*** lucasagomes has quit IRC | 00:00 | |
*** hoangcx2 has quit IRC | 00:01 | |
*** lucasagomes has joined #openstack-nova | 00:02 | |
*** ssurana has quit IRC | 00:04 | |
*** tblakes has quit IRC | 00:04 | |
*** thorst_ has joined #openstack-nova | 00:04 | |
melwitt | mriedem: a-ha you're here! I thought you should look at this fix for nova-manage db archive_deleted_rows. I tagged it as newton-rc-potential, let me know if it shouldn't be https://review.openstack.org/#/c/323684/ | 00:05 |
---|---|---|
*** Swami has quit IRC | 00:06 | |
*** ducttape_ has joined #openstack-nova | 00:07 | |
*** lucasagomes has quit IRC | 00:07 | |
*** tblakes has joined #openstack-nova | 00:08 | |
*** penick has quit IRC | 00:08 | |
*** zul has quit IRC | 00:08 | |
mriedem | i don't think that's a regression | 00:09 |
*** lucasagomes has joined #openstack-nova | 00:10 | |
*** thorst_ has quit IRC | 00:11 | |
melwitt | mriedem: oh. I thought it had been operational until this | 00:13 |
mriedem | melwitt: similar https://review.openstack.org/#/c/345191/4 | 00:13 |
melwitt | oh, okay. my bad | 00:13 |
*** lucasagomes has quit IRC | 00:14 | |
*** ociuhandu has quit IRC | 00:15 | |
*** ducttape_ has quit IRC | 00:16 | |
*** lucasagomes has joined #openstack-nova | 00:17 | |
*** MikeG451 has quit IRC | 00:17 | |
*** stewie925 has quit IRC | 00:17 | |
*** cmartin has quit IRC | 00:17 | |
*** xenogear has quit IRC | 00:17 | |
*** sdake_ has quit IRC | 00:17 | |
*** sdake has joined #openstack-nova | 00:17 | |
*** jinli has quit IRC | 00:18 | |
openstackgerrit | Takashi NATSUME proposed openstack/nova: Add a hacking rule for string interpolation at logging https://review.openstack.org/334844 | 00:18 |
*** sai has quit IRC | 00:19 | |
*** sai has joined #openstack-nova | 00:19 | |
*** syjulian has joined #openstack-nova | 00:22 | |
*** gongysh has quit IRC | 00:26 | |
*** sai has quit IRC | 00:27 | |
*** lucasagomes has quit IRC | 00:27 | |
*** sai has joined #openstack-nova | 00:30 | |
*** krtaylor has joined #openstack-nova | 00:30 | |
*** lucasagomes has joined #openstack-nova | 00:30 | |
mriedem | melwitt: do we delete migrations when we delete instances? | 00:30 |
mriedem | or is the problem that we don't have a foreign key between instances and migrations? | 00:30 |
melwitt | mriedem: I doubt it | 00:30 |
melwitt | mriedem: I think there is a FK which causes the problem. it's that migrations are never deleted but instances are. similar to instance_actions was my assumption | 00:31 |
mriedem | yeah, so https://github.com/openstack/nova/blob/8211d3fade1b44173a8d6d54d1918b9f0bc26d4e/nova/db/sqlalchemy/api.py#L1801 | 00:32 |
mriedem | we should probably also delete migrations when the instance is deleted | 00:32 |
*** MikeG451 has joined #openstack-nova | 00:32 | |
mriedem | at some point we should just make https://github.com/openstack/nova/blob/8211d3fade1b44173a8d6d54d1918b9f0bc26d4e/nova/db/sqlalchemy/api.py#L1801 smarter and look for referenced entries in other tables and delete those rather than hard-coding it | 00:32 |
*** kaisers1 has joined #openstack-nova | 00:32 | |
*** kaisers2 has joined #openstack-nova | 00:33 | |
*** kaisers has quit IRC | 00:34 | |
*** lucasagomes has quit IRC | 00:34 | |
*** 32NABYQA1 has quit IRC | 00:35 | |
melwitt | mriedem: makes sense | 00:35 |
*** lixiaoy1 has joined #openstack-nova | 00:37 | |
*** lucasagomes has joined #openstack-nova | 00:38 | |
*** itlinux has quit IRC | 00:40 | |
*** itlinux has joined #openstack-nova | 00:41 | |
*** gongysh has joined #openstack-nova | 00:41 | |
pumaranikar | I am checking os-vif logging in unit tests. https://bugs.launchpad.net/nova/+bug/1615676. I don't find logging config in os-vif repo. any pointers? | 00:42 |
openstack | Launchpad bug 1615676 in OpenStack Compute (nova) "os-vif log messages corrupt and pollute test output" [Wishlist,New] - Assigned to Pushkar Umaranikar (pushkar-umaranikar) | 00:42 |
*** vladikr has quit IRC | 00:42 | |
*** browne has quit IRC | 00:43 | |
*** dave-mccowan has joined #openstack-nova | 00:43 | |
*** smatzek has joined #openstack-nova | 00:44 | |
*** rrecio_ has quit IRC | 00:44 | |
*** raunak has quit IRC | 00:45 | |
*** jinli has joined #openstack-nova | 00:45 | |
*** lucasagomes has quit IRC | 00:46 | |
*** sai has quit IRC | 00:47 | |
*** lucasagomes has joined #openstack-nova | 00:48 | |
*** tuanla_ has joined #openstack-nova | 00:49 | |
openstackgerrit | Hieu LE proposed openstack/nova: Config options: improve libvirt help text (2) https://review.openstack.org/316550 | 00:52 |
*** ijw has joined #openstack-nova | 00:54 | |
*** tjones has left #openstack-nova | 00:55 | |
*** stewie925 has joined #openstack-nova | 00:56 | |
*** itlinux has quit IRC | 00:56 | |
*** lucasagomes has quit IRC | 00:57 | |
*** jfding has quit IRC | 00:58 | |
*** cheneydc has joined #openstack-nova | 00:59 | |
*** devananda is now known as devananda|dinner | 00:59 | |
*** ijw has quit IRC | 00:59 | |
*** lucasagomes has joined #openstack-nova | 01:00 | |
*** itlinux has joined #openstack-nova | 01:00 | |
*** gyee has quit IRC | 01:03 | |
*** tbachman has quit IRC | 01:03 | |
*** lucasagomes has quit IRC | 01:04 | |
*** smatzek has quit IRC | 01:07 | |
*** lucasagomes has joined #openstack-nova | 01:07 | |
*** sai has joined #openstack-nova | 01:08 | |
*** thorst_ has joined #openstack-nova | 01:09 | |
*** ducttape_ has joined #openstack-nova | 01:09 | |
*** cmartin has joined #openstack-nova | 01:09 | |
*** xenogear has joined #openstack-nova | 01:10 | |
*** Sukhdev has quit IRC | 01:10 | |
*** thorst_ has quit IRC | 01:12 | |
*** lucasagomes has quit IRC | 01:12 | |
*** thorst_ has joined #openstack-nova | 01:14 | |
*** lucasagomes has joined #openstack-nova | 01:14 | |
*** yuntongjin has joined #openstack-nova | 01:15 | |
*** gouthamr has quit IRC | 01:17 | |
*** lucasagomes has quit IRC | 01:19 | |
*** baoli has joined #openstack-nova | 01:20 | |
*** lucasagomes has joined #openstack-nova | 01:21 | |
*** yassou has quit IRC | 01:22 | |
*** hoangcx has joined #openstack-nova | 01:26 | |
*** jamielennox is now known as jamielennox|away | 01:26 | |
*** jdurgin has joined #openstack-nova | 01:27 | |
*** cheneydc has quit IRC | 01:27 | |
*** moshele has joined #openstack-nova | 01:29 | |
*** lucasagomes has quit IRC | 01:30 | |
*** sai has quit IRC | 01:32 | |
*** mingyu has joined #openstack-nova | 01:32 | |
openstackgerrit | huan proposed openstack/nova: xenapi: support the hotplug of a neutron port https://review.openstack.org/365505 | 01:33 |
*** lucasagomes has joined #openstack-nova | 01:33 | |
*** hoangcx has quit IRC | 01:33 | |
*** tri2sing has quit IRC | 01:34 | |
*** sai has joined #openstack-nova | 01:34 | |
*** mingyu has quit IRC | 01:36 | |
*** artom has quit IRC | 01:38 | |
*** itlinux has quit IRC | 01:39 | |
*** artom has joined #openstack-nova | 01:39 | |
*** salv-orlando has joined #openstack-nova | 01:41 | |
*** tblakes has quit IRC | 01:41 | |
*** sai has quit IRC | 01:42 | |
*** nic has quit IRC | 01:42 | |
*** salv-orl_ has quit IRC | 01:43 | |
*** sai has joined #openstack-nova | 01:44 | |
*** penick has joined #openstack-nova | 01:44 | |
*** sdake_ has joined #openstack-nova | 01:45 | |
*** sdake_ has quit IRC | 01:45 | |
*** sdake_ has joined #openstack-nova | 01:45 | |
*** itlinux has joined #openstack-nova | 01:46 | |
*** sdake has quit IRC | 01:47 | |
*** mriedem has quit IRC | 01:49 | |
*** hoangcx has joined #openstack-nova | 01:50 | |
*** Apoorva_ has joined #openstack-nova | 01:51 | |
*** ducttape_ has quit IRC | 01:52 | |
*** thorst_ has quit IRC | 01:52 | |
*** yuntongjin has quit IRC | 01:53 | |
*** thorst_ has joined #openstack-nova | 01:53 | |
*** jaypipes has joined #openstack-nova | 01:54 | |
*** annegentle has joined #openstack-nova | 01:54 | |
*** yamahata has quit IRC | 01:55 | |
*** Apoorva has quit IRC | 01:55 | |
*** ijw has joined #openstack-nova | 01:56 | |
*** Apoorva_ has quit IRC | 01:56 | |
*** itlinux has quit IRC | 01:56 | |
*** hoangcx has quit IRC | 01:57 | |
*** sai has quit IRC | 01:59 | |
*** sai has joined #openstack-nova | 02:00 | |
*** thorst_ has quit IRC | 02:02 | |
*** unicell has quit IRC | 02:02 | |
*** ijw has quit IRC | 02:02 | |
*** ducttape_ has joined #openstack-nova | 02:04 | |
*** hoangcx has joined #openstack-nova | 02:04 | |
*** ducttape_ has quit IRC | 02:04 | |
*** lucasagomes has quit IRC | 02:06 | |
*** lucasagomes has joined #openstack-nova | 02:07 | |
*** yuntongjin has joined #openstack-nova | 02:10 | |
*** sai has quit IRC | 02:13 | |
*** gjayavelu has quit IRC | 02:14 | |
*** gongysh has quit IRC | 02:14 | |
*** lucasagomes has quit IRC | 02:15 | |
*** sai has joined #openstack-nova | 02:15 | |
*** lucasagomes has joined #openstack-nova | 02:17 | |
*** yuntongjin has quit IRC | 02:17 | |
*** hoangcx has quit IRC | 02:18 | |
*** ijw has joined #openstack-nova | 02:19 | |
*** woodster_ has quit IRC | 02:19 | |
*** thorst_ has joined #openstack-nova | 02:20 | |
*** thorst_ has quit IRC | 02:21 | |
*** lucasagomes has quit IRC | 02:22 | |
*** jdurgin has quit IRC | 02:22 | |
*** ayoung has joined #openstack-nova | 02:23 | |
*** lucasagomes has joined #openstack-nova | 02:23 | |
*** ijw has quit IRC | 02:23 | |
*** zul has joined #openstack-nova | 02:27 | |
*** lucasagomes has quit IRC | 02:28 | |
*** zul has quit IRC | 02:28 | |
*** zul has joined #openstack-nova | 02:28 | |
*** lucasagomes has joined #openstack-nova | 02:29 | |
*** lucasagomes has quit IRC | 02:34 | |
*** thorst_ has joined #openstack-nova | 02:36 | |
*** lucasagomes has joined #openstack-nova | 02:37 | |
*** dave-mccowan has quit IRC | 02:38 | |
*** lucasagomes has quit IRC | 02:41 | |
alex_xu | jaypipes: thanks for review https://review.openstack.org/#/c/369147/ | 02:41 |
*** moshele has quit IRC | 02:42 | |
*** ociuhandu has joined #openstack-nova | 02:42 | |
*** lucasagomes has joined #openstack-nova | 02:43 | |
*** sai has quit IRC | 02:45 | |
*** sai has joined #openstack-nova | 02:47 | |
*** akshai has joined #openstack-nova | 02:48 | |
*** lucasagomes has quit IRC | 02:48 | |
*** akshai_ has joined #openstack-nova | 02:49 | |
*** stevemar_ is now known as stevemar | 02:49 | |
*** lucasagomes has joined #openstack-nova | 02:49 | |
*** akshai has quit IRC | 02:53 | |
*** mdrabe has quit IRC | 02:53 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_security_group https://review.openstack.org/307135 | 02:53 |
*** sai has quit IRC | 02:54 | |
*** ssurana has joined #openstack-nova | 02:55 | |
*** sai has joined #openstack-nova | 02:55 | |
*** lucasagomes has quit IRC | 02:57 | |
*** itlinux has joined #openstack-nova | 02:57 | |
*** lucasagomes has joined #openstack-nova | 02:58 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_instance_info_cache https://review.openstack.org/307155 | 03:00 |
*** jfding has joined #openstack-nova | 03:01 | |
*** annegentle has quit IRC | 03:02 | |
*** lucasagomes has quit IRC | 03:03 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_pci_device https://review.openstack.org/307615 | 03:03 |
*** zul has quit IRC | 03:05 | |
*** ducttape_ has joined #openstack-nova | 03:05 | |
*** lucasagomes has joined #openstack-nova | 03:05 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_compute_node https://review.openstack.org/307625 | 03:06 |
*** gjayavelu has joined #openstack-nova | 03:07 | |
*** itlinux has quit IRC | 03:07 | |
*** NightKhaos has quit IRC | 03:08 | |
*** mingyu has joined #openstack-nova | 03:08 | |
*** sai has quit IRC | 03:08 | |
*** sai has joined #openstack-nova | 03:10 | |
*** lucasagomes has quit IRC | 03:10 | |
*** raunak has joined #openstack-nova | 03:10 | |
*** lucasagomes has joined #openstack-nova | 03:13 | |
*** akshai_ has quit IRC | 03:19 | |
*** lucasagomes has quit IRC | 03:21 | |
*** lucasagomes has joined #openstack-nova | 03:23 | |
*** sai has quit IRC | 03:26 | |
*** sai has joined #openstack-nova | 03:27 | |
*** lucasagomes has quit IRC | 03:28 | |
*** lucasagomes has joined #openstack-nova | 03:29 | |
*** Sukhdev has joined #openstack-nova | 03:36 | |
*** ducttape_ has quit IRC | 03:38 | |
*** baoli has quit IRC | 03:39 | |
*** hoangcx has joined #openstack-nova | 03:40 | |
*** lucasagomes has quit IRC | 03:41 | |
*** lucasagomes has joined #openstack-nova | 03:43 | |
*** itlinux has joined #openstack-nova | 03:46 | |
*** takedakn has joined #openstack-nova | 03:47 | |
*** sai has quit IRC | 03:47 | |
*** lucasagomes has quit IRC | 03:48 | |
*** sai has joined #openstack-nova | 03:49 | |
*** lucasagomes has joined #openstack-nova | 03:50 | |
*** raunak has quit IRC | 03:50 | |
*** thorst_ has quit IRC | 03:50 | |
*** jdurgin has joined #openstack-nova | 03:51 | |
*** lucasagomes has quit IRC | 03:54 | |
*** shangxdy has joined #openstack-nova | 03:54 | |
*** itlinux has quit IRC | 03:56 | |
*** lucasagomes has joined #openstack-nova | 03:56 | |
*** sai has quit IRC | 03:58 | |
*** sai has joined #openstack-nova | 03:59 | |
*** lucasagomes has quit IRC | 04:01 | |
*** thorst_ has joined #openstack-nova | 04:02 | |
*** takedakn has quit IRC | 04:03 | |
*** shangxdy has quit IRC | 04:04 | |
*** lucasagomes has joined #openstack-nova | 04:04 | |
*** Sukhdev has quit IRC | 04:05 | |
*** sai has quit IRC | 04:05 | |
*** sai_ has joined #openstack-nova | 04:08 | |
*** vishwanathj is now known as vishwanathj_zzz | 04:09 | |
*** lucasagomes has quit IRC | 04:12 | |
*** lucasagomes has joined #openstack-nova | 04:13 | |
*** phuongnh has joined #openstack-nova | 04:15 | |
*** lucasagomes has quit IRC | 04:17 | |
*** lucasagomes has joined #openstack-nova | 04:19 | |
*** thorst_ has quit IRC | 04:20 | |
*** tbachman has joined #openstack-nova | 04:23 | |
*** Sukhdev has joined #openstack-nova | 04:24 | |
*** lucasagomes has quit IRC | 04:24 | |
*** takashin has joined #openstack-nova | 04:24 | |
*** lucasagomes has joined #openstack-nova | 04:26 | |
*** diga has joined #openstack-nova | 04:28 | |
*** thorst_ has joined #openstack-nova | 04:28 | |
*** armax has quit IRC | 04:29 | |
*** thorst_ has quit IRC | 04:33 | |
*** lucasagomes has quit IRC | 04:33 | |
*** lucasagomes has joined #openstack-nova | 04:35 | |
*** itlinux has joined #openstack-nova | 04:37 | |
*** ducttape_ has joined #openstack-nova | 04:39 | |
*** pbandark has joined #openstack-nova | 04:40 | |
*** raunak has joined #openstack-nova | 04:41 | |
*** kaisers has joined #openstack-nova | 04:43 | |
*** ducttape_ has quit IRC | 04:43 | |
*** jaosorior has joined #openstack-nova | 04:44 | |
*** yamahata has joined #openstack-nova | 04:44 | |
*** sai_ has quit IRC | 04:44 | |
*** lucasagomes has quit IRC | 04:44 | |
*** sai_ has joined #openstack-nova | 04:45 | |
*** itlinux has quit IRC | 04:47 | |
*** lucasagomes has joined #openstack-nova | 04:47 | |
*** Sukhdev has quit IRC | 04:47 | |
*** mingyu has quit IRC | 04:50 | |
*** lucasagomes has quit IRC | 04:51 | |
*** thorst_ has joined #openstack-nova | 04:52 | |
*** lucasagomes has joined #openstack-nova | 04:53 | |
*** lixiaoy1 has quit IRC | 04:53 | |
*** moshele has joined #openstack-nova | 04:54 | |
*** thorst_ has quit IRC | 04:56 | |
*** lucasagomes has quit IRC | 04:58 | |
*** lucasagomes has joined #openstack-nova | 04:58 | |
*** sai_ has quit IRC | 04:59 | |
*** Jeffrey4l_ has joined #openstack-nova | 05:00 | |
*** jaosorior has quit IRC | 05:00 | |
*** claudiub has joined #openstack-nova | 05:01 | |
*** jaosorior has joined #openstack-nova | 05:01 | |
*** penick_ has joined #openstack-nova | 05:02 | |
*** sai has joined #openstack-nova | 05:03 | |
*** lucasagomes has quit IRC | 05:03 | |
*** thorst_ has joined #openstack-nova | 05:03 | |
*** Jeffrey4l has quit IRC | 05:04 | |
*** lucasagomes has joined #openstack-nova | 05:04 | |
*** penick has quit IRC | 05:04 | |
*** penick_ is now known as penick | 05:04 | |
*** sai has quit IRC | 05:07 | |
*** jaypipes has quit IRC | 05:08 | |
*** thorst_ has quit IRC | 05:08 | |
*** lucasagomes has quit IRC | 05:09 | |
*** sai_ has joined #openstack-nova | 05:09 | |
*** edand has joined #openstack-nova | 05:11 | |
*** lucasagomes has joined #openstack-nova | 05:11 | |
*** scsnow has quit IRC | 05:14 | |
*** armax has joined #openstack-nova | 05:15 | |
*** ratailor has joined #openstack-nova | 05:15 | |
*** thorst_ has joined #openstack-nova | 05:15 | |
*** phschwartz has quit IRC | 05:15 | |
*** mordred has quit IRC | 05:16 | |
*** gongysh has joined #openstack-nova | 05:16 | |
*** sai_ has quit IRC | 05:17 | |
*** lpetrut has joined #openstack-nova | 05:18 | |
*** raunak has quit IRC | 05:18 | |
*** sai_ has joined #openstack-nova | 05:19 | |
*** phschwartz has joined #openstack-nova | 05:19 | |
*** takedakn has joined #openstack-nova | 05:19 | |
*** lucasagomes has quit IRC | 05:19 | |
*** ssurana has quit IRC | 05:20 | |
*** mordred has joined #openstack-nova | 05:20 | |
*** thorst_ has quit IRC | 05:20 | |
*** lucasagomes has joined #openstack-nova | 05:23 | |
*** sai_ has quit IRC | 05:25 | |
*** sai_ has joined #openstack-nova | 05:27 | |
*** lucasagomes has quit IRC | 05:27 | |
*** takedakn has quit IRC | 05:27 | |
*** lucasagomes has joined #openstack-nova | 05:28 | |
*** rcernin has joined #openstack-nova | 05:33 | |
*** lucasagomes has quit IRC | 05:33 | |
*** lucasagomes has joined #openstack-nova | 05:35 | |
*** sai_ has quit IRC | 05:35 | |
*** thorst_ has joined #openstack-nova | 05:37 | |
*** sai has joined #openstack-nova | 05:37 | |
*** giri has joined #openstack-nova | 05:40 | |
*** karthiks has joined #openstack-nova | 05:40 | |
*** jamielennox|away is now known as jamielennox | 05:40 | |
*** ducttape_ has joined #openstack-nova | 05:40 | |
*** gjayavelu has quit IRC | 05:44 | |
*** sai has quit IRC | 05:44 | |
*** ducttape_ has quit IRC | 05:45 | |
*** lucasagomes has quit IRC | 05:45 | |
*** armax has quit IRC | 05:45 | |
*** sai has joined #openstack-nova | 05:45 | |
*** lucasagomes has joined #openstack-nova | 05:47 | |
*** unicell has joined #openstack-nova | 05:47 | |
*** itlinux has joined #openstack-nova | 05:47 | |
*** lucasagomes has quit IRC | 05:51 | |
*** nkrinner_afk is now known as nkrinner | 05:52 | |
*** sai has quit IRC | 05:52 | |
*** sai_ has joined #openstack-nova | 05:55 | |
*** lucasagomes has joined #openstack-nova | 05:55 | |
*** JoseMello has joined #openstack-nova | 05:55 | |
*** raunak has joined #openstack-nova | 05:55 | |
*** lucasagomes has quit IRC | 05:59 | |
*** mjura has joined #openstack-nova | 06:00 | |
*** itlinux has quit IRC | 06:01 | |
*** adisky has quit IRC | 06:01 | |
*** lucasagomes has joined #openstack-nova | 06:02 | |
*** sai_ has quit IRC | 06:02 | |
*** artom has quit IRC | 06:02 | |
*** sai_ has joined #openstack-nova | 06:03 | |
*** artom has joined #openstack-nova | 06:03 | |
*** gcb has joined #openstack-nova | 06:04 | |
*** mingyu has joined #openstack-nova | 06:07 | |
*** paul-carlton2 has joined #openstack-nova | 06:08 | |
*** lucasagomes has quit IRC | 06:09 | |
*** lucasagomes has joined #openstack-nova | 06:10 | |
*** mingyu has quit IRC | 06:12 | |
*** sai_ has quit IRC | 06:16 | |
*** abalutoiu has quit IRC | 06:18 | |
*** sai_ has joined #openstack-nova | 06:18 | |
*** salv-orlando has quit IRC | 06:18 | |
*** lucasagomes has quit IRC | 06:18 | |
*** sridharg has joined #openstack-nova | 06:19 | |
*** lucasagomes has joined #openstack-nova | 06:20 | |
*** nherciu has joined #openstack-nova | 06:22 | |
*** unicell1 has joined #openstack-nova | 06:25 | |
*** unicell has quit IRC | 06:26 | |
*** lpetrut has quit IRC | 06:28 | |
*** raunak has quit IRC | 06:29 | |
*** jaosorior has quit IRC | 06:29 | |
*** jaosorior has joined #openstack-nova | 06:30 | |
*** pcaruana has joined #openstack-nova | 06:32 | |
*** hoangcx has left #openstack-nova | 06:34 | |
*** aswadr_ has joined #openstack-nova | 06:34 | |
*** liverpooler has joined #openstack-nova | 06:35 | |
*** itlinux has joined #openstack-nova | 06:35 | |
*** dimtruck is now known as zz_dimtruck | 06:36 | |
*** prateek has joined #openstack-nova | 06:38 | |
*** code-R has joined #openstack-nova | 06:38 | |
*** sahid has joined #openstack-nova | 06:39 | |
*** sai_ has quit IRC | 06:40 | |
*** code-R_ has joined #openstack-nova | 06:41 | |
*** sai has joined #openstack-nova | 06:42 | |
*** andreas_s has joined #openstack-nova | 06:42 | |
*** code-R has quit IRC | 06:44 | |
*** lucasagomes has quit IRC | 06:44 | |
*** itlinux has quit IRC | 06:46 | |
openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Use assertEqual() instead of assertDictEqual() https://review.openstack.org/346609 | 06:46 |
*** lucasagomes has joined #openstack-nova | 06:46 | |
*** namnh has joined #openstack-nova | 06:48 | |
*** salv-orlando has joined #openstack-nova | 06:50 | |
*** lucasagomes has quit IRC | 06:51 | |
*** unicell has joined #openstack-nova | 06:52 | |
*** sai has quit IRC | 06:52 | |
*** lucasagomes has joined #openstack-nova | 06:53 | |
*** sai has joined #openstack-nova | 06:53 | |
*** unicell1 has quit IRC | 06:54 | |
*** EinstCrazy has joined #openstack-nova | 06:55 | |
*** tesseract- has joined #openstack-nova | 06:57 | |
*** zz_dimtruck is now known as dimtruck | 06:57 | |
*** EinstCra_ has joined #openstack-nova | 06:58 | |
*** EinstCrazy has quit IRC | 06:58 | |
*** lucasagomes has quit IRC | 07:00 | |
*** lucasagomes has joined #openstack-nova | 07:00 | |
*** sai has quit IRC | 07:01 | |
*** abalutoiu has joined #openstack-nova | 07:01 | |
*** ManishD has joined #openstack-nova | 07:02 | |
*** FL1SK has quit IRC | 07:02 | |
*** ManishD has quit IRC | 07:03 | |
*** sai has joined #openstack-nova | 07:03 | |
*** ManishD has joined #openstack-nova | 07:03 | |
*** hoangcx has joined #openstack-nova | 07:04 | |
*** EinstCra_ has quit IRC | 07:04 | |
*** trams_ has quit IRC | 07:04 | |
*** lucasagomes has quit IRC | 07:05 | |
*** trams has joined #openstack-nova | 07:05 | |
*** jsheeren has joined #openstack-nova | 07:05 | |
openstackgerrit | Ghanshyam Mann proposed openstack/nova: Complete verification for os-cloudpipe.inc https://review.openstack.org/369182 | 07:06 |
*** lucasagomes has joined #openstack-nova | 07:06 | |
*** jdurgin has quit IRC | 07:06 | |
*** garyk has joined #openstack-nova | 07:06 | |
*** jistr is now known as jistr|mtgs | 07:09 | |
*** jistr|mtgs is now known as jistr|mtg | 07:09 | |
*** garyk has quit IRC | 07:10 | |
*** dimtruck is now known as zz_dimtruck | 07:10 | |
*** sai has quit IRC | 07:10 | |
openstackgerrit | Ghanshyam Mann proposed openstack/nova: Correct image.inc for heading https://review.openstack.org/369141 | 07:11 |
gmann | alex_xu: hope this is fine now - https://review.openstack.org/#/c/369182/ | 07:11 |
gmann | alex_xu: also very small one - https://review.openstack.org/#/c/369141/1 | 07:11 |
gmann | https://review.openstack.org/#/c/369141/ | 07:11 |
openstackgerrit | Nguyen Phuong An proposed openstack/nova: Prevent use filter() to meet python2,3 https://review.openstack.org/369862 | 07:12 |
alex_xu | gmann: let me check it | 07:12 |
openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: Ensure that periodic reclaim cleans DB deleted instances https://review.openstack.org/323250 | 07:12 |
*** thorst_ has quit IRC | 07:12 | |
gmann | alex_xu: Thanks | 07:12 |
*** sai has joined #openstack-nova | 07:12 | |
gmann | alex_xu: next week i will be in qa code sprint so thinking to fix api-ref one as many i can | 07:13 |
*** garyk has joined #openstack-nova | 07:13 | |
alex_xu | gmann: cool | 07:13 |
*** lucasagomes has quit IRC | 07:13 | |
alex_xu | gmann: will try to review those asap | 07:13 |
alex_xu | gmann: should we say list, show, create, update image and image metadata? | 07:14 |
gmann | alex_xu: ah, yea for metadata | 07:14 |
gmann | ll fix | 07:14 |
*** lucasagomes has joined #openstack-nova | 07:15 | |
*** hoangcx has quit IRC | 07:15 | |
alex_xu | gmann: thanks | 07:15 |
*** wendar has quit IRC | 07:15 | |
gmann | alex_xu: oh you mean to merge both? | 07:15 |
*** wendar has joined #openstack-nova | 07:15 | |
gmann | alex_xu: actually there is no create, update image thing so it will be explicitly for that | 07:16 |
alex_xu | gmann: yea, just think of maybe we should metion metadata? | 07:16 |
alex_xu | gmann: yea, agree | 07:16 |
alex_xu | gmann: oops, there is second line, it mentioned metadata | 07:17 |
gmann | yea | 07:17 |
alex_xu | gmann: sorry, my fault, let me +2 | 07:17 |
gmann | alex_xu: np. Thanks | 07:17 |
alex_xu | gmann: np | 07:17 |
*** paul-carlton2 has quit IRC | 07:19 | |
*** unicell1 has joined #openstack-nova | 07:19 | |
*** unicell has quit IRC | 07:19 | |
*** mingyu has joined #openstack-nova | 07:20 | |
*** lucasagomes has quit IRC | 07:20 | |
*** paul-carlton2 has joined #openstack-nova | 07:20 | |
*** lucasagomes has joined #openstack-nova | 07:21 | |
*** rmart04 has joined #openstack-nova | 07:21 | |
*** gongysh has quit IRC | 07:24 | |
*** markus_z has joined #openstack-nova | 07:24 | |
*** jaosorior has quit IRC | 07:27 | |
*** EinstCrazy has joined #openstack-nova | 07:27 | |
*** jaosorior has joined #openstack-nova | 07:27 | |
*** NightKhaos has joined #openstack-nova | 07:28 | |
*** jaosorior has quit IRC | 07:29 | |
bauzas | markus_z: around ? | 07:29 |
bauzas | markus_z: I need some of your console skills for https://bugs.launchpad.net/nova/+bug/1621257 | 07:29 |
openstack | Launchpad bug 1621257 in OpenStack Compute (nova) "VNC console keeps reporting "setkeycodes 00" exception" [Undecided,Invalid] | 07:29 |
markus_z | bauzas: yep, I'm here, let me have a look | 07:29 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Remove redundant check from code https://review.openstack.org/350122 | 07:29 |
*** sai has quit IRC | 07:30 | |
*** jlanoux has joined #openstack-nova | 07:30 | |
*** rosmaita has quit IRC | 07:30 | |
*** jaosorior has joined #openstack-nova | 07:30 | |
*** rosmaita has joined #openstack-nova | 07:30 | |
bauzas | markus_z: basically, using the serial console works fine, but when using novnc, there is a keymap issue | 07:31 |
bauzas | markus_z: I try to find whether it's due to novnc or nova | 07:31 |
*** lucasagomes has quit IRC | 07:31 | |
bauzas | yesterday, I was thinking it was due to cirros, but the old image is also impacted | 07:31 |
*** EinstCrazy has quit IRC | 07:32 | |
markus_z | bauzas: IIRC there was a pending libvirt driver change regarding keymaps. | 07:32 |
*** paul-carlton2 has quit IRC | 07:33 | |
markus_z | I have a meeting in 30 min, let's see what I can do until then. | 07:33 |
alex_xu | Kevin_Zheng: re: https://review.openstack.org/#/c/359134/11 looks like we are still missing a lot of fields compare to db schema | 07:33 |
*** inara` has quit IRC | 07:34 | |
*** lucasagomes has joined #openstack-nova | 07:34 | |
bauzas | markus_z: but the serial console uses the same libvirt driver for keymaps, right? | 07:34 |
*** phuongnh has quit IRC | 07:34 | |
markus_z | yes | 07:34 |
*** inara has joined #openstack-nova | 07:34 | |
*** oanson has joined #openstack-nova | 07:34 | |
*** phuongnh has joined #openstack-nova | 07:35 | |
Kevin_Zheng | alex_xu: yes, I thought about it | 07:35 |
bauzas | and the serial console uses the same proxy code for reaching the websockets agent ? | 07:35 |
bauzas | markus_z: ^ | 07:35 |
bauzas | if so, that sounds a novnc issue | 07:35 |
Kevin_Zheng | alex_xu: looks like some of them are not good to use as sort key | 07:36 |
markus_z | bauzas: there is a "serialproxy" but it uses the same "baseproxy" as the "novncproxy" | 07:36 |
markus_z | so many proxies... | 07:36 |
Kevin_Zheng | such as vcpu | 07:36 |
bauzas | markus_z: yeah that's what I meant | 07:37 |
*** fragatina has quit IRC | 07:37 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve consistency in serial_console https://review.openstack.org/348902 | 07:37 |
markus_z | bauzas: The latest novnc release was in July, it would fit into the timeline | 07:37 |
bauzas | yeah | 07:37 |
markus_z | https://github.com/kanaka/noVNC/commit/da82b3426c27bf1a79f671c5825d68ab8c0c5d9f | 07:37 |
*** jlanoux has quit IRC | 07:38 | |
*** jlanoux has joined #openstack-nova | 07:38 | |
*** lucasagomes has quit IRC | 07:38 | |
*** sai has joined #openstack-nova | 07:39 | |
*** lucasagomes has joined #openstack-nova | 07:39 | |
alex_xu | Kevin_Zheng: but I think it is part of API contract, I prefer keep it as before. we may remove some of them, but that should be with Microversions | 07:39 |
bauzas | markus_z: either way, it's out of nova scope | 07:39 |
bauzas | markus_z: I can ask the bug reporter to test a former release | 07:39 |
*** EinstCrazy has joined #openstack-nova | 07:40 | |
*** zz_dimtruck has quit IRC | 07:40 | |
Kevin_Zheng | alex_xu: Ok then I will add all fields to the list | 07:40 |
*** salv-orl_ has joined #openstack-nova | 07:40 | |
*** zenoway has joined #openstack-nova | 07:40 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Move notification_format and delete rpc.py https://review.openstack.org/349600 | 07:41 |
*** EinstCrazy has quit IRC | 07:41 | |
markus_z | bauzas: I'll check a little more in Nova itself to be sure. | 07:41 |
*** leifz has quit IRC | 07:42 | |
*** zz_dimtruck has joined #openstack-nova | 07:42 | |
*** salv-orlando has quit IRC | 07:42 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for service options. https://review.openstack.org/332861 | 07:43 |
*** leifz has joined #openstack-nova | 07:43 | |
*** zz_dimtruck is now known as dimtruck | 07:43 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for ldap_dns_opts https://review.openstack.org/341984 | 07:44 |
*** lucasagomes has quit IRC | 07:44 | |
bauzas | markus_z: I just did | 07:44 |
bauzas | markus_z: nova just runs a separate service that ships a baseproxy which runs a websockify handler for listening on websockets and connecting to TCP services | 07:45 |
*** paul-carlton2 has joined #openstack-nova | 07:45 | |
bauzas | markus_z: all that code is shared between the serial console and novnc as far as I can see | 07:46 |
openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Make list->dict conversion more specific https://review.openstack.org/359338 | 07:46 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Reorder scheduler options https://review.openstack.org/350085 | 07:46 |
*** lucasagomes has joined #openstack-nova | 07:47 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Reorder scheduler options https://review.openstack.org/350085 | 07:47 |
markus_z | bauzas: Yes. Trying an earlier version of novnc is the best choice I think. | 07:47 |
*** sai has quit IRC | 07:47 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Make list->dict conversion more specific https://review.openstack.org/359338 | 07:48 |
*** sai has joined #openstack-nova | 07:48 | |
*** phuongnh has quit IRC | 07:48 | |
*** phuongnh has joined #openstack-nova | 07:49 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Add 'deprecated_reason' to osapi opts https://review.openstack.org/357061 | 07:49 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Consider deprecating isolated_images https://review.openstack.org/271354 | 07:50 |
*** lyarwood is now known as lyarwood_ | 07:50 | |
*** lyarwood_ is now known as lyarwood | 07:50 | |
*** hoangcx has joined #openstack-nova | 07:51 | |
*** RuiChen has quit IRC | 07:51 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Move consoleauth options to a group https://review.openstack.org/356611 | 07:52 |
*** lucasagomes has quit IRC | 07:52 | |
*** lucasagomes has joined #openstack-nova | 07:52 | |
*** RuiChen has joined #openstack-nova | 07:52 | |
jsheeren | hi all, is it possible to auto assign a floating ip to a newly booted instance? i see in the config reference for nova an option auto_assign_floating_ip, but this does not seem to work with Neutron? | 07:52 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Add related options to floating ip config options https://review.openstack.org/355561 | 07:53 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for xenapi_session_opts https://review.openstack.org/322322 | 07:53 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Group scheduler options https://review.openstack.org/349959 | 07:54 |
openstackgerrit | Nguyen Phuong An proposed openstack/nova: Prevent use filter() to meet python2,3 https://review.openstack.org/369862 | 07:54 |
openstackgerrit | Maciej Szankin proposed openstack/nova: config options: Improve help for SPICE https://review.openstack.org/317046 | 07:55 |
*** itlinux has joined #openstack-nova | 07:55 | |
*** aunnam has quit IRC | 07:55 | |
*** sai has quit IRC | 07:55 | |
openstackgerrit | Mikhail Feoktistov proposed openstack/nova: Check hypervisor version while getting bus type https://review.openstack.org/282398 | 07:55 |
*** thorst_ has joined #openstack-nova | 07:56 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for glance options https://review.openstack.org/324007 | 07:56 |
*** lucasagomes has quit IRC | 07:57 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve consistency in libvirt https://review.openstack.org/353679 | 07:57 |
*** sai has joined #openstack-nova | 07:58 | |
*** gjayavelu has joined #openstack-nova | 07:58 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for upgrade_levels options https://review.openstack.org/329154 | 07:58 |
*** thorst_ has quit IRC | 07:59 | |
*** lucasagomes has joined #openstack-nova | 07:59 | |
*** zzzeek has quit IRC | 08:00 | |
*** zzzeek has joined #openstack-nova | 08:00 | |
*** artom has quit IRC | 08:00 | |
*** takashin has left #openstack-nova | 08:00 | |
*** thorst_ has joined #openstack-nova | 08:01 | |
*** artom has joined #openstack-nova | 08:01 | |
*** artom has quit IRC | 08:02 | |
*** artom has joined #openstack-nova | 08:02 | |
*** ManishD has quit IRC | 08:02 | |
openstackgerrit | Zhenyu Zheng proposed openstack/nova: Should not be able to sort instances based on joined tables https://review.openstack.org/359134 | 08:03 |
*** permalac has joined #openstack-nova | 08:03 | |
*** CristinaPauna has quit IRC | 08:03 | |
*** lucasagomes has quit IRC | 08:03 | |
*** paul-carlton2 has quit IRC | 08:03 | |
*** danpb has joined #openstack-nova | 08:05 | |
*** sai has quit IRC | 08:05 | |
*** itlinux has quit IRC | 08:05 | |
*** sai has joined #openstack-nova | 08:07 | |
*** lucasagomes has joined #openstack-nova | 08:08 | |
alex_xu | Kevin_Zheng: thanks | 08:10 |
*** sai has quit IRC | 08:11 | |
jsheeren | nevermind, just read this http://lists.openstack.org/pipermail/openstack-dev/2016-September/103056.html | 08:11 |
*** sai has joined #openstack-nova | 08:12 | |
*** derekh has joined #openstack-nova | 08:12 | |
*** kevinz has quit IRC | 08:12 | |
*** code-R_ has quit IRC | 08:13 | |
*** kevinz has joined #openstack-nova | 08:13 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Remove deprecated manager classes in service.py https://review.openstack.org/338553 | 08:15 |
*** thorst_ has quit IRC | 08:16 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve consistency in scheduler opts https://review.openstack.org/349666 | 08:17 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Fix opt description for network.py https://review.openstack.org/341688 | 08:18 |
*** gjayavelu has quit IRC | 08:19 | |
*** giri has quit IRC | 08:19 | |
*** jdurgin has joined #openstack-nova | 08:20 | |
*** thorst_ has joined #openstack-nova | 08:21 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Fix opt indentation for scheduler.py https://review.openstack.org/339736 | 08:21 |
*** yassine has joined #openstack-nova | 08:22 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Config options consistency of ephemeral_storage.py https://review.openstack.org/348487 | 08:22 |
*** yassine is now known as Guest79543 | 08:22 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Move PCI options to a PCI group https://review.openstack.org/356604 | 08:23 |
openstackgerrit | Maciej Szankin proposed openstack/nova: Fix check_config_option_in_central_place https://review.openstack.org/355597 | 08:24 |
*** liverpoo1er has joined #openstack-nova | 08:25 | |
*** gjayavelu has joined #openstack-nova | 08:25 | |
*** gjayavelu1 has joined #openstack-nova | 08:25 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improve help text for interval_opts https://review.openstack.org/331761 | 08:25 |
*** liverpooler has quit IRC | 08:26 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Deprecate all topic conf options https://review.openstack.org/351608 | 08:27 |
*** lucasagomes has quit IRC | 08:27 | |
*** sai has quit IRC | 08:27 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: conf: Remove invalid floating_ips TODO https://review.openstack.org/356572 | 08:28 |
*** matrohon has joined #openstack-nova | 08:28 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Check Config Options Consistency for xenserver.py https://review.openstack.org/350284 | 08:28 |
*** gjayavelu1 has quit IRC | 08:29 | |
*** gjayavelu has quit IRC | 08:29 | |
*** paul-carlton2 has joined #openstack-nova | 08:29 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Add deprecated_since parameter https://review.openstack.org/353631 | 08:30 |
*** sai has joined #openstack-nova | 08:30 | |
*** priteau has joined #openstack-nova | 08:30 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Improving help text for xenapi_vmops_opts https://review.openstack.org/347085 | 08:31 |
*** gjayavelu has joined #openstack-nova | 08:31 | |
*** yamahata has quit IRC | 08:31 | |
*** gjayavelu1 has joined #openstack-nova | 08:31 | |
openstackgerrit | Prateek Arora proposed openstack/nova: Delete traces of in-progress snapshot on VM being deleted https://review.openstack.org/294513 | 08:31 |
*** oomichi has quit IRC | 08:31 | |
*** lucasagomes has joined #openstack-nova | 08:32 | |
*** oomichi has joined #openstack-nova | 08:32 | |
openstackgerrit | Maciej Szankin proposed openstack/nova: Fixed deprecation status for xvp options https://review.openstack.org/353658 | 08:32 |
*** gjayavelu1 has quit IRC | 08:33 | |
*** gjayavelu has quit IRC | 08:33 | |
*** andreykurilin has left #openstack-nova | 08:35 | |
*** andreykurilin has joined #openstack-nova | 08:36 | |
*** garyk has quit IRC | 08:38 | |
*** thorst_ has quit IRC | 08:38 | |
*** thorst_ has joined #openstack-nova | 08:39 | |
*** sai has quit IRC | 08:41 | |
namnh | johnthetubaguy, Hello Mr.Garbutt> I am who discussed rolling-upgrade with you last time. I already thought your command again. And you are quite right. | 08:42 |
johnthetubaguy | namnh: hello, hello, I totally have better docs on the upgrade process on my team's TODO list | 08:42 |
*** RuiChen has quit IRC | 08:43 | |
*** RuiChen has joined #openstack-nova | 08:43 | |
*** thorst_ has quit IRC | 08:44 | |
*** thorst_ has joined #openstack-nova | 08:46 | |
*** lucasagomes has quit IRC | 08:46 | |
*** lucasagomes has joined #openstack-nova | 08:46 | |
namnh | johnthetubaguy, :)) would you mind sharing something about the docs with me. I am really appreciate that. :))) | 08:48 |
*** thorst_ has quit IRC | 08:48 | |
johnthetubaguy | namnh: they are currently in my head, the plan is to get those ones you were reading updated with something a bit more real | 08:48 |
johnthetubaguy | namnh: just checking did you see this blog post: http://superuser.openstack.org/articles/upgrading-nova-to-kilo-with-minimal-downtime/ | 08:49 |
johnthetubaguy | namnh: its not totally up to date, we have made things easier since then, but gives you some good background | 08:49 |
*** NightKhaos has quit IRC | 08:52 | |
*** gjayavelu has joined #openstack-nova | 08:53 | |
*** NightKhaos has joined #openstack-nova | 08:53 | |
*** thorst_ has joined #openstack-nova | 08:54 | |
namnh | johnthetubaguy, Hoping that I can your archive/blog about that soon. Yes, I did. One more thing, did you see this archive: http://www.danplanet.com/blog/2015/10/07/upgrades-in-nova-database-migrations/ | 08:56 |
namnh | johnthetubaguy, that's why I asked you about --expand, --contract options. :) | 08:57 |
johnthetubaguy | namnh: yeah, I helped review some of that before dan posted it | 08:57 |
johnthetubaguy | basically we do everything that happens before upgrade in "db sync" (and "api_db syc"), and everything after you are upgraded in "db online_data_migrations" | 08:58 |
*** thorst_ has quit IRC | 08:59 | |
*** hoangcx has quit IRC | 08:59 | |
*** hoangcx has joined #openstack-nova | 09:02 | |
*** FL1SK has joined #openstack-nova | 09:02 | |
*** prateek has quit IRC | 09:03 | |
namnh | johnthetubaguy, Great :). Oh. I have encountered a bug: https://bugs.launchpad.net/devstack/+bug/1619554 after I finished nova. My log is similar the log of the bug. | 09:04 |
openstack | Launchpad bug 1621257 in OpenStack Compute (nova) "duplicate for #1619554 VNC console keeps reporting "setkeycodes 00" exception" [Undecided,Invalid] | 09:04 |
*** lucasagomes has quit IRC | 09:05 | |
openstackgerrit | javeme proposed openstack/nova: Missing to update task state for confirm_resize https://review.openstack.org/244433 | 09:05 |
*** itlinux has joined #openstack-nova | 09:06 | |
*** lucasagomes has joined #openstack-nova | 09:08 | |
*** claudiub has quit IRC | 09:08 | |
*** sai has joined #openstack-nova | 09:09 | |
*** thorst_ has joined #openstack-nova | 09:09 | |
*** coolsvap_ is now known as coolsvap | 09:09 | |
*** dimtruck is now known as zz_dimtruck | 09:11 | |
*** thorst__ has joined #openstack-nova | 09:12 | |
*** thorst_ has quit IRC | 09:13 | |
*** sai has quit IRC | 09:13 | |
*** lucasagomes has quit IRC | 09:14 | |
*** itlinux has quit IRC | 09:16 | |
*** lucasagomes has joined #openstack-nova | 09:17 | |
openstackgerrit | Alberto Planas proposed openstack/nova: Add migrate_uri for invoking the migration https://review.openstack.org/356558 | 09:19 |
*** mvk has quit IRC | 09:19 | |
*** thorst__ has quit IRC | 09:20 | |
*** efoley has joined #openstack-nova | 09:21 | |
*** lucasagomes has quit IRC | 09:21 | |
*** prateek has joined #openstack-nova | 09:21 | |
*** efoley_ has joined #openstack-nova | 09:22 | |
*** lucasagomes_ has joined #openstack-nova | 09:23 | |
*** lucasagomes_ has quit IRC | 09:23 | |
*** jsheeren has quit IRC | 09:24 | |
*** gjayavelu has quit IRC | 09:25 | |
csatari | Can anyone help in novaclient functional testing? | 09:25 |
*** efoley has quit IRC | 09:25 | |
*** achinm has joined #openstack-nova | 09:25 | |
*** cdent has joined #openstack-nova | 09:26 | |
*** jsheeren has joined #openstack-nova | 09:26 | |
*** hoangcx has quit IRC | 09:27 | |
*** code-R has joined #openstack-nova | 09:27 | |
*** sdake_ has quit IRC | 09:28 | |
*** thorst_ has joined #openstack-nova | 09:28 | |
*** code-R_ has joined #openstack-nova | 09:29 | |
*** hoangcx has joined #openstack-nova | 09:30 | |
*** hoangcx has quit IRC | 09:31 | |
*** lucasagomes has joined #openstack-nova | 09:32 | |
*** code-R has quit IRC | 09:32 | |
*** efoley_ is now known as efoley | 09:34 | |
*** thorst_ has quit IRC | 09:35 | |
*** sai has joined #openstack-nova | 09:37 | |
*** permalac has quit IRC | 09:38 | |
*** hoangcx has joined #openstack-nova | 09:38 | |
*** mingyu_ has joined #openstack-nova | 09:41 | |
*** hoangcx has quit IRC | 09:45 | |
*** mingyu has quit IRC | 09:45 | |
*** sai has quit IRC | 09:46 | |
*** daemontool has joined #openstack-nova | 09:46 | |
*** sai has joined #openstack-nova | 09:47 | |
sfinucan | melwitt: damn it, yeah that's ignored since 1.6.0. So much for random internet articles. I'll revert it. | 09:50 |
openstackgerrit | Stephen Finucane proposed openstack/nova: Revert "tox: Don't create '.pyc' files" https://review.openstack.org/369986 | 09:50 |
*** JoseMello has quit IRC | 09:51 | |
sfinucan | melwitt: and done ^ | 09:51 |
*** mvk has joined #openstack-nova | 09:52 | |
*** hoangcx has joined #openstack-nova | 09:53 | |
*** sai has quit IRC | 09:54 | |
*** jaosorior is now known as jaosorior_lunch | 09:54 | |
*** takedakn has joined #openstack-nova | 09:55 | |
*** takedakn has quit IRC | 09:55 | |
*** rfolco has joined #openstack-nova | 09:56 | |
*** lpetrut has joined #openstack-nova | 09:57 | |
*** sai has joined #openstack-nova | 09:57 | |
*** alexchadin has joined #openstack-nova | 09:59 | |
alexchadin | hi | 09:59 |
alexchadin | can someone help me with booting instance? | 09:59 |
sfinucan | alexchadin: #openstack might be a better place for that question | 10:01 |
alexchadin | sfinucan: ok, thank you! | 10:02 |
sfinucan | alexchadin: np :) | 10:02 |
*** thorst_ has joined #openstack-nova | 10:02 | |
*** efoley has quit IRC | 10:04 | |
*** hoangcx has quit IRC | 10:04 | |
*** takedakn has joined #openstack-nova | 10:04 | |
*** jdurgin has quit IRC | 10:05 | |
*** acabot has quit IRC | 10:06 | |
*** acabot has joined #openstack-nova | 10:07 | |
*** thorst_ has quit IRC | 10:11 | |
*** efoley has joined #openstack-nova | 10:12 | |
*** zenoway has quit IRC | 10:12 | |
*** thorst_ has joined #openstack-nova | 10:12 | |
*** itlinux has joined #openstack-nova | 10:17 | |
*** liverpooler has joined #openstack-nova | 10:17 | |
*** liverpoo1er has quit IRC | 10:19 | |
openstackgerrit | Nguyen Phuong An proposed openstack/nova: Prevent use filter(lambda ...) to meet python2,3 https://review.openstack.org/369862 | 10:19 |
*** thorst_ has quit IRC | 10:20 | |
*** jdurgin has joined #openstack-nova | 10:20 | |
*** mingyu_ has quit IRC | 10:20 | |
*** prateek has quit IRC | 10:22 | |
*** priteau has quit IRC | 10:23 | |
*** gcb has quit IRC | 10:25 | |
*** jaosorior_lunch is now known as jaosorior | 10:27 | |
*** itlinux has quit IRC | 10:28 | |
*** takedakn has quit IRC | 10:28 | |
*** takedakn has joined #openstack-nova | 10:29 | |
*** EinstCrazy has joined #openstack-nova | 10:30 | |
*** jdurgin has quit IRC | 10:32 | |
*** alexchadin has quit IRC | 10:33 | |
*** prateek has joined #openstack-nova | 10:33 | |
*** alexchadin has joined #openstack-nova | 10:34 | |
*** gszasz has joined #openstack-nova | 10:36 | |
*** alexchadin has quit IRC | 10:36 | |
*** alexchadin has joined #openstack-nova | 10:36 | |
*** gszasz has quit IRC | 10:38 | |
*** thorst_ has joined #openstack-nova | 10:39 | |
openstackgerrit | Rikimaru Honjo proposed openstack/nova: Add checks for preventing unexpected detaching volume https://review.openstack.org/354617 | 10:41 |
*** garyk has joined #openstack-nova | 10:42 | |
*** tbachman has quit IRC | 10:45 | |
*** EinstCrazy has quit IRC | 10:48 | |
openstackgerrit | javeme proposed openstack/nova: Missing to update task state for confirm_resize https://review.openstack.org/244433 | 10:48 |
*** takedakn has quit IRC | 10:48 | |
*** thorst_ has quit IRC | 10:51 | |
*** gszasz has joined #openstack-nova | 10:52 | |
*** phuongnh has quit IRC | 10:54 | |
*** CristinaPauna has joined #openstack-nova | 10:54 | |
*** itlinux has joined #openstack-nova | 10:58 | |
*** thorst_ has joined #openstack-nova | 10:59 | |
*** cdelatte has joined #openstack-nova | 11:00 | |
*** zz_dimtruck is now known as dimtruck | 11:01 | |
*** mfedosin has joined #openstack-nova | 11:02 | |
*** ekuris has joined #openstack-nova | 11:03 | |
*** sai has quit IRC | 11:08 | |
*** sdague has joined #openstack-nova | 11:08 | |
*** itlinux has quit IRC | 11:09 | |
johnthetubaguy | sahid: is it possible to quickly add a release note to this patch? https://review.openstack.org/#/c/367909/3 | 11:09 |
mdbooth | Can anybody tell me why we don't setup_network_on_host(dest) *before* starting a live migration? | 11:09 |
*** sai has joined #openstack-nova | 11:10 | |
johnthetubaguy | mdbooth: you mean for neutron or nova-network? | 11:11 |
*** thorst_ has quit IRC | 11:11 | |
mdbooth | johnthetubaguy: If there's a difference between the semantics of the 2, that may be the answer :) | 11:11 |
*** dimtruck is now known as zz_dimtruck | 11:11 | |
mdbooth | I'm just talking about the api call | 11:11 |
mdbooth | In fact, we already do exactly this in pre_live_migration() | 11:13 |
mdbooth | I wonder why we do it again | 11:13 |
*** sahid has quit IRC | 11:13 | |
* mdbooth asks git history | 11:13 | |
*** ihrachys has joined #openstack-nova | 11:13 | |
*** adisky has joined #openstack-nova | 11:14 | |
johnthetubaguy | mdbooth: most of it is about setting up the iptables or ovs rules on the destination, so the network traffic flows there as soon as the instance goes active | 11:14 |
johnthetubaguy | mdbooth: the neutron case has only just had this added for DVR, where we tell neutron where we are heading towards | 11:14 |
mdbooth | johnthetubaguy: That's what I'd have assumed, so it sounds like something you'd want to do early as it's time consuming | 11:15 |
*** rtheis has joined #openstack-nova | 11:15 | |
*** alexchadin has quit IRC | 11:15 | |
johnthetubaguy | mdbooth: honestly, that method really needs splitting into three as I think it does prep, setup and teardown, depending on when its called | 11:15 |
mdbooth | There's also migrate_instance_start() and migration_instance_finish() | 11:15 |
mdbooth | We currently do: | 11:15 |
mdbooth | migration_instance_start(), setup_networks_on_host(), migrate_instance_finish() | 11:15 |
mdbooth | migrate_instance_start() is a no-op in neutron | 11:16 |
mdbooth | setup_networks_on_host seems to do what you just described | 11:16 |
mdbooth | and migrate_instance_finish() seems to swap the ports over | 11:16 |
*** thorst_ has joined #openstack-nova | 11:16 | |
johnthetubaguy | yeah, the finish is more about "we are done" | 11:16 |
mdbooth | So intuitively I'd say we should do the first 2 *before* migration, and only _finish() afterwards. | 11:17 |
johnthetubaguy | I think we call setup_networks_on_host on the source after we are finsihed too, but I could be miss-remembering | 11:17 |
mdbooth | However, intuition tends to get me in trouble in Nova. | 11:17 |
johnthetubaguy | mdbooth: +1 | 11:17 |
johnthetubaguy | all I remember is, its a bit freeky | 11:17 |
mdbooth | We do call it both before and after, just looking to see if git can tell me why. | 11:17 |
*** diga has quit IRC | 11:18 | |
*** bvanhav has joined #openstack-nova | 11:19 | |
johnthetubaguy | so the after is the teardown bit, removing the ip tables rules | 11:20 |
johnthetubaguy | I believe | 11:20 |
johnthetubaguy | or its adding in floating ips if its on the destination | 11:21 |
*** thorst_ has quit IRC | 11:21 | |
lyarwood | dansmith: Morning, I've updated https://review.openstack.org/#/c/290793/ (duplicate bdm bug) with a few notes if you have a chance to take a look once you're online. Still work to do but hopefully straight forward now. | 11:22 |
mdbooth | johnthetubaguy: The answer is in a comment :) Who thought to put useful information in a comment?!?!? | 11:25 |
mdbooth | # NOTE(tr3buchet): setup networks on destination host | 11:25 |
mdbooth | # this is called a second time because | 11:25 |
mdbooth | # multi_host does not create the bridge in | 11:25 |
mdbooth | # plug_vifs | 11:25 |
mdbooth | I wonder if that is still tru | 11:25 |
mdbooth | e | 11:25 |
* mdbooth wonders what multi_host is | 11:26 | |
sdague | mdbooth: it's a specific nova-network mode, but the one that most run in | 11:27 |
mdbooth | Other than 'OpenStack', that is | 11:27 |
mdbooth | sdague: I guessed it meant something specific. I'll dig there for anything relating to the above. | 11:28 |
sdague | where ever node dumps straight onto the physical interface for network traffice, instead of going through a network node | 11:28 |
*** namnh has quit IRC | 11:29 | |
mdbooth | Nova networks not quite dead yet, is it? | 11:29 |
sdague | yeh, not quiet yet | 11:31 |
*** tuanla_ has quit IRC | 11:32 | |
*** sai has quit IRC | 11:32 | |
*** vladikr has joined #openstack-nova | 11:32 | |
*** sai has joined #openstack-nova | 11:33 | |
*** openstackgerrit has quit IRC | 11:34 | |
mdbooth | tr3buchet: Do you happen to have an insight into whether the above comment is still true? | 11:34 |
*** openstackgerrit has joined #openstack-nova | 11:35 | |
mdbooth | danpb: Actually, looking at the libvirt code, does ^^^ intersect os-vif? | 11:35 |
danpb | mdbooth: what in particular ? | 11:36 |
mdbooth | I posted a comment just above from tr3buchet | 11:36 |
mdbooth | I'm wondering if the thing it describes is covered by os-vif | 11:36 |
mdbooth | And even better, if you happen to know anything about it personally | 11:37 |
*** gjayavelu has joined #openstack-nova | 11:37 | |
*** efoley_ has joined #openstack-nova | 11:39 | |
*** efoley has quit IRC | 11:39 | |
danpb | mdbooth: whre does that comment even come from ? | 11:41 |
mdbooth | danpb: post_live_migration_at_destination | 11:41 |
*** yassine has joined #openstack-nova | 11:41 | |
*** gjayavelu has quit IRC | 11:41 | |
mdbooth | It's explaining why we call setup_networks_on_host(dest) twice | 11:41 |
*** yassine is now known as Guest8550 | 11:42 | |
*** sai has quit IRC | 11:43 | |
danpb | mdbooth: the explanation doesn't make alot of sense to me | 11:43 |
danpb | mdbooth: it is saying that this method we call doesn't create the bridge | 11:43 |
danpb | mdbooth: but somehow this method does create the bridge if you call it a second time, which makes no sense at all | 11:43 |
mdbooth | sdague: If I delete it, do we have test coverage of live migration using multi_host? | 11:43 |
*** sai has joined #openstack-nova | 11:44 | |
mdbooth | danpb: That's what I thought, but I don't know the networking code well enough to second guess. | 11:44 |
*** claudiub has joined #openstack-nova | 11:44 | |
*** Guest79543 has quit IRC | 11:45 | |
mdbooth | danpb: I think that's the slowest operation during network switch-over, btw | 11:45 |
mdbooth | If it's not required, that would be good | 11:45 |
danpb | damned if i know - as the comment makes zero sense | 11:46 |
mdbooth | danpb: Ok, thanks | 11:46 |
*** claudiub has quit IRC | 11:46 | |
mdbooth | https://bugs.launchpad.net/nova/+bug/939060 | 11:47 |
openstack | Launchpad bug 939060 in OpenStack Compute (nova) "live migrations do not update dnsmasq entries or setup networking on destination node when using multi_host" [Critical,Fix released] - Assigned to Vish Ishaya (vishvananda) | 11:47 |
*** dave-mccowan has joined #openstack-nova | 11:47 | |
*** woodster_ has joined #openstack-nova | 11:49 | |
*** claudiub has joined #openstack-nova | 11:50 | |
*** lucasagomes is now known as lucas-hungry | 11:50 | |
*** ekuris has quit IRC | 11:52 | |
*** edmondsw has joined #openstack-nova | 11:55 | |
*** markus_zz has joined #openstack-nova | 11:55 | |
*** markus_z has quit IRC | 11:55 | |
*** takedakn has joined #openstack-nova | 11:56 | |
*** artom has quit IRC | 11:56 | |
*** artom has joined #openstack-nova | 11:57 | |
*** lbeliveau has quit IRC | 11:58 | |
*** achinm has quit IRC | 12:00 | |
*** lbeliveau has joined #openstack-nova | 12:00 | |
*** ratailor has quit IRC | 12:01 | |
*** maticue_ has joined #openstack-nova | 12:03 | |
*** efoley__ has joined #openstack-nova | 12:03 | |
openstackgerrit | Stephen Finucane proposed openstack/nova: Update references for split up hardware module https://review.openstack.org/368968 | 12:04 |
*** efoley_ has quit IRC | 12:05 | |
*** Kevin_Zheng has quit IRC | 12:06 | |
*** itlinux has joined #openstack-nova | 12:09 | |
*** gouthamr has joined #openstack-nova | 12:09 | |
*** thorst_ has joined #openstack-nova | 12:10 | |
*** sahid has joined #openstack-nova | 12:10 | |
*** sai has quit IRC | 12:10 | |
openstackgerrit | Stephen Finucane proposed openstack/nova: conf: Make list->dict conversion more specific https://review.openstack.org/359338 | 12:10 |
openstackgerrit | Stephen Finucane proposed openstack/nova: conf: Add 'deprecated_reason' to osapi opts https://review.openstack.org/357061 | 12:10 |
*** sai has joined #openstack-nova | 12:11 | |
*** oanson has quit IRC | 12:11 | |
*** ducttape_ has joined #openstack-nova | 12:12 | |
*** sudipto has joined #openstack-nova | 12:13 | |
*** sudipto_ has joined #openstack-nova | 12:13 | |
*** alexchadin has joined #openstack-nova | 12:13 | |
*** ducttape_ has quit IRC | 12:13 | |
*** sudipto has quit IRC | 12:13 | |
*** sudipto_ has quit IRC | 12:13 | |
*** baoli has joined #openstack-nova | 12:14 | |
*** sudipto has joined #openstack-nova | 12:14 | |
*** sudipto has quit IRC | 12:14 | |
*** ducttape_ has joined #openstack-nova | 12:14 | |
*** sudipto has joined #openstack-nova | 12:14 | |
*** baoli has quit IRC | 12:14 | |
*** sudipto_ has joined #openstack-nova | 12:15 | |
*** baoli has joined #openstack-nova | 12:15 | |
openstackgerrit | Stephen Finucane proposed openstack/nova: Doc fix for Nova API Guide, added missing word. https://review.openstack.org/369061 | 12:15 |
openstackgerrit | Stephen Finucane proposed openstack/nova: Doc fix for Nova API Guide, added missing word. https://review.openstack.org/369061 | 12:15 |
*** lamt has quit IRC | 12:17 | |
*** baoli_ has joined #openstack-nova | 12:18 | |
*** maticue has joined #openstack-nova | 12:18 | |
openstackgerrit | Matthew Booth proposed openstack/nova: Don't call setup_networks_on_host(dest) multiple times https://review.openstack.org/370089 | 12:19 |
*** itlinux has quit IRC | 12:20 | |
*** tbachman has joined #openstack-nova | 12:21 | |
*** baoli has quit IRC | 12:21 | |
*** annegentle has joined #openstack-nova | 12:21 | |
*** jsheeren has quit IRC | 12:22 | |
openstackgerrit | sahid proposed openstack/nova: libvirt: add ps2mouse in choice for pointer_model https://review.openstack.org/367909 | 12:23 |
*** alexchadin has quit IRC | 12:23 | |
*** alexchadin has joined #openstack-nova | 12:23 | |
openstackgerrit | Matthew Booth proposed openstack/nova: Initiate all network changes from source during live migration https://review.openstack.org/369423 | 12:26 |
*** roeyc has joined #openstack-nova | 12:27 | |
*** mriedem has joined #openstack-nova | 12:27 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: add ps2mouse in choice for pointer_model https://review.openstack.org/367909 | 12:31 |
openstackgerrit | Radoslav Gerganov proposed openstack/nova: Fix API doc for os-console-auth-tokens https://review.openstack.org/340969 | 12:31 |
*** efoley__ has quit IRC | 12:32 | |
*** annegentle has quit IRC | 12:34 | |
mriedem | for any stable cores out there, we have quite a few stable/mitaka backports that could use some review https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/mitaka | 12:36 |
*** xyang1 has joined #openstack-nova | 12:37 | |
*** sudipto has quit IRC | 12:39 | |
*** tbachman has quit IRC | 12:39 | |
sdague | mriedem: I think I have a new theory on your unicode bug | 12:39 |
*** sudipto has joined #openstack-nova | 12:39 | |
sdague | https://bugs.launchpad.net/nova/+bug/1580728 | 12:39 |
openstack | Launchpad bug 1580728 in OpenStack Compute (nova) "UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 386: ordinal not in range(128) in nova.virt.libvirt.vif:unplug with unicode instance.display_name" [Medium,Triaged] | 12:39 |
*** sudipto_ has quit IRC | 12:39 | |
*** sudipto_ has joined #openstack-nova | 12:40 | |
mriedem | sdague: ah weird, that would make sense because i could not recreate the failure | 12:40 |
sdague | yeh | 12:41 |
mriedem | and kept getting that formatted Instance str | 12:41 |
sdague | plus the base images run a weird C.utf8 locale, because they aren't built by distros | 12:41 |
*** ociuhandu has quit IRC | 12:42 | |
*** ducttape_ has quit IRC | 12:43 | |
*** lpetrut has quit IRC | 12:43 | |
*** gabor_antal has quit IRC | 12:43 | |
*** lpetrut has joined #openstack-nova | 12:43 | |
*** ftersin has joined #openstack-nova | 12:43 | |
ftersin | mdbooth: hi, are you around? | 12:45 |
mdbooth | ftersin: Yup. Wassup? | 12:45 |
ftersin | about fix https://review.openstack.org/#/c/355415 | 12:45 |
*** sai has quit IRC | 12:46 | |
mdbooth | ftersin: Yes. | 12:46 |
mdbooth | Does it cause a regression? | 12:46 |
ftersin | why not self.get_disk_size(self.path)? | 12:47 |
ftersin | not yet | 12:47 |
ftersin | but i'm wondering how this is work with resizing | 12:47 |
mdbooth | It's to do with resize of an image | 12:48 |
ftersin | particularly with the second and the next resizings | 12:48 |
*** lamt has joined #openstack-nova | 12:48 | |
mdbooth | Although, yeah, given that we haven't resized by that point we could have just used path instead. | 12:48 |
ftersin | when the base image is 1Gb, and flavor gives 2Gb, this code resizes initial disk to 2Gb | 12:48 |
johnthetubaguy | mriedem: I was thinking this should have a reno note, or should we just push that in and put the note in a separate patch? https://review.openstack.org/#/c/367909/3 | 12:49 |
mdbooth | Yes | 12:49 |
mdbooth | ftersin: That's intentional though, right? | 12:49 |
ftersin | but when i resize the instance to 10 Gb flavor, how the disk will be resized? | 12:49 |
mriedem | johnthetubaguy: it has a reno now | 12:49 |
mriedem | johnthetubaguy: i'm +2 | 12:49 |
* johnthetubaguy thinking about getting an eye test | 12:49 | |
*** zz_dimtruck is now known as dimtruck | 12:50 | |
alex_xu | nova api subteam meeting is in 10 mins at #openstack-meeting-4 | 12:50 |
*** tbachman has joined #openstack-nova | 12:50 | |
johnthetubaguy | mriedem: thanks, my spelling is terrible otherwise I would have pushed a reno | 12:50 |
mriedem | heh np | 12:50 |
sdague | mriedem: yeh, it's totally locale environment | 12:51 |
mdbooth | ftersin: Let me just check finish_migration again | 12:51 |
mdbooth | ftersin: Although... I think this *might* have caused a regression with Rbd :( | 12:51 |
paul-carlton2 | pkoniszewski, re post copy implementation, is the switch to post copy mode only manually initiated via force-complete or will it occur automatically when qemu thinks it is time? | 12:51 |
sdague | so... you get some really odd garbage in the display in dumping instances if it's not a utf8 locale, though it doesn't always break | 12:51 |
mdbooth | Because Rbd overrides get_disk_size() and ignores its argument :/ | 12:51 |
mdbooth | And therefore would have previously worked correctly | 12:52 |
sdague | though, interestingly enough, placement api didn't seem to start when I make this a utf8 domain | 12:52 |
ftersin | mdbooth: the idea of this code is to resize a disk, when its size is smaller than requested one | 12:53 |
*** jheroux has joined #openstack-nova | 12:53 | |
*** sai has joined #openstack-nova | 12:53 | |
mdbooth | ftersin: Resize still works here, because size is passed in | 12:53 |
ftersin | but cheking base image, this code can work with some crutches only | 12:54 |
mdbooth | ftersin: This code is a disaster, nobody's going to argue anything else | 12:54 |
mdbooth | This was just a bandaid | 12:54 |
pkoniszewski | paul-carlton2: so there isnt any mechanism in QEMU to trigger post-copy automatically | 12:54 |
*** MVenesio has joined #openstack-nova | 12:54 | |
mdbooth | ftersin: I don't think it broke resize in the general case, because size is passed in correctly from the flavor | 12:54 |
*** MVenesio has quit IRC | 12:55 | |
ftersin | resize_image is never called for resize operation. is it correct? | 12:55 |
*** kylek3h has joined #openstack-nova | 12:55 | |
alex_xu | mriedem: sdague, maybe this patch is worth for rc1 https://review.openstack.org/359134 | 12:55 |
mdbooth | ftersin: see _create_and_inject_local_root(), which is called from _create_image(), which is called from finish_migration() | 12:55 |
mdbooth | ftersin: size comes from the flavor | 12:55 |
paul-carlton2 | pkoniszewski, but nova automatically trigger it when we decide it is time? | 12:55 |
pkoniszewski | paul-carlton2: yes | 12:55 |
sdague | alex_xu: that patch feels kind of high risk right now | 12:56 |
mdbooth | ftersin: So we'll be saying: if (size from flavor) > (size of backing file): resize_image() | 12:56 |
mdbooth | which is correct | 12:56 |
sdague | alex_xu: I think it should probably wait until ocata opens up | 12:56 |
alex_xu | sdague: ah, yeah, have risk break the api contract | 12:56 |
paul-carlton2 | pkoniszewski, ta | 12:56 |
pkoniszewski | paul-carlton2: so nova might change pre-copy to post-copy, currently switch occurs when percentage increase is less than 10% | 12:56 |
mdbooth | *except* for Rbd, which has its own disgusting hack, which I forgot about :( | 12:56 |
sdague | alex_xu: or just other fallout | 12:56 |
sdague | given that it's doing things with db query filtering | 12:57 |
alex_xu | sdague: yea, agree | 12:57 |
sdague | don't get me wrong, I think it's a good fix | 12:57 |
ftersin | mdbooth: hmm... need to think again... | 12:57 |
sdague | I just think it might have fallout, and don't want to hurt the release | 12:57 |
*** lucas-hungry is now known as lucasagomes | 12:57 | |
mdbooth | mriedem: I think https://review.openstack.org/#/c/355415/ has a critical bug | 12:58 |
alex_xu | sdague: yea, I see you | 12:58 |
mdbooth | Blocker | 12:58 |
* mdbooth will propose a revert and write a new patch | 12:58 | |
mdbooth | Actually, wait... | 12:58 |
mdbooth | Maybe not | 12:58 |
* mdbooth reads the Rbd backend again | 12:58 | |
mdbooth | mriedem: Nope, it's a bug, because the prepare function passed in from driver has Rbd-specific gubbins which will ignore the path it was given if it's cloning from a ceph glance store | 13:00 |
*** Kevin_Zheng has joined #openstack-nova | 13:00 | |
alex_xu | nova api meeting is running | 13:01 |
*** jaosorior has quit IRC | 13:01 | |
mdbooth | The bug is that when creating an Rbd instance using Rbd glance, the instance disk won't be resized to the flavor size. | 13:01 |
mdbooth | Sorry about that. | 13:01 |
*** jaosorior has joined #openstack-nova | 13:01 | |
mdbooth | ftersin's solution was better: just check path | 13:02 |
*** pradk has quit IRC | 13:03 | |
*** efoley__ has joined #openstack-nova | 13:04 | |
*** rmart04_ has joined #openstack-nova | 13:04 | |
*** rmart04 has quit IRC | 13:04 | |
*** rmart04_ is now known as rmart04 | 13:04 | |
*** sai has quit IRC | 13:05 | |
ftersin | mdbooth: yeah, you're right - this code will resize disks (except rbd), but also will try to do it when resizing is not required | 13:05 |
mdbooth | ftersin: That last part isn't a regression, though. | 13:05 |
mdbooth | ftersin: I'm just writing up a bug for the Rbd thing, though, and I'll propose a revert and a new patch | 13:06 |
mdbooth | RC1 is Friday :/ | 13:06 |
bauzas | moshele: lbeliveau: hola, is that a regression ? https://bugs.launchpad.net/nova/+bug/1623089 | 13:06 |
openstack | Launchpad bug 1623089 in OpenStack Compute (nova) "Fail to suspend guest with direct-physical ports" [Undecided,New] | 13:06 |
bauzas | mdbooth: s/Friday/Thur EOB | 13:06 |
lbeliveau | bauzas: no it's been there forever, or at least not covered when introducing direct-physical | 13:07 |
*** sai has joined #openstack-nova | 13:07 | |
bauzas | lbeliveau: okay, thanks for that | 13:09 |
mriedem | mdbooth: hmm, dansmith asked about rbd paths in that one | 13:09 |
mriedem | mdbooth: rc1 is really tomorrow eod | 13:09 |
mriedem | PST | 13:09 |
mriedem | mdbooth: that's pretty annoying given all of the people saying it was fixing their issues and how it was just right and didn't need tests or an explanation | 13:10 |
*** rmart04 has quit IRC | 13:10 | |
mdbooth | mriedem: I forgot about this: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L3064-L3070 | 13:11 |
mdbooth | mriedem: Point taken. | 13:11 |
* alex_xu won't be online too much next two days, it is holiday in china | 13:11 | |
mriedem | just surprising since i thought thouse ops that were pushing it were all rbd users | 13:11 |
mriedem | oh well, shit happens | 13:12 |
mdbooth | Perhaps they didn't notice? | 13:12 |
*** efoley__ has quit IRC | 13:12 | |
mdbooth | I'll get a new patch out today. | 13:12 |
mdbooth | And a revert. | 13:12 |
mriedem | just tag the bug with newton-rc-potential and we can throw it into the etherpad | 13:12 |
mriedem | mdbooth: probably want to get those guys to test the new change if possible | 13:12 |
*** takedakn has quit IRC | 13:12 | |
alex_xu | dansmith: morning, when you wakup, hope you can take a look at https://review.openstack.org/369147, if you happy with rename the method, then I will update the patch. | 13:12 |
*** mdrabe has joined #openstack-nova | 13:14 | |
*** tbachman_ has joined #openstack-nova | 13:14 | |
*** efoley__ has joined #openstack-nova | 13:14 | |
*** tbachman has quit IRC | 13:15 | |
*** tbachman_ is now known as tbachman | 13:15 | |
mdbooth | mriedem: That code really is total spaghetti, though :/ The way it's structured you can't touch it unless you retain all these gross hacks in your head simultaneously. | 13:15 |
* mdbooth remembered 2, but forgot 2 more. | 13:16 | |
mriedem | yeah that's why i was asking for an explanation in the commit message | 13:16 |
mriedem | and tests etc | 13:16 |
mdbooth | That was only going to be picked up by functional or CI. The call path length is enormous. | 13:16 |
bauzas | mriedem: FWIW, we still have 31 open bugs that haven't been triaged yet, trying to reduce that number today | 13:17 |
mriedem | we do have a ceph ci job, | 13:17 |
mriedem | bauzas: i've gone through some of those, no red flags as of last night | 13:17 |
mdbooth | Yup. I don't think tempest checks disk sizes though, does it? diana_clarke hit that one trying to write a test for a resize bug. | 13:18 |
mriedem | i've got to get my kid on the bus for school so will be back on later | 13:18 |
mriedem | mdbooth: nope | 13:18 |
mriedem | we need to talk about in-tree functional/integration testing for ocata at the summit i think | 13:18 |
mriedem | for stuff like this | 13:18 |
*** lamt has quit IRC | 13:18 | |
bauzas | mriedem: as well, I did a few of those this (EU) morning | 13:19 |
*** itlinux has joined #openstack-nova | 13:20 | |
*** rmart04 has joined #openstack-nova | 13:21 | |
*** rmart04 has quit IRC | 13:21 | |
moshele | bauzas: no, I would say missing pieces of PF pass-through implementation (it never worked before) | 13:21 |
ftersin | mdbooth: i've spent a day about to implement ephemeral resizing... there is a bit more work than 'update bdms and write tests' | 13:21 |
mdbooth | ftersin: Cool. Got patches? | 13:22 |
ftersin | mdbooth: heh, not yet | 13:22 |
ftersin | mdbooth: at the first - we need to support revert resize operation | 13:22 |
mdbooth | ftersin: We're just using copies for that, right? | 13:23 |
ftersin | mdbooth: i.e. we have to save old ephemeral sizes somewhere to restore bdms | 13:23 |
ftersin | mdbooth: copies of what? of bdms? | 13:23 |
mdbooth | Copies of disks | 13:23 |
*** hamzy has quit IRC | 13:24 | |
ftersin | mdbooth: disks stored on files are copied, root disk on rbd is copied as well | 13:24 |
*** jistr|mtg is now known as jistr | 13:24 | |
mdbooth | Right. Don't they all do that? | 13:24 |
ftersin | mdbooth: this is another task - to snapshot (and restore) ephemerals on ceph | 13:25 |
*** mriedem has quit IRC | 13:25 | |
*** sudipto_ has quit IRC | 13:25 | |
*** sudipto has quit IRC | 13:25 | |
ftersin | mdbooth: but now i'm stopped on _create_image method. i updated bdms, but resize does not work | 13:25 |
*** sudipto has joined #openstack-nova | 13:26 | |
ftersin | mdbooth: because _create_image calls imagebackend.cache with new ephemeral base image | 13:26 |
*** sudipto_ has joined #openstack-nova | 13:26 | |
mdbooth | ftersin: So, getting rid of ephemeral base images is also on my todo list. | 13:27 |
mdbooth | They're a bug. | 13:27 |
*** ducttape_ has joined #openstack-nova | 13:27 | |
ftersin | mdbooth: _create_image gets new bdms and knows nothing about resizing and original ephemeral sizes | 13:27 |
mdbooth | Ah.. hence you're wondering why not (path)? | 13:28 |
mdbooth | Yes, that check doesn't make sense for swap or ephemerals, which is partly why I was ok with the change I'm about to revert | 13:28 |
ftersin | mdbooth: it was my first think. moreover i thought to fix ephemeral bug on flat backend with this (path) | 13:29 |
ftersin | mdbooth: but currently i understand that it is not correct - imagebackend must handle real base image, i.e. _create_image must pass original ephemeral base image | 13:31 |
*** adisky has quit IRC | 13:31 | |
ftersin | mdbooth: but you're right - it was the reason why i went to look on that fix | 13:32 |
mdbooth | ftersin: Of course... cache_name is passed in | 13:32 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Acquire TCP ports for console during live migration https://review.openstack.org/275801 | 13:32 |
*** markus_zz is now known as markus_z | 13:32 | |
mdbooth | And it's dependent on size | 13:32 |
*** burgerk has joined #openstack-nova | 13:32 | |
mdbooth | Eurgh, that's nasty | 13:32 |
openstackgerrit | Matthew Treinish proposed openstack/nova: Use tempest tox with regex first https://review.openstack.org/368566 | 13:33 |
mdbooth | ftersin: In all seriousness, why don't you work on removing ephemeral and swap base images first? | 13:33 |
*** tblakes has joined #openstack-nova | 13:33 | |
*** itlinux has quit IRC | 13:34 | |
mdbooth | The goal of that refactor series, btw, is to persist storage metadata. | 13:34 |
mdbooth | I haven't got that far yet, but that's where it's going. | 13:35 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: Add functional regression test for bug 1595962 https://review.openstack.org/338416 | 13:35 |
openstack | bug 1615613 in OpenStack Compute (nova) "duplicate for #1595962 Live migration always fails when VNC/SPICE is listening at non-local, non-catch-all address" [High,Fix released] https://launchpad.net/bugs/1615613 - Assigned to John Garbutt (johngarbutt) | 13:35 |
mdbooth | With persistent storage metadata, you'd have somewhere to stash the info you need to stash. | 13:35 |
dansmith | alex_xu: +1 | 13:35 |
mdbooth | Without it, I'm not sure where it would go. It's currently inferred from host config, image metadata and bdms. | 13:35 |
mdbooth | I think it would be easier, and better to get rid of it. | 13:35 |
lyarwood | . | 13:37 |
*** hemanthm|away is now known as hemanthm | 13:40 | |
*** artom has quit IRC | 13:40 | |
*** salv-orlando has joined #openstack-nova | 13:40 | |
*** cleong has joined #openstack-nova | 13:41 | |
*** artom has joined #openstack-nova | 13:41 | |
*** ducttape_ has quit IRC | 13:41 | |
*** artom has quit IRC | 13:42 | |
*** artom has joined #openstack-nova | 13:42 | |
*** markvoelker has joined #openstack-nova | 13:43 | |
ftersin | mdbooth: what do you mean with 'removing base images'? | 13:43 |
*** salv-orl_ has quit IRC | 13:43 | |
mdbooth | ftersin: Ephemeral and swap disks shouldn't use the image cache. | 13:43 |
*** tongli has joined #openstack-nova | 13:43 | |
mdbooth | They should create the target disk, and run the template function to initialize it | 13:44 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: WIP: live-migration test hook for serial console https://review.openstack.org/347471 | 13:44 |
mdbooth | i.e. They should be flat in all cases. | 13:44 |
ftersin | mdbooth: got it. but... recently you mention huge time requred to format an empty ephemeral... | 13:44 |
*** mgoddard_ has joined #openstack-nova | 13:45 | |
mdbooth | ftersin: That's not for all formats. e.g. ext4 is very quick. ext3 isn't. | 13:45 |
mdbooth | ftersin: However, we also don't create them for Lvm. | 13:45 |
*** marst has quit IRC | 13:45 | |
mdbooth | And also, flat copies the whole thing | 13:46 |
mdbooth | And Rbd copies it over the wire, which is very inefficient | 13:46 |
ftersin | mdbooth: yes. probably it's a good idea. for the future | 13:46 |
*** burgerk has quit IRC | 13:46 | |
ftersin | mdbooth: though it will require support of existing instances, i.e. we cannot just drop base ephemerals | 13:47 |
mdbooth | ftersin: And it's a data corruptor bug during both live and cold migration. | 13:47 |
mdbooth | ftersin: Yup, that would be the biggest issue to address. We either continue to support them having base images, or flatten them when we find them. | 13:48 |
mdbooth | Latter would probably be simplest, tbh, as it only affects Qcow2 | 13:48 |
*** mgoddard has quit IRC | 13:49 | |
*** jistr is now known as jistr|mtg | 13:49 | |
*** mtanino has joined #openstack-nova | 13:49 | |
openstackgerrit | Matthew Booth proposed openstack/nova: Revert "Fix resizing in imagebackend.cache()" https://review.openstack.org/370135 | 13:49 |
*** lamt has joined #openstack-nova | 13:52 | |
*** mgoddard_ has quit IRC | 13:52 | |
*** alexchadin has left #openstack-nova | 13:52 | |
*** mgoddard has joined #openstack-nova | 13:52 | |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: WIP: live-migration test hook for serial console https://review.openstack.org/347471 | 13:52 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Acquire TCP ports for console during live migration https://review.openstack.org/275801 | 13:52 |
*** mmedvede has joined #openstack-nova | 13:53 | |
ftersin | mdbooth: as the result we have: a) horrible code; b) some ideas to refactor it, wich require a lot of time (one's already ate 1.5 version cycles); c) disability to improve current state of code (including new features) due to its horribleness; d) customers still wait for the features and bug fixes. All this is sad | 13:55 |
*** penick has quit IRC | 13:56 | |
mdbooth | ftersin: This is all true | 13:57 |
openstackgerrit | Alex Xu proposed openstack/nova: Clean up allocation when update available resources https://review.openstack.org/369147 | 13:57 |
alex_xu | dansmith: ^ done | 13:57 |
*** sdake has joined #openstack-nova | 13:58 | |
*** marst has joined #openstack-nova | 13:59 | |
*** anush__ has joined #openstack-nova | 14:00 | |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: Add functional regression test for bug 1595962 https://review.openstack.org/338416 | 14:00 |
openstack | bug 1615613 in OpenStack Compute (nova) "duplicate for #1595962 Live migration always fails when VNC/SPICE is listening at non-local, non-catch-all address" [High,Fix released] https://launchpad.net/bugs/1615613 - Assigned to John Garbutt (johngarbutt) | 14:00 |
*** akshai has joined #openstack-nova | 14:01 | |
*** takedakn has joined #openstack-nova | 14:01 | |
*** hamzy has joined #openstack-nova | 14:01 | |
*** Swami has joined #openstack-nova | 14:03 | |
*** Swami has quit IRC | 14:03 | |
*** Swami has joined #openstack-nova | 14:03 | |
*** pradk has joined #openstack-nova | 14:05 | |
*** pradk- has joined #openstack-nova | 14:05 | |
*** pradk- has quit IRC | 14:05 | |
*** rodrigods has quit IRC | 14:06 | |
*** rodrigods has joined #openstack-nova | 14:06 | |
*** sajeesh has joined #openstack-nova | 14:06 | |
*** lamt has quit IRC | 14:06 | |
*** eharney has joined #openstack-nova | 14:07 | |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: WIP: live-migration test hook for serial console https://review.openstack.org/347471 | 14:07 |
*** ayoung has quit IRC | 14:08 | |
openstackgerrit | Alex Xu proposed openstack/nova: Clean up allocation when update available resources https://review.openstack.org/369147 | 14:08 |
*** hshiina has joined #openstack-nova | 14:09 | |
*** claudiub has quit IRC | 14:09 | |
*** READ10 has joined #openstack-nova | 14:11 | |
*** ducttape_ has joined #openstack-nova | 14:11 | |
*** efoley__ has quit IRC | 14:11 | |
*** mriedem has joined #openstack-nova | 14:11 | |
*** mriedem has quit IRC | 14:11 | |
*** rmart04 has joined #openstack-nova | 14:12 | |
*** Daviey_ is now known as Daviey | 14:12 | |
*** matrohon has quit IRC | 14:14 | |
*** artom has quit IRC | 14:14 | |
*** artom has joined #openstack-nova | 14:16 | |
*** artom has quit IRC | 14:16 | |
*** artom has joined #openstack-nova | 14:17 | |
*** code-R_ has quit IRC | 14:17 | |
*** code-R has joined #openstack-nova | 14:17 | |
*** rlrossit has joined #openstack-nova | 14:18 | |
*** rmart04_ has joined #openstack-nova | 14:20 | |
dansmith | sdague: a bunch of things just failed a bunch of tests.. should I recheck or is some known thing going on? | 14:20 |
dansmith | heh | 14:20 |
*** moshele has quit IRC | 14:20 | |
dansmith | nevermind then | 14:20 |
openstackgerrit | Chris Dent proposed openstack/nova: [placement] prevent a KeyError in webob.dec.wsgify https://review.openstack.org/370154 | 14:20 |
openstackgerrit | Sergey Nikitin proposed openstack/nova-specs: Enable to share PCI devices between numa nodes https://review.openstack.org/361140 | 14:21 |
*** rmart04 has quit IRC | 14:21 | |
*** rmart04_ is now known as rmart04 | 14:21 | |
sdague | yeh, something is odd with the cloud8 osic region | 14:21 |
dansmith | okay | 14:21 |
sdague | the fails are all over the place there, but may be related to things being slow due to a non optimal ceph setup | 14:21 |
sdague | infra is doing a temporary disable on that region until it can be debugged | 14:21 |
dansmith | sdague: okay I got two fails, only one on ceph | 14:22 |
*** tri2sing has joined #openstack-nova | 14:23 | |
dansmith | grenade fail on the old side it looks like | 14:23 |
*** tonytan4ever has joined #openstack-nova | 14:24 | |
dansmith | http://logs.openstack.org/33/365633/9/gate/gate-grenade-dsvm-neutron-ubuntu-trusty/92cb41c/logs/grenade.sh.txt.gz#_2016-09-14_12_44_45_681 | 14:24 |
*** mriedem has joined #openstack-nova | 14:25 | |
mriedem | i'm back! | 14:25 |
*** anush__ has quit IRC | 14:26 | |
mriedem | danpb: i thought you might enjoy this bug https://bugs.launchpad.net/nova/+bug/1621076 | 14:26 |
openstack | Launchpad bug 1621076 in OpenStack Compute (nova) "Can't detach interface from VM (if VM has two interface with same mac addresses)" [Undecided,New] - Assigned to leehom (feli5) | 14:26 |
melwitt | o/ | 14:26 |
mriedem | danpb: you can attach multiple interfaces with the same mac to a libvirt guest but can't detach them because libvirt can't find the unique device to detach i guess | 14:26 |
openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: WIP: live-migration test hook for serial console https://review.openstack.org/347471 | 14:28 |
danpb | mriedem: this is still nova's fault | 14:29 |
*** yamahata has joined #openstack-nova | 14:29 | |
danpb | mriedem: the API contract for virDomainDetachDevice says you must pass the full XML for the device shown in the current guest XML | 14:29 |
danpb | mriedem: nova doesn't do this - it re-creates the minimal <interface> XML from scratch | 14:30 |
*** raunak has joined #openstack-nova | 14:30 | |
danpb | mriedem: as a result it is missing the PCI device ID libvirt would normally use to uniquely identify the device | 14:30 |
*** mingyu has joined #openstack-nova | 14:30 | |
danpb | mriedem: so libvirt has no option but to try and find it using mac address | 14:30 |
mriedem | ok, do we have any existing code to build that interface xml from the vif? | 14:30 |
mriedem | seems like a simple fix | 14:30 |
mriedem | it's a latent bug, so not an rc issue, just wanted your ack on it | 14:31 |
danpb | yeah, its been broken forever | 14:31 |
*** pbandark has quit IRC | 14:31 | |
*** jaosorior has quit IRC | 14:31 | |
*** psachin has joined #openstack-nova | 14:33 | |
*** pwnall has joined #openstack-nova | 14:33 | |
*** pwnall is now known as Guest27617 | 14:33 | |
*** fragatina has joined #openstack-nova | 14:34 | |
*** itlinux has joined #openstack-nova | 14:35 | |
*** Swami_ has joined #openstack-nova | 14:35 | |
*** nherciu has quit IRC | 14:36 | |
sdague | dansmith: not our ceph | 14:36 |
*** matrohon has joined #openstack-nova | 14:36 | |
sdague | that cloud is on ceph | 14:36 |
dansmith | sdague: ah, heh | 14:36 |
dansmith | gotcha | 14:36 |
*** rfolco has quit IRC | 14:37 | |
*** brian_price has joined #openstack-nova | 14:37 | |
*** nkrinner is now known as nkrinner_afk | 14:37 | |
*** fragatina has quit IRC | 14:38 | |
*** Swami has quit IRC | 14:38 | |
*** Swami_ has quit IRC | 14:39 | |
*** jamesden_ has joined #openstack-nova | 14:41 | |
*** pbandark has joined #openstack-nova | 14:41 | |
*** mlavalle has joined #openstack-nova | 14:42 | |
mriedem | jwcroppe: did you (PRS) have a scheduler filter that took network bandwidth into account and if a host was flooded would not schedule to it? | 14:42 |
mriedem | i remember you trying to get some network info into the scheduler filter info awhile back | 14:43 |
*** raunak has quit IRC | 14:43 | |
*** jamesde__ has joined #openstack-nova | 14:43 | |
*** claudiub has joined #openstack-nova | 14:43 | |
*** claudiub has quit IRC | 14:43 | |
*** claudiub has joined #openstack-nova | 14:44 | |
bauzas | mriedem: I remember some change about proposing to integrate a new monitor for giving network bandwidth information, maybe from IBM but I need to remember that one | 14:44 |
*** amotoki has joined #openstack-nova | 14:45 | |
bauzas | mriedem: nevermind, it was for memory bandwidth | 14:45 |
mriedem | yeah that was sudipto's | 14:45 |
bauzas | yeah the pcp stuff | 14:45 |
*** jamesden_ has quit IRC | 14:45 | |
sudipto | yeah | 14:45 |
*** GordonTX has joined #openstack-nova | 14:46 | |
*** anush__ has joined #openstack-nova | 14:46 | |
sudipto | Not for network bandwidth though. | 14:46 |
sudipto | It was for memory bandwidth. | 14:46 |
alaski | dansmith: bauzas melwitt mriedem thoughts on skipping the cells meeting today? Since nothing there affects RC I don't want to take up peoples time. | 14:47 |
mriedem | alaski: sounds fine to me | 14:47 |
dansmith | alaski: +1 | 14:47 |
bauzas | alaski: nope, we merged the cellsv2 related bugfix, so I'm all clear | 14:47 |
melwitt | +1 | 14:47 |
mriedem | alaski: my only thing was to start thinking about a summit session | 14:47 |
bauzas | unless some bug pops up | 14:47 |
* alaski slams gavel | 14:47 | |
alaski | it's done | 14:47 |
alaski | mriedem: the wheels are already turning | 14:48 |
openstackgerrit | Matthew Booth proposed openstack/nova: Fix resizing in imagebackend.cache() https://review.openstack.org/370180 | 14:48 |
*** rmart04_ has joined #openstack-nova | 14:48 | |
*** Swami has joined #openstack-nova | 14:48 | |
*** rmart04 has quit IRC | 14:49 | |
*** rmart04_ is now known as rmart04 | 14:49 | |
*** tblakes has quit IRC | 14:49 | |
*** burgerk has joined #openstack-nova | 14:49 | |
*** erlon has joined #openstack-nova | 14:50 | |
*** mingyu has quit IRC | 14:53 | |
*** mingyu has joined #openstack-nova | 14:54 | |
*** roeyc has quit IRC | 14:54 | |
*** rrecio has joined #openstack-nova | 14:55 | |
*** rrecio_ has joined #openstack-nova | 14:56 | |
*** rfolco has joined #openstack-nova | 14:58 | |
*** tbachman has quit IRC | 14:58 | |
*** jamesde__ has quit IRC | 14:58 | |
mdbooth | ftersin: Could you please review https://review.openstack.org/#/c/370180/ ? | 14:58 |
mdbooth | ftersin: This is an rc-potential with imminent deadline :) I think you're best placed to spot any new bugs in there. | 14:59 |
*** akshai has quit IRC | 14:59 | |
*** rrecio has quit IRC | 14:59 | |
mdbooth | ftersin: Also, I've argued in the commit message against adding a test for the Rbd case because it's too hard. This is going to make mriedem understandably sad, so if you can think of a good way to test that I'm sure he'd be very happy. | 15:00 |
*** mjura has quit IRC | 15:01 | |
*** akshai has joined #openstack-nova | 15:01 | |
ftersin | mdbooth: i've started. btw, why do you use resolves-bug tag? closes-bug is obsolete? | 15:01 |
*** jistr|mtg is now known as jistr | 15:01 | |
mdbooth | ftersin: Ah, it's because our internal gerrit uses Resolves. | 15:02 |
mdbooth | I'll edit it. | 15:02 |
*** BjoernT has joined #openstack-nova | 15:03 | |
ftersin | do not worry, i just wondered | 15:03 |
openstackgerrit | Matthew Booth proposed openstack/nova: Fix resizing in imagebackend.cache() https://review.openstack.org/370180 | 15:03 |
*** lykinsbd has joined #openstack-nova | 15:04 | |
*** pbandark has quit IRC | 15:05 | |
*** david-lyle_ has joined #openstack-nova | 15:06 | |
*** gjayavelu has joined #openstack-nova | 15:06 | |
*** david-lyle_ has quit IRC | 15:06 | |
bauzas | mdbooth: re: https://bugs.launchpad.net/nova/+bug/1623497 are you able to reproduce some regression ? | 15:06 |
openstack | Launchpad bug 1623497 in OpenStack Compute (nova) "Booting Ceph instance using Ceph glance doesn't resize root disk to flavor size" [Undecided,New] | 15:06 |
*** tjones has joined #openstack-nova | 15:06 | |
kashyap | mdbooth: The commit summary message is same as the one you cite (I46b5658e) - is that intentional? | 15:07 |
ftersin | mdbooth: as for rbd test - as i remember diana_clakre was going to write a test, which verifies that disk is resized via ssh, iianm | 15:07 |
ftersin | mdbooth: does this test exist? | 15:07 |
mdbooth | kashyap: Yep. It's still a fix for the same bug. | 15:08 |
*** jlanoux has quit IRC | 15:08 | |
kashyap | Maybe mark it part-2, please, if you re-spin. Sorry for the OCD | 15:08 |
mdbooth | ftersin: I don't think this has to involve ssh, as it's not exclusive to resize. | 15:08 |
bauzas | mdbooth: mriedem: tagging some bugreport as RC blocker needs us to make sure there *is* a regression, so I wonder if we should possibly try to find some way to reproduce that like in https://review.openstack.org/#/c/369645/ ? | 15:08 |
bauzas | d/?/ | 15:08 |
mdbooth | bauzas: I'm moderately confident, but I take your point. Does anybody have a ceph devstack to hand? This is trivial to reproduce. | 15:09 |
bauzas | mdbooth: we have a ceph job | 15:09 |
mriedem | the ceph job only runs tempest | 15:09 |
bauzas | mdbooth: so we could write some DNM patch for canary testing it | 15:10 |
mdbooth | Yeah, but it doesn't test disks sizes. | 15:10 |
bauzas | oh f* | 15:10 |
mdbooth | Doesn't, and won't. | 15:10 |
mriedem | whatever happened to jbernard? | 15:10 |
bauzas | graaaah | 15:10 |
*** hemnafk is now known as hemna | 15:10 | |
*** rmart04 has quit IRC | 15:11 | |
melwitt | I think he was recently replicated | 15:11 |
bauzas | mriedem: sec, what does the ceph job ? running a couple of tempest scenarios against a tuned devstack that uses ceph, right? | 15:12 |
mdbooth | bauzas: Yes. | 15:12 |
mriedem | bauzas: it runs full tempest but with ceph for storage | 15:12 |
mdbooth | The bug in question won't cause anything to fail, though. | 15:12 |
mriedem | cinder and glance are backed by ceph | 15:12 |
mdbooth | It'll just cause the root disk to be too small. | 15:12 |
*** armax has joined #openstack-nova | 15:12 | |
mriedem | and the nova imagebackend is rbd | 15:13 |
bauzas | sec, sec, finding | 15:13 |
diana_clarke | ftersin: I abandoned my attempt to write that tempest test b/c it wasn't going to merge. We'll need a different framework for tests like those, I suspect. | 15:13 |
bauzas | the real problem is that we aren't testing that path, right? | 15:13 |
mdbooth | bauzas: What diana_clarke said ^^^. That's the 'won't' part. | 15:13 |
dansmith | mriedem: is this the thing you pointed danpb at recently? http://logs.openstack.org/33/365633/9/check/gate-nova-python27-db-ubuntu-xenial/5a6acb4/console.html#_2016-09-14_14_41_07_869145 | 15:13 |
mriedem | dansmith: i don't remember that | 15:14 |
dansmith | okay | 15:14 |
*** lamt has joined #openstack-nova | 15:15 | |
*** moshele has joined #openstack-nova | 15:16 | |
openstackgerrit | Kashyap Chamarthy proposed openstack/nova: libvirt: fix detach volume race condition https://review.openstack.org/290510 | 15:17 |
*** gjayavelu has quit IRC | 15:17 | |
*** dnalezyty has joined #openstack-nova | 15:19 | |
diana_clarke | Here's that first attempt at a tempest test, perhaps it'll be a useful jumping off point: https://review.openstack.org/#/c/338411/ | 15:19 |
diana_clarke | (I'm only half paying attention, so perhaps not) | 15:19 |
openstackgerrit | Chris Dent proposed openstack/nova: [placement] Full request and response representation gabbit https://review.openstack.org/370204 | 15:19 |
*** dnalezyty has quit IRC | 15:19 | |
cdent | dansmith, bauzas, mriedem that's ^ a start at something at least some of your were asking about. I'm not sure it is a total success, but gets it going. | 15:20 |
bauzas | cdent: cool, if that doesn't mind you, will defer that to post-RC1 | 15:20 |
cdent | bauzas: oh yeah, definitely | 15:20 |
bauzas | but I totally appreciate the change | 15:21 |
cdent | just wanted to be sure people were aware it was there as a referece | 15:21 |
cdent | reference | 15:21 |
bauzas | kewl | 15:21 |
bauzas | diana_clarke: yeah, mtreinish's answer totally makes sense, I just wonder if we could test that in the nova tree | 15:23 |
diana_clarke | bauzas: I don't think so. IIUC that important stuff is stubbed out currently. | 15:24 |
bauzas | in the functional tests tree ? | 15:24 |
diana_clarke | s/that/the | 15:24 |
diana_clarke | bauzas: yes, but I could be wrong. I asked but I don't think I ever got an answer | 15:24 |
diana_clarke | (and then I moved on) | 15:25 |
bauzas | okay | 15:25 |
*** lykinsbd has quit IRC | 15:25 | |
*** rmcall has joined #openstack-nova | 15:25 | |
*** tblakes has joined #openstack-nova | 15:25 | |
*** Kevin_Zheng has quit IRC | 15:26 | |
mtreinish | bauzas: I made sense? That can't be right :) | 15:26 |
*** lykinsbd has joined #openstack-nova | 15:26 | |
bauzas | mtreinish: heh, was talking of https://review.openstack.org/#/c/338411 | 15:26 |
mriedem | bauzas: diana_clarke: the functional tests in nova use the fake virt driver | 15:26 |
mriedem | and no real services external to nova, like cinder or glance or neutron | 15:26 |
diana_clarke | yup, that's what I figured | 15:27 |
mriedem | functional tests only have the api and db | 15:27 |
bauzas | mriedem: that, I know | 15:27 |
mriedem | there has been talk of making an in-tree integration test bucket that's backed by devstack | 15:27 |
mriedem | so we can hit actual things | 15:27 |
bauzas | mriedem: I just need to figure the best approach, and if we could mock out some specific stuff | 15:27 |
bauzas | mriedem: yeah I remember the functional devstack talk | 15:27 |
*** zenoway has joined #openstack-nova | 15:27 | |
mdbooth | Eurgh | 15:28 |
mdbooth | devstack's ceph stuff has moved | 15:28 |
*** edand has quit IRC | 15:28 | |
openstackgerrit | Chris Dent proposed openstack/nova: [placement] Full request and response representation gabbit https://review.openstack.org/370204 | 15:28 |
*** diana_clarke is now known as diana-meeting | 15:28 | |
mriedem | mdbooth: it's in devstack-plugin-ceph | 15:29 |
mriedem | https://github.com/openstack/devstack-plugin-ceph | 15:29 |
*** claudiub has quit IRC | 15:29 | |
mdbooth | Yeah, I've still got old config here | 15:29 |
* mdbooth hates new things | 15:29 | |
*** sridharg has quit IRC | 15:29 | |
*** psachin has quit IRC | 15:30 | |
mriedem | mdbooth: http://logs.openstack.org/15/355415/7/check/gate-tempest-dsvm-full-devstack-plugin-ceph/5a2985a/logs/reproduce.sh | 15:33 |
mriedem | i haven't tried the reproduce.sh script yet | 15:33 |
mriedem | but, if those work it'd be sweet | 15:33 |
mdbooth | I think I've got it. It's a one-liner in local.conf | 15:33 |
mdbooth | enable_plugin devstack-plugin-ceph git://git.openstack.org/openstack/devstack-plugin-ceph | 15:33 |
mdbooth | If that works, I can't complain :) | 15:34 |
mriedem | i don't think that's everything | 15:34 |
mriedem | well maybe | 15:34 |
*** rcernin has quit IRC | 15:34 | |
*** ducttape_ has quit IRC | 15:35 | |
melwitt | if you're only running one replica, that should work | 15:35 |
*** ducttape_ has joined #openstack-nova | 15:35 | |
*** jlanoux has joined #openstack-nova | 15:36 | |
*** jlanoux has quit IRC | 15:36 | |
*** moshele has quit IRC | 15:37 | |
*** jlanoux has joined #openstack-nova | 15:38 | |
cfriesen | bauzas: live migration resource tracking question for you...in _post_live_migration() we do a cast to post_live_migration_at_destination(), then do some stuff and then set the migration status to 'completed'. How do we know that the destination has run the post_live_migration_at_destination() code yet? | 15:39 |
cfriesen | If it hasn't and the dest runs the resource audit code, don't we run the risk of "losing" the instance resources? | 15:40 |
*** pcaruana has quit IRC | 15:40 | |
bauzas | cfriesen: I'm sorry, I'm not expert of that area but I can read code :) | 15:41 |
sdague | cdent / dansmith there is a real unit test fail in here now - http://logs.openstack.org/33/365633/9/check/gate-nova-python27-db-ubuntu-xenial/5a6acb4/console.html | 15:41 |
bauzas | cfriesen: from a pure resource standpoint, we only account on resource usage every periodic call to update_avail_res() since we haven't merged the live-mig claim change | 15:41 |
mdbooth | mriedem: Seems it's just that one line | 15:42 |
mdrabe | jaypipes: I'm taking a look at https://github.com/openstack/nova/blob/master/nova/pci/stats.py#L209-L211, which is _filter_pools_for_spec in the pci stats. This doesn't appear to be correct to me because it's calling utils.pci_device_prop_match with the pool and request specs, but that method expects a pci device and request specs | 15:42 |
dansmith | sdague: it's not real, AFAICT.. I don't see it locally | 15:42 |
dansmith | sdague: and doesn't look related | 15:42 |
sdague | dansmith: well, maybe a merge issue? | 15:43 |
*** rlrossit has quit IRC | 15:43 | |
*** code-R has quit IRC | 15:43 | |
cfriesen | bauzas: ah, I forgot that the claim-over-migration wasn't merged yet. | 15:43 |
dansmith | sdague: I just pulled and ran it and I don't see it locally (a few minutes ago) | 15:43 |
mriedem | mdbooth: good to know, i'm testing out reproduce.sh since i've been meaning to try that anyway | 15:43 |
dansmith | sdague: I guess maybe I need to rebase on master and try, just a sec | 15:44 |
sdague | dansmith: yeh, I don't know. It odd enough though that we really shouldn't have races there. If some other bad code snuck in we should try to get that sorted | 15:44 |
*** slaweq_ has joined #openstack-nova | 15:44 | |
dansmith | sdague: yeah, which is why I went to run it locally and rechecked to see if it happens again, but it's far far away from the content of any of those patches | 15:45 |
sdague | yeh | 15:45 |
* cdent will try too | 15:45 | |
cfriesen | sahid: are you still planning on pushing https://review.openstack.org/#/c/244489 and friends in N? | 15:45 |
dansmith | sdague: cdent: re-running rebased on master now | 15:45 |
*** daemontool_ has joined #openstack-nova | 15:45 | |
dansmith | locally, I mean | 15:45 |
*** BjoernT has quit IRC | 15:45 | |
dansmith | sdague: this is what I asked mriedem about above, because I thought he had said something suck in related to libvirt and os-vif, which is this near | 15:46 |
sdague | dansmith: ok, I was otp, so I guess I missed the backscroll | 15:46 |
sdague | I thought the os-vif thing he was looking at was the unicode issue | 15:47 |
sdague | which I think is an artifact of our test env | 15:47 |
mdrabe | Would someone be able to help me out with a concern I have related to pci stats? | 15:48 |
*** devananda|dinner is now known as devananda | 15:48 | |
dansmith | sdague: maybe I dunno, I just recalled a similar "vif-y" relation | 15:48 |
*** daemontool has quit IRC | 15:48 | |
sdague | yeh, it was showing up during vif-unplug | 15:49 |
sdague | but it was message processing and printing out the instance | 15:49 |
sdague | in a debug message | 15:49 |
dansmith | okay, well, whatever, not related apparently ;) | 15:50 |
*** mriedem has quit IRC | 15:51 | |
*** mriedem1 has joined #openstack-nova | 15:51 | |
*** code-R has joined #openstack-nova | 15:52 | |
*** BjoernT has joined #openstack-nova | 15:52 | |
sdague | cdent: on the back burner plate for you - http://logs.openstack.org/90/369490/1/check/gate-nova-tox-db-functional-ubuntu-xenial/085ac3e/console.html#_2016-09-13_14_37_49_882469 - gabbi fixture setups don't seem to be doing the right thing with log / output capture so they are printing inline from 8 workers simultaneously and making the output stream a bit messy | 15:53 |
cdent | sdague: yeah, I've seen that but don't know what the stories is on how it is supposed to work? | 15:54 |
sahid | cfriesen: yes it something needed, just feel demotivated to keep it up to date so far | 15:54 |
cdent | Is stdout and stderr getting eaten somewhere in the other tests? | 15:54 |
openstackgerrit | Dave Johnston proposed openstack/nova-specs: Implements: blueprint multi-tenant-exclusion-filter https://review.openstack.org/370243 | 15:54 |
cdent | and then replayed in some fashion? | 15:54 |
*** ihrachys has quit IRC | 15:54 | |
*** prateek has quit IRC | 15:55 | |
cdent | sdague: those fixtures never did anything with any logging because there never was any until recently :) | 15:55 |
dansmith | sdague: cdent passes locally still | 15:55 |
*** burt has quit IRC | 15:55 | |
* cdent is still waiting | 15:55 | |
cdent | I did a limited run and it was fine, now doing everything | 15:56 |
sdague | dansmith: ok, I guess just recheck and hope for the best | 15:56 |
cdent | sdague: anyway, I'll create a bug so it doesn't get lost | 15:56 |
*** Sukhdev has joined #openstack-nova | 15:56 | |
dansmith | sdague: dun did that already | 15:56 |
ftersin | mdbooth: some comments looks obsolete, iianm. see inside | 15:57 |
*** rama_y has joined #openstack-nova | 15:57 | |
*** ayoung has joined #openstack-nova | 15:57 | |
*** andreas_s has quit IRC | 15:58 | |
mdbooth | ftersin: Ah, yes. | 15:58 |
mdbooth | The code is still good, but the comment are indeed out of date. | 15:58 |
sdague | cdent: https://github.com/openstack/nova/blob/04e94df9112a03727f040fd55e2cb2cdc8840d60/nova/test.py#L181-L191 is where all that is sorted | 15:59 |
*** kfarr has joined #openstack-nova | 16:00 | |
cdent | sdague: yeah, just found it | 16:00 |
*** mgoddard_ has joined #openstack-nova | 16:01 | |
*** zul has joined #openstack-nova | 16:01 | |
ftersin | mdbooth: should we ask diana to restore her review with the test to let us run it against master and patched code? i tried to run ci, but it's not started probably because the review is abandoned | 16:02 |
*** marst has quit IRC | 16:02 | |
*** tbachman has joined #openstack-nova | 16:03 | |
mdbooth | diana-meeting: ^^^ ? | 16:03 |
*** marst has joined #openstack-nova | 16:03 | |
*** browne has joined #openstack-nova | 16:03 | |
*** mfedosin has quit IRC | 16:04 | |
*** mgoddard has quit IRC | 16:04 | |
*** ayoung has quit IRC | 16:04 | |
*** efoley has joined #openstack-nova | 16:04 | |
*** Dinesh_Bhor has quit IRC | 16:07 | |
mdbooth | mriedem1 ftersin: Hah, it works anyway because Rbd implements this resize itself :) | 16:07 |
mdbooth | So it was previously doing the resize twice | 16:07 |
mdbooth | Now it's only doing it once | 16:07 |
mdbooth | Which is nice | 16:07 |
*** mriedem1 is now known as mriedem | 16:08 | |
mriedem | mdbooth: | 16:08 |
mdbooth | So, meh | 16:08 |
mriedem | ? | 16:08 |
mriedem | so, the bug isn't an rc blocker | 16:08 |
mdbooth | It's up to you whether to take the new patch or not | 16:08 |
mriedem | it's an optimization | 16:08 |
markus_z | sdague: macsz: I created https://github.com/sdague/rst-burndown/pull/1 for the config opts burndown. Maybe I can work tomorrow on that to make it more generic. I'm on vacation for 2 weeks starting on Friday. | 16:08 |
mdbooth | mriedem: I guess you could spin it that way :) | 16:09 |
mriedem | well when you say now it's only doing it once, you mean with your patch? | 16:09 |
mdbooth | No, I mean the bug I reported is real | 16:09 |
mdbooth | And it wasn't doing the resize there | 16:09 |
mdbooth | However, it was doing an additional resize anyway | 16:10 |
mdbooth | So the 'bug' makes it more efficient | 16:10 |
ftersin | mdbooth: rbd backend implements resizing in create_image, but it may be not called | 16:10 |
*** GordonTX has quit IRC | 16:10 | |
*** krtaylor has quit IRC | 16:10 | |
mdbooth | ftersin: When would it not be called? | 16:11 |
mriedem | mdbooth: and your change would introduce a redundant explicit resize? | 16:11 |
mdbooth | mriedem: My change would re-introduce the redundant explicit resize | 16:11 |
diana-meeting | ftersin: sorry about that (was in a meeting) | 16:11 |
*** diana-meeting is now known as dinaa | 16:11 | |
mriedem | ok | 16:11 |
dinaa | ftersin: I've restored the abandoned patch | 16:11 |
*** dinaa has quit IRC | 16:12 | |
mdbooth | mriedem: TBH, I'm slightly happier with the new one than the old, but give that I view both versions as hacks I'm not bothered either way. | 16:12 |
*** diana_clarke has joined #openstack-nova | 16:12 | |
*** moshele has joined #openstack-nova | 16:12 | |
mdbooth | If it's not actually on fire, lets leave it alone | 16:12 |
mriedem | yeah i agree, i'd like to have nic take a look at it | 16:13 |
*** lykinsbd has quit IRC | 16:13 | |
mriedem | i'm assuming you'll want to update the commit message? | 16:13 |
mdbooth | Which one? | 16:13 |
mriedem | your new patch | 16:13 |
mriedem | mdbooth: can you update the bug and remove the rc-potential tag? | 16:13 |
openstackgerrit | STEW TY proposed openstack/nova: Complete verification for os-floating-ips https://review.openstack.org/329704 | 16:13 |
mdbooth | mriedem: Sure | 16:13 |
mriedem | thanks | 16:13 |
melwitt | do we need to merge the revert before rc1? | 16:14 |
*** slaweq_ has quit IRC | 16:14 | |
mriedem | no | 16:14 |
mdbooth | melwitt: no | 16:14 |
melwitt | okay, thanks. was a bit confused | 16:14 |
ftersin | mdbooth: ahh, it's called always since base image never exists | 16:15 |
ftersin | dinaa: thx | 16:15 |
dansmith | sdague: recheck passed those tests | 16:17 |
*** mfedosin has joined #openstack-nova | 16:17 | |
openstackgerrit | STEW TY proposed openstack/nova: Body Verification of api-ref os-volume-attachments.inc https://review.openstack.org/336334 | 16:17 |
*** markus_z has quit IRC | 16:20 | |
mdbooth | mriedem: Compare and contrast: https://review.openstack.org/#/c/317694/20/nova/virt/libvirt/imagebackend.py create_from_image() | 16:21 |
*** Apoorva has joined #openstack-nova | 16:21 | |
*** code-R has quit IRC | 16:21 | |
mdbooth | There are no other calls: that's everything that it does. | 16:22 |
*** Guest27617 has quit IRC | 16:22 | |
mdbooth | All in 1 place | 16:22 |
*** nic has joined #openstack-nova | 16:22 | |
*** lucasagomes has quit IRC | 16:22 | |
mdbooth | The cloning behaviour has complete test coverage: https://review.openstack.org/#/c/317694/20/nova/tests/unit/virt/libvirt/test_imagebackend.py | 16:23 |
*** lucasagomes has joined #openstack-nova | 16:23 | |
openstackgerrit | Dave Johnston proposed openstack/nova-specs: Implements: blueprint multi-tenant-exclusion-filter https://review.openstack.org/370243 | 16:24 |
*** raunak has joined #openstack-nova | 16:25 | |
*** amotoki has quit IRC | 16:26 | |
mriedem | why is that abandoned? | 16:27 |
mriedem | that series is going to continue into ocata | 16:27 |
*** abalutoiu has quit IRC | 16:29 | |
*** cdent has quit IRC | 16:32 | |
*** jlanoux has quit IRC | 16:32 | |
*** matrohon has quit IRC | 16:32 | |
*** marst has quit IRC | 16:33 | |
*** ssurana has joined #openstack-nova | 16:34 | |
*** tbachman has quit IRC | 16:34 | |
*** rcernin has joined #openstack-nova | 16:34 | |
dansmith | sdague: but it failed on that osic timeout again :( | 16:34 |
diana_clarke | mriedem: I abandoned my initial contributions because I'm no longer working on this. | 16:35 |
diana_clarke | mriedem: I figured any parts that are still relevant when there's review bandwidth for them can be pulled in fresh (because code rots). | 16:35 |
diana_clarke | mriedem: There were things I had hoped to discuss during reviews, but in the months since I wrote that code, I've swapped most of that context out. | 16:35 |
*** rlrossit has joined #openstack-nova | 16:35 | |
*** tesseract- has quit IRC | 16:35 | |
diana_clarke | mriedem: Old code like this makes me nervous. Anyhoo, that's why I abandoned it, but I can restore them if you like. | 16:35 |
mriedem | diana_clarke: ok...i thought this was all part of a series so it doesn't rot if it's actively rebased and worked on | 16:35 |
*** tbachman has joined #openstack-nova | 16:36 | |
diana_clarke | mriedem: Except I no longer recall the bits I thought might be suspect. That context has since been lost. | 16:36 |
mriedem | alright | 16:37 |
mriedem | i guess i'd leave TODOs or something in code i wasn't sure about while working on it | 16:37 |
mriedem | idk | 16:37 |
mriedem | then someone else can pick it up | 16:37 |
mriedem | cdent does that quite often | 16:37 |
mriedem | anyway, i suspect mdbooth will pick it up in ocata | 16:38 |
diana_clarke | I have pretty extensive notes, but I had hoped for more of a back and forth with a subject matter expert. | 16:38 |
*** garyk has quit IRC | 16:38 | |
*** kylek3h is now known as kylek3h_away | 16:38 | |
mriedem | ok i'm honestly not sure who that is, mikal maybe, or danpb, or now just mdbooth | 16:39 |
diana_clarke | To see this refactor through to completion, I suspect I nova core will need to be assigned to work closely with mdbooth. | 16:40 |
mriedem | i go to nic for rbd imagebackend questions | 16:40 |
diana_clarke | That is, it's pretty crazy code. The reviewers need to be knee deep in it as well. | 16:40 |
*** tbachman_ has joined #openstack-nova | 16:41 | |
*** Matias has quit IRC | 16:41 | |
mriedem | yeah i don't disagree there | 16:41 |
diana_clarke | Anyhoo, everyone is really busy. I don't mean that in a bad way. Just that it's a really hard area of the code to change and review. | 16:41 |
diana_clarke | typo: "I suspect I nova core" -> "I suspect *a* nova core" | 16:42 |
*** tbachman has quit IRC | 16:42 | |
*** tbachman_ is now known as tbachman | 16:42 | |
dansmith | mriedem: do you know what this osic ceph issue is that's causing fails? | 16:43 |
mriedem | link? | 16:43 |
dansmith | mriedem: the one sdague was talking about | 16:44 |
dansmith | mriedem: he said an infra-related osic backend ceph misconfiguration or something that was causing spurious fails | 16:44 |
*** yamahata has quit IRC | 16:44 | |
*** Swami has quit IRC | 16:44 | |
mriedem | uh | 16:45 |
mriedem | link? | 16:45 |
*** ircuser-1 has quit IRC | 16:45 | |
*** takedakn1 has joined #openstack-nova | 16:46 | |
*** takedakn has quit IRC | 16:47 | |
*** takedakn1 is now known as takedakn | 16:47 | |
*** lpetrut has quit IRC | 16:47 | |
dansmith | mriedem: I'd have to link to eavesdrop, but jsut a sec | 16:47 |
*** mvk has quit IRC | 16:47 | |
dansmith | mriedem: trying to figure out if this is that failure: http://logs.openstack.org/33/365633/9/check/gate-tempest-dsvm-cells-ubuntu-xenial/71ae727/console.html#_2016-09-14_16_31_32_343128 | 16:47 |
mriedem | that's cells v1 | 16:48 |
mriedem | you said ceph above | 16:48 |
dansmith | mriedem: http://eavesdrop.openstack.org/irclogs/%23openstack-nova/%23openstack-nova.2016-09-14.log.html#t2016-09-14T14:21:11 | 16:48 |
dansmith | mriedem: he said ceph | 16:48 |
diana_clarke | mriedem: I've restore my abandoned patches. cc: mdbooth | 16:49 |
diana_clarke | mdbooth: please change the author if you end up using those patches but change them significantly (which I suspect you'll need to). Thanks! | 16:49 |
mriedem | dansmith: oh b/c osic is ceph backed, | 16:49 |
mriedem | which might make the image snapshot time out? | 16:49 |
*** diogogmt has joined #openstack-nova | 16:49 | |
dansmith | diana_clarke: you can't change the author in gerrit | 16:49 |
dansmith | diana_clarke: so if you want that, abandon and let him re-propose them | 16:49 |
dansmith | mriedem: yes because it uses ceph on the backend, but not sure if that's the related bit here or not | 16:49 |
*** jaypipes has joined #openstack-nova | 16:50 | |
mriedem | i think that's this http://status.openstack.org//elastic-recheck/index.html#1620761 | 16:50 |
dansmith | mriedem: I don't see a ton of issues in the glance logs | 16:50 |
dansmith | okay | 16:51 |
diana_clarke | dansmith: Oh, I thought we had this discussion and it was just the gerrit owner (vs author) that couldn't be changed. | 16:51 |
*** ircuser-1 has joined #openstack-nova | 16:51 | |
melwitt | dansmith, diana_clarke: I think that's the case. owner can't be changed but author and committer can | 16:51 |
* diana_clarke shrugs, I'll leave to mdbooth to play with when the time comes | 16:51 | |
*** gszasz has quit IRC | 16:51 | |
dansmith | melwitt: I thought only committer can change? I know that the gerrit owner can't change, but thought that informs the author, but maybe not | 16:52 |
*** burt has joined #openstack-nova | 16:52 | |
dansmith | either way, not sure why it matters :) | 16:52 |
melwitt | dansmith: author can be changed but it's a manual thing you have to do in git before you upload again | 16:52 |
mriedem | that logstash hit is only 4 in 10 days | 16:52 |
mriedem | all on rax-ord which i'm not sure if that's osic or not | 16:52 |
dansmith | melwitt: okay I thought gerrit refuses upload if you --reset-author? | 16:52 |
dansmith | maybe I'm thinking of something else | 16:53 |
diana_clarke | dansmith: I dislike having my name on code I didn't write. And I can usually tell I didn't write it my looking at it :) | 16:53 |
sdague | mriedem: osic-cloud8 was the ceph backed cloud with issues | 16:53 |
diana_clarke | a/my/by | 16:53 |
melwitt | dansmith: I might have old info but I've seen someone change the author line before and it worked. it's just owner can never change | 16:53 |
dansmith | melwitt: okay | 16:53 |
*** derekh has quit IRC | 16:53 | |
*** mdrabe has quit IRC | 16:53 | |
dansmith | sdague: what was the failure symptom of that? | 16:53 |
sdague | but there wasn't any particular signature, mostly things being quite slow | 16:53 |
dansmith | okay | 16:54 |
sdague | and some jobs fully timing out | 16:54 |
dansmith | sdague: yeah, i first read this as a job timeout because I wasn't paying enough attention | 16:54 |
sdague | and it has to do with image format on top of ceph IIRC that was making performance tank | 16:54 |
dansmith | sdague: the bottom placement test seems to have poor karma and just hits everything | 16:54 |
dansmith | s/test/patch/ | 16:54 |
*** lamt has quit IRC | 16:54 | |
sdague | anyway, they are working on it, and pulled the cloud of out rotation until post RC1 to reduce variability for patch landing | 16:55 |
dansmith | oh okay | 16:55 |
*** Sukhdev has quit IRC | 16:56 | |
*** kfarr has quit IRC | 16:56 | |
*** kfarr has joined #openstack-nova | 16:57 | |
*** krtaylor has joined #openstack-nova | 16:57 | |
*** mdrabe has joined #openstack-nova | 16:59 | |
*** jed56 has joined #openstack-nova | 16:59 | |
*** hshiina has quit IRC | 16:59 | |
*** karthiks has quit IRC | 16:59 | |
dansmith | melwitt: my last experience with that might've been wrapped up with any of these: https://gerrit-review.googlesource.com/Documentation/error-invalid-author.html | 17:01 |
*** pwnall has joined #openstack-nova | 17:01 | |
*** tblakes has quit IRC | 17:01 | |
*** pwnall is now known as Guest32546 | 17:01 | |
melwitt | dansmith: I was just about to link that. I was searching around and found whether or not it will work is based on the gerrit access config | 17:02 |
pkholkin | mriedem: hi Matt, are you here? | 17:02 |
dansmith | yeah | 17:02 |
mriedem | pkholkin: i'm about to run out for lunch | 17:03 |
pkholkin | mriedem: ok, no problem, had a question | 17:03 |
melwitt | I just thought I've seen someone do it before. I might have dreamed it | 17:03 |
pkholkin | dansmith: hi Dan) | 17:04 |
mriedem | pkholkin: feel free to ask | 17:04 |
*** fragatina has joined #openstack-nova | 17:04 | |
pkholkin | mriedem: dansmith: today I played with metadata on aggregates and images/flavors | 17:04 |
pkholkin | I understand that if on aggregate we do 'os_distro=xxx,yyy,zzz' | 17:04 |
pkholkin | we will think that we have a set defaultdict(<type 'set'>, {u'os_distro': set([u'xxx', u'yyy', u'zzz'])}) | 17:05 |
mriedem | there is a certain frenchman that would probably love talking about this | 17:05 |
pkholkin | and I was confused about it, and have some other questions | 17:05 |
*** _gryf has quit IRC | 17:06 | |
dansmith | pkholkin: I'm not the right person to ask about that.. maybe jaypipes ? | 17:06 |
pkholkin | I am not sure) | 17:07 |
dansmith | I'd have to go read code to act like I know :) | 17:07 |
pkholkin | I can ask him | 17:07 |
dansmith | pkholkin: my understanding is that jaypipes knows everything, but he's only on the ground for short layovers between flights to important meetings, so... | 17:07 |
pkholkin | dansmith: :) | 17:07 |
dansmith | :P | 17:08 |
pkholkin | jaypipes: hello Jay, are you here)? | 17:08 |
*** brian_price has quit IRC | 17:08 | |
*** hemanthm is now known as hemanthm|lunch | 17:08 | |
*** Guest32546 has quit IRC | 17:08 | |
jaypipes | dansmith: hi. just reviewed alex_xu's patch that you had pushed a revision for yesterday. please see comments. | 17:08 |
jaypipes | pkholkin: hey Pavel, what's up? | 17:08 |
pkholkin | dansmith: mriedem: I was asking because I want to implement this spec https://review.openstack.org/#/c/85399/17 | 17:09 |
dansmith | jaypipes: okay, still trying to ramrod the last few of the other set through the gate :/ | 17:09 |
pkholkin | but also for flavors | 17:09 |
pkholkin | jaypipes: I have a question | 17:09 |
jaypipes | pkholkin: yes, that is correct. you will get set(...) of those values. | 17:09 |
jaypipes | dansmith: need reviews from me on things? | 17:10 |
pkholkin | jaypipes: what is the use case for this | 17:10 |
*** fragatina has quit IRC | 17:10 | |
*** _gryf has joined #openstack-nova | 17:10 | |
pkholkin | I want to implement the strict host isolation, but I always think that metadata is key=value | 17:10 |
*** Swami has joined #openstack-nova | 17:10 | |
*** fragatina has joined #openstack-nova | 17:11 | |
pkholkin | why not just Strings | 17:11 |
jaypipes | pkholkin: there is no "use case" for this :) it's just a giant pile of hacky poo to indicate that the aggregate has computes that can support operating system distros of XXX, YYY, or ZZZ | 17:11 |
*** danpb has quit IRC | 17:12 | |
dansmith | jaypipes: no, they're all locked and loaded, I just have to convince jenkins to let us through | 17:12 |
*** mgoddard_ has quit IRC | 17:13 | |
pkholkin | jaypipes: for flavors and images we can only set one key=value | 17:13 |
*** mgoddard has joined #openstack-nova | 17:13 | |
pkholkin | I thought that for aggregates it is the same | 17:13 |
pkholkin | jaypipes: do we have some documentation about this 'feature') ? | 17:14 |
pkholkin | also e.g. if we set 'os_distro=xxx,yyy,zzz' for image/flavor your filter will not pass | 17:15 |
*** takedakn has quit IRC | 17:16 | |
*** lamt has joined #openstack-nova | 17:17 | |
*** unicell has joined #openstack-nova | 17:18 | |
*** unicell1 has quit IRC | 17:19 | |
jaypipes | pkholkin: for flavors and images, you can only have one because the flavor and image represents the *requested* thing and the aggregate represents the *available* thing. It's like saying "I request an orange and the things that can fulfill my request are providers that include 'orange' in their list of fruit types." | 17:20 |
pkholkin | jaypipes: agree with you, but sometimes I want the provider that does only oranges :) | 17:21 |
*** daemontool_ has quit IRC | 17:21 | |
jaypipes | pkholkin: in other words, a flavor or image says "I am designed for Windows". It doesn't say "I am designed for Windows OR Linux"... | 17:21 |
pkholkin | jaypipes: I want to implement this spec for images and for flavors, too https://review.openstack.org/#/c/85399/17 | 17:21 |
pkholkin | jaypipes: os_distro was just an example, this could be any keys of course) | 17:22 |
openstackgerrit | STEW TY proposed openstack/nova: Complete verification of servers-action-fixed-ip.inc https://review.openstack.org/338606 | 17:22 |
jaypipes | pkholkin: the provider never "does only oranges". A provider has a list of accepted fruit types (the aggregate metadata value). However what *you* are looking for is not what the provider *allows* but rather what the provider *has already provided*. So, you need to look in an entirely different table (the instances table instead of hte aggregate_metadata table) for your answers. | 17:22 |
jaypipes | pkholkin: this is, incidentally, why server groups are a total pile of poo. | 17:23 |
pkholkin | ) | 17:24 |
pkholkin | jaypipes: do you look the spec above? maybe before? | 17:24 |
*** slaweq_ has joined #openstack-nova | 17:25 | |
*** lucasagomes is now known as lucas-dinner | 17:25 | |
*** annegentle has joined #openstack-nova | 17:25 | |
pkholkin | today I worked for PoC for this but these sets broke everything) | 17:26 |
*** unicell has quit IRC | 17:26 | |
pkholkin | seems that if some set has at least two values we will not get strict isolation | 17:26 |
jaypipes | pkholkin: correct. | 17:28 |
*** mdrabe has quit IRC | 17:29 | |
*** mdrabe_ has joined #openstack-nova | 17:29 | |
*** hemanthm|lunch is now known as hemanthm | 17:29 | |
*** kfarr has quit IRC | 17:31 | |
*** sudipto_ has quit IRC | 17:31 | |
*** sudipto has quit IRC | 17:31 | |
jaypipes | biab | 17:32 |
*** jaypipes has quit IRC | 17:32 | |
clarkb | hello nova. managed to get libvirt info on https://bugs.launchpad.net/nova/+bug/1558807 so I went ahead and changed the status from expired to new | 17:35 |
openstack | Launchpad bug 1558807 in OpenStack Compute (nova) "Volume attached at different path than reported by nova/cinder when using config drive" [Undecided,New] | 17:35 |
*** jaypipes has joined #openstack-nova | 17:36 | |
*** claudiub has joined #openstack-nova | 17:36 | |
*** kfarr has joined #openstack-nova | 17:38 | |
*** yamahata has joined #openstack-nova | 17:38 | |
*** itlinux has quit IRC | 17:39 | |
*** rhagarty has joined #openstack-nova | 17:39 | |
*** mingyu has quit IRC | 17:41 | |
*** thorst_ has quit IRC | 17:45 | |
*** sahid has quit IRC | 17:45 | |
*** brian_price has joined #openstack-nova | 17:45 | |
*** thorst_ has joined #openstack-nova | 17:46 | |
ftersin | mdbooth: diana's test's passed on master against ceph: http://logs.openstack.org/11/338411/2/check/gate-tempest-dsvm-full-devstack-plugin-ceph-ubuntu-xenial/499f2b0/console.html#_2016-09-14_17_11_14_583044 | 17:50 |
*** READ10 has quit IRC | 17:50 | |
auggy | clarkb: did you update the bug with the libvirt info? i saw a comment from another person providing some info, is it the same as that? | 17:50 |
clarkb | auggy: its not my libvirt info its theirs | 17:50 |
auggy | clarkb: right yeah, are they going to update the bug with that info? or is the comment from Jim the same? | 17:51 |
*** baoli_ has quit IRC | 17:51 | |
clarkb | I am saying the comment from jim is why I unexpired the review | 17:51 |
clarkb | er bug | 17:51 |
*** tblakes has joined #openstack-nova | 17:51 | |
clarkb | since the lacking piece of info has been provided | 17:51 |
*** unicell has joined #openstack-nova | 17:52 | |
auggy | clarkb: cool, wfm, just wanted to make sure that was the missing info you were referring to :) thanks | 17:52 |
*** baoli has joined #openstack-nova | 17:52 | |
*** READ10 has joined #openstack-nova | 17:52 | |
diana_clarke | ftersin: And by "pass" we actually mean "fail" b/c that test demonstrates a bug. PS. I suspect mdbooth is gone for the day. | 17:52 |
*** thorst_ has quit IRC | 17:54 | |
ftersin | diana_clarke: why? it resizes to 2Gb and verifies that the instance get 2Gb | 17:54 |
*** zenoway has quit IRC | 17:54 | |
diana_clarke | ftersin: I mean the ephemeral resize part. | 17:56 |
ftersin | diana_clarke: there is a bug of ephemeral resizing, but today matthew spoke about resizing of root | 17:56 |
* diana_clarke nods | 17:56 | |
ftersin | ahh, yes. it does not work at all | 17:57 |
*** pwnall has joined #openstack-nova | 17:59 | |
*** edand has joined #openstack-nova | 17:59 | |
*** pwnall is now known as Guest2984 | 17:59 | |
*** vishwanathj_zzz is now known as vishwanathj | 18:01 | |
*** BjoernT has quit IRC | 18:03 | |
*** mvk has joined #openstack-nova | 18:03 | |
*** moshele has quit IRC | 18:08 | |
*** brian_price has quit IRC | 18:11 | |
*** brian_price has joined #openstack-nova | 18:13 | |
openstackgerrit | Sean Dague proposed openstack/nova: DNM: test hacking os-api-ref-src target https://review.openstack.org/370354 | 18:14 |
*** Sukhdev has joined #openstack-nova | 18:16 | |
dansmith | effing finally | 18:27 |
*** Guest8550 has quit IRC | 18:27 | |
dansmith | clarkb: so, libvirt's notion of where devices are is a total guess | 18:28 |
*** READ10 has quit IRC | 18:28 | |
dansmith | clarkb: and we even recently stopped trying so hard | 18:29 |
dansmith | because we can't enforce anything | 18:29 |
dansmith | clarkb: the notion of where something is attached is a xenism | 18:29 |
clarkb | the problem is there isn't any other reasonable way to know where your volume is | 18:29 |
clarkb | because kvm and xen and different and nova is wrong | 18:29 |
clarkb | its actually a hugely important use case for people trying to use volumes | 18:29 |
dansmith | clarkb: there is now, with device metadata | 18:29 |
clarkb | dansmith: so I am supposed to ask cinder for the location now? | 18:30 |
* clarkb doesn't know what device metadata is or where to find it | 18:30 | |
dansmith | clarkb: no, cinder has no idea | 18:30 |
dansmith | clarkb: device metadata is in configdrive or metadata server and should give you information you can use, like volume serial number or something | 18:30 |
clarkb | dansmith: but that won't update because config drive is static | 18:31 |
clarkb | which means its only useful if I have all my volumes attached at boot and never chagne them | 18:31 |
dansmith | clarkb: yeah if you use configdrive | 18:31 |
dansmith | clarkb: my point is, a bug that says the api shows the wrong device path is somewhat pointless at this point I think | 18:32 |
clarkb | except users still need it | 18:32 |
clarkb | is my point | 18:32 |
clarkb | none of the other options come close | 18:32 |
dansmith | clarkb: but we really can't ever make that correct for anything but xen | 18:32 |
dansmith | clarkb: we can't control where your udev decides to put a thing | 18:32 |
clarkb | dansmith: you don't have to control it though you just have to report it back accurately | 18:33 |
dansmith | clarkb: we can give you *other* information perhaps, but the pathing is not under our control | 18:33 |
clarkb | I don't care where it goes just that I am tiold where it went | 18:33 |
clarkb | and that I am told accurately | 18:33 |
*** mgoddard_ has joined #openstack-nova | 18:33 | |
dansmith | clarkb: the only thing that knows that is the guest OS | 18:33 |
*** artom has quit IRC | 18:33 | |
clarkb | ok then we need something else (and static data on config drive doesn't cut it) that tells users what volumes map to which devices | 18:34 |
clarkb | it turns out that we add and remove volumes all the time and its a fairly normal thing to do | 18:34 |
*** baoli has quit IRC | 18:34 | |
dansmith | clarkb: so despite it not helping you for configdrive, this is the device metadata: https://github.com/openstack/nova/blob/master/nova/api/metadata/base.py#L378-L437 | 18:34 |
*** artom has joined #openstack-nova | 18:35 | |
*** mgoddard has quit IRC | 18:36 | |
*** ducttape_ has quit IRC | 18:36 | |
clarkb | looks like serial is optional? and either it or path are the things that would help you find the volume? | 18:36 |
*** ducttape_ has joined #openstack-nova | 18:37 | |
dansmith | clarkb: they're optional in that we get some things from some virt drivers and other things from others, AFAIK | 18:37 |
dansmith | clarkb: meaning, on libvirt we might be able to give you serial, and on xen we might only be able to give you device node | 18:37 |
*** mgoddard_ has quit IRC | 18:38 | |
clarkb | but we should get at least one of them? | 18:38 |
dansmith | clarkb: yeah, and I think bus/address is the most reliable | 18:38 |
clarkb | dansmith: the thing we get on kvm with /dev/disk/by-id/$uuidprefixhere is actually pretty nice too but xen can't do that? | 18:39 |
*** lpetrut has joined #openstack-nova | 18:40 | |
dansmith | clarkb: I think we can't shove the full uuid in or something though right? | 18:40 |
dansmith | clarkb: dunno about xen | 18:40 |
clarkb | ya there is a length limit and I want to say full uuid goes past that | 18:40 |
clarkb | something like that at least | 18:40 |
dansmith | clarkb: vish did a bunch of work trying to get the volume uuid into or something and couldn't because of length | 18:41 |
dansmith | yeah | 18:41 |
dansmith | clarkb: correlating a tag you provide with a bus, controller, device is pretty accurate, | 18:41 |
dansmith | and cloud-init could do all the nice stuff for you, modulo the synchronization problem of configdrive of course | 18:42 |
dansmith | but configdrive is kinda well-understood to be static, so.. | 18:42 |
clarkb | ya our problem is our volumes have never really been static | 18:42 |
clarkb | we semi frequently add new ones to increase disk availability or swap out slow volumes for fast ones and so on | 18:42 |
dansmith | clarkb: so I dunno what to tell you.. I'm not sure I'd want to expose this out the public API, but I guess we could.. only helps if you're instrumenting this from outside, and this was really added for things inside the instance to consume | 18:42 |
dansmith | clarkb: sure, I'm not discounting your use case of course | 18:43 |
clarkb | ya the reason we use the external public api is because it appeared to work the same across clouds (but doesnt and took a while to figure that out) | 18:43 |
clarkb | whereas we knew pretty early on the on instance info was spotty and not consistent across clouds | 18:43 |
dansmith | clarkb: yeah it totally did when all public clouds were RAX on Xen :) | 18:43 |
*** sajeesh has quit IRC | 18:44 | |
*** mgoddard has joined #openstack-nova | 18:44 | |
dansmith | we talked about deprecating and removing that field from the api in a microversion, but not sure we actually did it | 18:45 |
clarkb | what would be really neat is if the cinder uuid could match the uuid reported by blkid | 18:46 |
clarkb | I wonder if thats something that could be forced over iscis | 18:46 |
dansmith | clarkb: right but that blkid uuid is on the filesystem right? | 18:46 |
dansmith | -U uuid | 18:47 |
dansmith | Look up the device that uses this filesystem uuid. | 18:47 |
*** lpetrut has quit IRC | 18:47 | |
dansmith | looks like if you use a fancy partition table maybe there's some correlation there, but either way, not a block thing AFAIK | 18:47 |
clarkb | I have partuuid and uuid | 18:47 |
clarkb | partuuid looks to belong to partition table | 18:48 |
dansmith | right, but a completely blank block device should have neither I think | 18:48 |
clarkb | ah | 18:48 |
dansmith | a DOS partition table won't have partuuid and uuid is on the filesystem I think | 18:48 |
dansmith | serial is the only thing that belongs to the actual block device as far as I know, and that's the short one | 18:48 |
dansmith | short based on 1980s ATA specs or something I think :P | 18:49 |
*** lpetrut has joined #openstack-nova | 18:49 | |
dansmith | clarkb: but yes, I agree with your assertion about what /would/ be nice :P | 18:49 |
openstackgerrit | Ludovic Beliveau proposed openstack/nova: PCI: Release PCI devices on drop_move_claim() https://review.openstack.org/370374 | 18:50 |
*** tonytan4ever has quit IRC | 18:51 | |
*** thorst_ has joined #openstack-nova | 18:53 | |
*** edand has quit IRC | 18:53 | |
*** zul_ has quit IRC | 18:55 | |
*** liverpooler has quit IRC | 18:58 | |
dansmith | sdague: I think the convo with clarkb just now has me pretty convinced we need to go ahead and deprecate that piece of the api to avoid confusion going forward | 18:58 |
dansmith | sdague: thoughts? | 18:58 |
*** zul has quit IRC | 18:59 | |
*** thorst_ has quit IRC | 19:00 | |
*** moshele has joined #openstack-nova | 19:00 | |
*** ayoung has joined #openstack-nova | 19:01 | |
openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Fix wrong call of _is_booted_from_volume https://review.openstack.org/334609 | 19:01 |
openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Delete duplicate check when live-migrating https://review.openstack.org/337056 | 19:01 |
openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Simplify internal usage of get_instance_disk_info https://review.openstack.org/334901 | 19:01 |
*** fragatina has quit IRC | 19:02 | |
*** efoley has quit IRC | 19:02 | |
mriedem | this reminded me that https://review.openstack.org/#/c/305120/ isn't merged yet | 19:02 |
mriedem | heh, i read 'piece of the api' as something else | 19:02 |
dansmith | heh | 19:02 |
*** moshele has quit IRC | 19:03 | |
*** zul_ has joined #openstack-nova | 19:06 | |
*** pcaruana has joined #openstack-nova | 19:07 | |
mriedem | alex_xu: are you working through the night to update this? https://review.openstack.org/#/c/369147/ | 19:10 |
openstackgerrit | Tom Patzig proposed openstack/nova: refresh instances_path when shared storage used https://review.openstack.org/366857 | 19:10 |
dansmith | mriedem: I can update it, I was mostly waiting for the other few things that were in the queue | 19:11 |
mriedem | works for me | 19:11 |
mriedem | we have 2 things left in the rc bugs etherpad | 19:11 |
mriedem | https://etherpad.openstack.org/p/nova-newton-rc-potential | 19:11 |
*** kfarr has quit IRC | 19:17 | |
dansmith | jaypipes: can you see my comments just now on alex_xu's patch? https://review.openstack.org/#/c/369147/ | 19:17 |
jaypipes | dansmith: reading now. | 19:18 |
*** kylek3h_away is now known as kylek3h | 19:19 | |
jaypipes | dansmith: done | 19:21 |
dansmith | jaypipes: thanks | 19:21 |
dansmith | mriedem: I have edleafe's nit corrected and will rebase when the stuff in the gate makes it through, as this conflicts with those | 19:22 |
mriedem | ok | 19:22 |
*** kaisers has quit IRC | 19:22 | |
dansmith | I would do it now, but it is at the bottom of a stack from alex and I wouldn't want him pushing that stack again and messing up the ones in the gate | 19:22 |
*** prateek has joined #openstack-nova | 19:24 | |
*** prateek has quit IRC | 19:24 | |
*** cdent has joined #openstack-nova | 19:25 | |
*** jed56 has quit IRC | 19:25 | |
mriedem | oh i see it's ready to go, just needs the rebase when the time comes | 19:27 |
dansmith | yup | 19:27 |
cdent | diana_clarke: [t 9jB] <- I had similar hopes and expectations. I'm wondering what we can do (down the road, next cycle, not during rc1, etc) to make more of that happen. | 19:28 |
purplerbot | <diana_clarke> I have pretty extensive notes, but I had hoped for more of a back and forth with a subject matter expert. [2016-09-14 16:38:36] [n 9jB] | 19:28 |
*** akshai has quit IRC | 19:28 | |
*** sdake_ has joined #openstack-nova | 19:29 | |
*** ducttape_ has quit IRC | 19:30 | |
dansmith | cdent: really? jaypipes was holding out on you? :) | 19:31 |
*** matrohon has joined #openstack-nova | 19:31 | |
*** aswadr_ has quit IRC | 19:32 | |
*** bvanhav has quit IRC | 19:32 | |
*** lpetrut has quit IRC | 19:32 | |
cdent | dansmith: heh. not just him. I think much of the problem was on my side: I didn't manage my expectations. I was hoping for a lot more iteration, earlier. What's happened in the past week and a half has been great (in the way it has proved things out) but it would have been nice a lot earlier. | 19:33 |
*** sdake has quit IRC | 19:33 | |
*** Apoorva_ has joined #openstack-nova | 19:33 | |
dansmith | cdent: wow, I think you got an incredible amount of back and forth early on, but.. maybe that's just me | 19:33 |
dansmith | definitely more hands-on than most people get with their features :) | 19:33 |
cdent | dansmith: that's what I'm saying, my expectations were out of wack with reality | 19:34 |
*** Sukhdev has quit IRC | 19:34 | |
dansmith | heh, okay | 19:34 |
mriedem | cdent: just interested, how was velocity like this kind of thing in ceilometer? i think working in different projects gives different expectations | 19:34 |
cdent | on the other hand if the attention I got was way above average then it sucks to be other people and my comment about making it better, for everyone, down the road is even more important | 19:35 |
mriedem | well, you got more attention b/c you're working on a big ass priority item | 19:35 |
mriedem | big ass as in, big change | 19:36 |
mriedem | a new api isn't a trivial vendor bp thingy | 19:36 |
*** lpetrut has joined #openstack-nova | 19:36 | |
*** Sukhdev has joined #openstack-nova | 19:36 | |
dansmith | yeah, we've all been banking up attention, hesitant to spend it... | 19:36 |
*** Apoorva has quit IRC | 19:36 | |
cdent | review latency in ceilo, for important stuff was on the order of hours, not days or weeks | 19:37 |
cdent | but the volume of stuff was obviously _way_ less | 19:37 |
mriedem | yeah i was going to say | 19:37 |
mriedem | we had 100 approved bps | 19:37 |
cdent | and the ratio of cores to contributions was more balanced | 19:37 |
mriedem | with half the core team driving several of them | 19:37 |
diana_clarke | mriedem: Do you think a nova core can be assigned to the image backend refactor for the next cycle? | 19:38 |
cdent | In case I'm not being clear here, I'm not trying to bitch, I'm merely trying to support what diana_clarke was saying and wondering how we can make it better for everyone | 19:38 |
mriedem | diana_clarke: i can't assign anyone to anything | 19:38 |
diana_clarke | mriedem: :) | 19:38 |
mriedem | i've been reminded a few times this release that people are going to work on and review what they want to work on and review | 19:38 |
mriedem | i think we'll have fewer priorities for starters in ocata | 19:39 |
mriedem | it's a shorter cycle | 19:39 |
mriedem | so i think it's going to be a continuation of newton for some of the things that didn't get done | 19:39 |
dansmith | yeah | 19:40 |
*** salv-orl_ has joined #openstack-nova | 19:40 | |
mriedem | except ocata we have the py3 thing... | 19:40 |
*** lpetrut has quit IRC | 19:40 | |
mriedem | a core might need to be literally assigned to reviewing those | 19:41 |
dansmith | lol | 19:41 |
diana_clarke | There are only like 1000 mox tests, at an hours each... plus reviews. lol | 19:42 |
mriedem | yes super fun | 19:42 |
*** salv-orlando has quit IRC | 19:43 | |
*** baoli has joined #openstack-nova | 19:43 | |
* cdent shrugs | 19:44 | |
*** akshai has joined #openstack-nova | 19:45 | |
cdent | I'm getting the impression that there's not much enthusiasm explore ways to change things? Is that because now is the wrong time to be talking about it, or because there's no need to change things, or because of something else? | 19:45 |
*** akshai_ has joined #openstack-nova | 19:46 | |
*** Apoorva_ has quit IRC | 19:47 | |
diana_clarke | cdent: I'm nervous about discussing it b/c I don't want to burn any goodwill. If I have any left.., lol. | 19:47 |
*** Apoorva has joined #openstack-nova | 19:47 | |
* mriedem gets out his folder of permanent records | 19:48 | |
mriedem | cdent: there is a chunk in the ocata summit ideas etherpad about a newton retrospective in barcelona | 19:49 |
mriedem | i was going to bring it up in the nova meeting tomorrow too | 19:49 |
diana_clarke | cdent: I think delegating some of the things like reviewing mox patches to non-cores is one thing that might help. | 19:49 |
diana_clarke | cdent: but I gather that's been discussed and declined may times before | 19:49 |
mriedem | subteams haven't been declined | 19:49 |
*** akshai has quit IRC | 19:49 | |
* dansmith literally smashes his face into the table | 19:49 | |
mriedem | that's what the review priorities etherpad was mainly for, like 60% of it was for subteams | 19:50 |
mriedem | granted, the etherpad grows stale | 19:50 |
* cdent gives dansmith a cold cloth | 19:50 | |
mriedem | but it was meant as a place for subteams to post things that need to be reviewed by the subteam and then put in a queue for cores to look at | 19:50 |
mriedem | markus_z (i think) floated the idea of a gerrit dashboard for subteams but i think that was dependent on gerrit tag support or something | 19:51 |
diana_clarke | I mean more like, "you get to +2 only mox patches. 3 bad calls and your lose +2. proceed with caution." | 19:51 |
mriedem | basically star/tag reviews that are deemed ready by a subteam for review by cores | 19:51 |
diana_clarke | man... there's a typo in everything I type today. baby is keeping me up most nights recently, sorry. | 19:52 |
mriedem | i'm not sure if gerrit can even grant +2 on something like that w/o just having +2 on the entire repo | 19:52 |
*** tonytan4ever has joined #openstack-nova | 19:52 | |
mriedem | subtrees maybe | 19:52 |
mriedem | i'm not crazy about +2s like that anyway because it's easier to give someone +2 than it is to take it away | 19:53 |
mriedem | it gets messy | 19:53 |
diana_clarke | mriedem: I don't think gerrit necessarily needs to enforce it. | 19:53 |
*** hoonetorg has quit IRC | 19:53 | |
openstackgerrit | edan david proposed openstack/nova: Remove mox in network/test_linux_net.py https://review.openstack.org/316427 | 19:53 |
diana_clarke | But anyhoo, I know this has been discussed to death, so I won't make you re-hash it. | 19:53 |
*** tonytan4ever has quit IRC | 19:53 | |
*** tonytan4ever has joined #openstack-nova | 19:54 | |
*** moshele has joined #openstack-nova | 19:54 | |
*** moshele has quit IRC | 19:55 | |
*** can8dnSix has joined #openstack-nova | 19:55 | |
sdague | dansmith: the device setting part? yeh, agreed. | 19:57 |
dansmith | sdague: yeah, cool, lets NFO that early in ocata | 19:57 |
mriedem | cdent: question in https://review.openstack.org/#/c/370154/1 | 19:57 |
*** garyk has joined #openstack-nova | 19:58 | |
*** ijw has joined #openstack-nova | 19:58 | |
*** thorst_ has joined #openstack-nova | 19:58 | |
cdent | mriedem: https://www.python.org/dev/peps/pep-0333/#environ-variables | 19:58 |
*** garyk has quit IRC | 19:58 | |
cdent | the _values_ are unpredictable, but the key names for the incoming environ have a spec | 19:59 |
mriedem | ok, your upper threw me off | 19:59 |
mriedem | oh | 19:59 |
mriedem | yeah | 19:59 |
mriedem | derp, mixed that up | 19:59 |
*** pcaruana has quit IRC | 20:00 | |
*** garyk has joined #openstack-nova | 20:01 | |
*** lamt has quit IRC | 20:01 | |
*** rhagarty has quit IRC | 20:02 | |
*** zul has joined #openstack-nova | 20:02 | |
*** jvgrant has quit IRC | 20:02 | |
cdent | mriedem: I'm away for the next 5 days so if you or others would like that to move sooner, please feel free to have at | 20:02 |
mriedem | is it just me or does this not work? tox -e functional -- nova.tests.functional.api.openstack.placement | 20:02 |
mriedem | i'm guessing that doesn't fly with gabbi? | 20:03 |
cdent | mriedem: that does not work because of the magical way gabbi names tests, so it busts discover | 20:03 |
*** rama_y has quit IRC | 20:03 | |
cdent | tox -efunctional placement will work | 20:03 |
cdent | so will tox -efunctional gabbi | 20:03 |
*** ijw has quit IRC | 20:03 | |
*** thorst_ has quit IRC | 20:04 | |
*** rhagarty has joined #openstack-nova | 20:04 | |
sdague | mriedem: it's using the test loader standard, which testr uses, so yeh, it all works | 20:06 |
openstackgerrit | Matt Riedemann proposed openstack/nova: [placement] prevent a KeyError in webob.dec.wsgify https://review.openstack.org/370154 | 20:06 |
*** gouthamr has quit IRC | 20:07 | |
*** garyk has quit IRC | 20:07 | |
*** lamt has joined #openstack-nova | 20:07 | |
*** hoonetorg has joined #openstack-nova | 20:08 | |
*** can8dnSix has quit IRC | 20:08 | |
*** ftersin has quit IRC | 20:08 | |
cdent | mriedem, sdague: the gabbi tests end up with names like this: gabbi.suitemaker.test_placement_api_allocations_check_allocations_by_consumer_id | 20:08 |
*** baoli has quit IRC | 20:09 | |
cdent | that's because the loader is letting a default prefix happen. that can be changed to use the package path to the loader if it is helpful | 20:09 |
*** baoli has joined #openstack-nova | 20:09 | |
*** zul has quit IRC | 20:11 | |
*** zul has joined #openstack-nova | 20:12 | |
*** jvgrant has joined #openstack-nova | 20:12 | |
*** ducttape_ has joined #openstack-nova | 20:12 | |
*** can8dnSix has joined #openstack-nova | 20:13 | |
*** Apoorva has quit IRC | 20:15 | |
*** Apoorva has joined #openstack-nova | 20:16 | |
cdent | good night all | 20:16 |
*** cdent has quit IRC | 20:16 | |
*** fragatina has joined #openstack-nova | 20:17 | |
*** ijw has joined #openstack-nova | 20:19 | |
*** tonytan4ever has quit IRC | 20:22 | |
*** ijw has quit IRC | 20:24 | |
openstackgerrit | Matt Riedemann proposed openstack/nova: [placement] Mark HTTP error responses for translation https://review.openstack.org/369035 | 20:24 |
*** mfedosin has quit IRC | 20:25 | |
*** ducttape_ has quit IRC | 20:27 | |
*** kfox1111 has joined #openstack-nova | 20:28 | |
kfox1111 | anyone ever see this out of libvirt: qemu-system-x86_64: -chardev pty,id=charserial1: Fai | 20:28 |
kfox1111 | led to create chardev | 20:28 |
*** ducttape_ has joined #openstack-nova | 20:33 | |
*** rcernin has quit IRC | 20:36 | |
kfox1111 | a strace shows: chown("/dev/pts/4", 107, 5) = -1 EPERM (Operation not permitted) | 20:37 |
kfox1111 | weird... | 20:37 |
*** tongli has quit IRC | 20:38 | |
*** rcernin has joined #openstack-nova | 20:40 | |
*** rcernin has quit IRC | 20:41 | |
*** rcernin has joined #openstack-nova | 20:41 | |
*** rcernin has quit IRC | 20:42 | |
*** rcernin has joined #openstack-nova | 20:42 | |
*** gouthamr has joined #openstack-nova | 20:43 | |
*** hemna is now known as hemnafk | 20:44 | |
openstackgerrit | Sujitha proposed openstack/nova: Improve help text for glance options https://review.openstack.org/324007 | 20:45 |
*** kfarr has joined #openstack-nova | 20:48 | |
*** mgoddard has quit IRC | 20:48 | |
*** akshai_ has quit IRC | 20:51 | |
*** coolsvap has quit IRC | 20:52 | |
*** ijw has joined #openstack-nova | 20:52 | |
*** can8dnSix has quit IRC | 20:59 | |
*** thorst_ has joined #openstack-nova | 21:03 | |
*** jheroux has quit IRC | 21:03 | |
*** matrohon has quit IRC | 21:04 | |
*** ijw has quit IRC | 21:07 | |
*** slaweq_ has quit IRC | 21:08 | |
*** thorst_ has quit IRC | 21:09 | |
*** xinliang has quit IRC | 21:10 | |
*** kfarr has quit IRC | 21:14 | |
*** ducttape_ has quit IRC | 21:20 | |
*** edmondsw has quit IRC | 21:21 | |
*** xinliang has joined #openstack-nova | 21:23 | |
*** ijw has joined #openstack-nova | 21:28 | |
openstackgerrit | Sujitha proposed openstack/nova: Improve help text for network options. https://review.openstack.org/341203 | 21:30 |
*** rtheis has quit IRC | 21:31 | |
*** ijw has quit IRC | 21:32 | |
*** cleong has quit IRC | 21:34 | |
*** rhagarty has quit IRC | 21:39 | |
*** fragatin_ has joined #openstack-nova | 21:39 | |
*** mriedem has quit IRC | 21:40 | |
*** fragatina has quit IRC | 21:42 | |
*** ijw has joined #openstack-nova | 21:44 | |
*** rlrossit has quit IRC | 21:47 | |
*** dave-mccowan has quit IRC | 21:48 | |
*** ijw has quit IRC | 21:49 | |
*** Swami has quit IRC | 21:49 | |
*** mikal has quit IRC | 21:51 | |
*** mikal has joined #openstack-nova | 21:51 | |
*** tonytan4ever has joined #openstack-nova | 21:53 | |
*** mikal has quit IRC | 21:53 | |
*** mikal has joined #openstack-nova | 21:56 | |
*** claudiub has quit IRC | 21:57 | |
*** tonytan4ever has quit IRC | 21:58 | |
*** fragatin_ has quit IRC | 21:58 | |
*** fragatina has joined #openstack-nova | 21:59 | |
*** ijw has joined #openstack-nova | 22:01 | |
*** krtaylor has quit IRC | 22:02 | |
*** burgerk has quit IRC | 22:04 | |
*** mikal has quit IRC | 22:05 | |
*** lpetrut has joined #openstack-nova | 22:06 | |
*** mikal has joined #openstack-nova | 22:07 | |
*** thorst has joined #openstack-nova | 22:08 | |
jaypipes | dansmith: seems we're in recheck hel... | 22:09 |
dansmith | jaypipes: well, I merged one thing | 22:09 |
dansmith | was about to recheck the others | 22:09 |
dansmith | been watching like a large bird | 22:09 |
*** eharney has quit IRC | 22:11 | |
*** browne has quit IRC | 22:13 | |
*** rama_y has joined #openstack-nova | 22:14 | |
*** thorst has quit IRC | 22:15 | |
*** mdrabe_ has quit IRC | 22:15 | |
*** TravT has joined #openstack-nova | 22:16 | |
openstackgerrit | Dan Smith proposed openstack/nova: Clean up allocation when update available resources https://review.openstack.org/369147 | 22:17 |
*** ducttape_ has joined #openstack-nova | 22:17 | |
*** lpetrut has quit IRC | 22:19 | |
*** yamahata has quit IRC | 22:20 | |
*** yamahata has joined #openstack-nova | 22:21 | |
openstackgerrit | Pavel Kholkin proposed openstack/nova: [WIP] scheduler: strictly isolate group of hosts for image/flavor https://review.openstack.org/370458 | 22:22 |
*** mtanino has quit IRC | 22:23 | |
*** lamt has quit IRC | 22:29 | |
*** baoli has quit IRC | 22:30 | |
*** baoli has joined #openstack-nova | 22:31 | |
*** ducttape_ has quit IRC | 22:32 | |
*** ducttape_ has joined #openstack-nova | 22:32 | |
*** hemanthm is now known as hemanthm|away | 22:32 | |
*** ociuhandu has joined #openstack-nova | 22:33 | |
*** Apoorva_ has joined #openstack-nova | 22:36 | |
*** ducttape_ has quit IRC | 22:37 | |
*** Apoorva has quit IRC | 22:39 | |
*** sdake_ has quit IRC | 22:40 | |
*** rcernin has quit IRC | 22:43 | |
*** mlavalle has quit IRC | 22:43 | |
*** zul has quit IRC | 22:44 | |
*** Swami has joined #openstack-nova | 22:45 | |
*** Swami has quit IRC | 22:46 | |
*** Swami has joined #openstack-nova | 22:46 | |
*** pradk has quit IRC | 22:47 | |
*** diogogmt has quit IRC | 22:47 | |
*** jaypipes has quit IRC | 22:50 | |
*** tblakes has quit IRC | 22:56 | |
*** takashin has joined #openstack-nova | 22:56 | |
openstackgerrit | Anusha Unnam proposed openstack/nova: Improve help text for interval_opts https://review.openstack.org/331761 | 22:59 |
*** slagle has quit IRC | 23:01 | |
openstackgerrit | Anusha Unnam proposed openstack/nova: Improve help text for interval_opts https://review.openstack.org/331761 | 23:01 |
*** sdague has quit IRC | 23:02 | |
alex_xu | dansmith: thanks for take care the patch | 23:06 |
dansmith | alex_xu: np | 23:06 |
*** lamt has joined #openstack-nova | 23:06 | |
*** baoli has quit IRC | 23:06 | |
* alex_xu can prepare to go to the zoo with baby now \o/ | 23:07 | |
*** tblakes has joined #openstack-nova | 23:08 | |
*** kuzko has quit IRC | 23:09 | |
*** ducttape_ has joined #openstack-nova | 23:10 | |
*** xyang1 has quit IRC | 23:10 | |
*** kuzko has joined #openstack-nova | 23:11 | |
*** thorst has joined #openstack-nova | 23:12 | |
*** browne has joined #openstack-nova | 23:13 | |
*** nic has quit IRC | 23:19 | |
*** thorst has quit IRC | 23:20 | |
*** Swami has quit IRC | 23:21 | |
*** Swami has joined #openstack-nova | 23:21 | |
*** jamielennox is now known as jamielennox|away | 23:22 | |
openstackgerrit | Anusha Unnam proposed openstack/nova: Remove deprecated flag in neutron.py https://review.openstack.org/370475 | 23:22 |
*** tri2sing has quit IRC | 23:22 | |
*** takedakn has joined #openstack-nova | 23:24 | |
*** ababich has quit IRC | 23:27 | |
*** ababich has joined #openstack-nova | 23:27 | |
*** kylek3h has quit IRC | 23:27 | |
*** ijw has quit IRC | 23:27 | |
*** ijw has joined #openstack-nova | 23:28 | |
*** ijw has quit IRC | 23:29 | |
*** Swami_ has joined #openstack-nova | 23:30 | |
*** Swami has quit IRC | 23:32 | |
openstackgerrit | Vu Tran proposed openstack/nova: libvirt: incorrectly create nova/instances/<uuid>/<uuid>_disk with lvm backend https://review.openstack.org/370480 | 23:33 |
*** rfolco has quit IRC | 23:36 | |
*** ducttape_ has quit IRC | 23:36 | |
*** Apoorva_ has quit IRC | 23:36 | |
*** Apoorva has joined #openstack-nova | 23:37 | |
clarkb | dansmith: http://logs.openstack.org/42/370442/1/check/gate-tempest-dsvm-multinode-full-ubuntu-xenial/13109f3/logs/screen-n-cpu.txt.gz?level=WARNING I think I fixed one issue with nova net multinode testing then ran into that. Is that a known thing? | 23:38 |
*** annegentle has quit IRC | 23:38 | |
clarkb | dansmith: almost looks like nova computes are just not registering themselves properly? | 23:38 |
*** gjayavelu has joined #openstack-nova | 23:38 | |
dansmith | clarkb: mmm, nope | 23:39 |
clarkb | I have rechecked to see if thats a consistent thing. also oddly doesn't seem to affect the live migration job itself | 23:39 |
*** maticue_ has quit IRC | 23:40 | |
*** maticue has quit IRC | 23:40 | |
clarkb | dansmith: both computes have the same complaints too "this is another computes instance" | 23:40 |
*** ducttape_ has joined #openstack-nova | 23:41 | |
*** ducttape_ has quit IRC | 23:44 | |
*** ducttape_ has joined #openstack-nova | 23:44 | |
*** browne has quit IRC | 23:45 | |
*** mriedem has joined #openstack-nova | 23:48 | |
*** ociuhandu has quit IRC | 23:49 | |
mriedem | o/ | 23:50 |
*** ijw has joined #openstack-nova | 23:50 | |
*** ijw has quit IRC | 23:50 | |
*** ijw has joined #openstack-nova | 23:51 | |
pumaranikar | hieulq, | 23:51 |
pumaranikar | I have a question regarding libvirt config options patches that you submitted in gerrit, I can see few options were left in that file, do you have patches for them? | 23:51 |
*** ducttape_ has quit IRC | 23:51 | |
dansmith | clarkb: both complain about the same instance/ | 23:56 |
*** hieulq has quit IRC | 23:56 | |
*** mmedvede has quit IRC | 23:57 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!