 | |
| *** GB21 has quit IRC | 07:38 | |
| *** shaohe_feng has joined #openstack-nova | 07:38 | |
| *** thorst_ has quit IRC | 07:41 | |
| *** shaohe_feng has quit IRC | 07:48 | |
| *** shaohe_feng has joined #openstack-nova | 07:48 | |
| *** ljxiash has quit IRC | 07:51 | |
| *** wangqun has joined #openstack-nova | 07:53 | |
| *** GB21 has joined #openstack-nova | 07:53 | |
| *** roeyc has joined #openstack-nova | 07:58 | |
| *** shaohe_feng has quit IRC | 07:58 | |
| *** dimtruck is now known as zz_dimtruck | 07:59 | |
| *** shaohe_feng has joined #openstack-nova | 07:59 | |
| *** wangqun has quit IRC | 07:59 | |
| *** ljxiash has joined #openstack-nova | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** sdake has joined #openstack-nova | 08:01 | |
| *** zzzeek has joined #openstack-nova | 08:02 | |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/nova: Clean imports in code In some part in the code we import objects. In the Openstack style guidelines they recommend to import only modules. We need to fix that. https://review.openstack.org/355359 | 08:02 |
| *** tonytan4ever has joined #openstack-nova | 08:03 | |
| *** ducttape_ has joined #openstack-nova | 08:06 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Remove user_id policy enforcement from instance_actions https://review.openstack.org/354726 | 08:07 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Remove user_id policy enforcement from unlock_override https://review.openstack.org/354737 | 08:07 |
| *** tonytan4ever has quit IRC | 08:08 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Fix policy tests for project_id enforcement https://review.openstack.org/354690 | 08:08 |
| *** shaohe_feng has quit IRC | 08:08 | |
| *** shaohe_feng has joined #openstack-nova | 08:08 | |
| *** Hosam has joined #openstack-nova | 08:09 | |
| *** ducttape_ has quit IRC | 08:11 | |
| *** sdake has quit IRC | 08:15 | |
| *** shaohe_feng has quit IRC | 08:18 | |
| *** shaohe_feng has joined #openstack-nova | 08:19 | |
| *** lyarwood has joined #openstack-nova | 08:20 | |
| *** danpb has joined #openstack-nova | 08:21 | |
| markus_z | Anyone from Intel NFV CI here? | 08:26 |
| markus_z | It looks like the recheck command doesn't work: https://review.openstack.org/#/c/323765/21 (12. Aug 09:27 without response) | 08:27 |
| *** mnestratov has joined #openstack-nova | 08:27 | |
| *** shaohe_feng has quit IRC | 08:29 | |
| *** shaohe_feng has joined #openstack-nova | 08:29 | |
| *** houming has quit IRC | 08:30 | |
| *** Thelo has joined #openstack-nova | 08:34 | |
| *** sdake has joined #openstack-nova | 08:35 | |
| *** shaohe_feng has quit IRC | 08:39 | |
| *** thorst_ has joined #openstack-nova | 08:39 | |
| *** shaohe_feng has joined #openstack-nova | 08:40 | |
| *** houming has joined #openstack-nova | 08:40 | |
| *** markvoelker has joined #openstack-nova | 08:41 | |
| *** tuanla_ has joined #openstack-nova | 08:42 | |
| *** sdake has quit IRC | 08:43 | |
| *** markvoelker has quit IRC | 08:45 | |
| *** thorst_ has quit IRC | 08:47 | |
| *** lpetrut has quit IRC | 08:47 | |
| *** shaohe_feng has quit IRC | 08:49 | |
| *** houming has quit IRC | 08:50 | |
| *** shaohe_feng has joined #openstack-nova | 08:52 | |
| *** Thelo has quit IRC | 08:52 | |
| *** Gibi has joined #openstack-nova | 08:53 | |
| *** hshiina has quit IRC | 08:55 | |
| *** mvk has joined #openstack-nova | 08:56 | |
| *** houming has joined #openstack-nova | 08:57 | |
| *** lpetrut has joined #openstack-nova | 08:58 | |
| *** shaohe_feng has quit IRC | 08:59 | |
| *** shaohe_feng has joined #openstack-nova | 09:00 | |
| *** ducttape_ has joined #openstack-nova | 09:07 | |
| *** shaohe_feng has quit IRC | 09:10 | |
| *** andrey-mp has joined #openstack-nova | 09:10 | |
| *** shaohe_feng has joined #openstack-nova | 09:10 | |
| openstackgerrit | Nguyen Hung Phuong proposed openstack/nova: Clean imports in code https://review.openstack.org/355359 | 09:10 |
| *** ducttape_ has quit IRC | 09:11 | |
| *** lixiaoy1 has joined #openstack-nova | 09:12 | |
| tdurakov | markus_z: hi, have you tried mails, mentioned here https://wiki.openstack.org/wiki/ThirdPartySystems/Intel_NFV_CI ? | 09:15 |
| *** ljxiash has quit IRC | 09:16 | |
| *** ljxiash has joined #openstack-nova | 09:17 | |
| *** ljxiash_ has joined #openstack-nova | 09:19 | |
| *** shaohe_feng has quit IRC | 09:20 | |
| *** shaohe_feng has joined #openstack-nova | 09:20 | |
| *** ljxiash has quit IRC | 09:22 | |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Stop using mox from compute/test_volumes.py https://review.openstack.org/323257 | 09:23 |
| *** oanson has joined #openstack-nova | 09:28 | |
| *** armax has joined #openstack-nova | 09:28 | |
| *** lpetrut has quit IRC | 09:30 | |
| *** shaohe_feng has quit IRC | 09:30 | |
| *** shaohe_feng has joined #openstack-nova | 09:31 | |
| *** ljxiash_ has quit IRC | 09:32 | |
| *** armax has quit IRC | 09:33 | |
| *** jichen has joined #openstack-nova | 09:33 | |
| *** efoley has joined #openstack-nova | 09:35 | |
| markus_z | tdurakov: thanks, will do exactly that in 1h in case nobody is listening here. | 09:35 |
| tdurakov | markus_z: yeah, just in case) | 09:36 |
| *** jaosorior is now known as jaosorior_brb | 09:38 | |
| *** lixiaoy1 has quit IRC | 09:39 | |
| *** lixiaoy1_ has joined #openstack-nova | 09:39 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove unused code in nova/api/validator.py https://review.openstack.org/355392 | 09:39 |
| *** shaohe_feng has quit IRC | 09:40 | |
| *** shaohe_feng has joined #openstack-nova | 09:41 | |
| *** hoangcx has quit IRC | 09:41 | |
| markus_z | FYI: trivial patch: ^ | 09:42 |
| *** roeyc has quit IRC | 09:43 | |
| *** thorst_ has joined #openstack-nova | 09:45 | |
| *** sdake has joined #openstack-nova | 09:46 | |
| *** roeyc has joined #openstack-nova | 09:47 | |
| *** ftersin has joined #openstack-nova | 09:48 | |
| *** hoangcx has joined #openstack-nova | 09:49 | |
| *** shaohe_feng has quit IRC | 09:51 | |
| *** shaohe_feng has joined #openstack-nova | 09:51 | |
| *** thorst_ has quit IRC | 09:52 | |
| *** roeyc has quit IRC | 09:52 | |
| *** wangqun has joined #openstack-nova | 09:57 | |
| *** oanson has quit IRC | 09:58 | |
| sfinucan | markus_z: try wznoinsk | 09:58 |
| tdurakov | markus_z: heh, about the patch above, I wonder where other method from validator.py are used? | 09:59 |
| wznoinsk | markus_z: checking | 10:00 |
| *** shaohe_feng has quit IRC | 10:01 | |
| *** shaohe_feng has joined #openstack-nova | 10:02 | |
| *** prithiv has joined #openstack-nova | 10:03 | |
| andrey-mp | danpb: could you look to this bug - https://bugs.launchpad.net/nova/+bug/1612953 ? | 10:04 |
| openstack | Launchpad bug 1612953 in OpenStack Compute (nova) "fail to boot instance with port of specfic subnet (regression)" [Undecided,New] | 10:04 |
| *** tonytan4ever has joined #openstack-nova | 10:04 | |
| *** wangqun has quit IRC | 10:05 | |
| wznoinsk | markus_z: it commented on 10:33 | 10:05 |
| wznoinsk | (~32min ago) | 10:05 |
| *** ducttape_ has joined #openstack-nova | 10:08 | |
| *** tonytan4ever has quit IRC | 10:08 | |
| *** hoangcx has quit IRC | 10:10 | |
| *** zenoway has joined #openstack-nova | 10:10 | |
| *** shaohe_feng has quit IRC | 10:11 | |
| *** prithiv has quit IRC | 10:11 | |
| danpb | andrey-mp: that;s already reported and fixed | 10:12 |
| *** ducttape_ has quit IRC | 10:12 | |
| *** shaohe_feng has joined #openstack-nova | 10:12 | |
| *** edand has joined #openstack-nova | 10:13 | |
| wznoinsk | would someone know which iptables are relevant (except for REDIRECT in nat) to make the metadata service working (neutron proxy -> nova metadata) when not using routed network? (no qrouter netns etc.) ? | 10:14 |
| *** prithiv has joined #openstack-nova | 10:16 | |
| *** zz_dimtruck is now known as dimtruck | 10:18 | |
| *** EinstCra_ has quit IRC | 10:20 | |
| *** jwcroppe has joined #openstack-nova | 10:20 | |
| *** shaohe_feng has quit IRC | 10:21 | |
| *** shaohe_feng has joined #openstack-nova | 10:22 | |
| andrey-mp | danpb: could you point me to fix? | 10:25 |
| danpb | see the bug i marked it dup of | 10:26 |
| andrey-mp | thanks | 10:27 |
| *** jwcroppe has quit IRC | 10:27 | |
| kevinz | danpb: Nice meeting you~ If you have time,Could you give some comments to https://review.openstack.org/#/c/329403/ | 10:27 |
| *** dimtruck is now known as zz_dimtruck | 10:28 | |
| danpb | kevinz: plesae stop pinging me every day | 10:29 |
| danpb | if i had time i would have commented already | 10:29 |
| openstackgerrit | bailin.zhang proposed openstack/nova: Unplug vifs when VM create failed https://review.openstack.org/331113 | 10:30 |
| *** jaosorior_brb is now known as jaosorior | 10:31 | |
| *** sdague has joined #openstack-nova | 10:31 | |
| kevinz | danpb: OK, sorry for disturb you :-) I will wait | 10:31 |
| *** shaohe_feng has quit IRC | 10:32 | |
| *** shaohe_feng has joined #openstack-nova | 10:32 | |
| *** prithiv has quit IRC | 10:35 | |
| *** prithiv has joined #openstack-nova | 10:35 | |
| markus_z | wznoinsk: yep, that's for the recheck I triggered today, which is fine. Out of curiosity: is your Zuul publicly accessible? | 10:37 |
| markus_z | tdurakov: Do you mean the other "validate_*" methods? | 10:37 |
| wznoinsk | markus_z: it's not | 10:40 |
| markus_z | wznoinsk: OK, thanks | 10:40 |
| wznoinsk | markus_z: I see zuul had issues reporting results back on 12 Aug, connectivity issue | 10:41 |
| *** shaohe_feng has quit IRC | 10:42 | |
| markus_z | wznoinsk: That explains that it didn't vote again after the recheck command. no biggie. Thanks for being reachable here in #nova | 10:42 |
| *** markvoelker has joined #openstack-nova | 10:42 | |
| *** shaohe_feng has joined #openstack-nova | 10:42 | |
| wznoinsk | markus_z: no problem | 10:43 |
| *** hieulq has quit IRC | 10:44 | |
| *** lpetrut has joined #openstack-nova | 10:45 | |
| *** prithiv has quit IRC | 10:46 | |
| *** markvoelker has quit IRC | 10:47 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Remove nova/api/validator.py https://review.openstack.org/355392 | 10:48 |
| openstackgerrit | Rikimaru Honjo proposed openstack/nova: Add checks for preventing unexpected detaching volume https://review.openstack.org/354617 | 10:51 |
| *** shaohe_feng has quit IRC | 10:52 | |
| *** shaohe_feng has joined #openstack-nova | 10:53 | |
| *** andrey-mp has left #openstack-nova | 10:53 | |
| *** phuongnh has quit IRC | 10:58 | |
| *** roeyc has joined #openstack-nova | 10:59 | |
| openstackgerrit | Tina Kevin proposed openstack/nova: Instance can not resize ephemeral disk https://review.openstack.org/320759 | 11:02 |
| *** shaohe_feng has quit IRC | 11:02 | |
| *** shaohe_feng has joined #openstack-nova | 11:03 | |
| *** thorst_ has joined #openstack-nova | 11:04 | |
| *** lpetrut has quit IRC | 11:06 | |
| *** Hosam_ has joined #openstack-nova | 11:07 | |
| *** ducttape_ has joined #openstack-nova | 11:08 | |
| *** Hosam has quit IRC | 11:10 | |
| *** shaohe_feng has quit IRC | 11:13 | |
| *** ducttape_ has quit IRC | 11:13 | |
| *** shaohe_feng has joined #openstack-nova | 11:14 | |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Fix wrong call of _is_booted_from_volume https://review.openstack.org/334609 | 11:15 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Delete duplicate check when live-migrating https://review.openstack.org/337056 | 11:15 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Simplify internal usage of get_instance_disk_info https://review.openstack.org/334901 | 11:15 |
| *** lucasagomes is now known as lucas-hungry | 11:16 | |
| *** zerda2 has quit IRC | 11:17 | |
| *** bvanhav has joined #openstack-nova | 11:22 | |
| *** shaohe_feng has quit IRC | 11:23 | |
| *** shaohe_feng has joined #openstack-nova | 11:24 | |
| openstackgerrit | Jens Rosenboom proposed openstack/nova: Fix ephemaral disk creation https://review.openstack.org/355415 | 11:24 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: live-migration test hook for serial console https://review.openstack.org/347471 | 11:27 |
| *** armax has joined #openstack-nova | 11:30 | |
| *** amotoki_ has quit IRC | 11:31 | |
| *** shaohe_feng has quit IRC | 11:33 | |
| *** amotoki has joined #openstack-nova | 11:34 | |
| *** jaosorior has quit IRC | 11:34 | |
| *** armax has quit IRC | 11:35 | |
| *** jaosorior has joined #openstack-nova | 11:35 | |
| *** shaohe_feng has joined #openstack-nova | 11:36 | |
| *** amotoki has quit IRC | 11:37 | |
| tdurakov | markus_z: left comment on that patch, so looks like nova is not using validator.py anymore | 11:38 |
| *** rodrigods has quit IRC | 11:38 | |
| *** rodrigods has joined #openstack-nova | 11:38 | |
| markus_z | tdurakov: yeah, I have seen it right now, good catch! Less code to maintain. | 11:38 |
| *** shaohe_feng has quit IRC | 11:43 | |
| *** shaohe_feng has joined #openstack-nova | 11:44 | |
| *** gcb has joined #openstack-nova | 11:45 | |
| *** amotoki has joined #openstack-nova | 11:48 | |
| openstackgerrit | Andrey Volkov proposed openstack/nova: Tests: use fakes.HTTPRequest in compute tests https://review.openstack.org/353326 | 11:51 |
| openstackgerrit | Andrey Volkov proposed openstack/nova: Wrap servers detail API call in single transaction https://review.openstack.org/352950 | 11:51 |
| *** rfolco has joined #openstack-nova | 11:51 | |
| *** shaohe_feng has quit IRC | 11:54 | |
| *** shaohe_feng has joined #openstack-nova | 11:54 | |
| openstackgerrit | Sean Dague proposed openstack/python-novaclient: Use neutron for network name -> id resolution https://review.openstack.org/354981 | 11:57 |
| *** baoli_ has joined #openstack-nova | 12:00 | |
| *** sdake has quit IRC | 12:02 | |
| *** shaohe_feng has quit IRC | 12:04 | |
| *** edmondsw has joined #openstack-nova | 12:04 | |
| *** shaohe_feng has joined #openstack-nova | 12:05 | |
| *** tonytan4ever has joined #openstack-nova | 12:05 | |
| *** smatzek has joined #openstack-nova | 12:06 | |
| *** sdake has joined #openstack-nova | 12:06 | |
| *** JoseMello has joined #openstack-nova | 12:08 | |
| *** ducttape_ has joined #openstack-nova | 12:09 | |
| *** tonytan4ever has quit IRC | 12:09 | |
| *** gouthamr has joined #openstack-nova | 12:10 | |
| tdurakov | johnthetubaguy: hi, are you around? https://review.openstack.org/#/c/350480/3/releasenotes/notes/libvirt-change-default-value-of-live-migration-tunnelled-4248cf76df605fdf.yaml - do you mean that it's should be explicitly mentioned that it's about encryption? | 12:11 |
| *** gcb has quit IRC | 12:12 | |
| *** Hosam has joined #openstack-nova | 12:12 | |
| *** ducttape_ has quit IRC | 12:12 | |
| *** gcb has joined #openstack-nova | 12:13 | |
| *** ducttape_ has joined #openstack-nova | 12:13 | |
| *** tonytan4ever has joined #openstack-nova | 12:13 | |
| *** shaohe_feng has quit IRC | 12:14 | |
| *** shaohe_feng has joined #openstack-nova | 12:15 | |
| *** jwcroppe has joined #openstack-nova | 12:15 | |
| *** Hosam_ has quit IRC | 12:16 | |
| *** amotoki has quit IRC | 12:16 | |
| *** zz_dimtruck is now known as dimtruck | 12:18 | |
| *** Zara has quit IRC | 12:20 | |
| *** amotoki has joined #openstack-nova | 12:23 | |
| *** alineb has joined #openstack-nova | 12:23 | |
| *** shaohe_feng has quit IRC | 12:24 | |
| *** shaohe_feng has joined #openstack-nova | 12:25 | |
| *** jwcroppe has quit IRC | 12:26 | |
| *** takashin has left #openstack-nova | 12:26 | |
| *** lucas-hungry is now known as lucasagomes | 12:27 | |
| *** ftersin has quit IRC | 12:27 | |
| *** amotoki has quit IRC | 12:28 | |
| *** dimtruck is now known as zz_dimtruck | 12:28 | |
| *** markvoelker has joined #openstack-nova | 12:30 | |
| *** woodster_ has joined #openstack-nova | 12:32 | |
| *** pradk has joined #openstack-nova | 12:33 | |
| *** shaohe_feng has quit IRC | 12:35 | |
| *** GB21 has quit IRC | 12:35 | |
| *** shaohe_feng has joined #openstack-nova | 12:35 | |
| *** mtanino has joined #openstack-nova | 12:35 | |
| *** ducttape_ has quit IRC | 12:35 | |
| *** tuanla_ has quit IRC | 12:36 | |
| *** sdake has quit IRC | 12:38 | |
| openstackgerrit | garyk proposed openstack/nova: VIF: add in missing translation https://review.openstack.org/355447 | 12:43 |
| *** kevinz has quit IRC | 12:44 | |
| *** shaohe_feng has quit IRC | 12:45 | |
| *** jaypipes has joined #openstack-nova | 12:45 | |
| *** shaohe_feng has joined #openstack-nova | 12:46 | |
| *** rkrum has joined #openstack-nova | 12:47 | |
| *** Zara has joined #openstack-nova | 12:47 | |
| *** rlrossit has joined #openstack-nova | 12:49 | |
| *** hshiina has joined #openstack-nova | 12:52 | |
| *** vladikr has joined #openstack-nova | 12:53 | |
| *** gcb has quit IRC | 12:55 | |
| *** shaohe_feng has quit IRC | 12:55 | |
| *** shaohe_feng has joined #openstack-nova | 12:56 | |
| *** xyang1 has joined #openstack-nova | 12:56 | |
| *** PaulMurray has joined #openstack-nova | 12:59 | |
| openstackgerrit | Jay Pipes proposed openstack/nova-specs: Resource providers: custom resource classes https://review.openstack.org/312696 | 13:01 |
| jaypipes | edleafe, cdent: ^^ | 13:02 |
| *** ftersin has joined #openstack-nova | 13:03 | |
| *** kylek3h has joined #openstack-nova | 13:03 | |
| *** zz_dimtruck is now known as dimtruck | 13:04 | |
| alex_xu | jaypipes: good morning, quick question, whether LIVE_MIGRATION, LIVE_RESIZE can be a capabilities? | 13:04 |
| *** houming has quit IRC | 13:04 | |
| *** shaohe_feng has quit IRC | 13:05 | |
| jaypipes | alex_xu: no. those aren't things that a user would request (since users should have no idea which compute host their guest is on nor do users ever trigger live migration themselves. | 13:05 |
| jaypipes | alex_xu: those are, of course, things that we can/should expose to administrators via the "capabilities REST API" that alaski and others have been discussing (the thing that is more alike to the old GET /extensions API call) | 13:07 |
| alex_xu | jaypipes: ok, thanks. I'm still thinking the case people want to boot instance it can support LIVE_RESIZE. And the capabilities discovery API may need to know whether th.... | 13:07 |
| alex_xu | jaypipes: :) I just type on the halfway... | 13:08 |
| jaypipes | alex_xu: which brings up a good point that we should settle on the name of these different things :) We've been calling both of those different things "capabilities". | 13:08 |
| *** mdrabe has joined #openstack-nova | 13:08 | |
| jaypipes | alex_xu: users don't boot instances and request that they can live-resize it... it's just not something cloud users do, really. | 13:08 |
| alaski | alex_xu: I think LIVE_MIGRATION/LIVE_RESIZE might depend on host "capabilities" but not something called LIVE_MIGRATION. It might be something like cpu_arch or whatever is relevant | 13:09 |
| jaypipes | alex_xu: just like users never issue a live-migrate command themselves. | 13:09 |
| jaypipes | alaski: ++ | 13:09 |
| *** shaohe_feng has joined #openstack-nova | 13:10 | |
| alex_xu | alaski: ah, i got you, learned :) | 13:10 |
| *** rkrum has quit IRC | 13:10 | |
| PaulMurray | jaypipes, resize is issued by users | 13:11 |
| alex_xu | I just say no LIVE_MIGRATION/LIVE_RESIZE won't depend on any hardware caps | 13:11 |
| *** rkrum has joined #openstack-nova | 13:11 | |
| jaypipes | PaulMurray: users don't request to launch an instance that can only land on a host that supports hot-resizing, though... what use case does that serve? | 13:12 |
| edleafe | jaypipes: +1'd the spec | 13:12 |
| jaypipes | edleafe: cheerio. | 13:12 |
| alex_xu | jaypipes: alaski but the reality is we have many different virt driver, those virt driver have different caps to support different action. so that is fault of our API | 13:12 |
| alex_xu | s/fault of our API/fault of API?/ | 13:12 |
| PaulMurray | jaypipes, fare enough, just pointing out that users initiate resize - admins to migrate | 13:13 |
| jaypipes | PaulMurray: understood. | 13:13 |
| PaulMurray | s/fare/fair/ | 13:13 |
| *** ducttape_ has joined #openstack-nova | 13:14 | |
| * alex_xu back to think about those | 13:15 | |
| jaypipes | alex_xu: the capabilities part of the placement API isn't intended to determine what *actions* a user can take. It's intended to determine what *features* or *attributes* a resource provider has that make it a good fit for a requested workload. | 13:15 |
| *** shaohe_feng has quit IRC | 13:16 | |
| *** rkrum has quit IRC | 13:16 | |
| jaypipes | alex_xu: while the discovery API that alaski has been working on is intended to answer the question of "can I perform this *action* against this particular OpenStack cloud?" | 13:17 |
| alex_xu | jaypipes: actions need resource provider have the live_resize feature | 13:17 |
| alex_xu | jaypipes: yes, whether I can perform this action, depend on the policy and whether hypervisor support this action | 13:17 |
| johnthetubaguy | tdurakov: I basically mean it should mention the behaviour change, that by default live-migrate traffic will no longer be encrypted. | 13:18 |
| jaypipes | alex_xu: think of the difference like this: the capabilities part of the placement API describes *adjectives* (i.e. is this apple red? is it tasty?) whereas the capabilities discovery API answers the question "what can I do with this?" (i.e. can I pick apples at this orchard?) | 13:18 |
| *** shaohe_feng has joined #openstack-nova | 13:19 | |
| *** jwcroppe has joined #openstack-nova | 13:19 | |
| PaulMurray | alex_xu, do you expect to have a mix of hosts, some that can and some that can't do live resize presented to one tenant? | 13:19 |
| *** amotoki has joined #openstack-nova | 13:19 | |
| jaypipes | alex_xu: obviously we need both of those things in the REST API. we just need to be very careful not to mix policy stuff with "features" or "adjectives" that describe the resources that a resource provider provides :) | 13:19 |
| alex_xu | jaypipes: yes, so how can we answer the question of "what can i do with this" | 13:20 |
| alex_xu | PaulMurray: good point | 13:20 |
| jaypipes | alex_xu: we need a REST API for that :) it just shouldn't be the same REST API as the proposed placement API capabilities stuff. Something like GET /policies would be good with me, though policy is an overloaded term already (as is capabilities). Guh, naming is *hard*! :) | 13:21 |
| alex_xu | jaypipes: yea, policy discovery, capabilities discovery, I hate those also :) | 13:23 |
| jaypipes | alex_xu: I was discussing the naming of these things with alaski the other day. We couldn't come up with good terms for them. | 13:24 |
| alex_xu | heh :) | 13:24 |
| *** GB21 has joined #openstack-nova | 13:24 | |
| *** ravikama_ has joined #openstack-nova | 13:25 | |
| *** shaohe_feng has quit IRC | 13:26 | |
| *** jamesden_ has joined #openstack-nova | 13:27 | |
| *** shaohe_feng has joined #openstack-nova | 13:27 | |
| alaski | just sent a message to the ML soliciting naming assistance | 13:27 |
| PaulMurray | alaski, DNS - distributed naming solicitation | 13:31 |
| *** belmoreira has quit IRC | 13:31 | |
| alaski | PaulMurray: exactly, just a simple DNS query | 13:31 |
| jaypipes | hehe | 13:32 |
| *** ducttape_ has quit IRC | 13:33 | |
| *** tonytan4ever has quit IRC | 13:33 | |
| *** mriedem has joined #openstack-nova | 13:34 | |
| mriedem | rar! | 13:34 |
| jaypipes | mriedem: grawr! | 13:34 |
| mriedem | sdague: on the network proxy thing in novaclient, i was thinking over the weekend that we should merge that in with dansmith's change to fallback to 2.35 | 13:36 |
| sdague | mriedem: on the novaclient front, how about this other direction for networks - https://review.openstack.org/#/c/354981/ | 13:36 |
| *** shaohe_feng has quit IRC | 13:36 | |
| sdague | mriedem: I think we want to take a different approach | 13:36 |
| sdague | which is to always go to neutron if we can | 13:36 |
| *** shaohe_feng has joined #openstack-nova | 13:36 | |
| *** jwcroppe has quit IRC | 13:37 | |
| mriedem | sdague: ok, i was having similar thoughts also after working on making the functional tests work with neutron | 13:38 |
| mriedem | doing the catalog lookup that is | 13:38 |
| sdague | the functional tests there exploded because of running out of time | 13:39 |
| sdague | we may need to extend the timeout if we're putting more tests into that bucket | 13:39 |
| mriedem | how about saving the has_neutron result as an attr in the client? | 13:41 |
| *** GB21 has quit IRC | 13:41 | |
| sdague | also... ug - https://github.com/openstack/python-novaclient/blob/cbef3db003deb2bafb0cd3beaa993950a7185d3e/novaclient/tests/unit/v2/test_shell.py#L674-L676 test mocking out other part of other test | 13:41 |
| mriedem | so we only have to look it up once? | 13:41 |
| sdague | mriedem: it's not a network call | 13:42 |
| mriedem | nice, mock the mock | 13:42 |
| mriedem | what's not a network call? | 13:42 |
| sdague | get_endpoint | 13:42 |
| mriedem | right, in a given session is that going to change? | 13:43 |
| mriedem | i guess we recreate the client on each call, | 13:43 |
| mriedem | so storing as an attr in the client won't buy much | 13:43 |
| sdague | it would just add a caching complexity | 13:43 |
| sdague | when all you are saving is a dozen instructions | 13:43 |
| mriedem | so the functional timeouts aren't just me | 13:44 |
| mriedem | i was noticing the same when running my use-neutron functional job patch | 13:44 |
| *** pwnall has joined #openstack-nova | 13:44 | |
| sdague | yeh, I think we just grew the test bucket | 13:45 |
| sdague | and given that the tests run serially | 13:45 |
| *** tblakes has joined #openstack-nova | 13:45 | |
| mriedem | ok, i also noticed over the weekend that we're leaking resources in the functional run | 13:46 |
| *** shaohe_feng has quit IRC | 13:46 | |
| *** shaohe_feng has joined #openstack-nova | 13:47 | |
| mriedem | sdague: so when we get to dansmith's 2.36 change, i guess we'd have to just special case the fallback nova-net proxy call if we're not using neutron, and use 2.35 | 13:47 |
| mriedem | to be consistent with the other 2.35 network CLI fallback | 13:47 |
| mriedem | but hopefully with your change that won't be as many things | 13:48 |
| *** rlrossit has quit IRC | 13:48 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: Refactor unit test for trackable migrations https://review.openstack.org/352610 | 13:48 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: Create multiple resize claim unit test https://review.openstack.org/352625 | 13:49 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt: delete test_resource_tracker.py https://review.openstack.org/353800 | 13:49 |
| *** belmoreira has joined #openstack-nova | 13:50 | |
| sdague | mriedem: why? | 13:50 |
| mriedem | sdague: are you going to hack on this today or do you want me to? i also have to sort out what i'm doing with the get-me-a-network REST API last minute change request, because i'm on pager duty this week and on vacation next week | 13:50 |
| mriedem | so need to prioritize | 13:50 |
| sdague | mriedem: I'm trying to get unit tests passing right now | 13:50 |
| mriedem | sdague: because in dan's change we fallback to 2.35 for network CLIs, i'd think we'd do the same here if not using neutron but the request is >=2.36 | 13:51 |
| mriedem | otherwise it's kind of weird, we'll list networks for you but you can't boot with those networks, you'll get a 404 | 13:51 |
| mriedem | not here as in your change, here as in the use case but in dan's change | 13:51 |
| sdague | mriedem: oh, because the fallback code is only on the shell | 13:51 |
| sdague | right | 13:52 |
| mriedem | right | 13:52 |
| mriedem | ok | 13:52 |
| *** rlrossit_ has joined #openstack-nova | 13:52 | |
| mriedem | if we're good with that, then i think we can land this change and the image proxy change and move forward with dan's | 13:52 |
| sdague | yeh | 13:52 |
| mriedem | wait why would the functional tests now be timing out on your change that's checking to see if we're using neutron? the functional job is using nova-network | 13:53 |
| sdague | because, they are pretty close to the timing line | 13:54 |
| mriedem | the extra endpoint lookup pushes that over? | 13:54 |
| sdague | no | 13:54 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: verify cpu pinning with prefer policy https://review.openstack.org/355475 | 13:54 |
| sdague | test runs have variance of 20 - 30% | 13:54 |
| *** burt has joined #openstack-nova | 13:54 | |
| sdague | given cloud providers, noisy neighbors, network perf on mirrors | 13:54 |
| *** ayoung has joined #openstack-nova | 13:55 | |
| edleafe | Scheduler subteam meeting in 5 minutes in #openstack-meeting-alt | 13:55 |
| sdague | so if we're getting close to the line, sometimes we'll go over it regardless of the contents of the patch | 13:55 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: verify cpu pinning with prefer policy https://review.openstack.org/355475 | 13:55 |
| sdague | has_neutron probably takes about 0.002s to execute | 13:55 |
| *** rtheis has joined #openstack-nova | 13:56 | |
| *** shaohe_feng has quit IRC | 13:57 | |
| mriedem | ok | 13:57 |
| *** shaohe_feng has joined #openstack-nova | 13:57 | |
| mriedem | sdague: when you get a minute this morning, i could use your input on this http://lists.openstack.org/pipermail/openstack-dev/2016-August/101499.html since if i'm going to change it i have to get that done this week. | 13:58 |
| openstackgerrit | Sean Dague proposed openstack/python-novaclient: Use neutron for network name -> id resolution https://review.openstack.org/354981 | 13:58 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: adding debug info for pinning calculation https://review.openstack.org/355150 | 13:59 |
| sdague | mriedem: ok, I honestly haven't looked at your change in a while, so I can switch gears over there while these tests run | 13:59 |
| *** dims has quit IRC | 13:59 | |
| *** anush_ has joined #openstack-nova | 14:00 | |
| *** ftersin_ has joined #openstack-nova | 14:00 | |
| *** oanson has joined #openstack-nova | 14:01 | |
| *** ftersin has quit IRC | 14:02 | |
| *** ecelik has quit IRC | 14:03 | |
| *** ducttape_ has joined #openstack-nova | 14:04 | |
| *** venusf has joined #openstack-nova | 14:04 | |
| mriedem | fyi this will unblock neutronclient 5.1.0 from getting into upper-constraints http://lists.openstack.org/pipermail/openstack-dev/2016-August/101499.html | 14:05 |
| *** dims has joined #openstack-nova | 14:05 | |
| *** shaohe_feng has quit IRC | 14:07 | |
| *** mingyu has joined #openstack-nova | 14:07 | |
| *** shaohe_feng has joined #openstack-nova | 14:07 | |
| *** GB21 has joined #openstack-nova | 14:07 | |
| *** su_zhang has joined #openstack-nova | 14:08 | |
| *** venusf has left #openstack-nova | 14:08 | |
| *** venusf has joined #openstack-nova | 14:09 | |
| *** mtanino has quit IRC | 14:09 | |
| mriedem | andrewbogott: https://review.openstack.org/#/c/355210/ | 14:09 |
| mriedem | for when you get to mitaka | 14:09 |
| *** Hosam has quit IRC | 14:10 | |
| *** sdake has joined #openstack-nova | 14:10 | |
| *** Hosam has joined #openstack-nova | 14:10 | |
| venusf | Hi ,guys. As we know, there are more than one nova-api thread for processing restful api request, but how can I know which nova-api thread process the reqeust I send? | 14:11 |
| *** annegentle has joined #openstack-nova | 14:14 | |
| *** kevinz has joined #openstack-nova | 14:14 | |
| *** Hosam has quit IRC | 14:15 | |
| *** jaypipes has quit IRC | 14:17 | |
| *** shaohe_feng has quit IRC | 14:17 | |
| sdague | mriedem: so, from a code structure perspective on - https://review.openstack.org/#/c/316398/35/nova/api/openstack/compute/servers.py ... | 14:18 |
| sdague | it's a bit convoluted to pass that third parameter down multiple layers in the stack to cause different behavior | 14:18 |
| *** rrecio has joined #openstack-nova | 14:18 | |
| sdague | why not just blast those out as different functions? | 14:19 |
| *** brian_price has joined #openstack-nova | 14:20 | |
| *** shaohe_feng has joined #openstack-nova | 14:20 | |
| *** fragatina has joined #openstack-nova | 14:20 | |
| *** jaypipes has joined #openstack-nova | 14:21 | |
| *** fragatina has quit IRC | 14:21 | |
| *** fragatina has joined #openstack-nova | 14:21 | |
| *** tri2sing has joined #openstack-nova | 14:23 | |
| *** burgerk has joined #openstack-nova | 14:23 | |
| *** tonytan4ever has joined #openstack-nova | 14:23 | |
| mriedem | sdague: that's the way it worked out, given the existing requested_networks list allowed non-uuid values for the network id and such. plus port/fixed ip handling etc. i mean i could have probably at the top level avoided calling _get_requested_networks and just check for auto/none and fail if anything else is specified | 14:24 |
| *** lpetrut has joined #openstack-nova | 14:24 | |
| mriedem | sdague: but i think that's where ken'ichi was asking, why even allow the auto/none as the network uuid in a list if it's just a single string value in the request will work | 14:24 |
| mriedem | it would simplify the validation logic, basically networks would be a string or a list | 14:24 |
| mriedem | and if it's a string, it has to be auto/none | 14:24 |
| mriedem | if it's a list, it's whatever is supported today | 14:25 |
| *** rrecio_ has joined #openstack-nova | 14:26 | |
| *** ametts has joined #openstack-nova | 14:26 | |
| *** annegentle has quit IRC | 14:27 | |
| *** shaohe_feng has quit IRC | 14:27 | |
| *** shaohe_feng has joined #openstack-nova | 14:28 | |
| *** rrecio has quit IRC | 14:28 | |
| sdague | yeh, it's just really dense to unpack and walk through, I'd much rather divert a lot earlier with the special case | 14:28 |
| *** burt has quit IRC | 14:29 | |
| mriedem | i can look at doing that, if we stick with this request format | 14:29 |
| mriedem | if we change how server.networks is represented, then... | 14:29 |
| sdague | I'm massively suspicious of boolean parameters that are "do things differently" | 14:29 |
| sdague | ok, so what's actually the request format question? | 14:29 |
| *** dimtruck is now known as zz_dimtruck | 14:29 | |
| mriedem | for auto/none, change from server.networks = [{'uuid': 'auto'}] to server.networks = 'auto' | 14:30 |
| mriedem | same for 'none' | 14:30 |
| *** oanson has quit IRC | 14:30 | |
| sdague | oh, yeh, that makes a ton more sense to me | 14:30 |
| *** pcaruana has quit IRC | 14:30 | |
| sdague | +1 on that | 14:30 |
| mriedem | if you want multiple nics, or specific networks, you use the list format | 14:30 |
| mriedem | for auto/none you doe the direct string one | 14:30 |
| mriedem | ok, i'll start working on that change then. it would be awesome if i could hide some of the validation for that in jsonschema, but i'm not super familiar with being fancy in jsonschema | 14:31 |
| sdague | yeh, alex_xu / oomichi are the ones that probably would know best how to do that | 14:32 |
| sdague | I'm just not sure how we change out the type on a field | 14:32 |
| *** burt has joined #openstack-nova | 14:32 | |
| sdague | there is definitely already an enum mechanism | 14:33 |
| *** armax has joined #openstack-nova | 14:33 | |
| * alex_xu reading the log | 14:33 | |
| *** devkulkarni has joined #openstack-nova | 14:33 | |
| *** zz_dimtruck is now known as dimtruck | 14:34 | |
| *** edand has quit IRC | 14:34 | |
| mriedem | alex_xu: basically, https://review.openstack.org/#/c/316398/35/nova/api/openstack/compute/schemas/servers.py@37 | 14:34 |
| mriedem | can we define server.networks as an array or a string? | 14:34 |
| mriedem | well, array or enum? | 14:34 |
| mriedem | maybe using oneOf? | 14:35 |
| mriedem | like here https://review.openstack.org/#/c/316398/35/nova/api/openstack/compute/schemas/servers.py@89 | 14:35 |
| alex_xu | mriedem: i'm thinking of oneOf also | 14:35 |
| mriedem | ok, i'll play with that i guess | 14:35 |
| mriedem | that's going to be my day today | 14:36 |
| *** armax has quit IRC | 14:37 | |
| *** mdrabe has quit IRC | 14:37 | |
| *** shaohe_feng has quit IRC | 14:38 | |
| *** shaohe_feng has joined #openstack-nova | 14:38 | |
| *** rtheis has quit IRC | 14:38 | |
| openstackgerrit | Carl Baldwin proposed openstack/nova: Abort on HostNotCompatibleWithFixedIpsClient https://review.openstack.org/346278 | 14:39 |
| *** jdurgin1 has joined #openstack-nova | 14:39 | |
| *** roeyc has quit IRC | 14:39 | |
| *** vishwanathj_zzz is now known as vishwanathj | 14:40 | |
| *** JoseMello has quit IRC | 14:40 | |
| sdague | mriedem: ok, I'll try to run with your test fixes | 14:40 |
| sdague | to get the functional tests working on the neutron side | 14:40 |
| alex_xu | mriedem: we have example like this https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/schemas/user_data.py#L23 I think pretty close | 14:41 |
| sdague | mriedem: before you go completely down that rabbit hole. The os-brick locking issue remains in a nebulous state | 14:41 |
| *** kevinz has quit IRC | 14:41 | |
| *** devkulkarni has quit IRC | 14:42 | |
| syjulian | jaypipes, mikal: hi michael and jay. If you guys are available could you review this third part of my change https://review.openstack.org/#/c/354277/ ? | 14:42 |
| *** krtaylor has quit IRC | 14:42 | |
| *** devkulkarni has joined #openstack-nova | 14:43 | |
| mriedem | sdague: this? https://review.openstack.org/#/c/355215/ | 14:43 |
| mriedem | i started ^ | 14:43 |
| mriedem | but there are still like 10 failures | 14:43 |
| *** andrearosa has quit IRC | 14:44 | |
| *** ducttape_ has quit IRC | 14:44 | |
| sdague | mriedem: yeh, I'll run with that, I've got the environment up locally anyway | 14:44 |
| mriedem | sdague: on the lock_path thing, i think the best we can do for newton is (1) use the same lock_path for nova/cinder in devstack if we're going to recommend that in release notes, (2) add the release notes to nova/cinder - for whatever that's worth, and (3) update the lock_path option help text in oslo | 14:44 |
| *** mdrabe has joined #openstack-nova | 14:45 | |
| mriedem | we're not going to start using tooz at this point | 14:45 |
| mriedem | in newton i mean | 14:45 |
| sdague | ok, so I had blocked 1 until 2 was sorted | 14:45 |
| mriedem | and i blocked 2 until 1 and 3 were done :) | 14:45 |
| mriedem | but i think patrickeast and i are on the same page | 14:45 |
| sdague | no, 2 has to go first | 14:45 |
| sdague | we do this all the time on the grenade side | 14:45 |
| mriedem | well, i didn't block, but i asked for 1 and 3 | 14:46 |
| sdague | you don't get to hack the gate with non default config unless it's explained in projects | 14:46 |
| mriedem | sure, 3 is what i really wanted | 14:46 |
| mriedem | the actual config option help text getting updated | 14:46 |
| sdague | what do you envision 3 looks like? | 14:46 |
| mriedem | https://review.openstack.org/#/c/355269/1/oslo_concurrency/lockutils.py | 14:47 |
| *** shaohe_feng has quit IRC | 14:48 | |
| sdague | mriedem: so... this also means that cinder & nova services have to run as the same users selinux context, right? | 14:48 |
| *** shaohe_feng has joined #openstack-nova | 14:48 | |
| *** sdake has quit IRC | 14:48 | |
| sdague | dansmith / danpb - do you know what the rhel story is there? | 14:48 |
| mriedem | sdague: just commented on something like that in the patch | 14:49 |
| mriedem | since it says, 'restrict this for the user that needs to write to the dir' but now you'd have 2 | 14:49 |
| danpb | sdague: well they wouldn't need to runthe same context, but you'd need to grant both of them permission to write to those files | 14:50 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add ObjectListBase concat methods https://review.openstack.org/352587 | 14:50 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Fix migration list + MigrationList operation https://review.openstack.org/353694 | 14:50 |
| sdague | danpb: on rhel do the nova / cinder processes run under dedicated service users? or under some combinded one? | 14:51 |
| *** venusf has left #openstack-nova | 14:51 | |
| dansmith | sdague: different users | 14:52 |
| dansmith | sdague: but that's not an selinux thing | 14:52 |
| *** jbernard1 has joined #openstack-nova | 14:52 | |
| dansmith | sdague: so they need unix perms and selinux policy to touch the same things | 14:52 |
| sdague | dansmith: ok, so setting the lock_path the same will also mean they have to be in the same group, and that g+w bits have to exist on the lock_path and all files | 14:52 |
| sdague | for it to work under rhel | 14:53 |
| mriedem | sdague: so alternatively, | 14:53 |
| dansmith | sdague: no different for ubuntu, AFAIK | 14:53 |
| mriedem | this is latent and we punt | 14:53 |
| mriedem | and say, it's a race condition and you might hit it and might need to retry | 14:53 |
| sdague | dansmith: ok, sure, I don't have an ubuntu stack running right now | 14:53 |
| dansmith | sdague: just confirmed, cinder and nova run as different users on ubuntu | 14:53 |
| dansmith | sdague: selinux is easier to fix than the group-write perms, AFAIK | 14:54 |
| sdague | yeh, so this seems like a wreck for distros | 14:54 |
| dansmith | because it's not really user specific | 14:54 |
| sdague | dansmith: right | 14:54 |
| dansmith | you can directly say "nova can write cinder lock types" or whatever | 14:54 |
| sdague | I knew that rhel had other selinux policy here, I wasn't sure if that made it better or worse | 14:54 |
| danpb | you could use POSIX ACLs too | 14:54 |
| danpb | (to avoid group write + shared group ) | 14:55 |
| dansmith | sdague: I certainly agree that them having to share a file lock is kinda icky | 14:55 |
| danpb | but agree this is all just revolting | 14:55 |
| *** ociuhandu has joined #openstack-nova | 14:55 | |
| danpb | whats the actual thing cinder+nova are racing on ? | 14:55 |
| dansmith | danpb: thread on the ML | 14:56 |
| dansmith | sdague: if you run cinder and nova inside service containers, which is common and becoming more common, this won't be a possibility anyway | 14:56 |
| sdague | danpb the crux of the issue is os-brick uses @lockutils.synchronized and expects it to work, but it's being called from 2 different services | 14:56 |
| dansmith | well, depending on your container, but... | 14:56 |
| danpb | i saw the thread, but didn't see specific example of the commands hitting the problem | 14:56 |
| sdague | dansmith: ah, right | 14:56 |
| sdague | danpb: yeh, I'm not sure all the comands in question, I just saw the devstack patch which I was told was needed, which raised a red flag in my mind | 14:57 |
| danpb | is there a concrete example of something tha tbreaks ? | 14:57 |
| mriedem | there was an example, i'm trying to find it | 14:58 |
| *** shaohe_feng has quit IRC | 14:58 | |
| *** shaohe_feng has joined #openstack-nova | 14:59 | |
| mriedem | oh it's a comment in the review https://review.openstack.org/#/c/354502/ | 14:59 |
| mriedem | "As far as specific issues, we've seen races with attaching/detaching iscsi volumes where os-brick needs to determine whether it was the last LUN being used for an iscsi session. You can have problems where one detach removes a device while the attach process is running scans on the bus. Things like that." | 14:59 |
| *** kylek3h_ has joined #openstack-nova | 15:00 | |
| dansmith | X( | 15:00 |
| danpb | i can't help thinking some of that should be solved at a higher level | 15:00 |
| dansmith | you mean lower level? | 15:00 |
| mriedem | so i think everyone is in agreement that the current situation sucks, | 15:01 |
| mriedem | the question is what can be done, if anything, in newton | 15:01 |
| danpb | dansmith: no, as in avoid the need for os-brick to decide whether its the last LUN or not | 15:01 |
| mriedem | and if recommending that nova/cinder share a lock_path on the same host in the docs and release notes | 15:01 |
| dansmith | danpb: that's the point of brick to some degree I think | 15:02 |
| danpb | dansmith: by Nova keeping track of whether the iSCSI target is used by any VMs or not | 15:02 |
| mriedem | danpb: cinder should track that, and i think jgriffith is actually working on that | 15:02 |
| danpb | os-brick is stateless though afaik ? | 15:02 |
| *** kylek3h has quit IRC | 15:02 | |
| mriedem | os-brick is stateless correct | 15:02 |
| danpb | if it wasn't stateless, it wouldn't have to run iscsiadm commands to figure out if it was the last LUN, and thus wouldn't hit the race | 15:02 |
| *** hshiina has quit IRC | 15:02 | |
| jgriffith | mriedem: +1 | 15:03 |
| dansmith | mriedem: the service containers thing breaks any sort of file-based lock I think | 15:03 |
| jgriffith | dansmith: yes brick is stateless | 15:03 |
| jgriffith | and should be IMO | 15:03 |
| mriedem | dansmith: how about venvs? | 15:03 |
| mriedem | same situation? | 15:03 |
| dansmith | jgriffith: I know, I don't disagree, I meant that brick is supposed to insulate nova from knowing those things | 15:03 |
| dansmith | mriedem: no, venvs are fine | 15:03 |
| danpb | dansmith: well you can have shared folders between different containers, but again it all sucks | 15:03 |
| jgriffith | dansmith: yeah, I get ya | 15:04 |
| dansmith | danpb: sure, but that makes it extra terrible :) | 15:04 |
| jgriffith | dansmith: TBH that was what I thought the entire point of Brick was :) | 15:04 |
| dansmith | jgriffith: :) | 15:04 |
| jgriffith | otherwise we don't gain anything from it but another project to maintain | 15:04 |
| dansmith | that's a gain? :P | 15:04 |
| openstackgerrit | Raghuveer Shenoy proposed openstack/nova-specs: Support SR-IOV and PCI statistics through Nova API https://review.openstack.org/342001 | 15:05 |
| jgriffith | dansmith: you can't see my sarcasm and rolling eyes from there? | 15:05 |
| dansmith | jgriffith: I totally can, just playing along :) | 15:05 |
| jgriffith | Unfortunately none of what i'm working on will go in N | 15:05 |
| *** su_zhang has quit IRC | 15:05 | |
| jgriffith | dansmith: haha | 15:05 |
| *** devkulkarni has quit IRC | 15:05 | |
| *** devkulkarni has joined #openstack-nova | 15:06 | |
| *** devkulkarni has quit IRC | 15:06 | |
| sdague | mriedem: it's not stateless if it allows lock files | 15:06 |
| jgriffith | In retrospect I should've just tackled moving all of the tracking before trying to reimplement attach/detach... but honestly that code is so messy at this point hacking in state tracking was just plane awful | 15:06 |
| *** lpetrut has quit IRC | 15:06 | |
| *** rcernin has quit IRC | 15:06 | |
| jgriffith | sdague: it should be stateless... but Nova and Cinder aren't. And we have this weird thing now of both of those consuming/sharing that module at the same time | 15:07 |
| jgriffith | sdague: Our design is flawed | 15:07 |
| jgriffith | we're trying to pretend that brick and be a static module and be stateless, but it can't because we don't have a central entity tracking the connection info | 15:08 |
| jgriffith | sometimes it's in Nova, sometimes it's in Cinder, and even worse some times we just don't worry about it :( | 15:08 |
| johnthetubaguy | jgriffith: so the new plan is to add all that tracking in Cinder? | 15:08 |
| *** shaohe_feng has quit IRC | 15:08 | |
| jgriffith | johnthetubaguy: yes | 15:08 |
| *** shaohe_feng has joined #openstack-nova | 15:09 | |
| johnthetubaguy | jgriffith: cool, just checking I am still keeping track | 15:09 |
| jgriffith | johnthetubaguy: no more split brain or confusing hacks trying to guess what's going on and making assumptions | 15:09 |
| jgriffith | and for the record, most of the badness is on the cinder side | 15:09 |
| johnthetubaguy | jgriffith: yeah, it sounds good, I am basically talking to neutron about doing the same thing with port bindings | 15:09 |
| jgriffith | johnthetubaguy: ahh.. that makes sense | 15:09 |
| jgriffith | similar problems/challenges | 15:10 |
| jgriffith | definitely a pattern we could use more broadly I think | 15:10 |
| johnthetubaguy | jgriffith: they have the added complexity of needing to track the *active* binding, although its driver specific | 15:10 |
| *** kevinz has joined #openstack-nova | 15:10 | |
| jgriffith | yeah, that's EXACTLY what led to the hackiness in Cinder | 15:10 |
| *** mvk has quit IRC | 15:10 | |
| jgriffith | lots of *clever* adds to the Cinder manager to force it to specific drivers will | 15:11 |
| jgriffith | and that's kinda why things are such a mess now | 15:11 |
| johnthetubaguy | yeah, for neutron I am assuming all drivers need the active tracking | 15:11 |
| johnthetubaguy | in my current draft of things, anyways | 15:11 |
| jgriffith | it became the least understood code in Cinder... although now it no longer has that distinction for better or worse | 15:11 |
| johnthetubaguy | cinder its more the shared per host connection issue I guess? | 15:11 |
| sdague | ok, well regardless of how we got here, how do we come up with something that's not a completely fail for newton? | 15:12 |
| jgriffith | sdague: for newton? | 15:12 |
| sdague | yes | 15:12 |
| sdague | because this shared lock_path I don't think is even viable | 15:12 |
| dansmith | sdague: oh I have a great idea for newton | 15:12 |
| sdague | given the unix user issues | 15:12 |
| johnthetubaguy | sdague: the brick and lock files bit? | 15:13 |
| sdague | yes | 15:13 |
| dansmith | sdague: time.sleep(random.random()*100) | 15:13 |
| dansmith | instalock ^ | 15:13 |
| jgriffith | sdague: the problem is that I can't come up with a way to address it that doesn't include API changes | 15:13 |
| jgriffith | volume-attach/volume-detach | 15:13 |
| jgriffith | and that ship sailed a few weeks ago as I understood it | 15:13 |
| sdague | is smcginnis up and about yet? | 15:14 |
| smcginnis | sdague: Yep | 15:14 |
| johnthetubaguy | so I understand someone is going to shoot me for saying this, did we rule out using sudo/root to acquire the shared lock file, as a short term hack? | 15:14 |
| * smcginnis reads backscroll | 15:14 | |
| danpb | the ideal way to solve it would be to have a privileged linux iscsi service with API you could just call out to perform tasks | 15:14 |
| dansmith | johnthetubaguy: that doesn't fix containers | 15:15 |
| danpb | that service would take care of locking itself | 15:15 |
| sdague | danpb: yeh, and that for every driver | 15:15 |
| dansmith | johnthetubaguy: service containers I mean, without having to have a special piece of shared filesystem mapped in | 15:15 |
| johnthetubaguy | dansmith: yes, dang it. | 15:15 |
| sdague | danpb: the same pattern for iscsi seems to be repeated for others | 15:15 |
| danpb | sdague: in some ways, we'd actually be doing just that with privsep | 15:16 |
| sdague | the race windows might be smaller given how the backends work, but I don't think we can write them off | 15:16 |
| johnthetubaguy | thats basically the privsep pattern, I guess | 15:16 |
| danpb | and since privsep runs privileged, it doesn't have the user/group perm problem | 15:16 |
| *** kevinz has quit IRC | 15:16 | |
| *** kevinz has joined #openstack-nova | 15:17 | |
| danpb | so you could say os-brick could use lock.synchronized, provided it used privsep, and annotated the privileged methods, not the unprivileged ones | 15:17 |
| jgriffith | has anybody considered how valuable the locks in brick actually are? | 15:17 |
| johnthetubaguy | does privsep communication work between containers? | 15:17 |
| danpb | johnthetubaguy: don't think anyone's figured that out yet :-) | 15:18 |
| jgriffith | sdague: kinda to your point about the race window.... | 15:18 |
| sdague | ok, wait, lets hold off on containers for a second, because the scope of the problem might be different than we were considering | 15:18 |
| dansmith | johnthetubaguy: if you're using filesystem locks, then no | 15:18 |
| danpb | dansmith: not neccessarily | 15:18 |
| *** shaohe_feng has quit IRC | 15:19 | |
| dansmith | danpb: lock.synchronized only works single-process unless you use external=True | 15:19 |
| danpb | dansmith: service containers already have bucket loads of shared filesystem paths | 15:19 |
| jgriffith | we're really just talking about two calls right? Connect, Disconnect and extend? | 15:19 |
| smcginnis | Maths :) | 15:19 |
| danpb | so it potentially only means adding another fielsystem sub-dir passthrough for the service container | 15:19 |
| sdague | jgriffith: well, I think that's the overall question | 15:19 |
| dansmith | danpb: but they depend on how your containers are set up | 15:19 |
| *** shaohe_feng has joined #openstack-nova | 15:19 | |
| dansmith | danpb: it just means yet another thing, agreed, I just think it's terrible to do that | 15:20 |
| jgriffith | sdague: well my point being that I've never thought connect and disconnect should have them at all | 15:20 |
| sdague | " | 15:20 |
| sdague | As far as specific issues, we've seen races with attaching/detaching iscsi volumes where os-brick needs to determine whether it was the last LUN being used for an iscsi session. You can have problems where one detach removes a device while the attach process is running scans on the bus. Things like that. | 15:20 |
| sdague | " | 15:20 |
| *** phschwartz has quit IRC | 15:20 | |
| jgriffith | sdague: we already provide reserve/unreserve on the Cinder side, we should just use those | 15:20 |
| smcginnis | jgriffith: I think we can have rescan collisions if we don't lock. | 15:20 |
| johnthetubaguy | jgriffith: long term I think we can, but for newton? | 15:21 |
| jgriffith | smcginnis: on you mean the initiator specific side? | 15:21 |
| smcginnis | jgriffith: Yes. Although if we are serializing things outside of the attach flow we shouldn't need to be concerned with that in brick. | 15:21 |
| johnthetubaguy | so... if every brick user had its own lock, what breaks? cinder rescan while nova is attaching? | 15:21 |
| smcginnis | johnthetubaguy: I believe that is the concern from my understanding. | 15:22 |
| jgriffith | cinder rescan? | 15:22 |
| dansmith | is the problem when we have cinder co-located with n-cpu, or only between competing n-cpu things? | 15:22 |
| sdague | smcginnis: just in the iscsi case? | 15:22 |
| sdague | dansmith: yes, I believe so | 15:22 |
| dansmith | sdague: which? | 15:23 |
| sdague | the colocation | 15:23 |
| smcginnis | jgriffith: Err.. | 15:23 |
| dansmith | okay | 15:23 |
| dansmith | thought so, but wanted to make sure | 15:23 |
| smcginnis | Yeah, I believe this is only a colocation issue. | 15:23 |
| sdague | that's what patrickeast said in the devstack review, which is why he didn't think it was a big deal to make the change - https://review.openstack.org/#/c/341744/ - because a lot of people would be isolated fromit | 15:24 |
| dansmith | in that case, | 15:25 |
| dansmith | I think I'd rather not introduce some half-baked hack that potentially has wide implications, and just document it | 15:25 |
| *** moshele has quit IRC | 15:25 | |
| dansmith | the race is either leaking targets or potentially causing a build to fail, right? | 15:25 |
| johnthetubaguy | the work around is run both services as the same user, with access to a shared filesystem? | 15:26 |
| dansmith | we have known issues in release notes for a reason.. "if you run these two things on the same box, they may compete in the following way" | 15:26 |
| mriedem | i'm fine with a release note with a known issue, but i think we also want something in the lock_path option description | 15:27 |
| smcginnis | I like the idea of also putting it in the help text. Just to try to raise visibility. | 15:27 |
| dansmith | johnthetubaguy: not sure the existing locks protect us, otherwise we'd fail to ever grab locks when they're owned by other processes | 15:27 |
| mriedem | because anyone new to opentsack in ocata isn't going to see the newton release note | 15:27 |
| dansmith | johnthetubaguy: we'd fail instead of block it hink | 15:27 |
| dansmith | mriedem: new people don't read release notes? | 15:27 |
| prometheanfire | mriedem: https://review.openstack.org/355526 you might want to combine this with your review | 15:27 |
| mriedem | dansmith: for newton? | 15:27 |
| dansmith | mriedem: we keep this as a known issue until we fix it | 15:28 |
| mriedem | dansmith: release notes are based on tag | 15:28 |
| sdague | ok, so we'd just have to remember to carry forward | 15:28 |
| dansmith | mriedem: I'm just saying don't rush this for newton, | 15:28 |
| sdague | mriedem: yeh, but we can manually carry forward | 15:28 |
| dansmith | maybe jgriffith's tracking stuff will make it not a problem in ocata, right? | 15:28 |
| openstackgerrit | Carl Baldwin proposed openstack/nova: Abort on HostNotCompatibleWithFixedIpsClient https://review.openstack.org/346278 | 15:28 |
| mriedem | dansmith: well, unless you're using icehouse level cinder | 15:28 |
| mriedem | :) | 15:28 |
| mriedem | cough rax cough | 15:28 |
| johnthetubaguy | dansmith: I suspect thats a bit of code where we just throw away exceptions again? I remember something about that in the past | 15:29 |
| dansmith | mriedem: in which case you already know :) | 15:29 |
| prometheanfire | mriedem: hi :P | 15:29 |
| *** shaohe_feng has quit IRC | 15:29 | |
| jgriffith | dansmith: I think so...but frankly I'm trying to get some clarification from folks on why those locks are even there... still no real concrete answers | 15:29 |
| sdague | smcginnis: is there a narrower retry path we could put into place to mitigate in os-brick | 15:29 |
| dansmith | johnthetubaguy: I hope not, because we're hoping to block on the lock, but.. | 15:29 |
| prometheanfire | guess what version of openstack osuosl has deployed | 15:29 |
| sdague | so we're not pretending to rely on those locks | 15:29 |
| * jgriffith is chatting with authors in cinder channel | 15:29 | |
| prometheanfire | I'll give a hint, it's not maintained anymore | 15:29 |
| mriedem | prometheanfire: he has a depends-on to https://review.openstack.org/#/c/355526/ | 15:29 |
| mriedem | which is the same thing you were saying to combine it with | 15:29 |
| mriedem | the depends-on should be fine | 15:29 |
| smcginnis | sdague: We're kind of discussing that now. | 15:29 |
| *** shaohe_feng has joined #openstack-nova | 15:29 | |
| smcginnis | Need to understand exactly when this is an issue and why. | 15:30 |
| prometheanfire | mriedem: updating gr without uc will break the world | 15:30 |
| sdague | ok, I'll go over there | 15:30 |
| mriedem | prometheanfire: the g-r change depends on the u-c change which depends on the nova change | 15:30 |
| mriedem | prometheanfire: the chaining is all good | 15:30 |
| prometheanfire | also, do people know that osuosl has an openstack cluster running on openpower | 15:30 |
| johnthetubaguy | sdague: I am good with the known issue, I am just curious if there is a way to reduce the sting | 15:30 |
| prometheanfire | mriedem: oh, I totally missed the depends on at the bottom | 15:30 |
| mriedem | :) | 15:31 |
| *** pwnall has quit IRC | 15:31 | |
| *** su_zhang has joined #openstack-nova | 15:31 | |
| prometheanfire | ya, used to it being spaced on it's own line | 15:31 |
| prometheanfire | anyway, ack'd | 15:31 |
| *** diogogmt has joined #openstack-nova | 15:33 | |
| *** armax has joined #openstack-nova | 15:33 | |
| *** sameo has joined #openstack-nova | 15:34 | |
| *** mdrabe has quit IRC | 15:35 | |
| *** mdrabe has joined #openstack-nova | 15:35 | |
| *** baoli_ has quit IRC | 15:35 | |
| *** kfarr has joined #openstack-nova | 15:37 | |
| *** armax has quit IRC | 15:37 | |
| *** belmoreira has quit IRC | 15:38 | |
| *** shaohe_feng has quit IRC | 15:39 | |
| *** harlowja_at_home has joined #openstack-nova | 15:40 | |
| *** shaohe_feng has joined #openstack-nova | 15:40 | |
| *** haplo37__ has joined #openstack-nova | 15:41 | |
| *** devkulkarni has joined #openstack-nova | 15:41 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for service options. https://review.openstack.org/332861 | 15:43 |
| mriedem | alex_xu: while you're around, i posted a question to the ML on this get me a network server.networks thing, i was fixing the legacy support for passing a network uuid of br-uuid form, but no longer really need to do that | 15:43 |
| mriedem | so wondering if i should continue fixing that in the 2.37 microversion or just drop it to avoid additional complexity in the change | 15:43 |
| mriedem | i'm kind of leaning toward the latter | 15:43 |
| *** pwnall has joined #openstack-nova | 15:43 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Improve help text for glance options https://review.openstack.org/324007 | 15:43 |
| mriedem | if we wanted to fix that at some point it would have to be a new microversion though | 15:43 |
| *** sameo has quit IRC | 15:44 | |
| mriedem | unless we consider passing non-uuids for networks as a bug | 15:44 |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for service options. https://review.openstack.org/332861 | 15:44 |
| *** sameo has joined #openstack-nova | 15:44 | |
| *** rlrossit_ has quit IRC | 15:44 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: DO NOT MERGE: get instance diagnostic https://review.openstack.org/355540 | 15:47 |
| *** PaulMurray has quit IRC | 15:47 | |
| *** ekuris has quit IRC | 15:47 | |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: placement: add filtering by attrs to resource_providers https://review.openstack.org/343046 | 15:49 |
| *** shaohe_feng has quit IRC | 15:49 | |
| *** shaohe_feng has joined #openstack-nova | 15:50 | |
| *** pbandark has quit IRC | 15:50 | |
| *** hshiina has joined #openstack-nova | 15:50 | |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: ironic_host_manager: fix population of instances info on schedule https://review.openstack.org/346966 | 15:50 |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: ironic_host_manager: fix population of instances info on start https://review.openstack.org/347948 | 15:50 |
| *** Apoorva has joined #openstack-nova | 15:51 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Deprecate all topic conf options https://review.openstack.org/351608 | 15:51 |
| *** jichen has quit IRC | 15:53 | |
| *** Sukhdev has joined #openstack-nova | 15:53 | |
| openstackgerrit | Julian proposed openstack/nova: Add unit tests for nova.virt.firewall.IpTablesFirewallDriver (Part 4) https://review.openstack.org/355550 | 15:56 |
| raj_singh | sfinucan: When you get a chance can you comment on https://review.openstack.org/#/c/353580 | 15:56 |
| *** Apoorva has quit IRC | 15:57 | |
| *** brig-man has joined #openstack-nova | 15:57 | |
| *** mtanino has joined #openstack-nova | 15:57 | |
| sfinucan | raj_singh: Hey :) I think my comments still stand? | 15:58 |
| sfinucan | Those markers at the top of the file are to indicate that work still needs to be done on the file. No point removing them and generating burndown if it's not actually true, no? | 15:58 |
| *** shaohe_feng has quit IRC | 16:00 | |
| raj_singh | sfinucan: Well for other cases yes, but for grouping we are removing the flags and adding todo's. | 16:00 |
| sfinucan | raj_singh: Can I ask why? The marker seemed like a good approach? | 16:00 |
| raj_singh | sfinucan: And for some political reasons too | 16:00 |
| raj_singh | :) | 16:00 |
| *** shaohe_feng has joined #openstack-nova | 16:01 | |
| sfinucan | raj_singh: ah | 16:01 |
| *** yamahata has joined #openstack-nova | 16:01 | |
| *** BjoernT has joined #openstack-nova | 16:02 | |
| *** mat128 is now known as mat128|afk | 16:02 | |
| *** GB21 has quit IRC | 16:05 | |
| *** kylek3h_ has quit IRC | 16:06 | |
| *** markus_z has quit IRC | 16:07 | |
| *** kylek3h_ has joined #openstack-nova | 16:08 | |
| *** shaohe_feng has quit IRC | 16:10 | |
| *** ihrachys has quit IRC | 16:10 | |
| *** dims has quit IRC | 16:10 | |
| *** shaohe_feng has joined #openstack-nova | 16:11 | |
| kfarr | sdague thanks for your comments on https://review.openstack.org/#/c/338518/ . Do you think a google hangout would still be worthwhile? Thanks!! | 16:11 |
| openstackgerrit | Andrey Volkov proposed openstack/nova: Eliminate additional DB queries in nova lists https://review.openstack.org/355558 | 16:11 |
| sdague | kfarr: is there anything in newton that you are blocked on here? | 16:11 |
| *** efried has quit IRC | 16:12 | |
| sdague | because it wasn't entirely clear from that spec, and if the answer to that is no, I'd like to push this until after freeze to talk about | 16:12 |
| sdague | and if the answer is yes, I'd like to understand that one more | 16:12 |
| *** raj_singh has left #openstack-nova | 16:13 | |
| *** raj_singh has joined #openstack-nova | 16:13 | |
| mriedem | sdague: i'm going to keep the change in 2.37 for enforcing server.networks.uuid to be a uuid rather than just a string, | 16:14 |
| mriedem | we should close that door while we're in here | 16:14 |
| sdague | mriedem: +1 | 16:15 |
| mriedem | i've already waffled over that 3 times in the last hour :) | 16:15 |
| kfarr | sdague, I was just concerned that the original patch would be reverted if I did not attempt to address concerns this cycle | 16:15 |
| *** zenoway has quit IRC | 16:15 | |
| *** zenoway has joined #openstack-nova | 16:16 | |
| kfarr | sdague but if the code is currently fine as is for Newton, I'm also ok with waiting until freeze to discuss it | 16:16 |
| sdague | yeh, I think we're ok for newton. | 16:16 |
| *** lucasagomes is now known as lucas-afk | 16:17 | |
| sdague | honestly, I think the post freeze conversation needs to be a wider one about could we make barbican required at the iaas layer, because there are a lot of encryption features that would be good to be there out of the box that are all wonky with barbican not being in the fold | 16:18 |
| *** ducttape_ has joined #openstack-nova | 16:18 | |
| mdbooth | dansmith: So, the gotFiles and imported_files thing | 16:18 |
| dansmith | mdbooth: the __init__ method of the new fixture is not called by any existing tests | 16:18 |
| mdbooth | dansmith: Right, but the existing tests already call these things gotFiles and imported_files | 16:18 |
| dansmith | mdbooth: sure, but none will break if you name that one properly | 16:18 |
| mdbooth | I'm not a fan of the way these tests work, btw, which is why I wrote the disks dict | 16:18 |
| dansmith | mdbooth: I think the previous pass you got was because you had to slowly evolve that name | 16:19 |
| mdbooth | I consider them legacy, and imminently doomed | 16:19 |
| dansmith | so keep anything named so that you don't have to break tests, but for new things I think you should be moving to the proper name | 16:19 |
| kfarr | sdague, ok thanks. Agreed that it would be good to have that conversation. Would be fantastic to have barbican as a dependency. I'll be back in touch at the beginning of September then? | 16:19 |
| sdague | kfarr: sounds great | 16:19 |
| mdbooth | Calling them the same in both places is less confusing imho, even if the name is horrible | 16:19 |
| mdbooth | This is purely for interoperability until the users have been rewritten | 16:20 |
| mdbooth | It's not an interface which will hang around, because the fixture provides a better one | 16:20 |
| *** shaohe_feng has quit IRC | 16:20 | |
| *** zenoway has quit IRC | 16:20 | |
| mdbooth | My new and rewritten tests exclusively use the new interfaces | 16:20 |
| mdbooth | dansmith: Anyway, I can rename them, it's not a big deal | 16:21 |
| *** shaohe_feng has joined #openstack-nova | 16:21 | |
| dansmith | mdbooth: you know I'm talking only about the gotFiles in the __init__ on the constructor and on the class, right? those are used in like four places, and all new | 16:21 |
| *** su_zhang has quit IRC | 16:21 | |
| mdbooth | Sure, they're new because the constructor is new. But all the callers have an existing thing called gotFiles which they used to update themselves. | 16:22 |
| mdbooth | Now the pass it to the fixture, and the fixture updates it. | 16:22 |
| mdbooth | When they're rewritten, they will no longer have a thing called gotFiles. | 16:22 |
| dansmith | whatever, don't rename it if you want and if jaypipes is fine then whatever.. however, it's all new code, it's a thing that is used internally to a few things with the fixture, and should not be named like that | 16:23 |
| mdbooth | This is purely for interoperability so I don't have to rewrite those tests completely. | 16:23 |
| dansmith | it is NOT for interop :) | 16:23 |
| dansmith | nothing existing calls it :) | 16:23 |
| jaypipes | dansmith: yeah, 345427 removes gotFiles entirely. currently reviewing that patch now. | 16:24 |
| *** lbeliveau has quit IRC | 16:24 | |
| jaypipes | dansmith: though I agree with you it's new code. | 16:24 |
| dansmith | yeah. | 16:25 |
| jaypipes | dansmith: but five patches later, it goes bye-bye | 16:25 |
| mdbooth | jaypipes: Incidentally, I didn't especially like those new generator functions in driver. I have a patch locally which pulls them all into a new module. | 16:25 |
| mdbooth | The new module basically works the same way, but has a cleaner calling interface. | 16:25 |
| mdbooth | It will have its own tests, and this, I think will address your real concerns about the unitiness of the driver tests | 16:26 |
| mdbooth | The new interface is a StoragePolicy object with methods like local_disk() and ephemerals() | 16:26 |
| mdbooth | And it returns objects on which you just call create() | 16:26 |
| jaypipes | mdbooth: yeah, I'm pretty deep into the reviews of this series now. I'm liking the end state. | 16:26 |
| mdbooth | So tests can easily mock at that level, and just check that we called create() | 16:27 |
| mdbooth | No need to worry about what ephemerals there should be, or whether there was a legacy backing file or whatever | 16:27 |
| *** lbeliveau has joined #openstack-nova | 16:28 | |
| *** harlowja_at_home has quit IRC | 16:28 | |
| dansmith | mdbooth: so I just made the change in like 60 seconds | 16:28 |
| mdbooth | Anyway, as I say it's basically just code motion. | 16:28 |
| dansmith | if you're respinning I don't think it's a large ask | 16:28 |
| *** ametts_ has joined #openstack-nova | 16:28 | |
| dansmith | but if jaypipes is going to +2 it anyway, then do what you want | 16:28 |
| mdbooth | dansmith: You rewrote the tests which use gotFiles to assert on disks instead? | 16:28 |
| dansmith | mdbooth: nothing outside the fixture references the fixture's gotFiles | 16:29 |
| mdbooth | dansmith: Yeah, that change is absolutely no problem. | 16:29 |
| *** yamahata has quit IRC | 16:29 | |
| mdbooth | I was just proposing to use the same, wrong name in both places :) | 16:29 |
| mdbooth | You could rename it in the callers too, I guess. | 16:30 |
| *** ametts has quit IRC | 16:30 | |
| mdbooth | Is that what you did? | 16:30 |
| mdbooth | dansmith: Anyway, you had at least 1 thing which should have been a pep8 failure imho | 16:30 |
| *** shaohe_feng has quit IRC | 16:30 | |
| mdbooth | And you ask for some re-indentation | 16:30 |
| cmartin | I am trying to work on some mox to mock conversion in test_virt_drivers and I am unsure of how to convert self.mox.UnsetStubs() can anybody help or point me to some documentation I can refer to? | 16:30 |
| mdbooth | So I was going to respin anyway | 16:30 |
| mdbooth | Not a problem | 16:30 |
| openstackgerrit | Pushkar Umaranikar proposed openstack/nova: Add related options to floating ip config options https://review.openstack.org/355561 | 16:31 |
| *** shaohe_feng has joined #openstack-nova | 16:31 | |
| dansmith | mdbooth: yes, you could fix it with git show and sed :) | 16:31 |
| mdbooth | Meh. I try to keep my patches small... :P | 16:31 |
| mdbooth | It's np, I'll do it first thing in the morning. | 16:31 |
| dansmith | mdbooth: it won't change the line count at all | 16:31 |
| *** Apoorva has joined #openstack-nova | 16:31 | |
| dansmith | mdbooth: I'll do it for you in a bit | 16:31 |
| mdbooth | dansmith: Ok, cool. Thanks. I'll check in the morning and do it anyway if you haven't. | 16:32 |
| openstackgerrit | John Garbutt proposed openstack/nova: Improve consistency in libvirt https://review.openstack.org/353679 | 16:32 |
| *** Jeffrey4l_ has quit IRC | 16:33 | |
| dansmith | mdbooth: whatever, it'll make conflicts on the following patches | 16:33 |
| dansmith | mdbooth: I'll take my meds and shut up for five patches | 16:33 |
| *** bwallis has quit IRC | 16:36 | |
| *** kevinz has quit IRC | 16:36 | |
| mriedem | cmartin: the mox->mock conversion is on hold as an effort until ocata | 16:38 |
| cmartin | mriedem: oh ok then I'll find a different blueprint to work on | 16:39 |
| *** nic has joined #openstack-nova | 16:39 | |
| mriedem | cmartin: fwiw, we're really closing down for the release, so at this point, bug triage, reviews and bug fixes are most valuable | 16:40 |
| mriedem | we've got 3 weeks until feature freeze | 16:40 |
| jaypipes | mdbooth: k, I'm clear through reviews of the largest series of your code. let me know if you have any questions on my comments. | 16:40 |
| *** shaohe_feng has quit IRC | 16:41 | |
| *** devkulkarni1 has joined #openstack-nova | 16:41 | |
| *** shaohe_feng has joined #openstack-nova | 16:41 | |
| *** dims has joined #openstack-nova | 16:42 | |
| cmartin | mriedem: ok I'll just do that then | 16:43 |
| *** devkulkarni has quit IRC | 16:44 | |
| *** oomichi has joined #openstack-nova | 16:45 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Deprecate all topic conf options https://review.openstack.org/351608 | 16:45 |
| mriedem | cmartin: something else that's easy to hack on https://bugs.launchpad.net/python-novaclient/+bug/1613105 | 16:45 |
| openstack | Launchpad bug 1613105 in python-novaclient "functional tests leak cinder volumes" [Undecided,New] | 16:45 |
| mriedem | there are a few places in the novaclient functional tests that create a volume for testing but don't cleanup the volume | 16:46 |
| *** dims has quit IRC | 16:47 | |
| *** anush__ has joined #openstack-nova | 16:47 | |
| jaypipes | syjulian: done. | 16:48 |
| *** shaohe_feng has quit IRC | 16:51 | |
| *** anush_ has quit IRC | 16:51 | |
| *** shaohe_feng has joined #openstack-nova | 16:52 | |
| *** dims has joined #openstack-nova | 16:53 | |
| *** tonytan4ever has quit IRC | 16:54 | |
| *** rama_y has joined #openstack-nova | 16:54 | |
| *** GB21 has joined #openstack-nova | 16:55 | |
| *** lucas-afk is now known as lucasagomes | 16:58 | |
| sdague | mriedem: you want to +1 this - https://review.openstack.org/355566 ? | 16:59 |
| mriedem | done | 17:00 |
| *** anush__ has quit IRC | 17:00 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Deprecate all topic conf options https://review.openstack.org/351608 | 17:00 |
| *** sdake has joined #openstack-nova | 17:00 | |
| *** shaohe_feng has quit IRC | 17:01 | |
| *** shaohe_feng has joined #openstack-nova | 17:02 | |
| *** bwallis has joined #openstack-nova | 17:02 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/354459 | 17:03 |
| sfinucan | BobBall: johnthetubaguy: I'm going to jump offline soon, but before I forget: is the XenAPI code still stuck with Python 2.4? | 17:05 |
| johnthetubaguy | sfinucan: for the plugins, yes | 17:05 |
| sfinucan | johnthetubaguy: drat - noqa time so | 17:05 |
| BobBall | So yes, in dom0 | 17:05 |
| BobBall | non-py3 compatible bits? | 17:05 |
| ildikov | johnthetubaguy: mriedem: we have the meeting if there's anything from your side you would like to hear or say | 17:05 |
| *** jaosorior has quit IRC | 17:06 | |
| johnthetubaguy | ildikov: did we cover most bits in channel already? not sure if we got got the bottom of the locks stuff | 17:06 |
| ildikov | johnthetubaguy: I haven't read the last bits, but that's covered and nothing else regarding the API interactions stuff, then it should be fine | 17:07 |
| *** mat128|afk is now known as mat128 | 17:07 | |
| *** harlowja has joined #openstack-nova | 17:08 | |
| *** yamahata has joined #openstack-nova | 17:09 | |
| *** shaohe_feng has quit IRC | 17:11 | |
| *** edand has joined #openstack-nova | 17:12 | |
| *** shaohe_feng has joined #openstack-nova | 17:12 | |
| *** ihrachys has joined #openstack-nova | 17:12 | |
| sfinucan | BobBall: Yeah, I'm working on that .py extension thing again | 17:13 |
| sfinucan | Can't fix 'Exception, exc' -> 'Exception as exc' | 17:14 |
| *** danpb has quit IRC | 17:15 | |
| beagles | vladikr, quick question about PCI device lists... say someone has a PF linked to an OVS bridge and whitelists it, would nova be able to filter it out of the pci_stats? | 17:15 |
| *** oanson has joined #openstack-nova | 17:16 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: plugins/xenserver: Add '.py' extension https://review.openstack.org/289431 | 17:17 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: plugins/xenserver: Resolve PEP8 issues https://review.openstack.org/355574 | 17:17 |
| *** su_zhang has joined #openstack-nova | 17:17 | |
| beagles | vladikr, on one hand I don't really want it to be because I don't like the idea of what could happen if a guest actually boots and pulls it in | 17:18 |
| vladikr | beagles, hmm... I doubt it. if the device is online and has a capability of a physical function , I think we would count it . | 17:19 |
| vladikr | beagle, the pf will disappear from the host while the guest is using it, iirc | 17:20 |
| vladikr | beagles, ^^ | 17:21 |
| beagles | vladikr, ack | 17:21 |
| *** su_zhang has quit IRC | 17:21 | |
| *** yuelongguang has quit IRC | 17:21 | |
| beagles | vladikr, and at that point, if the interface was being used for something important ... kaboom :) | 17:21 |
| vladikr | beagles, do you think we should filter it out? | 17:22 |
| *** shaohe_feng has quit IRC | 17:22 | |
| *** su_zhang has joined #openstack-nova | 17:22 | |
| vladikr | beagles, true | 17:22 |
| *** krtaylor has joined #openstack-nova | 17:22 | |
| beagles | vladikr, also a confirmation - a device won't be recognized by the resource tracker *unless* it has virtual functions associated with it? | 17:22 |
| *** shaohe_feng has joined #openstack-nova | 17:22 | |
| vladikr | beagles, it doesn't has to have active virtual functions, just a capability | 17:23 |
| beagles | vladikr, regarding filtering it... I don't know how we can do that really and really, this is the admin's responsibility to make sure that they don't something silly like that :) | 17:23 |
| *** yuelongguang has joined #openstack-nova | 17:23 | |
| beagles | vladikr, that's what the whitelist is for after all | 17:24 |
| vladikr | beagles, https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L5072 | 17:24 |
| vladikr | beagles, yea, true | 17:24 |
| *** mingyu has quit IRC | 17:24 | |
| *** lucasagomes is now known as lucas-dinner | 17:25 | |
| jaypipes | alaski: https://review.openstack.org/#/c/352610/4 | 17:25 |
| *** ayoung has quit IRC | 17:25 | |
| beagles | vladikr, in the nodedev-dumpxml, virt_functions doesn't show up unless there numvfs > 0 | 17:26 |
| beagles | vladikr, I don't know if it does the same with the API though | 17:26 |
| jaypipes | alaski: https://review.openstack.org/#/c/312696/ | 17:28 |
| *** ihrachys has quit IRC | 17:29 | |
| *** shaohe_feng has quit IRC | 17:32 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add REST API support for get me a network https://review.openstack.org/316398 | 17:33 |
| mriedem | sdague: alex_xu: oomichi: thar she blar ^ the new format makes the validation a lot simpler | 17:33 |
| vladikr | beagles, hmm.. you're right, I don't see it | 17:34 |
| oomichi | mriedem: thanks, will check it later. how about tempest side? | 17:34 |
| mriedem | oomichi: that's next | 17:34 |
| mriedem | and the novaclient change | 17:34 |
| mriedem | and the spec | 17:34 |
| oomichi | mriedem: ok, will check them also later | 17:35 |
| *** shaohe_feng has joined #openstack-nova | 17:35 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Return None in get_instance_id_by_floating_address https://review.openstack.org/351960 | 17:37 |
| *** rcernin has joined #openstack-nova | 17:37 | |
| *** kfarr has quit IRC | 17:37 | |
| beagles | vladikr, oah.. there's a comment in the code right below that moshele added that mentions it is dependent on libvirt version | 17:40 |
| *** TravT has quit IRC | 17:40 | |
| *** ayoung has joined #openstack-nova | 17:41 | |
| *** shaohe_feng has quit IRC | 17:42 | |
| *** shaohe_feng has joined #openstack-nova | 17:43 | |
| *** tbachman has joined #openstack-nova | 17:44 | |
| *** tonytan4ever has joined #openstack-nova | 17:45 | |
| *** dbite has joined #openstack-nova | 17:45 | |
| *** JoseMello has joined #openstack-nova | 17:47 | |
| beagles | vladikr, mmmm. does the pci whitelisting mechanism work for PFs? | 17:47 |
| *** raunak has joined #openstack-nova | 17:48 | |
| vladikr | beagles, ah, I was puzzled for a moment, why does it work on my workstation.. :) | 17:48 |
| vladikr | beagles, it supposed to | 17:48 |
| beagles | vladikr, I'm not sure how the match operation is supposed to work | 17:49 |
| vladikr | beagles, are you using an alias or are you attaching the pf as a neutron port? | 17:50 |
| beagles | vladikr, wait a sec.. misreading this a bit | 17:50 |
| beagles | vladikr, well, actually that begs the next question, are PFs stored in the pci_devices table/ | 17:51 |
| beagles | vladikr, it isn't from what I'm seeing on this particular host - I'm just trying to rule out misconfiguration :) | 17:51 |
| *** shaohe_feng has quit IRC | 17:52 | |
| vladikr | beagles, yes, afaik, they should be stored. | 17:52 |
| *** hshiina has quit IRC | 17:53 | |
| *** shaohe_feng has joined #openstack-nova | 17:53 | |
| oomichi | mriedem: dansmith pointed strictly on https://review.openstack.org/#/c/316398 | 17:54 |
| mriedem | gdi | 17:55 |
| dansmith | oomichi: mriedem appreciates typo fixes, as he is a fellow lover of proper spelling and grammar | 17:55 |
| *** GB21 has quit IRC | 17:56 | |
| *** edand has quit IRC | 17:56 | |
| dansmith | oomichi: other than that nit, I think it all makes sense to me.. are you close to +2 on that/ | 17:56 |
| vladikr | beagles, btw, if you are using an alias, you should add "device_type":"type-PF" to it, in addition to the vendor/product ids | 17:56 |
| *** ijw has joined #openstack-nova | 17:56 | |
| beagles | vladikr, mmmm... okay | 17:56 |
| beagles | vladikr, was using a devname and type-PF is specified | 17:56 |
| beagles | vladikr, it's kinda weird because the VFs are getting in | 17:57 |
| oomichi | dansmith: heh, I know. mriedem sometime put typos on his patches | 17:57 |
| *** Sukhdev has quit IRC | 17:58 | |
| mriedem | it's a test to see if anyone notices | 17:58 |
| dansmith | oomichi: I think he puts them in on purpose sometimes to see if people are paying attention | 17:58 |
| dansmith | hah | 17:58 |
| beagles | vladikr, the thing that caught my eye was https://github.com/openstack/nova/blob/master/nova/pci/devspec.py#L163 | 17:58 |
| mriedem | wow | 17:58 |
| beagles | vladikr, but then I realized I needed to check what PciAddress actually does ;) | 17:58 |
| vladikr | :) | 17:59 |
| oomichi | dansmith: I was ready to put my +2 because I missed it. I can avoid higher my disagreement ratio(that is already high | 18:00 |
| dansmith | oomichi: heh, okay, well, if you can hit it again once he fixes that typo then I'll circle back with it and we can be done :) | 18:00 |
| mriedem | i'm also going to move the 'br-x' uuid thing in the release note to the upgrade or other section | 18:01 |
| mriedem | since that's not really a feature | 18:01 |
| vladikr | beagles, if you'll switch to a debug mode you should see all the available device being reported | 18:02 |
| beagles | vladikr, yeah, they are appearing the hypervisor's list but I suspect they are being filtered out by the whitelist. I just need to work out what the "gotcha" is :) | 18:02 |
| *** shaohe_feng has quit IRC | 18:03 | |
| *** electrocucaracha has joined #openstack-nova | 18:03 | |
| *** shaohe_feng has joined #openstack-nova | 18:03 | |
| electrocucaracha | jaypipes: ping? | 18:04 |
| vladikr | beagles, there is this : https://github.com/openstack/nova/blob/master/nova/pci/stats.py#L156 | 18:04 |
| *** pwnall has quit IRC | 18:04 | |
| vladikr | beagles, maybe it doesn't work well if you are whitelisting these by devname, as oppose to using a alias in the extra_specs or a direct-physical binding? | 18:06 |
| beagles | vladikr, yeah, that's what I'm wondering | 18:07 |
| vladikr | beagles, maybe whitelisting devname doesn | 18:08 |
| vladikr | beagles, doesn't whitelist itself | 18:08 |
| vladikr | only anything that has a parent that matches the pci_address of that dev.. | 18:09 |
| *** nic has quit IRC | 18:10 | |
| vladikr | we should probably log the pool spec somewhere .. | 18:10 |
| beagles | vladikr, yup that's kind of where my head is at.. | 18:12 |
| electrocucaracha | johnthetubaguy: ping? | 18:12 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add REST API support for get me a network https://review.openstack.org/316398 | 18:12 |
| beagles | vladikr, PciAddress::match()'s 2nd parameter is that of the parent_address (apparently) which isn't going to match properly | 18:12 |
| *** shaohe_feng has quit IRC | 18:13 | |
| *** ihrachys has joined #openstack-nova | 18:13 | |
| *** shaohe_feng has joined #openstack-nova | 18:14 | |
| openstackgerrit | Guillaume Espanel proposed openstack/nova: Reduce time and memory usage of tenant usage. https://review.openstack.org/343734 | 18:15 |
| beagles | vladikr, I wonder if we are getting bad synergy with an older version of libvirt | 18:19 |
| beagles | vladikr, as in it might just all work with libvirtd >= 1.3.0 | 18:19 |
| *** tblakes has quit IRC | 18:19 | |
| beagles | vladikr, unfortunately, I can't look deeper at the moment :( | 18:20 |
| sdague | mriedem: the new version of this looks so much better | 18:20 |
| sdague | yay for pushing the validation back to the json schema | 18:20 |
| *** Apoorva_ has joined #openstack-nova | 18:20 | |
| *** dbite has quit IRC | 18:20 | |
| *** dbite has joined #openstack-nova | 18:21 | |
| vladikr | beagles, why do you think so? Is it because of the capability that will be exposed in 1.3? | 18:21 |
| beagles | vladikr, idk, just spitballing as to why it's coming up now | 18:22 |
| *** xyang1 has quit IRC | 18:22 | |
| vladikr | beagles, I tend to think that it's just how the devname resolution in devspec works .. | 18:22 |
| beagles | vladikr, :) yeah | 18:23 |
| *** shaohe_feng has quit IRC | 18:23 | |
| *** shaohe_feng has joined #openstack-nova | 18:23 | |
| *** tblakes has joined #openstack-nova | 18:23 | |
| *** Apoorva has quit IRC | 18:24 | |
| beagles | vladikr, going to try vendor/product/pci address whitelist | 18:24 |
| *** mingyu has joined #openstack-nova | 18:25 | |
| vladikr | beagles, ok, I'll try to see how to improve the devspec to include the device itself.. | 18:25 |
| beagles | vladikr, cool thanks | 18:25 |
| *** xyang1 has joined #openstack-nova | 18:26 | |
| *** dbite has quit IRC | 18:30 | |
| *** mingyu has quit IRC | 18:30 | |
| *** jichen has joined #openstack-nova | 18:30 | |
| *** ftersin_ has left #openstack-nova | 18:31 | |
| mriedem | sdague: yeah | 18:33 |
| *** shaohe_feng has quit IRC | 18:33 | |
| mriedem | oh and +W even | 18:33 |
| mriedem | nice | 18:33 |
| *** neophy has joined #openstack-nova | 18:33 | |
| *** shaohe_feng has joined #openstack-nova | 18:34 | |
| sdague | mriedem: you want to rereview - https://review.openstack.org/#/c/354981, I assume it will pass functional, but not while running on rax. And still waiting for the timeout increase to land. | 18:35 |
| *** efoley has quit IRC | 18:36 | |
| mriedem | yeah | 18:36 |
| *** ayoung has quit IRC | 18:37 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Remove deprecated manager classes in service.py https://review.openstack.org/338553 | 18:39 |
| *** tbachman has quit IRC | 18:40 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Remove deprecated manager classes in service.py https://review.openstack.org/338553 | 18:40 |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Remove deprecated manager classes in service.py https://review.openstack.org/338553 | 18:41 |
| jaypipes | electrocucaracha: pong | 18:41 |
| *** rcernin has quit IRC | 18:42 | |
| electrocucaracha | hey jaypipes, we're working on the implementation of OVO in neutron and I noticed that you submitted a bp in juno for adding a validation for StringFields | 18:42 |
| electrocucaracha | https://blueprints.launchpad.net/nova/+spec/string-field-max-length | 18:43 |
| electrocucaracha | jaypipes: ^ | 18:43 |
| *** ijw has quit IRC | 18:43 | |
| *** shaohe_feng has quit IRC | 18:44 | |
| *** ijw has joined #openstack-nova | 18:44 | |
| *** marst has joined #openstack-nova | 18:44 | |
| *** tbachman has joined #openstack-nova | 18:44 | |
| *** shaohe_feng has joined #openstack-nova | 18:44 | |
| jaypipes | electrocucaracha: is there a question in there somewhere? :) | 18:45 |
| electrocucaracha | jaypipes: sorry, I forgot to make the question, hahaa | 18:46 |
| *** Apoorva_ has quit IRC | 18:46 | |
| electrocucaracha | jaypipes: the question is, I noticed that oslo-versionedobjects doesn't have that validation implemented | 18:46 |
| electrocucaracha | https://github.com/openstack/oslo.versionedobjects/blob/master/oslo_versionedobjects/fields.py#L711-L712 | 18:46 |
| *** Apoorva has joined #openstack-nova | 18:47 | |
| electrocucaracha | jaypipes: I would like to know if there were something to block the implementation of that? | 18:47 |
| openstackgerrit | Danil Akhmetov proposed openstack/nova: Handle user-defined ephemeral disk size if provided https://review.openstack.org/352522 | 18:47 |
| electrocucaracha | jaypipes: or it was only implemented for nova | 18:47 |
| jaypipes | electrocucaracha: it wasn't implemented in Nova, no. I will obselete that blueprint. | 18:47 |
| *** pwnall has joined #openstack-nova | 18:48 | |
| electrocucaracha | jaypipes: I think that's still valuable to have it, maybe we can retake that effort | 18:48 |
| jaypipes | electrocucaracha: I wouldn't mind adding that functionality to ovo, however IIRC the pushback I received from dansmith is that we would not be able to control the versioning of the field if the max length changed over time... | 18:48 |
| *** ijw has quit IRC | 18:49 | |
| * electrocucaracha is thinking | 18:49 | |
| jaypipes | electrocucaracha: but it was a long time ago, so honestly, I don't really know at this point :) | 18:49 |
| dansmith | I don't know that I said that, fwiw.. if you add a new string type that embeds the length into the representation used for the versioning then it's probably fine | 18:50 |
| jaypipes | electrocucaracha: there are lots of fields that can change their schema over time (e.g. Enum or StateMachine-based fields) that we don't have a mechanism that I know of to control version bumps of the contained objects if the definitions of the fields change. | 18:50 |
| *** haplo37__ has quit IRC | 18:50 | |
| *** ijw has joined #openstack-nova | 18:50 | |
| dansmith | jaypipes: we just don't change them in incompatible ways, or we handle it in the backporting routines | 18:50 |
| jaypipes | dansmith: right, I just meant there's no code-based control or enforcement of that. | 18:51 |
| dansmith | jaypipes: electrocucaracha the length getting longer is the only real problem, and you'd just have to truncate or fail during backport somehow | 18:51 |
| jaypipes | dansmith: you mean the length getting shorter I presume? | 18:51 |
| dansmith | the version hash can _detect_ a violation, but otherwise, correct yeah | 18:51 |
| dansmith | jaypipes: well either I guess | 18:52 |
| *** efried has joined #openstack-nova | 18:52 | |
| dansmith | jaypipes: if you send something too long to an older node with a shorter limit, then that's a problem | 18:52 |
| jaypipes | dansmith: ah, right indeed. | 18:52 |
| mriedem | sdague: aren't the magic callbacks in https://review.openstack.org/#/c/354981/6/novaclient/tests/unit/v2/fakes.py@1912 awesome? | 18:52 |
| sdague | mriedem: yeh, but at least they are predictable | 18:53 |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Fix check_config_option_in_central_place https://review.openstack.org/355597 | 18:53 |
| mriedem | right, once you get it down in your head why things fail it's not so bad | 18:53 |
| electrocucaracha | jaypipes: dansmith I didn't think about that scenario, but it makes sense | 18:54 |
| *** shaohe_feng has quit IRC | 18:54 | |
| *** shaohe_feng has joined #openstack-nova | 18:55 | |
| *** su_zhang has quit IRC | 18:56 | |
| *** rcrit has joined #openstack-nova | 18:56 | |
| rcrit | mikal, ping re auth and the new vendor data REST API | 18:57 |
| mriedem | sdague: is anything testing the has_neutron=True case? | 18:57 |
| sdague | mriedem: yeh, that's why the new get_v2_0_networks call exists | 19:04 |
| *** haplo37__ has joined #openstack-nova | 19:04 | |
| *** shaohe_feng has quit IRC | 19:04 | |
| sdague | apparently we're default True on that in the tests | 19:04 |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Improve help text for glance options https://review.openstack.org/324007 | 19:05 |
| mriedem | sdague: oh because the fake get_endpoint() is returning "http://nova-api:8774/v2.1/190a755eef2e4aac9f06aa6be9786385" | 19:05 |
| mriedem | and the has_neutron sees that and assumes it's cool | 19:06 |
| sdague | ah, right | 19:06 |
| *** shaohe_feng has joined #openstack-nova | 19:06 | |
| *** fifieldt has quit IRC | 19:07 | |
| *** TravT has joined #openstack-nova | 19:09 | |
| *** Swami has joined #openstack-nova | 19:10 | |
| *** haplo37__ has quit IRC | 19:10 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Fix check_config_option_in_central_place https://review.openstack.org/355597 | 19:10 |
| *** kfarr has joined #openstack-nova | 19:10 | |
| *** ametts_ has quit IRC | 19:13 | |
| *** shaohe_feng has quit IRC | 19:14 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Remove redundant check from code https://review.openstack.org/350122 | 19:14 |
| *** shaohe_feng has joined #openstack-nova | 19:15 | |
| *** tblakes has quit IRC | 19:15 | |
| *** TravT_ has joined #openstack-nova | 19:15 | |
| mriedem | sdague: you want me to hack on https://review.openstack.org/#/c/354981/ while you're in infra looking at why the job is slow? | 19:16 |
| *** devkulkarni has joined #openstack-nova | 19:17 | |
| *** devkulkarni1 has quit IRC | 19:17 | |
| sdague | mriedem: I'm about to swing back of to it. It might make the most sense if you want to work on the testing piece. Unless you want to swap and give me the testing and you do this one. | 19:17 |
| *** ametts_ has joined #openstack-nova | 19:18 | |
| mriedem | i can poke on the testing one, or i'll work on rebasing dan's change on top of this | 19:18 |
| *** ametts_ has quit IRC | 19:18 | |
| *** fifieldt has joined #openstack-nova | 19:19 | |
| *** ametts has joined #openstack-nova | 19:19 | |
| *** TravT has quit IRC | 19:19 | |
| mriedem | Vek: WIP :) https://review.openstack.org/#/c/355215/ | 19:21 |
| *** haplo37__ has joined #openstack-nova | 19:22 | |
| mriedem | sdague: also, https://review.openstack.org/#/c/355148/ would be helpful in case the things i'm seeing locally are not the same we'd hit upstream | 19:22 |
| *** Hosam has joined #openstack-nova | 19:23 | |
| *** tblakes has joined #openstack-nova | 19:24 | |
| *** shaohe_feng has quit IRC | 19:25 | |
| sdague | mriedem: yeh, though I'd rather get a local working stack first before we do that instead of burning nodes when we know it's borked | 19:25 |
| *** Sukhdev has joined #openstack-nova | 19:25 | |
| mriedem | sure | 19:25 |
| mriedem | i love the smell of borked burning nodes in the morning | 19:25 |
| *** shaohe_feng has joined #openstack-nova | 19:25 | |
| *** su_zhang has joined #openstack-nova | 19:26 | |
| mriedem | alaski: dansmith: uh oh http://logs.openstack.org/98/316398/37/check/gate-tempest-dsvm-cells/61db698/logs/devstacklog.txt.gz#_2016-08-15_19_18_06_089 | 19:28 |
| mriedem | cells v1 job | 19:29 |
| dansmith | mriedem: we broke ironic too | 19:29 |
| dansmith | mriedem: revert is in the pipeline | 19:29 |
| dansmith | it's an ordering thing and I have an adjustment up to test as well | 19:29 |
| mriedem | ok, can you link me to the thing so i can e-r it? | 19:30 |
| dansmith | I would love to | 19:30 |
| alaski | dansmith: we may need to not run the migration for cellsv1, or ignore the error. there are no hosts in the top level cell so it will always return that | 19:30 |
| dansmith | but jroll's username on gerrit is UNSEARCHABLE | 19:30 |
| jroll | wat | 19:30 |
| jroll | jim@jimrollenhagen.com | 19:30 |
| dansmith | alaski: ah, good point | 19:30 |
| mriedem | https://review.openstack.org/#/c/355599/ | 19:31 |
| alaski | I forget that we said v1 users will need to do the complicated thing | 19:31 |
| *** bvanhav has quit IRC | 19:31 | |
| alaski | slightly more complicated | 19:31 |
| mriedem | we don't run cells v1 on devstack changes? | 19:31 |
| mriedem | must be in the experimental queue | 19:31 |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Remove redundant check from code https://review.openstack.org/350122 | 19:32 |
| *** su_zhang has quit IRC | 19:34 | |
| *** oanson has quit IRC | 19:34 | |
| *** shaohe_feng has quit IRC | 19:35 | |
| mriedem | you can't find jroll in LP either | 19:35 |
| jroll | O_o | 19:35 |
| jroll | this is the downside of nobody knowing my actual name, eh | 19:35 |
| *** shaohe_feng has joined #openstack-nova | 19:35 | |
| jroll | mriedem: LP is jim-rollenhagen if you actually need it | 19:36 |
| mriedem | jimmy mcjam von rollenhaggen V? | 19:36 |
| jroll | s/V/IV/ | 19:37 |
| jroll | close though! | 19:37 |
| mriedem | apollo dies and you fought the russians in IV | 19:37 |
| mriedem | classic | 19:37 |
| *** tonytan4ever has quit IRC | 19:41 | |
| *** ankur-gupta-f has joined #openstack-nova | 19:42 | |
| sdague | mriedem: so for - https://review.openstack.org/#/c/354981/ your concern is there are no unit tests? | 19:44 |
| dansmith | alaski: wow, you really "touched" a lot of people with that five line change | 19:44 |
| sdague | for the find_network bits | 19:44 |
| *** shaohe_feng has quit IRC | 19:45 | |
| *** shaohe_feng has joined #openstack-nova | 19:46 | |
| mriedem | sdague: yeah it looks like we're not hitting any of that code somehow | 19:47 |
| mriedem | sdague: i'm not sure why though since i thought https://review.openstack.org/#/c/354981/6/novaclient/tests/unit/v2/fakes.py would handle it | 19:47 |
| mriedem | but maybe that was before you added the mock to just have client.has_neutron return False? | 19:47 |
| *** ijw has quit IRC | 19:48 | |
| mriedem | unless it's functional tests, i didn't run those | 19:48 |
| sdague | it will get run when the command is run for real | 19:48 |
| sdague | I don't think we have tests on this part of the shell at all | 19:48 |
| *** electrocucaracha has left #openstack-nova | 19:48 | |
| *** ayoung has joined #openstack-nova | 19:48 | |
| mriedem | we do for the ones that you added the mock to return False right? | 19:49 |
| *** yamahata has quit IRC | 19:49 | |
| mriedem | in https://review.openstack.org/#/c/354981/6/novaclient/tests/unit/v2/test_shell.py | 19:49 |
| *** yamahata has joined #openstack-nova | 19:49 | |
| alaski | dansmith: do I get a prize, or just get to be the target of ire? | 19:49 |
| dansmith | alaski: sadly the ire is the prize | 19:50 |
| openstackgerrit | Maciej Szankin proposed openstack/nova: config options: Improve help for SPICE https://review.openstack.org/317046 | 19:50 |
| sdague | mriedem: ah, right | 19:50 |
| jroll | pft, I'll totally buy alaski a beer, I like people that find fragile things | 19:50 |
| sdague | mriedem: ok, yeh, I'll pop some more into there | 19:51 |
| mriedem | cool | 19:51 |
| sdague | that's actually a bit easier than the thing I was going to do | 19:52 |
| sdague | fwiw, it works for reals, just not in the tests | 19:52 |
| *** prateek has joined #openstack-nova | 19:53 | |
| *** fragatina has quit IRC | 19:53 | |
| *** shaohe_feng has quit IRC | 19:55 | |
| *** shaohe_feng has joined #openstack-nova | 19:56 | |
| *** tonytan4ever has joined #openstack-nova | 19:58 | |
| *** oomichi_ has joined #openstack-nova | 20:02 | |
| *** oomichi has quit IRC | 20:03 | |
| *** oomichi_ is now known as oomichi | 20:03 | |
| *** shaohe_feng has quit IRC | 20:06 | |
| *** Apoorva has quit IRC | 20:06 | |
| mriedem | weird, do we run novnc in devstack when using neutron but not when using nova-net? | 20:06 |
| *** shaohe_feng has joined #openstack-nova | 20:06 | |
| mriedem | because novaclient.tests.functional.v2.legacy.test_consoles.TestConsolesNovaClient.test_vnc_console_get fails with neutron devstack, and you get a vnc console url back, which the test doesn't expect | 20:06 |
| *** liverpooler has quit IRC | 20:07 | |
| *** TravT_ is now known as TravT | 20:08 | |
| *** su_zhang has joined #openstack-nova | 20:08 | |
| *** ankur-gupta-f has left #openstack-nova | 20:12 | |
| mriedem | what i need is like a, oh idk, capabilities API or something | 20:13 |
| alaski | or a traits API | 20:13 |
| edleafe | mriedem: such silly thoughts | 20:13 |
| *** neophy has quit IRC | 20:14 | |
| mriedem | doffm: why do we check for CommandFailed twice here? https://review.openstack.org/#/c/230149/5/novaclient/tests/functional/v2/legacy/test_consoles.py@37 | 20:15 |
| mriedem | i mean, it's ran twice, expected to fail both times, is the 2nd call just to parse the error string? | 20:15 |
| mriedem | b/c we could that in a single call | 20:15 |
| *** whenry has joined #openstack-nova | 20:15 | |
| *** shaohe_feng has quit IRC | 20:16 | |
| *** shaohe_feng has joined #openstack-nova | 20:16 | |
| mriedem | huh, ok, | 20:17 |
| mriedem | so ENABLED_SERVICES in devstack defaults to include n-novnc | 20:17 |
| mriedem | but in the gate we don't set that | 20:17 |
| *** Apoorva has joined #openstack-nova | 20:17 | |
| mriedem | so these functional tests rely on how devstack-gate sets up the env it seems | 20:17 |
| mriedem | sdague: ^ any suggestion on how to handle that? tempest would make it configurable. | 20:18 |
| *** cargonza has joined #openstack-nova | 20:20 | |
| sdague | mriedem: ok, so that's a bad test if it's expecting that to not work | 20:22 |
| *** jichen has quit IRC | 20:22 | |
| sdague | I think we want to be careful about making this a thing that you have to do lots of manual config on | 20:23 |
| mriedem | yeah, me too | 20:23 |
| mriedem | but i don't think supported consoles are a thing i can discover | 20:23 |
| mriedem | for a given instance, until we have a capabilities api | 20:23 |
| *** tri2sing has quit IRC | 20:23 | |
| mriedem | i mean, i can skip this locally for now and realize it's just going to fail | 20:23 |
| mriedem | or i can kill the novnc service on my devstack to 'workaround' it | 20:24 |
| sdague | what's the actual failure? | 20:24 |
| *** pwnall has quit IRC | 20:24 | |
| sdague | because if this is failing if consoles actually exist, the test is just wrong | 20:24 |
| *** ijw has joined #openstack-nova | 20:25 | |
| mriedem | testtools.matchers._impl.MismatchError: <bound method TestConsolesNovaClient.nova of <novaclient.tests.functional.v2.legacy.test_consoles.TestConsolesNovaClient.test_vnc_console_get id=0x7f50b588c790>> returned '+-------+----------------------------------------------------------------------------------+\n| Type | Url |\n+-------+------------------------------- | 20:25 |
| mriedem | i get a response | 20:25 |
| mriedem | the test expects it to fail | 20:25 |
| sdague | ok, so that's just a wrong test | 20:25 |
| sdague | honestly, probably should just be deleted | 20:25 |
| mriedem | well, | 20:26 |
| sdague | or if you want to fix it, it should handle the success criteria | 20:26 |
| mriedem | we do have it to make sure the client picks the correct API b/c there is a 2.5 boundary here | 20:26 |
| *** shaohe_feng has quit IRC | 20:26 | |
| mriedem | 'this can pass or fail, let's just always consider it successful' :) | 20:26 |
| sdague | not in legacy | 20:26 |
| mriedem | sure | 20:26 |
| sdague | well there are specific outputs to expect | 20:26 |
| *** shaohe_feng has joined #openstack-nova | 20:26 | |
| mriedem | i'll work it as a separate bug | 20:27 |
| sdague | there is a range of legit behavior | 20:27 |
| sdague | it either gets good output like X, or it fails with a specific failure | 20:27 |
| *** tonytan4ever has quit IRC | 20:27 | |
| mriedem | yup | 20:27 |
| sdague | and, while that's not as good as there being only 1 correct response, it is success criteria | 20:28 |
| sdague | and, blame doffm, he wrote it :) | 20:28 |
| *** xyang1 has quit IRC | 20:29 | |
| *** marst has quit IRC | 20:29 | |
| *** xyang1 has joined #openstack-nova | 20:29 | |
| beagles | vladikr, for fun: https://bugs.launchpad.net/nova/+bug/1613434 | 20:30 |
| openstack | Launchpad bug 1613434 in OpenStack Compute (nova) "Whitelisted PFs aren't being recognized " [Undecided,New] | 20:30 |
| *** devkulkarni has quit IRC | 20:31 | |
| mriedem | https://bugs.launchpad.net/python-novaclient/+bug/1613435 | 20:31 |
| openstack | Launchpad bug 1613435 in python-novaclient "TestConsolesNovaClient.test_vnc_console_get fails if you're using default devstack" [Undecided,New] | 20:31 |
| *** ociuhandu has quit IRC | 20:31 | |
| mriedem | sdague: heh, look who +Wed that bad boy https://review.openstack.org/#/c/230149/ | 20:32 |
| *** sdake has quit IRC | 20:32 | |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: WIP: Skip nova-network-only tests if using Neutron https://review.openstack.org/355215 | 20:33 |
| sdague | gah, I'm so annoyed that testtools assertRaises is not a context manager like in unittest | 20:35 |
| mriedem | testtools has some expected exception context manager | 20:36 |
| sdague | mriedem: link? | 20:36 |
| *** shaohe_feng has quit IRC | 20:36 | |
| mriedem | sdague: https://github.com/openstack/tempest/blob/master/tempest/api/compute/servers/test_virtual_interfaces.py#L55 | 20:37 |
| *** shaohe_feng has joined #openstack-nova | 20:37 | |
| *** fragatina has joined #openstack-nova | 20:41 | |
| *** fragatina has quit IRC | 20:42 | |
| *** fragatina has joined #openstack-nova | 20:43 | |
| sdague | yeh, got it now. It's still annoying that this diverges from stdlib | 20:44 |
| sdague | ok, I've got new tests | 20:44 |
| *** tonytan4ever has joined #openstack-nova | 20:45 | |
| *** shaohe_feng has quit IRC | 20:47 | |
| *** shaohe_feng has joined #openstack-nova | 20:47 | |
| *** gouthamr has quit IRC | 20:51 | |
| *** marst has joined #openstack-nova | 20:51 | |
| mikal | Morning | 20:51 |
| mikal | rcrit: you still around? | 20:52 |
| rcrit | mikal, yup | 20:52 |
| mikal | rcrit: what's up? | 20:53 |
| rcrit | so question about how auth is expected to work with the vendor_data2.json | 20:53 |
| rcrit | there is a bit of auth code currently in the driver, and it works, but only if the auth filter is enabled | 20:54 |
| rcrit | if the request comes in from a guest, ala curl http://169.254.169.254/openstack/2016-10-06/vendor_data2.json | 20:54 |
| rcrit | there is no auth | 20:54 |
| rcrit | so I'm not sure how paste-wise is the best way to configure a REST service | 20:54 |
| mikal | Huh, yeah. I can see that -- the client doesn't pass in a keystone auth token, so we don't either | 20:54 |
| rcrit | and whether that service should get its own service user with which to obtain tokens to talk to other services, like glance | 20:54 |
| rcrit | not sure what best practice is here | 20:55 |
| mikal | I'm not either. Can we start an openstack-dev thread to talk that through? I think that might work better than asking people here. | 20:55 |
| rcrit | sure, I can start that. What tag(s) would you recommend, [nova] good enough or something else? | 20:56 |
| mikal | Perhaps [nova][keystone] ? | 20:56 |
| rcrit | works for me, thanks. | 20:56 |
| mikal | Thanks for all your testing by the way, I really appreciate it | 20:56 |
| *** shaohe_feng has quit IRC | 20:57 | |
| openstackgerrit | Sean Dague proposed openstack/python-novaclient: Use neutron for network name -> id resolution https://review.openstack.org/354981 | 20:57 |
| rcrit | we'll call it a team effort :-) | 20:57 |
| *** shaohe_feng has joined #openstack-nova | 20:57 | |
| *** sdake has joined #openstack-nova | 20:57 | |
| sdague | mriedem: ok... I think that addresses your concerns | 20:57 |
| jroll | doesn't metadata service auth by IP address? | 20:57 |
| jroll | mikal: ^ just out of curiousity | 20:57 |
| mikal | jroll: this is the auth to a new external service from the metadata service though | 20:58 |
| mikal | I naively thought I could just pass through the client keystone token, but that's not true | 20:58 |
| jroll | ahh | 20:59 |
| *** JoseMello has quit IRC | 21:00 | |
| *** lpetrut has joined #openstack-nova | 21:03 | |
| *** haplo37__ has quit IRC | 21:03 | |
| *** smatzek has quit IRC | 21:05 | |
| *** shaohe_feng has quit IRC | 21:07 | |
| cfriesen_ | anyone on here that knows something about PCIDevice class? | 21:07 |
| *** ihrachys has quit IRC | 21:08 | |
| cfriesen_ | or PciDevice rather | 21:08 |
| *** nic has joined #openstack-nova | 21:08 | |
| *** sdake has quit IRC | 21:08 | |
| mriedem | sdague: cool got the console func tests passing locally now | 21:09 |
| *** shaohe_feng has joined #openstack-nova | 21:10 | |
| *** lpetrut has quit IRC | 21:12 | |
| *** nic has quit IRC | 21:13 | |
| openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/nova: Remove context object in oslo.log method https://review.openstack.org/317109 | 21:15 |
| cfriesen_ | when serializing an object for RPC, how is 'nova_object.changes' supposed to handle a field which has been removed from 'nova_object.data'? Should something be automatically removing it from 'nova_object.changes'? | 21:15 |
| *** sdake has joined #openstack-nova | 21:17 | |
| *** shaohe_feng has quit IRC | 21:17 | |
| scottda | mriedem: For new Cinder volume-attach API, we no longer need nor care about instance mountpoint, do we? | 21:17 |
| *** shaohe_feng has joined #openstack-nova | 21:18 | |
| *** haplo37__ has joined #openstack-nova | 21:18 | |
| mriedem | scottda: correct, since like liberty when using libvirt | 21:18 |
| mriedem | it picks one for you | 21:18 |
| mriedem | scottda: mountpoint was always optional in the rest api actually | 21:18 |
| scottda | mriedem: That's what I thought. We might as well get rid of all that in Cinder while we have the chance... | 21:18 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Handle successful response in console functional tests https://review.openstack.org/355645 | 21:18 |
| mriedem | sdague: ^ | 21:18 |
| *** mvk has joined #openstack-nova | 21:19 | |
| *** pwnall has joined #openstack-nova | 21:20 | |
| *** lpetrut has joined #openstack-nova | 21:23 | |
| *** rstarmer_ has joined #openstack-nova | 21:24 | |
| rstarmer_ | hello, I'm trying to launch VMs on a shiny new Kolla installed environment, and I'm getting a new scheduler error: Require both a host and instance NUMA topology to fit instance on host. numa_fit_instance_to_host /usr/lib/python2.7/site-packages/nova/virt/hardware.py:1213 | 21:25 |
| rstarmer_ | Anyone seen this before? | 21:25 |
| *** pwnall has quit IRC | 21:26 | |
| *** pwnall has joined #openstack-nova | 21:27 | |
| *** shaohe_feng has quit IRC | 21:28 | |
| *** shaohe_feng has joined #openstack-nova | 21:28 | |
| *** ametts has quit IRC | 21:28 | |
| *** prateek has quit IRC | 21:29 | |
| *** ravikama_ has quit IRC | 21:35 | |
| openstackgerrit | Sivasathurappan Radhakrishnan proposed openstack/nova: Check for volume size https://review.openstack.org/348317 | 21:37 |
| *** shaohe_feng has quit IRC | 21:38 | |
| *** yamahata has quit IRC | 21:38 | |
| *** shaohe_feng has joined #openstack-nova | 21:39 | |
| *** sdake has quit IRC | 21:39 | |
| *** tbachman_ has joined #openstack-nova | 21:47 | |
| *** amotoki has quit IRC | 21:48 | |
| *** cfriesen_ has quit IRC | 21:48 | |
| *** shaohe_feng has quit IRC | 21:48 | |
| *** burgerk has quit IRC | 21:48 | |
| *** tbachman has quit IRC | 21:48 | |
| *** tbachman_ is now known as tbachman | 21:48 | |
| *** cfriesen_ has joined #openstack-nova | 21:49 | |
| *** shaohe_feng has joined #openstack-nova | 21:49 | |
| *** cfriesen_ is now known as cfriesen | 21:49 | |
| mriedem | sdague: another fun one https://bugs.launchpad.net/python-novaclient/+bug/1613454 - 60s is too fast, tempest defaults the server build timeout to 300s | 21:51 |
| openstack | Launchpad bug 1613454 in python-novaclient "novaclient.tests.functional.v2.test_trigger_crash_dump fails in default local devstack setup" [Undecided,New] | 21:51 |
| mriedem | i'm debugging it locally and seeing the guest os is booting, just not done in 60s | 21:51 |
| mriedem | i am testing with a smaller VM, 4 cpu and 8gb ram | 21:52 |
| *** devkulkarni has joined #openstack-nova | 21:52 | |
| mriedem | still, i shouldn't need an 8VCPU VM to run novaclient functional tests | 21:53 |
| *** sdake has joined #openstack-nova | 21:54 | |
| *** harlowja has quit IRC | 21:55 | |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Make wait_for_server_os_boot wait longer https://review.openstack.org/355658 | 21:56 |
| *** tonytan4ever has quit IRC | 21:57 | |
| *** burt has quit IRC | 21:58 | |
| *** shaohe_feng has quit IRC | 21:58 | |
| *** shaohe_feng has joined #openstack-nova | 21:59 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Make simple_cell_setup work when multiple nodes are present https://review.openstack.org/355659 | 21:59 |
| *** amotoki has joined #openstack-nova | 22:00 | |
| *** matrohon has quit IRC | 22:00 | |
| *** amotoki has quit IRC | 22:00 | |
| mriedem | and with that, i've got a clean local novaclient functional run | 22:01 |
| mriedem | and nova-net only tests are properly skipped | 22:01 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Skip nova-network-only tests if using Neutron https://review.openstack.org/355215 | 22:01 |
| *** yamahata has joined #openstack-nova | 22:01 | |
| *** mtanino has quit IRC | 22:03 | |
| cfriesen | for a nova object is it ever valid to have self.obj_what_changed() return a field that doesn't exist in that version of the object? | 22:04 |
| mriedem | we probably should have had 355599 bumped to the head of the gate queu | 22:04 |
| mriedem | since everything is backed up on | 22:04 |
| *** thorst_ has quit IRC | 22:08 | |
| *** shaohe_feng has quit IRC | 22:09 | |
| *** mdrabe has quit IRC | 22:09 | |
| *** mriedem has quit IRC | 22:09 | |
| *** shaohe_feng has joined #openstack-nova | 22:09 | |
| *** fragatin_ has joined #openstack-nova | 22:10 | |
| *** jamesden_ has quit IRC | 22:10 | |
| *** kfarr has quit IRC | 22:10 | |
| *** beagles is now known as beagles_brb | 22:11 | |
| *** lpetrut has quit IRC | 22:11 | |
| *** vladikr has quit IRC | 22:13 | |
| *** thorst_ has joined #openstack-nova | 22:13 | |
| *** fragatina has quit IRC | 22:13 | |
| *** fragatin_ has quit IRC | 22:14 | |
| *** fragatina has joined #openstack-nova | 22:15 | |
| *** haplo37__ has quit IRC | 22:15 | |
| *** bswartz has quit IRC | 22:15 | |
| *** thorst_ has quit IRC | 22:18 | |
| *** jistr has quit IRC | 22:18 | |
| *** ijw has quit IRC | 22:19 | |
| *** shaohe_feng has quit IRC | 22:19 | |
| *** jistr has joined #openstack-nova | 22:19 | |
| *** shaohe_feng has joined #openstack-nova | 22:20 | |
| *** sdake has quit IRC | 22:20 | |
| *** ijw has joined #openstack-nova | 22:22 | |
| *** shaohe_feng has quit IRC | 22:29 | |
| *** Hosam has quit IRC | 22:29 | |
| *** shaohe_feng has joined #openstack-nova | 22:30 | |
| *** tblakes has quit IRC | 22:32 | |
| *** kevinz has joined #openstack-nova | 22:33 | |
| *** ijw has quit IRC | 22:34 | |
| *** sdake has joined #openstack-nova | 22:35 | |
| *** kevinz has quit IRC | 22:37 | |
| *** krtaylor has quit IRC | 22:38 | |
| *** shaohe_feng has quit IRC | 22:39 | |
| *** ijw has joined #openstack-nova | 22:39 | |
| *** ducttape_ has quit IRC | 22:40 | |
| *** beagles_brb is now known as beagles | 22:42 | |
| *** shaohe_feng has joined #openstack-nova | 22:42 | |
| *** ravikama_ has joined #openstack-nova | 22:43 | |
| *** whenry has quit IRC | 22:45 | |
| *** efried has quit IRC | 22:49 | |
| *** shaohe_feng has quit IRC | 22:50 | |
| *** shaohe_feng has joined #openstack-nova | 22:50 | |
| *** ravikama_ has quit IRC | 22:50 | |
| *** edmondsw has quit IRC | 22:51 | |
| *** sdake has quit IRC | 22:57 | |
| *** mriedem has joined #openstack-nova | 22:57 | |
| *** tonytan4ever has joined #openstack-nova | 22:58 | |
| *** shaohe_feng has quit IRC | 23:00 | |
| *** nic has joined #openstack-nova | 23:00 | |
| *** shaohe_feng has joined #openstack-nova | 23:01 | |
| *** tonytan4ever has quit IRC | 23:03 | |
| *** devkulkarni has quit IRC | 23:04 | |
| *** markvoelker has quit IRC | 23:06 | |
| *** xyang1 has quit IRC | 23:08 | |
| oomichi | mriedem: https://review.openstack.org/#/c/351960/ is easy for you again (I catch you now;) | 23:10 |
| *** shaohe_feng has quit IRC | 23:10 | |
| *** shaohe_feng has joined #openstack-nova | 23:11 | |
| *** BjoernT has quit IRC | 23:13 | |
| *** baoli has joined #openstack-nova | 23:16 | |
| *** TravT has quit IRC | 23:18 | |
| *** shaohe_feng has quit IRC | 23:20 | |
| *** shaohe_feng has joined #openstack-nova | 23:21 | |
| *** devkulkarni has joined #openstack-nova | 23:29 | |
| *** shaohe_feng has quit IRC | 23:31 | |
| *** diogogmt has quit IRC | 23:31 | |
| *** shaohe_feng has joined #openstack-nova | 23:31 | |
| *** gjayavelu has joined #openstack-nova | 23:32 | |
| *** bwallis has quit IRC | 23:33 | |
| *** harlowja has joined #openstack-nova | 23:36 | |
| *** devkulkarni has quit IRC | 23:40 | |
| *** rama_y has quit IRC | 23:40 | |
| *** shaohe_feng has quit IRC | 23:41 | |
| *** rama_y has joined #openstack-nova | 23:41 | |
| *** shaohe_feng has joined #openstack-nova | 23:43 | |
| *** sdague has quit IRC | 23:44 | |
| *** rama_y has quit IRC | 23:46 | |
| *** shaohe_feng has quit IRC | 23:51 | |
| *** jaypipes has quit IRC | 23:51 | |
| *** shaohe_feng has joined #openstack-nova | 23:52 | |
| *** pwnall has quit IRC | 23:53 | |
| *** rrecio_ has quit IRC | 23:54 | |
| *** baoli has quit IRC | 23:54 | |
| *** bwallis has joined #openstack-nova | 23:55 | |
| *** takashin has joined #openstack-nova | 23:57 | |
| *** gjayavelu has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!