| *** tyrola has quit IRC | 00:01 | |
| *** annegentle has quit IRC | 00:01 | |
| *** lamt has quit IRC | 00:02 | |
| *** tyrola has joined #openstack-nova | 00:04 | |
| *** WillUK2010 has quit IRC | 00:05 | |
| *** baoli has quit IRC | 00:07 | |
| *** Guest70487 has quit IRC | 00:07 | |
| *** dtp has quit IRC | 00:08 | |
| *** mtanino has quit IRC | 00:10 | |
| *** asettle has joined #openstack-nova | 00:10 | |
| *** tbachman has joined #openstack-nova | 00:12 | |
| *** tjones has quit IRC | 00:14 | |
| *** asettle has quit IRC | 00:14 | |
| *** avolkov has quit IRC | 00:16 | |
| *** avolkov has joined #openstack-nova | 00:17 | |
| *** ducttape_ has joined #openstack-nova | 00:17 | |
| *** tjones has joined #openstack-nova | 00:17 | |
| openstackgerrit | Dan Smith proposed openstack/nova: WIP Move instance creation to conductor https://review.openstack.org/319379 | 00:18 |
|---|---|---|
| openstackgerrit | Dan Smith proposed openstack/nova: Fix non-parameterized service id in hypervisors sample tests https://review.openstack.org/410471 | 00:18 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make the SingleCellSimple fixture a little more comprehensive https://review.openstack.org/410472 | 00:18 |
| openstackgerrit | Dan Smith proposed openstack/nova: WIP: Make api_samples tests use simple cell environment https://review.openstack.org/410473 | 00:18 |
| *** rcernin has quit IRC | 00:22 | |
| *** hongbin has quit IRC | 00:22 | |
| *** tjones has quit IRC | 00:28 | |
| alex_xu | mriedem: yea, __wrapper__ is in the sqlalchemy | 00:29 |
| alex_xu | bauzas: I guess we should limit valid character for resource_provider | 00:30 |
| *** harlowja has joined #openstack-nova | 00:30 | |
| *** mlavalle has quit IRC | 00:34 | |
| *** tjones has joined #openstack-nova | 00:34 | |
| alex_xu | mriedem: strange, why you have 500 for sort on joined-table | 00:35 |
| *** diogogmt has quit IRC | 00:35 | |
| *** dimtruck is now known as zz_dimtruck | 00:36 | |
| *** yingjun has joined #openstack-nova | 00:36 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Stop allowing tags as empty string https://review.openstack.org/408879 | 00:37 |
| *** tjones has quit IRC | 00:38 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Enable cold migration with target host(1/3) https://review.openstack.org/357105 | 00:39 |
| *** ljxiash has quit IRC | 00:40 | |
| *** sreeram_v has joined #openstack-nova | 00:42 | |
| alex_xu | mriedem: emm..., i see now, one of model object attribute, another one if db table name. probably I test wrong name before | 00:43 |
| alex_xu | s/if/is | 00:44 |
| *** CristinaPauna has quit IRC | 00:44 | |
| *** CristinaPauna has joined #openstack-nova | 00:45 | |
| *** sreeram_v has quit IRC | 00:47 | |
| alex_xu | strange, __wrapper__ existed before | 00:48 |
| *** devkulkarni has quit IRC | 00:48 | |
| *** devkulkarni has joined #openstack-nova | 00:48 | |
| *** ducttape_ has quit IRC | 00:53 | |
| *** zz_dimtruck is now known as dimtruck | 00:55 | |
| *** sacharya has joined #openstack-nova | 00:58 | |
| *** hoangcx has joined #openstack-nova | 01:01 | |
| *** ljxiash has joined #openstack-nova | 01:02 | |
| *** tbachman has quit IRC | 01:02 | |
| *** tbachman has joined #openstack-nova | 01:02 | |
| *** sacharya has quit IRC | 01:03 | |
| *** Zer0Byte__ has quit IRC | 01:06 | |
| *** ducnc has joined #openstack-nova | 01:08 | |
| *** mingyu has joined #openstack-nova | 01:09 | |
| *** dimtruck is now known as zz_dimtruck | 01:11 | |
| *** asettle has joined #openstack-nova | 01:13 | |
| *** asettle has quit IRC | 01:18 | |
| *** tbachman has quit IRC | 01:19 | |
| *** zz_dimtruck is now known as dimtruck | 01:20 | |
| *** tbachman has joined #openstack-nova | 01:20 | |
| *** zhangjl has joined #openstack-nova | 01:22 | |
| *** esberglu has joined #openstack-nova | 01:29 | |
| *** markvoelker has quit IRC | 01:29 | |
| *** esberglu has quit IRC | 01:30 | |
| *** paw_ has quit IRC | 01:32 | |
| *** mingyu has quit IRC | 01:34 | |
| *** diogogmt has joined #openstack-nova | 01:36 | |
| *** ljxiash has quit IRC | 01:36 | |
| *** ljxiash has joined #openstack-nova | 01:37 | |
| *** tbachman has quit IRC | 01:37 | |
| *** avolkov` has joined #openstack-nova | 01:39 | |
| *** claudiub has quit IRC | 01:39 | |
| *** avolkov has quit IRC | 01:41 | |
| *** ljxiash has quit IRC | 01:41 | |
| *** sreeram_v has joined #openstack-nova | 01:41 | |
| *** tuanluong has joined #openstack-nova | 01:42 | |
| *** CristinaPauna has quit IRC | 01:43 | |
| *** sreeram_v has quit IRC | 01:46 | |
| *** tbachman has joined #openstack-nova | 01:46 | |
| *** yufei has joined #openstack-nova | 01:47 | |
| *** CristinaPauna has joined #openstack-nova | 01:53 | |
| *** ducttape_ has joined #openstack-nova | 01:53 | |
| *** ducttape_ has quit IRC | 01:55 | |
| *** ducttape_ has joined #openstack-nova | 01:55 | |
| openstackgerrit | John Haan proposed openstack/nova: VM with cross_az_attach raised InvaldBDM error https://review.openstack.org/409680 | 01:55 |
| *** zhs_ has joined #openstack-nova | 02:02 | |
| *** dimtruck is now known as zz_dimtruck | 02:04 | |
| *** zhs has quit IRC | 02:05 | |
| *** Jeffrey4l has quit IRC | 02:05 | |
| *** ducttape_ has quit IRC | 02:09 | |
| *** dave-mccowan has quit IRC | 02:10 | |
| *** dave-mccowan has joined #openstack-nova | 02:10 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Port network.test_manager.LdapDNSTestCase to Python 3 https://review.openstack.org/406425 | 02:12 |
| *** int32bit has joined #openstack-nova | 02:13 | |
| *** Jeffrey4l has joined #openstack-nova | 02:15 | |
| *** zz_dimtruck is now known as dimtruck | 02:15 | |
| *** yuntongjin has joined #openstack-nova | 02:17 | |
| openstackgerrit | pangliye proposed openstack/nova: Add Apache 2.0 license to source file https://review.openstack.org/409997 | 02:19 |
| *** hfu has joined #openstack-nova | 02:19 | |
| openstackgerrit | pangliye proposed openstack/nova: Add Apache 2.0 license to source file https://review.openstack.org/409996 | 02:19 |
| *** ducttape_ has joined #openstack-nova | 02:21 | |
| openstackgerrit | int32bit proposed openstack/nova: Fix error if free_disk_gb is None in CellStateManager https://review.openstack.org/409501 | 02:22 |
| *** yuntongjin has quit IRC | 02:22 | |
| *** fragatin_ has joined #openstack-nova | 02:23 | |
| *** yuntongjin has joined #openstack-nova | 02:23 | |
| alex_xu | ok...I tested in my new code | 02:23 |
| *** hfu has quit IRC | 02:25 | |
| *** fragatina has quit IRC | 02:27 | |
| *** hongbin has joined #openstack-nova | 02:27 | |
| *** yuntongjin has quit IRC | 02:28 | |
| *** fragatin_ has quit IRC | 02:28 | |
| *** yuntongjin has joined #openstack-nova | 02:28 | |
| *** kevinz has joined #openstack-nova | 02:31 | |
| int32bit | Hi, all. As we know, the "index()" method usally return object list and the "show()" method return an object for detail in api's controller. But I found in our InstanceUsageAuditLogController: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/instance_usage_audit_log.py#L43, the 'index' method return an object, I'm really not sure that index() is buggy, and I wonder what it is intended to do and what might be | 02:32 |
| int32bit | needed to fix it. | 02:32 |
| *** baoli has joined #openstack-nova | 02:32 | |
| *** kaisers__ has joined #openstack-nova | 02:36 | |
| *** mingyu has joined #openstack-nova | 02:38 | |
| *** raunak has joined #openstack-nova | 02:38 | |
| *** hfu has joined #openstack-nova | 02:40 | |
| *** kaisers_ has quit IRC | 02:40 | |
| *** thorst_ has joined #openstack-nova | 02:40 | |
| *** hieulq_ has joined #openstack-nova | 02:45 | |
| *** Swami_ has joined #openstack-nova | 02:46 | |
| *** hieulq_ has quit IRC | 02:46 | |
| *** thorst_ has quit IRC | 02:47 | |
| *** thorst_ has joined #openstack-nova | 02:47 | |
| *** Swami has quit IRC | 02:48 | |
| *** Swami__ has quit IRC | 02:48 | |
| *** mingyu has quit IRC | 02:48 | |
| *** Swami has joined #openstack-nova | 02:49 | |
| *** Kevin_Zheng has joined #openstack-nova | 02:49 | |
| *** tbachman has quit IRC | 02:50 | |
| *** alexpilo_ has quit IRC | 02:52 | |
| *** thorst_ has quit IRC | 02:52 | |
| *** liusheng has joined #openstack-nova | 02:55 | |
| *** ducttape_ has quit IRC | 02:56 | |
| *** mingyu has joined #openstack-nova | 02:59 | |
| *** sacharya has joined #openstack-nova | 02:59 | |
| *** sacharya has quit IRC | 03:04 | |
| openstackgerrit | Bin Zhou proposed openstack/nova: Avoid misremove of volume when live migrate fail https://review.openstack.org/387226 | 03:08 |
| *** tyrola has quit IRC | 03:18 | |
| *** nic has quit IRC | 03:18 | |
| *** gongysh has joined #openstack-nova | 03:19 | |
| *** EmilienM has quit IRC | 03:19 | |
| *** tyrola has joined #openstack-nova | 03:19 | |
| *** siva_krish has joined #openstack-nova | 03:19 | |
| *** EmilienM has joined #openstack-nova | 03:20 | |
| openstackgerrit | Huan Xie proposed openstack/nova: XenAPI Use os-xenapi lib for nova https://review.openstack.org/406059 | 03:22 |
| *** yamahata_ has quit IRC | 03:24 | |
| *** sacharya has joined #openstack-nova | 03:25 | |
| *** markvoelker has joined #openstack-nova | 03:30 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Simple tenant usage pagination https://review.openstack.org/386093 | 03:33 |
| *** markvoelker has quit IRC | 03:35 | |
| *** yuntongjin has quit IRC | 03:35 | |
| *** pbandark has joined #openstack-nova | 03:36 | |
| *** yingjun has quit IRC | 03:37 | |
| openstackgerrit | Diana Clarke proposed openstack/nova: Do not attempt volume swap when guest is stopped https://review.openstack.org/389798 | 03:39 |
| *** dave-mccowan has quit IRC | 03:39 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Enable cold migration with target host(2/3) https://review.openstack.org/408955 | 03:42 |
| openstackgerrit | Huan Xie proposed openstack/nova: XenAPI Use os-xenapi lib for nova https://review.openstack.org/406059 | 03:44 |
| *** pbandark is now known as pbandark`brb | 03:46 | |
| *** unicell has quit IRC | 03:46 | |
| *** prateek has joined #openstack-nova | 03:49 | |
| *** yingjun has joined #openstack-nova | 03:50 | |
| *** dikonoo has joined #openstack-nova | 03:51 | |
| *** dikonoor has joined #openstack-nova | 03:51 | |
| *** thorst_ has joined #openstack-nova | 03:54 | |
| openstackgerrit | Diana Clarke proposed openstack/python-novaclient: Microversion 2.40 - Simple tenant usage pagination https://review.openstack.org/407825 | 03:54 |
| *** Apoorva has joined #openstack-nova | 03:55 | |
| *** Cristina_ has joined #openstack-nova | 03:56 | |
| *** ducttape_ has joined #openstack-nova | 03:57 | |
| *** baoli has quit IRC | 03:57 | |
| *** Swami_ has quit IRC | 03:57 | |
| *** Swami_ has joined #openstack-nova | 03:57 | |
| *** thorst_ has quit IRC | 03:58 | |
| *** thorst_ has joined #openstack-nova | 03:59 | |
| *** ijw has quit IRC | 03:59 | |
| *** takashin has joined #openstack-nova | 03:59 | |
| *** CristinaPauna has quit IRC | 04:00 | |
| *** ducttape_ has quit IRC | 04:01 | |
| *** hoangcx has quit IRC | 04:04 | |
| *** hongbin has quit IRC | 04:05 | |
| *** nicolasbock has quit IRC | 04:05 | |
| *** Apoorva has quit IRC | 04:06 | |
| *** thorst_ has quit IRC | 04:07 | |
| *** hieulq has quit IRC | 04:07 | |
| *** hieulq has joined #openstack-nova | 04:08 | |
| *** yingjun has quit IRC | 04:12 | |
| *** yuntongjin has joined #openstack-nova | 04:13 | |
| *** sudipto has joined #openstack-nova | 04:18 | |
| *** psachin has joined #openstack-nova | 04:18 | |
| *** sudipto_ has joined #openstack-nova | 04:18 | |
| *** unicell has joined #openstack-nova | 04:19 | |
| *** mingyu has quit IRC | 04:20 | |
| *** sreeram_v has joined #openstack-nova | 04:21 | |
| *** mingyu has joined #openstack-nova | 04:21 | |
| *** bks has joined #openstack-nova | 04:22 | |
| *** ducnc has quit IRC | 04:22 | |
| *** ducnc has joined #openstack-nova | 04:24 | |
| *** mingyu has quit IRC | 04:26 | |
| *** unicell1 has joined #openstack-nova | 04:27 | |
| *** yingjun has joined #openstack-nova | 04:29 | |
| *** unicell has quit IRC | 04:29 | |
| *** prateek has quit IRC | 04:31 | |
| *** ducttape_ has joined #openstack-nova | 04:33 | |
| *** pbandark`brb is now known as pbandark | 04:36 | |
| *** thorst_ has joined #openstack-nova | 04:38 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: WIP: placement: RT now adds proper Ironic inventory https://review.openstack.org/404472 | 04:41 |
| *** yingjun has quit IRC | 04:43 | |
| *** ducttape_ has quit IRC | 04:44 | |
| *** ijw has joined #openstack-nova | 04:46 | |
| *** kaisers__ has quit IRC | 04:50 | |
| *** int32bit has quit IRC | 04:50 | |
| *** sudipto_ has quit IRC | 04:56 | |
| *** sudipto has quit IRC | 04:56 | |
| *** links has joined #openstack-nova | 04:57 | |
| *** hieulq_ has joined #openstack-nova | 05:00 | |
| *** thorst_ has quit IRC | 05:04 | |
| *** Swami_ has quit IRC | 05:06 | |
| *** Swami has quit IRC | 05:06 | |
| *** ducnc has quit IRC | 05:08 | |
| *** prateek has joined #openstack-nova | 05:09 | |
| *** prateek has quit IRC | 05:09 | |
| openstackgerrit | Wang Qing wu proposed openstack/nova: WIP(1): Introduce base PowerVM driver https://review.openstack.org/391288 | 05:10 |
| *** hieulq_ has quit IRC | 05:11 | |
| *** pradk has quit IRC | 05:12 | |
| *** ratailor has joined #openstack-nova | 05:13 | |
| *** mdnadeem has joined #openstack-nova | 05:13 | |
| *** prateek has joined #openstack-nova | 05:14 | |
| *** asettle has joined #openstack-nova | 05:14 | |
| *** lpetrut has joined #openstack-nova | 05:21 | |
| *** asettle has quit IRC | 05:21 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Enable cold migration with target host(3/3) https://review.openstack.org/408964 | 05:24 |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Add parameters in cold migrate action https://review.openstack.org/410042 | 05:25 |
| *** vi01et has joined #openstack-nova | 05:25 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Refactor a test method including 7 test cases https://review.openstack.org/395436 | 05:26 |
| *** tbachman has joined #openstack-nova | 05:26 | |
| *** yamahata_ has joined #openstack-nova | 05:28 | |
| *** markvoelker has joined #openstack-nova | 05:31 | |
| *** yingjun has joined #openstack-nova | 05:32 | |
| *** thorst_ has joined #openstack-nova | 05:34 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Add paging support in server groups list https://review.openstack.org/409282 | 05:35 |
| *** gcb has quit IRC | 05:35 | |
| *** markvoelker has quit IRC | 05:37 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Add Python 3.5 functional tests in tox.ini https://review.openstack.org/408561 | 05:39 |
| *** sudipto has joined #openstack-nova | 05:39 | |
| *** sudipto_ has joined #openstack-nova | 05:39 | |
| *** thorst_ has quit IRC | 05:41 | |
| *** ducttape_ has joined #openstack-nova | 05:44 | |
| *** ducttape_ has quit IRC | 05:49 | |
| *** kaisers_ has joined #openstack-nova | 05:50 | |
| *** jaosorior has joined #openstack-nova | 05:50 | |
| yingjun | ping gibi | 05:50 |
| *** ekuris has joined #openstack-nova | 05:51 | |
| *** gcb has joined #openstack-nova | 05:52 | |
| *** tbachman has quit IRC | 05:52 | |
| *** diogogmt has quit IRC | 05:58 | |
| *** diogogmt has joined #openstack-nova | 05:58 | |
| *** fragatina has joined #openstack-nova | 06:03 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Enble more tests on Python 3 https://review.openstack.org/410542 | 06:03 |
| *** gcb has quit IRC | 06:05 | |
| *** lpetrut has quit IRC | 06:05 | |
| *** fragatina has quit IRC | 06:08 | |
| *** amotoki has joined #openstack-nova | 06:12 | |
| *** takashin has left #openstack-nova | 06:13 | |
| *** kaisers_ has quit IRC | 06:13 | |
| *** amotoki has quit IRC | 06:14 | |
| *** tuanluong has quit IRC | 06:16 | |
| *** hieulq has quit IRC | 06:16 | |
| *** Fdaisuke has quit IRC | 06:17 | |
| *** amotoki has joined #openstack-nova | 06:17 | |
| *** mingyu has joined #openstack-nova | 06:18 | |
| *** Fdaisuke has joined #openstack-nova | 06:18 | |
| *** unicell1 has quit IRC | 06:19 | |
| *** hieulq has joined #openstack-nova | 06:20 | |
| *** tuanluong has joined #openstack-nova | 06:21 | |
| *** gcb has joined #openstack-nova | 06:22 | |
| *** unicell1 has joined #openstack-nova | 06:22 | |
| *** sridharg has joined #openstack-nova | 06:25 | |
| *** thorst_ has joined #openstack-nova | 06:28 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Support tag instances when boot https://review.openstack.org/394321 | 06:32 |
| *** thorst_ has quit IRC | 06:36 | |
| *** afazekas has quit IRC | 06:40 | |
| *** afazekas has joined #openstack-nova | 06:42 | |
| openstackgerrit | Wang Qing wu proposed openstack/nova: WIP(1): Introduce base PowerVM driver https://review.openstack.org/391288 | 06:43 |
| *** kumarb has joined #openstack-nova | 06:44 | |
| *** yingjun has quit IRC | 06:47 | |
| *** bks has quit IRC | 06:47 | |
| openstackgerrit | Huan Xie proposed openstack/nova: XenAPI Use os-xenapi lib for nova https://review.openstack.org/406059 | 06:47 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Fix urllib.urlencode issue in functional tests on Python 3 https://review.openstack.org/410554 | 06:50 |
| *** pooja_j has quit IRC | 06:50 | |
| *** yingjun has joined #openstack-nova | 06:54 | |
| *** Oku_OS-away is now known as Oku_OS | 06:57 | |
| *** moshele has joined #openstack-nova | 06:59 | |
| *** markvoelker has joined #openstack-nova | 07:00 | |
| *** moshele has quit IRC | 07:01 | |
| *** ijw has quit IRC | 07:02 | |
| *** nandal has joined #openstack-nova | 07:02 | |
| *** mjura has joined #openstack-nova | 07:02 | |
| *** nandal has left #openstack-nova | 07:02 | |
| *** nandal has joined #openstack-nova | 07:02 | |
| *** nandal has left #openstack-nova | 07:02 | |
| *** nandal has joined #openstack-nova | 07:03 | |
| *** moshele has joined #openstack-nova | 07:03 | |
| *** ducttape_ has joined #openstack-nova | 07:04 | |
| *** markvoelker has quit IRC | 07:06 | |
| *** maestropandy has joined #openstack-nova | 07:07 | |
| *** mjura has quit IRC | 07:08 | |
| *** mjura has joined #openstack-nova | 07:09 | |
| *** phuongnh has joined #openstack-nova | 07:09 | |
| *** kaisers_ has joined #openstack-nova | 07:09 | |
| *** nandal has quit IRC | 07:09 | |
| *** siva_krish has quit IRC | 07:11 | |
| *** maestropandy has left #openstack-nova | 07:14 | |
| *** siva_krish has joined #openstack-nova | 07:15 | |
| *** markus_z has joined #openstack-nova | 07:15 | |
| *** vi01et has quit IRC | 07:15 | |
| *** nherciu has joined #openstack-nova | 07:16 | |
| *** bks has joined #openstack-nova | 07:17 | |
| *** asettle has joined #openstack-nova | 07:18 | |
| *** bapalm has quit IRC | 07:18 | |
| openstackgerrit | pangliye proposed openstack/nova: Add Apache 2.0 license to source file https://review.openstack.org/409996 | 07:19 |
| openstackgerrit | pangliye proposed openstack/nova: Add Apache 2.0 license to source file https://review.openstack.org/409997 | 07:19 |
| *** kumarb has quit IRC | 07:19 | |
| openstackgerrit | pangliye proposed openstack/nova: [H235]fix assert to assertTrue https://review.openstack.org/410093 | 07:21 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Added missed "raises:" docstrings into numa_get_constraints() method https://review.openstack.org/410565 | 07:23 |
| *** thorst_ has joined #openstack-nova | 07:23 | |
| *** kaisers_ has quit IRC | 07:24 | |
| *** ducttape_ has quit IRC | 07:24 | |
| *** andreas_s has joined #openstack-nova | 07:25 | |
| *** asettle has quit IRC | 07:25 | |
| *** sacharya has quit IRC | 07:26 | |
| *** thorst_ has quit IRC | 07:31 | |
| *** yamahata_ has quit IRC | 07:37 | |
| *** mingyu has quit IRC | 07:37 | |
| *** bapalm has joined #openstack-nova | 07:38 | |
| *** zenoway has joined #openstack-nova | 07:41 | |
| *** gcb has quit IRC | 07:44 | |
| *** mingyu has joined #openstack-nova | 07:45 | |
| *** gcb has joined #openstack-nova | 07:45 | |
| *** Cristina_ is now known as CristinaPauna | 07:45 | |
| *** nherciu_ has joined #openstack-nova | 07:46 | |
| *** gongysh has quit IRC | 07:47 | |
| *** claudiub has joined #openstack-nova | 07:48 | |
| *** zenoway has quit IRC | 07:49 | |
| *** nherciu has quit IRC | 07:50 | |
| *** nandal has joined #openstack-nova | 07:50 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Fix misleading port delete description https://review.openstack.org/410582 | 07:50 |
| *** ralonsoh has joined #openstack-nova | 07:51 | |
| *** mingyu_ has joined #openstack-nova | 07:52 | |
| *** claudiub has quit IRC | 07:53 | |
| *** mingyu has quit IRC | 07:54 | |
| *** zhangjl1 has joined #openstack-nova | 07:55 | |
| *** mingyu_ has quit IRC | 07:55 | |
| *** rcernin has joined #openstack-nova | 07:56 | |
| *** mingyu has joined #openstack-nova | 07:56 | |
| *** rcernin has quit IRC | 07:56 | |
| *** zhangjl has quit IRC | 07:57 | |
| *** rcernin has joined #openstack-nova | 07:57 | |
| *** yamahata_ has joined #openstack-nova | 07:59 | |
| *** pcaruana has joined #openstack-nova | 07:59 | |
| *** sacharya has joined #openstack-nova | 08:00 | |
| *** raunak has quit IRC | 08:01 | |
| *** rmart04 has joined #openstack-nova | 08:03 | |
| bauzas | p!spy | 08:04 |
| gmann | alex_xu: can you have a look into this - https://review.openstack.org/#/c/408879/ | 08:05 |
| *** ociuhandu has quit IRC | 08:05 | |
| *** sacharya has quit IRC | 08:05 | |
| alex_xu | gmann: I'm thinking of that in the morning, the key point is whehter we should fix it without microversion. I need a time to think about it | 08:06 |
| gmann | alex_xu: ahh, sure. thats is valid point. I also thought many times then end up with bug fix but need feedback. | 08:07 |
| *** CristinaPauna has quit IRC | 08:07 | |
| *** ducnc has joined #openstack-nova | 08:07 | |
| bauzas | alex_xu: FWIW, just saw your highlight from yesterday early morning my time about Blazar | 08:08 |
| bauzas | alex_xu: FWIW, there is a new team resurrecting Blazar since last month | 08:08 |
| gmann | because this makes 200 to 400 but not sure if that can break anyone as empty tag does not make sense but we never knows | 08:09 |
| bauzas | can't find gongusyh on irc | 08:09 |
| *** ducnc1 has joined #openstack-nova | 08:10 | |
| *** gcb has quit IRC | 08:10 | |
| *** gcb has joined #openstack-nova | 08:11 | |
| *** zenoway has joined #openstack-nova | 08:11 | |
| *** ekuris has quit IRC | 08:12 | |
| *** ducnc has quit IRC | 08:12 | |
| *** ducnc1 is now known as ducnc | 08:12 | |
| *** jaosorior has quit IRC | 08:16 | |
| *** jaosorior has joined #openstack-nova | 08:16 | |
| *** ihrachys has joined #openstack-nova | 08:17 | |
| alex_xu | bauzas: I can ping him with another way | 08:18 |
| *** thorst_ has joined #openstack-nova | 08:19 | |
| alex_xu | johnthetubaguy: mriedem jaypipes looking for you can take a look at the comment at line 97 of https://review.openstack.org/#/c/393205/16/specs/ocata/approved/add-whitelist-for-server-list-filter-sort-parameters.rst, I need some help on the rule how we add db index. thanks in advance | 08:19 |
| *** kaisers_ has joined #openstack-nova | 08:20 | |
| *** ftersin_ has joined #openstack-nova | 08:20 | |
| *** ababich has quit IRC | 08:23 | |
| *** ababich has joined #openstack-nova | 08:23 | |
| ftersin_ | jbernard, mriedem: right, test_create_ebs_image_and_check_boot does not wait for 'active' state of the instance, and this looks bad for me (and for andrey-mp, the test author) | 08:24 |
| *** ducttape_ has joined #openstack-nova | 08:25 | |
| ftersin_ | i'm trying to change this on https://review.openstack.org/#/c/410338 | 08:26 |
| *** matrohon has joined #openstack-nova | 08:27 | |
| ftersin_ | if it doesn't help, we'll can use the workaround (commented there at the moment) | 08:27 |
| *** ekuris has joined #openstack-nova | 08:27 | |
| *** thorst_ has quit IRC | 08:27 | |
| ftersin_ | as for the test logic - it deletes the snapshot during cleanup process, since the snapshot is a part of the image, created in this test | 08:29 |
| *** ducttape_ has quit IRC | 08:30 | |
| *** roeyc has joined #openstack-nova | 08:31 | |
| *** yuntongjin has quit IRC | 08:33 | |
| ftersin_ | and probably the snapshot is busy because the volume based on this snapshot is still being created (i'm not sure if rbd driver flattens volumes created from snapshot, but if it does, this obviously takes time and makes the snapshot 'busy') | 08:33 |
| *** dinobot_ has joined #openstack-nova | 08:34 | |
| gibi | yingjun: hi! | 08:36 |
| yingjun | gibi: hi | 08:36 |
| yingjun | gibi, about this patch: https://review.openstack.org/#/c/398171/ | 08:37 |
| yingjun | gibi, i did some tests | 08:37 |
| yingjun | gibi, i didn't get the behaviour you mentioned in patch | 08:38 |
| yingjun | gibi, could you take another look at that? | 08:38 |
| gibi | yingjun: looking... | 08:38 |
| *** Zer0Byte__ has joined #openstack-nova | 08:39 | |
| gibi | yingjun: I commented quite late yesterday (in my timezone) so it is totally possible that I overlooked something. I will look deper now | 08:40 |
| yingjun | gibi, sure | 08:40 |
| *** ekuris has quit IRC | 08:40 | |
| *** daemontool has joined #openstack-nova | 08:44 | |
| *** kaisers_ has quit IRC | 08:44 | |
| *** moshele has quit IRC | 08:45 | |
| *** abalutoiu has joined #openstack-nova | 08:45 | |
| *** dimtruck is now known as zz_dimtruck | 08:46 | |
| ftersin_ | jbernard, mriedem: so if even the waiting for 'active' state helps, the minor problem exists in Ceph based Cinder, since it does not reject deletion of really busy snapshots right on api layer | 08:47 |
| *** moshele has joined #openstack-nova | 08:47 | |
| *** openstackgerrit has quit IRC | 08:48 | |
| *** mingyu has quit IRC | 08:49 | |
| *** mingyu has joined #openstack-nova | 08:50 | |
| *** zhangjl1 has left #openstack-nova | 08:51 | |
| *** dikonoor has quit IRC | 08:54 | |
| *** dikonoo has quit IRC | 08:54 | |
| *** dikonoor has joined #openstack-nova | 08:54 | |
| *** daemontool_ has joined #openstack-nova | 08:55 | |
| *** roeyc has quit IRC | 08:55 | |
| *** xinliang has quit IRC | 08:56 | |
| *** ekuris has joined #openstack-nova | 08:56 | |
| *** amoralej|off is now known as amoralej | 08:56 | |
| *** daemontool has quit IRC | 08:58 | |
| *** zzzeek has quit IRC | 09:00 | |
| *** zzzeek has joined #openstack-nova | 09:00 | |
| *** markvoelker has joined #openstack-nova | 09:02 | |
| *** tuanluong has quit IRC | 09:02 | |
| *** tuanluong has joined #openstack-nova | 09:03 | |
| *** mingyu has quit IRC | 09:04 | |
| *** mingyu has joined #openstack-nova | 09:04 | |
| *** markvoelker has quit IRC | 09:07 | |
| *** openstackgerrit has joined #openstack-nova | 09:07 | |
| openstackgerrit | Pavel Kholkin proposed openstack/python-novaclient: [proxy-api] microversion 2.39 deprecates image-metadata proxy API https://review.openstack.org/408166 | 09:07 |
| *** xinliang has joined #openstack-nova | 09:08 | |
| *** kaisers_ has joined #openstack-nova | 09:11 | |
| ftersin_ | jbernard, mriedem: fyi, i just removed workaround from the review, you may see it on the PS3 https://review.openstack.org/#/c/410338/3 | 09:12 |
| *** yassine has joined #openstack-nova | 09:13 | |
| *** yassine is now known as Guest89456 | 09:14 | |
| *** thorst_ has joined #openstack-nova | 09:18 | |
| *** gcb has quit IRC | 09:21 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Don't apply multi-queue to SRIOV ports https://review.openstack.org/397545 | 09:22 |
| openstackgerrit | Huan Xie proposed openstack/nova: XenAPI Use os-xenapi lib for nova https://review.openstack.org/406059 | 09:23 |
| *** thorst_ has quit IRC | 09:24 | |
| *** CristinaPauna has joined #openstack-nova | 09:25 | |
| *** mnasiadka has joined #openstack-nova | 09:26 | |
| lyarwood | mdbooth: https://review.openstack.org/#/c/409708/ - are you happy with this landing in stable/newton? mriedem has a concern that this only came from code inspection. | 09:27 |
| *** lucas-afk is now known as lucasagomes | 09:27 | |
| *** kevinz has quit IRC | 09:29 | |
| *** roeyc has joined #openstack-nova | 09:29 | |
| *** derekh has joined #openstack-nova | 09:32 | |
| *** int32bit has joined #openstack-nova | 09:32 | |
| *** tuanluong_ has joined #openstack-nova | 09:34 | |
| *** mingyu has quit IRC | 09:34 | |
| *** tuanluong has quit IRC | 09:36 | |
| *** mingyu has joined #openstack-nova | 09:36 | |
| *** yamahata_ has quit IRC | 09:37 | |
| *** gcb has joined #openstack-nova | 09:39 | |
| openstackgerrit | int32bit proposed openstack/nova: Fix error if free_disk_gb is None in CellStateManager https://review.openstack.org/409501 | 09:40 |
| *** kevinz has joined #openstack-nova | 09:41 | |
| *** Zer0Byte__ has quit IRC | 09:41 | |
| *** asettle has joined #openstack-nova | 09:44 | |
| *** kaisers_ has quit IRC | 09:45 | |
| *** yingjun has quit IRC | 09:46 | |
| *** karthiks has quit IRC | 09:48 | |
| mdbooth | lyarwood: I'm relatively comfortable with it. Do you want to reproduce first? | 09:52 |
| mdbooth | lyarwood: That said, this code is notoriously easy to miss edge cases in. | 09:53 |
| mdbooth | If nobody's actually complaining about broken LVM, it might be more 'stable' to leave it for a bit. | 09:53 |
| *** zhangyujun has joined #openstack-nova | 09:53 | |
| lyarwood | mdbooth: yeah, happy to just leave it until reported tbh | 09:53 |
| bauzas | alex_xu: thanks, the new people working on Blazar are now meeting every Tuesdays at 0900UTC | 09:54 |
| mdbooth | lyarwood: I'll comment on the backport to that effect. | 09:54 |
| bauzas | alex_xu: priteau is the main contributor | 09:54 |
| mdbooth | I think it's probably ok. | 09:54 |
| bauzas | alex_xu: and they discuss on #openstack-blazar | 09:54 |
| *** claudiub has joined #openstack-nova | 09:55 | |
| *** ducttape_ has joined #openstack-nova | 09:56 | |
| mdbooth | lyarwood: I gave it a +0 | 09:56 |
| alex_xu | bauzas: thanks, looks like gongsy not online, I can forward to him by other im tools. | 09:56 |
| bauzas | np | 09:57 |
| bauzas | alex_xu: the new contributors are mainly from NTT AFAIK | 09:57 |
| alex_xu | bauzas: got it | 09:57 |
| lyarwood | mdbooth: cool thanks | 09:57 |
| bauzas | some are coming from the OPNFV project, some are already using Blazar in their own production cloud | 09:57 |
| alex_xu | a | 09:57 |
| alex_xu | oops, my daughter press the keyboard | 09:58 |
| *** lpetrut has joined #openstack-nova | 09:59 | |
| alex_xu | bauzas: gongsy also ask whether we can implement reservation in nova | 09:59 |
| bauzas | I don't think so | 09:59 |
| bauzas | alex_xu: heh for your daughter :p | 09:59 |
| *** ducttape_ has quit IRC | 10:00 | |
| alex_xu | bauzas: :) | 10:00 |
| *** hfu has quit IRC | 10:02 | |
| *** mingyu has quit IRC | 10:02 | |
| *** kumarb has joined #openstack-nova | 10:02 | |
| *** gongysh has joined #openstack-nova | 10:04 | |
| johnthetubaguy | bauzas: longer term, I could see users passing in some placement_token that helps use some semi-reserved location | 10:05 |
| *** siva_krish has quit IRC | 10:05 | |
| johnthetubaguy | bauzas: still always comes back to spot instances for me, you need someone who can use the reserved space while the person who reserved it doesn't make use of it | 10:06 |
| *** bks has quit IRC | 10:06 | |
| bauzas | johnthetubaguy: long time ago, when we discussed in ATL about spot instances, we agreed on not having this for Nova but rather using Blazae | 10:06 |
| bauzas | Blazar | 10:06 |
| bauzas | johnthetubaguy: but it was before the Big Tent approach | 10:06 |
| bauzas | and Blazar never went incubated | 10:07 |
| bauzas | when the Big Tent became a thing, all of the existing Blazar contributors moved to other opportunities and the project became defunct | 10:07 |
| bauzas | now, the thing is that people are interested in resurrecting Blazar and hopefully have a stable release by Ocata, so they could ask for the Big Tent in Pike | 10:08 |
| johnthetubaguy | bauzas: I remember very different conversations when we defined the Nova scope, we left spot instances in scope, long term | 10:09 |
| johnthetubaguy | bauzas: but the dependency is on getting placement sorted, and split out of Nova | 10:09 |
| bauzas | by that time, a scheduler API wasn't a thing | 10:09 |
| bauzas | now, it's done and we need to think again on that, agreed | 10:09 |
| bauzas | but probably we could then discuss with the interested parties | 10:10 |
| johnthetubaguy | bauzas: did you see the phase we added in here: http://docs.openstack.org/developer/nova/project_scope.html#iaas-not-batch-processing | 10:10 |
| *** siva_krish has joined #openstack-nova | 10:10 | |
| johnthetubaguy | phrase | 10:10 |
| johnthetubaguy | I think I am still happy with that scope statement | 10:11 |
| bauzas | johnthetubaguy: well, not really, I just thought we considered those new API concepts as separate | 10:11 |
| *** brad[] has quit IRC | 10:11 | |
| *** paw_ has joined #openstack-nova | 10:12 | |
| *** paw has quit IRC | 10:12 | |
| bauzas | johnthetubaguy: anyway, we're not stuck with the old notion of programs | 10:13 |
| *** hshiina has joined #openstack-nova | 10:13 | |
| johnthetubaguy | I don't care about the organisation, its the architecture and code I believe that project scope is taking about | 10:13 |
| bauzas | johnthetubaguy: I mean, whatever the service type is, I still think that nova shouldn't handle the concept of leases | 10:14 |
| johnthetubaguy | well, I do care, but its something more fluid, I guess I should say | 10:14 |
| bauzas | johnthetubaguy: that said, now that we have a placement service, that's an open question if it should contract on that notion too | 10:14 |
| bauzas | johnthetubaguy: but that's long-term AFAICS | 10:15 |
| johnthetubaguy | once we have the key existing Nova use cases working, I think that needs a look, but yeah, thats at least a year out | 10:15 |
| bauzas | johnthetubaguy: while the Blazar project aims to deliver a very immediate way of managing those reservations | 10:15 |
| johnthetubaguy | the case I want to see us deal with is fair share of resources between departments in a private cloud | 10:15 |
| bauzas | johnthetubaguy: honestly, I'm pragmatic | 10:16 |
| johnthetubaguy | in my head, that means a "reserved" quota, and a spot instance quota so you can burst and use other folks capacity when they don't need it, but you do | 10:17 |
| bauzas | johnthetubaguy: if by one day, the placement API could envisage that, we could see how to ask Blazar to merge with that | 10:17 |
| johnthetubaguy | now I want things outside Nova to mange that (yeah, probably some placement and blazar combo), but thats a Nova API change | 10:17 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Adds support for versioned schema validation for query parameters https://review.openstack.org/389003 | 10:17 |
| bauzas | johnthetubaguy: agreed | 10:18 |
| bauzas | johnthetubaguy: don't know if you know the blazar workflow, but people pass a reservation uuid when booting | 10:18 |
| bauzas | that's a nova boot command | 10:18 |
| johnthetubaguy | yeah, that seems broken to me | 10:18 |
| *** paw_ has quit IRC | 10:18 | |
| johnthetubaguy | from a user point of view | 10:19 |
| johnthetubaguy | but anyways, it needs clear user cases, etc to clear that up | 10:19 |
| johnthetubaguy | my preferred one to worry about is above | 10:19 |
| johnthetubaguy | basically private cloud quota is very different to public cloud quota (which is what we currently have) | 10:20 |
| bauzas | johnthetubaguy: I agree | 10:20 |
| bauzas | johnthetubaguy: honestly, you have people wanting to work on that, it's probably worth getting in touch with them | 10:20 |
| bauzas | johnthetubaguy: because it seems they have resources for that | 10:21 |
| *** thorst_ has joined #openstack-nova | 10:21 | |
| johnthetubaguy | I would review the Nova spec discussing how that integrates | 10:21 |
| bauzas | if we say that Blazar isn't probably the best fit, then meh, okay, just wanting to make sure that we don't leave them work for something we disagree | 10:21 |
| johnthetubaguy | yeah, its a good call out | 10:21 |
| johnthetubaguy | I am not sure thats the correct approach | 10:21 |
| openstackgerrit | Béla Vancsics proposed openstack/nova: Reduce (notification) test duplication https://review.openstack.org/391428 | 10:22 |
| bauzas | johnthetubaguy: I think I see your point about the user API | 10:22 |
| bauzas | johnthetubaguy: the thing I'm still considering is that I don't think we should have Nova managing the admin-only APIs for setuping those rules | 10:23 |
| openstackgerrit | zhaolihui proposed openstack/nova: Fix keypair quota error https://review.openstack.org/403529 | 10:23 |
| *** tuanluong_ has quit IRC | 10:24 | |
| *** mvk has quit IRC | 10:24 | |
| *** bks has joined #openstack-nova | 10:28 | |
| *** thorst_ has quit IRC | 10:29 | |
| *** kumarb has quit IRC | 10:30 | |
| *** links has quit IRC | 10:38 | |
| *** links has joined #openstack-nova | 10:38 | |
| *** gszasz has joined #openstack-nova | 10:41 | |
| *** kaisers_ has joined #openstack-nova | 10:42 | |
| *** sreeram_v has quit IRC | 10:43 | |
| *** gongysh has quit IRC | 10:43 | |
| *** sreeram_v has joined #openstack-nova | 10:43 | |
| *** sreeram__ has joined #openstack-nova | 10:44 | |
| *** kevinz has quit IRC | 10:45 | |
| *** sreeram_v has quit IRC | 10:48 | |
| zhangyujun | Anyone has information about why server group doesn't provide a update method for change its name? | 10:48 |
| *** sreeram__ has quit IRC | 10:49 | |
| openstackgerrit | Pavel Kholkin proposed openstack/python-novaclient: [proxy-api] microversion 2.39 deprecates image-metadata proxy API https://review.openstack.org/408166 | 10:49 |
| *** kumarb has joined #openstack-nova | 10:52 | |
| *** yufei has quit IRC | 10:54 | |
| *** matrohon has quit IRC | 10:55 | |
| *** ftersin_ has quit IRC | 10:55 | |
| *** bks has quit IRC | 10:55 | |
| *** mvk has joined #openstack-nova | 10:56 | |
| *** markvoelker has joined #openstack-nova | 11:02 | |
| kashyap | lyarwood: Do you have a pointer to Newton backport of this one - https://review.openstack.org/#/c/408668/ | 11:07 |
| *** markvoelker has quit IRC | 11:07 | |
| lyarwood | kashyap: https://review.openstack.org/#/c/409774/ (click the change-id to get the full list) | 11:07 |
| kashyap | Ah, thanks | 11:08 |
| *** sdague has joined #openstack-nova | 11:08 | |
| *** sudipto_ has quit IRC | 11:09 | |
| *** sudipto has quit IRC | 11:09 | |
| *** tbachman has joined #openstack-nova | 11:09 | |
| *** CristinaPauna has quit IRC | 11:13 | |
| *** ekuris has quit IRC | 11:14 | |
| *** thorst_ has joined #openstack-nova | 11:16 | |
| *** jaosorior has quit IRC | 11:19 | |
| *** prateek_ has joined #openstack-nova | 11:21 | |
| *** prateek has quit IRC | 11:21 | |
| *** kaisers_ has quit IRC | 11:22 | |
| *** CristinaPauna has joined #openstack-nova | 11:22 | |
| *** thorst_ has quit IRC | 11:23 | |
| *** gongysh has joined #openstack-nova | 11:24 | |
| *** brad[] has joined #openstack-nova | 11:24 | |
| *** ducttape_ has joined #openstack-nova | 11:26 | |
| *** gmann_ has joined #openstack-nova | 11:27 | |
| gmann | sfinucan: can you check this now - https://review.openstack.org/#/c/406406/ | 11:28 |
| gmann | patch you mentioned merged now | 11:28 |
| *** ratailor has quit IRC | 11:28 | |
| *** mkoderer has joined #openstack-nova | 11:28 | |
| *** ekuris has joined #openstack-nova | 11:28 | |
| *** zhangyujun has quit IRC | 11:29 | |
| sfinucan | gmann: Done | 11:29 |
| gmann | sfinucan: thanks. | 11:29 |
| *** tbachman has quit IRC | 11:30 | |
| *** jaosorior has joined #openstack-nova | 11:30 | |
| *** ducttape_ has quit IRC | 11:31 | |
| *** nicolasbock has joined #openstack-nova | 11:32 | |
| *** kumarb has quit IRC | 11:33 | |
| *** phuongnh has quit IRC | 11:36 | |
| *** xinliang has quit IRC | 11:36 | |
| *** xinliang has joined #openstack-nova | 11:37 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Use JSON-Schema to validate query parameters for keypairs API https://review.openstack.org/398755 | 11:37 |
| openstackgerrit | Alex Xu proposed openstack/nova: Add query parameters white list for server list/detail https://review.openstack.org/408571 | 11:37 |
| openstackgerrit | Alex Xu proposed openstack/nova: Adds support for versioned schema validation for query parameters https://review.openstack.org/389003 | 11:37 |
| *** tbachman has joined #openstack-nova | 11:39 | |
| mdbooth | sfinucan: You've managed to get python 3.4 tests running on Fedora, right? | 11:40 |
| mdbooth | Did you ever encounter the error message: | 11:40 |
| sfinucan | mdbooth: Yup, works without issue on F25 | 11:40 |
| mdbooth | db type could not be determined | 11:40 |
| mdbooth | The test run didn't actually run any tests | 11:40 |
| * mdbooth is on F25 | 11:41 | |
| sfinucan | Yeah - read the tox.ini | 11:41 |
| mdbooth | I have python3.4 installed | 11:41 |
| sfinucan | That's a bug with testr | 11:41 |
| * sfinucan goes to find the actual text | 11:41 | |
| mdbooth | Ah, spotted | 11:41 |
| sfinucan | https://github.com/openstack/nova/blob/master/tox.ini#L18-L20 | 11:41 |
| sfinucan | mdbooth: Someone should really fix that bug upstream | 11:42 |
| sfinucan | ...or we should move to nose. Either-or | 11:42 |
| mdbooth | sfinucan: Hmm, is is possible to run a subset of py34 tests? | 11:42 |
| * mdbooth normally runs: tox -e py27 libvirt test_virt_driver | 11:43 | |
| *** bks has joined #openstack-nova | 11:43 | |
| alex_xu | gmann: re: https://review.openstack.org/#/c/408879/, so we can't delete the empty tag except delete all the tag | 11:43 |
| mdbooth | However, s/27/34/ runs no tests | 11:43 |
| alex_xu | gmann: if that is correct, I think that can be a point we fix it without microversion | 11:43 |
| *** tbachman has quit IRC | 11:43 | |
| sfinucan | Hmm, I usually give a class path like 'tox -e py34 nova.tests.unit.virt.libvirt' | 11:44 |
| sfinucan | I've never tried the other way so I can't say if that's a regression or not (though it seems like it might be) | 11:44 |
| mdbooth | sfinucan: Yeah, that works. Seems ostestr does prefix matching, whereas testr does contains test | 11:45 |
| gmann | alex_xu: yes that is also true, it will be 404 on delete when doing delete on single empty tag | 11:45 |
| *** abalutoiu has quit IRC | 11:46 | |
| sfinucan | Ah, they're using different tools. Didn't know that | 11:47 |
| * sfinucan also wishes he had a trick for converting classpaths -> paths/vi search hints | 11:47 | |
| mdbooth | Is there a trick for running a subset of py34 tests whilst also excluding tests from tests-py3.txt? | 11:48 |
| sdague | mdbooth: I think the exclude facility is pretty naive at this point | 11:48 |
| mdbooth | For the record, I care about python 3 tests only to the minimum extent required in order to get real work done | 11:48 |
| sdague | so it's either one or the other | 11:49 |
| openstackgerrit | preethi proposed openstack/nova: Added PRSM to HVType class for support PR/SM hypervisor https://review.openstack.org/410713 | 11:50 |
| *** zhangyujun has joined #openstack-nova | 11:50 | |
| mdbooth | sdague: Thanks. I'll take a look at the code of ostestr, but the alternative might be to take a sabatical to fix everything in libvirt first :/ | 11:51 |
| *** bks has quit IRC | 11:52 | |
| sfinucan | mdbooth: I'm pretty sure gcb and co have fixed mostly everything for Python 3 compatibility | 11:52 |
| mdbooth | sfinucan: This is current master | 11:52 |
| sfinucan | ...though they're slow getting in so I certainly wouldn't base your work on them | 11:52 |
| mdbooth | So if they have, it's not merged | 11:52 |
| sfinucan | Yeah, sorry - that's what I should have said | 11:52 |
| *** gszasz has quit IRC | 11:53 | |
| mdbooth | Well, if it prevents me from running a subset of tests locally it's kinda urgent, because I can't be running all tests against every patch in my series | 11:53 |
| * mdbooth will try to find their commits | 11:53 | |
| * sfinucan will try to get more of the py3 patches merged today | 11:54 | |
| alex_xu | gmann: I just prefer to update the commit message to add that point | 11:54 |
| gmann | alex_xu: sure, thanks ll do that | 11:56 |
| alex_xu | gmann: thanks | 11:56 |
| *** gszasz has joined #openstack-nova | 11:58 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Add support to display usage audits for all servers https://review.openstack.org/409413 | 12:00 |
| sdague | sfinucan: if you need a hand looking those through, drop me some links | 12:01 |
| *** sacharya has joined #openstack-nova | 12:02 | |
| gcb | mdbooth, you can find more details about whitelist and blacklist in https://github.com/openstack/os-testr/blob/master/os_testr/ostestr.py#L35 | 12:03 |
| sfinucan | sdague: Will do | 12:04 |
| *** dave-mccowan has joined #openstack-nova | 12:04 | |
| *** hfu has joined #openstack-nova | 12:04 | |
| sdague | gcb: https://review.openstack.org/#/c/403425/4/nova/virt/libvirt/storage/rbd_utils.py - that seems pretty repetitive, is there no oslo function to do that decode correctly regardless of python version? | 12:05 |
| *** sudipto has joined #openstack-nova | 12:06 | |
| *** sudipto_ has joined #openstack-nova | 12:06 | |
| gcb | sdague: ack ,let me double check | 12:07 |
| *** sacharya has quit IRC | 12:07 | |
| *** sapcc-bot has joined #openstack-nova | 12:07 | |
| *** tpatzig_ has joined #openstack-nova | 12:07 | |
| *** carthaca_ has joined #openstack-nova | 12:07 | |
| *** tommy_ has joined #openstack-nova | 12:07 | |
| *** mkoderer_ has joined #openstack-nova | 12:07 | |
| *** dgonzalez_ has joined #openstack-nova | 12:07 | |
| *** david_1 has joined #openstack-nova | 12:07 | |
| *** zenoway has quit IRC | 12:07 | |
| *** tommy_ is now known as Guest68836 | 12:07 | |
| *** kaisers_ has joined #openstack-nova | 12:07 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: libvirt: remove hack for dom.vcpus() returning None https://review.openstack.org/393253 | 12:08 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: libvirt: fix vCPU usage reporing for LXC/QEMU guests https://review.openstack.org/393254 | 12:08 |
| *** Guest68836 has quit IRC | 12:09 | |
| *** carthaca_ has quit IRC | 12:09 | |
| *** tpatzig_ has quit IRC | 12:09 | |
| *** mkoderer_ has quit IRC | 12:09 | |
| *** dgonzalez_ has quit IRC | 12:09 | |
| *** david_1 has quit IRC | 12:09 | |
| *** thorst_ has joined #openstack-nova | 12:10 | |
| *** rtheis has joined #openstack-nova | 12:12 | |
| *** kaisers_ has quit IRC | 12:14 | |
| *** tyrola has quit IRC | 12:14 | |
| sdague | andreykurilin: can you convince me that - https://review.openstack.org/#/c/350106/13 isn't a mistake. The projectid=project_name thing is making it hard for me to be convinced of that | 12:14 |
| *** prateek_ has quit IRC | 12:14 | |
| sdague | after that, I've +Aed up through the keystone session add | 12:14 |
| *** tyrola has joined #openstack-nova | 12:15 | |
| *** zhs__ has joined #openstack-nova | 12:16 | |
| *** lucasagomes is now known as lucas-hungry | 12:17 | |
| pkoniszewski | tdurakov: hey, can you take a look at my analysis of LM grenade job issue there https://bugs.launchpad.net/nova/+bug/1649586/comments/7 ? | 12:17 |
| openstack | Launchpad bug 1649586 in OpenStack Compute (nova) ""HTTP exception thrown: Cannot 'os-migrateLive' instance while it is in task_state migrating" in gate-grenade-dsvm-neutron-multinode-live-migration-nv" [Medium,Confirmed] | 12:17 |
| pkoniszewski | mriedem: ^ | 12:17 |
| tdurakov | pkoniszewski: sure | 12:18 |
| * tdurakov reading | 12:18 | |
| *** thorst_ has quit IRC | 12:18 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Stop allowing tags as empty string https://review.openstack.org/408879 | 12:18 |
| gmann | alex_xu: ^^ updated | 12:19 |
| alex_xu | gmann: thanks | 12:19 |
| *** gongysh has quit IRC | 12:19 | |
| *** bvanhav has joined #openstack-nova | 12:19 | |
| openstackgerrit | Matthew Booth proposed openstack/nova: Port libvirt.test_firewall to Python 3 https://review.openstack.org/403423 | 12:19 |
| *** gcb has quit IRC | 12:20 | |
| *** kaisers_ has joined #openstack-nova | 12:20 | |
| *** zhs_ has quit IRC | 12:20 | |
| tdurakov | pkoniszewski: thanks Pawel, https://github.com/openstack/tempest/blob/071e94c31ce2d1549ecb34df1fea7f5a6cd210a6/tempest/lib/common/http.py#L55 - great you found this, and yeah, need to figure out how to fix rabbit timeouts | 12:21 |
| *** smatzek has joined #openstack-nova | 12:21 | |
| alex_xu | gmann: I guess it should be "shouldn't" in the line 24 of https://review.openstack.org/#/c/408879/5//COMMIT_MSG | 12:21 |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Port libvirt.storage.test_rbd to Python 3 https://review.openstack.org/403425 | 12:23 |
| pkholkin | mriedem: alex_xu: hi folks, review needed https://review.openstack.org/#/c/408166, thanks ! | 12:23 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Stop allowing tags as empty string https://review.openstack.org/408879 | 12:23 |
| pkholkin | just two small questions there, other seems ok | 12:23 |
| gmann | alex_xu: opps, done | 12:23 |
| andreykurilin | sdague: hi | 12:23 |
| andreykurilin | sdague: see https://github.com/openstack/python-novaclient/blob/master/novaclient/v2/client.py#L127-L131 | 12:23 |
| alex_xu | pkholkin: thanks for the patch, add it to my list | 12:24 |
| *** gcb_ has joined #openstack-nova | 12:24 | |
| *** gcb has joined #openstack-nova | 12:25 | |
| *** gcb has quit IRC | 12:25 | |
| alex_xu | gmann: done | 12:25 |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Add support to display usage audits for all servers https://review.openstack.org/409413 | 12:25 |
| gcb_ | sdague, just update https://review.openstack.org/#/c/403425/4/nova/virt/libvirt/storage/rbd_utils.py , I copied that from https://review.openstack.org/#/c/261045/100/nova/virt/libvirt/storage/rbd_utils.py before | 12:25 |
| gmann | alex_xu: cool, thanks | 12:26 |
| andreykurilin | sdague: previously, in case of SessionClient, project_id meant project id(in terms of keystone), but in case of HTTPClient implementation, it was used as project_name. At least there is a specific note about it + additional variable for actual project id(in terms of keystone) - tenant_id. So we have now two vartiables - project_id and tenant_id :) | 12:26 |
| alex_xu | gmann: np | 12:26 |
| *** int32bit has quit IRC | 12:27 | |
| gcb_ | sfinucan, need look at https://review.openstack.org/#/c/408561/ before merging https://review.openstack.org/#/c/410554/ | 12:28 |
| gcb_ | thanks | 12:28 |
| gcb_ | gmann, I replied your comments on https://review.openstack.org/#/c/403419/ | 12:28 |
| andreykurilin | sdague: imo, `projectid=project_name` is not so important now, since one of last patches in the chain removes the whole HTTPClient implementation at all with this dirty compatibility workaround | 12:29 |
| *** ociuhandu has joined #openstack-nova | 12:29 | |
| *** gszasz has quit IRC | 12:30 | |
| sdague | andreykurilin: ok | 12:30 |
| *** gszasz_ has joined #openstack-nova | 12:30 | |
| sdague | that's good enough for me | 12:30 |
| gmann | gcb_: Thankd, i will check tomorrow morning. may be m missing something but vlan seems to be int all the time | 12:33 |
| gcb_ | gmann, ok | 12:35 |
| andreykurilin | sdague: here is removing of HTTPClient - https://review.openstack.org/#/c/408728 | 12:37 |
| andreykurilin | sdague: and thanks for review :) | 12:37 |
| sdague | sfinucan: is there a reason you only +1ed - https://review.openstack.org/#/c/199488/ | 12:37 |
| sdague | andreykurilin: yeh, lets get the rest of that merged, then I'll look at that patch | 12:37 |
| andreykurilin | ok, thanks | 12:37 |
| sdague | andreykurilin: the stack might need rebasing, lets see if gerrit will do that once the base patch goes in | 12:37 |
| sdague | gibi: if you are around today - https://review.openstack.org/#/c/409951/ has a question for you about mocking out locking in notification tests | 12:38 |
| sfinucan | sdague: I couldn't see anything wrong with it, but that code is a little tricky to parse. Wasn't happy +2'ing it personally | 12:38 |
| sdague | sfinucan: ok | 12:38 |
| gmann | sdague: as you are around, can you look into those - https://review.openstack.org/#/q/topic:tag-schema+status:open | 12:39 |
| sdague | sfinucan: I've stared at it a few times. I'm not super thrilled with the pci whitelist config getting *more* complex, but that seems to have been agreed to | 12:39 |
| openstackgerrit | jichenjc proposed openstack/nova: WIP: remove keypair server extension https://review.openstack.org/396611 | 12:39 |
| sfinucan | sdague: yeah, my thoughts exactly. jaypipes has used some choice terms in the past to describe the whole thing | 12:40 |
| sfinucan | ...but moshele has taken all the feedback on board and has maintained it thus far so I'm definitely ok to +1 | 12:41 |
| sfinucan | just not to +2 :) | 12:41 |
| sdague | ok | 12:41 |
| sdague | well, if it was team accepted direction, I don't want to hold it up too much. Ok, +A | 12:42 |
| sdague | the pci whitelist really never should have gone in via config and should have been part of the API to manage I think, but such it is | 12:42 |
| *** jamielennox is now known as jamielennox|away | 12:43 | |
| sdague | I'm sure they've got a bunch of ansible/puppet to manage large systems for stuff like this | 12:43 |
| *** kaisers_ has quit IRC | 12:44 | |
| alex_xu | nova api subteam meeting is in 15 mins@#openstack-meeting-4 | 12:45 |
| sdague | alex_xu: thanks for the reminder, I need to add that channel back into my ansible config for hexchat | 12:47 |
| alex_xu | sdague: np :) | 12:47 |
| *** nklenke has joined #openstack-nova | 12:47 | |
| gmann | sdague: Thanks | 12:48 |
| jaypipes | sfinucan: heh | 12:49 |
| sfinucan | sdague: I'd appreciate it if you could take a gawk at https://review.openstack.org/#/c/403425 - some unicode madness going on :/ | 12:49 |
| *** jamielennox|away is now known as jamielennox | 12:50 | |
| sdague | sfinucan: ok, lets see if I'm caffinated enough for unicode | 12:50 |
| *** sudipto_ has quit IRC | 12:51 | |
| *** sudipto has quit IRC | 12:51 | |
| sdague | oh, hey, this was the thing I was just asking about :) | 12:51 |
| *** gcb_ has quit IRC | 12:52 | |
| sdague | yeh, I was wondering why that was using encode in the first place | 12:52 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: hyperv: stop serial console workers while deleting vm files https://review.openstack.org/387831 | 12:52 |
| sdague | oh, is that because of expectations of the rbd library? | 12:52 |
| sdague | mdbooth: is this something you've got in working memory atm? | 12:53 |
| * sdague tries to remember all the folks that touched the rbd code | 12:53 | |
| ftersin | jbernard: could you look at the last comment on https://bugs.launchpad.net/cinder/+bug/1648885 and confirm that your patch is useful in that case? | 12:53 |
| openstack | Launchpad bug 1648885 in Cinder "ceph: TestVolumeBootPattern.test_create_ebs_image_and_check_boot times out failing to delete volume due to snapshot is busy" [Undecided,In progress] - Assigned to Jon Bernard (jbernard) | 12:53 |
| *** nandal has quit IRC | 12:53 | |
| alex_xu | pkholkin: the correct behaviour should be 'nova image-meta' return failed, right? | 12:54 |
| *** pbandark has quit IRC | 12:54 | |
| *** amoralej is now known as amoralej|lunch | 12:55 | |
| jbernard | ftersin: yes, if volume delete takes time (which is can on a gate setup) and snapshot_delete is allowed to race (which I believe it is), then this scenario (snapshot delete races and wins) becomes possible | 12:55 |
| jbernard | ftersin: i just updated the patch to log when this happens | 12:55 |
| *** ijw has joined #openstack-nova | 12:56 | |
| jbernard | ftersin: we see both operations starting | 12:57 |
| *** ducttape_ has joined #openstack-nova | 12:57 | |
| jbernard | ftersin: and we see snapshot delete fail before volume deletion completes | 12:57 |
| *** nklenke has quit IRC | 12:57 | |
| openstackgerrit | Hironori Shiina proposed openstack/nova: Ironic: Add soft reboot support to ironic driver https://review.openstack.org/403745 | 12:57 |
| jbernard | ftersin: so we can either serialize the tempest test or postpone in the driver (or change cinder api - but that's a larger topic) | 12:58 |
| sfinucan | sdague: Yeah, I'm looking at the code (linked in my comment) but it's not available on pip | 12:58 |
| sfinucan | ...so I can't just download to test myself. I'm guessing, however, that it requires byte strings | 12:59 |
| sfinucan | Do we have a ceph CI that we could fling that patch at? | 12:59 |
| *** yufei has joined #openstack-nova | 12:59 | |
| *** yufei has quit IRC | 12:59 | |
| pkholkin | alex_xu: 404 for >= 2.39, warning in other cases | 12:59 |
| sfinucan | ...or whatever we'd need for that | 12:59 |
| sdague | sfinucan: it should be running on that, no? | 12:59 |
| alex_xu | pkholkin: I see a strange case, I comment in the patch | 13:00 |
| sdague | sfinucan: gate-tempest-dsvm-full-devstack-plugin-ceph-ubuntu-xenial | 13:00 |
| sdague | yeh, just wait for that to return | 13:00 |
| openstackgerrit | Francesco Santoro proposed openstack/nova: [WIP] convert libvirt driver to use os-vif for fast path vhostuser https://review.openstack.org/385061 | 13:00 |
| openstackgerrit | Francesco Santoro proposed openstack/nova: os-vif-util: set vif_name for vhostuser ovs os-vif port https://review.openstack.org/410737 | 13:00 |
| sdague | jbernard: actually, you are one of the rbd experts, can you help look at that patch | 13:00 |
| *** ijw has quit IRC | 13:00 | |
| alex_xu | sdague: johnthetubaguy it's time for api meeting :) | 13:00 |
| ftersin | jbernard: i think that a correct way is to prevent deletion on api layer of Cinder. i.e. delete operation should not return 'ok' if the snapshot is busy | 13:00 |
| jbernard | sdague: which patch? | 13:00 |
| sdague | jbernard: https://review.openstack.org/#/c/403425 | 13:01 |
| sdague | trying to get rbd code python3-ified | 13:01 |
| *** yufei has joined #openstack-nova | 13:01 | |
| jbernard | sdague: ok | 13:01 |
| pkholkin | alex_xu: will check | 13:01 |
| *** ducttape_ has quit IRC | 13:01 | |
| pkholkin | thanks ! | 13:02 |
| *** pradk has joined #openstack-nova | 13:02 | |
| jbernard | ftersin: that's fair, but such a change has a larger impact on cinder and will necessarily take longer to reveiw and agree on | 13:02 |
| jbernard | ftersin: if we have such time, then im happy to drive that topic in cinder | 13:03 |
| *** jheroux has joined #openstack-nova | 13:03 | |
| jbernard | ftersin: and if not, im still happy to raise it anyway | 13:03 |
| sfinucan | sdague: Ah, yes. It passed so maybe we're good and can remove encode (and safe_decode) altogether | 13:03 |
| *** markvoelker has joined #openstack-nova | 13:03 | |
| sfinucan | ...and if it breaks someone's code, then take that as an excuse to improve the CI | 13:03 |
| *** erbadr has joined #openstack-nova | 13:04 | |
| ftersin | jbernard: i understand that correct fix takes a lot of time, so i propose to change the test | 13:04 |
| jbernard | ftersin: ok that's cool | 13:04 |
| *** thorst_ has joined #openstack-nova | 13:05 | |
| jbernard | ftersin: and we have the rbd driver patch if we need it | 13:05 |
| *** hfu has quit IRC | 13:05 | |
| ftersin | jbernard: but as i see on https://review.openstack.org/#/c/410338/ tempest team is not inclined to merge even pure test improvement | 13:06 |
| ftersin | jbernard: which improvement does not help to avoid the bug | 13:07 |
| *** lucas-hungry is now known as lucasagomes | 13:07 | |
| jbernard | ftersin: i saw that, the rbd driver patch might be the best short-term solution in that case | 13:07 |
| ftersin | jbernard: i'm going to add the workaround (mentioned above), but do not know how to get it merged | 13:08 |
| *** markvoelker has quit IRC | 13:08 | |
| *** ducttape_ has joined #openstack-nova | 13:09 | |
| ftersin | jbernard: yep, rbd driver patch is better. but as i understand, it just ignores busy snapshot, leaving it in ceph | 13:09 |
| *** moshele has quit IRC | 13:09 | |
| jbernard | ftersin: it doesn't ignore them exactly, it just 'marks' them | 13:10 |
| *** lpetrut has quit IRC | 13:10 | |
| jbernard | ftersin: and once the 'reference count' (final volume dependency) goes to 0, it deletes them | 13:10 |
| *** hfu has joined #openstack-nova | 13:10 | |
| jbernard | ftersin: additionaly, this behaviour is unique to the rbd driver, and confuses users as well | 13:11 |
| jbernard | ftersin: so it may be useful on its own | 13:11 |
| ftersin | jbernard: but this deletion occures when we delete the last volume derived from the snapshot, right? | 13:11 |
| jbernard | ftersin: right | 13:11 |
| *** kaisers_ has joined #openstack-nova | 13:11 | |
| jbernard | on volume delete, we look for any marked snapshots | 13:11 |
| jbernard | ftersin: and clean them up before returning | 13:11 |
| ftersin | jbernard: however in this case the last volume is being deleted when snapshot deletion come | 13:12 |
| *** thorst_ has quit IRC | 13:13 | |
| ftersin | jbernard: i mean that when the driver starts to delete the volume, the snapshot is not yet marked | 13:13 |
| jbernard | ftersin: ahh, yes i see your point | 13:14 |
| jbernard | ftersin: i can fix that ;) | 13:14 |
| jbernard | ftersin: but you're right, there is a test-set race there in that case | 13:15 |
| *** mvk has quit IRC | 13:15 | |
| *** kaisers_ has quit IRC | 13:15 | |
| diana_clarke | The nova side of usage paging is ready for review: https://review.openstack.org/#/c/386093/ (I'm on holiday starting Dec 20th, so it would be great to get reviews soon). Thanks! PS. It already has a +2. | 13:16 |
| jbernard | ftersin: i made a note on the patch and -1 | 13:17 |
| jbernard | ftersin: will get an update posted as soon as i can | 13:18 |
| *** ekuris has quit IRC | 13:18 | |
| *** yufei has quit IRC | 13:18 | |
| gibi | sdague: regarding the mocked out locking I only remember that something was broken with the test at some point but I'm not sure what was it. | 13:18 |
| gibi | sdague: as now every test passes with your patch I think the original problem went away in the meantime. | 13:19 |
| ftersin | jbernard: anyway lets get ci logs to figure out if the test passes with 'pending' snapshot deletion | 13:20 |
| *** edmondsw_ has joined #openstack-nova | 13:20 | |
| *** edmondsw_ has quit IRC | 13:20 | |
| gibi | sdague: so I'm +1 on that patch | 13:21 |
| *** yufei has joined #openstack-nova | 13:22 | |
| *** markvoelker has joined #openstack-nova | 13:24 | |
| *** brad[] has quit IRC | 13:28 | |
| *** yufei has quit IRC | 13:29 | |
| *** sreeram_v has joined #openstack-nova | 13:30 | |
| *** daemontool_ has quit IRC | 13:30 | |
| *** ekuris has joined #openstack-nova | 13:31 | |
| *** gmann_ has quit IRC | 13:34 | |
| sdague | gibi: well, I restored the REQUIRES_LOCKING in there, which the comment said it was supposed to avoid. But I think it's fine that we have locking :) | 13:34 |
| *** lamt has joined #openstack-nova | 13:35 | |
| gibi | sdague: honestly I failed to remember why I did the mocking and my original comment doesn't explain the reason unfortunately | 13:35 |
| *** lbeliveau has quit IRC | 13:36 | |
| sdague | gibi: ok, cool. We just wanted to double check with you for reasons before changing that | 13:37 |
| *** mdnadeem has quit IRC | 13:37 | |
| sdague | yeh, the change history wasn't easy to understand if that was a critical concept or just happened to slide in | 13:37 |
| *** tbachman has joined #openstack-nova | 13:37 | |
| *** lbeliveau has joined #openstack-nova | 13:38 | |
| *** lamt has quit IRC | 13:39 | |
| gibi | sdague: anyhow I'm running some test loops locally to have more confidence but I think it is OK to merge the patch | 13:39 |
| *** moshele has joined #openstack-nova | 13:39 | |
| *** hfu has quit IRC | 13:40 | |
| *** yufei has joined #openstack-nova | 13:43 | |
| *** ducttape_ has quit IRC | 13:45 | |
| openstackgerrit | Wang Qing wu proposed openstack/nova: WIP(1): Introduce base PowerVM driver https://review.openstack.org/391288 | 13:48 |
| *** lpetrut has joined #openstack-nova | 13:52 | |
| *** tlian has joined #openstack-nova | 13:52 | |
| *** brad[] has joined #openstack-nova | 13:53 | |
| *** abalutoiu has joined #openstack-nova | 13:53 | |
| *** baoli has joined #openstack-nova | 13:55 | |
| *** tongli has joined #openstack-nova | 13:56 | |
| *** krtaylor has joined #openstack-nova | 13:59 | |
| *** bvanhav_ has joined #openstack-nova | 14:02 | |
| *** slaweq_ has quit IRC | 14:02 | |
| Kevin_Zheng | johnthetubaguy: https://review.openstack.org/#/c/402372 are you suggesting make destination type mandantory in schema？ | 14:02 |
| *** sacharya has joined #openstack-nova | 14:03 | |
| johnthetubaguy | Kevin_Zheng: yeah, that seems to be the same thing, but I could be missing something big | 14:03 |
| *** bvanhav has quit IRC | 14:03 | |
| *** fragatina has joined #openstack-nova | 14:04 | |
| *** ayoung has quit IRC | 14:04 | |
| Kevin_Zheng | johnthetubaguy: what do you mean？ | 14:04 |
| johnthetubaguy | Kevin_Zheng: it just seems to be two ways to implement the same API behaviour | 14:05 |
| Kevin_Zheng | hmm yes | 14:06 |
| *** abalutoiu_ has joined #openstack-nova | 14:06 | |
| johnthetubaguy | Kevin_Zheng: given that, I would just rather we did it the JSON schema way, looks a bit like we just forgot to define which of those additional attributes are required | 14:06 |
| Kevin_Zheng | Ok, I will modify it | 14:07 |
| Kevin_Zheng | If it is better that way | 14:07 |
| *** amoralej|lunch is now known as amoralej | 14:07 | |
| johnthetubaguy | Kevin_Zheng: I just like getting the schema "more correct" I guess | 14:07 |
| johnthetubaguy | Kevin_Zheng: good catch by the way, thats a bit nasty | 14:07 |
| Kevin_Zheng | Thanks | 14:08 |
| *** sacharya has quit IRC | 14:08 | |
| Kevin_Zheng | We also find a problem about providing duplicate device name | 14:08 |
| *** fragatina has quit IRC | 14:09 | |
| *** abalutoiu has quit IRC | 14:09 | |
| openstackgerrit | Hironori Shiina proposed openstack/nova: Ironic: Add soft reboot support to ironic driver https://review.openstack.org/403745 | 14:09 |
| *** liverpooler has joined #openstack-nova | 14:09 | |
| *** liverpooler has quit IRC | 14:10 | |
| *** rwmjones has quit IRC | 14:10 | |
| *** liverpooler has joined #openstack-nova | 14:10 | |
| Kevin_Zheng | anyway I will try to do in schema way tahnks | 14:11 |
| *** kaisers_ has joined #openstack-nova | 14:12 | |
| *** liusheng has quit IRC | 14:13 | |
| jbernard | ftersin: i doesn't happen to often if im not mistaken, so you'd have to recheck it quite a few times to be sure | 14:14 |
| *** liusheng has joined #openstack-nova | 14:14 | |
| *** ducttape_ has joined #openstack-nova | 14:15 | |
| ftersin | jbernard: i do not see occurrences as well. if you're not ready yet to publish the new ps, i'll post recheck there | 14:17 |
| jbernard | ftersin: go ahead, ive got a few items this morning | 14:18 |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: raise exception if create Virtuozzo container with swap disk https://review.openstack.org/397829 | 14:18 |
| *** rwmjones has joined #openstack-nova | 14:18 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Merge v20 view builder for limit API https://review.openstack.org/406787 | 14:23 |
| *** tbachman has quit IRC | 14:23 | |
| mdbooth | sfinucan sdague: If I merge these 4 changes, libvirt tests pass on py34: https://review.openstack.org/#/c/403425/ https://review.openstack.org/#/c/403423/ https://review.openstack.org/#/c/403419/ https://review.openstack.org/#/c/403431/ | 14:26 |
| *** thorst_ has joined #openstack-nova | 14:26 | |
| *** unicell has joined #openstack-nova | 14:26 | |
| *** unicell1 has quit IRC | 14:27 | |
| *** yamahata_ has joined #openstack-nova | 14:28 | |
| sfinucan | mdbooth: Yeah, I've +2d two of them but there are issues with the other two. Hopefully gcb can take a look at them this week | 14:28 |
| *** erbadr has quit IRC | 14:28 | |
| *** hongbin has joined #openstack-nova | 14:29 | |
| mdbooth | I can't run python 3 tests against just libvirt without having these in my git history | 14:29 |
| mdbooth | Which is a pita, tbh | 14:29 |
| *** erbadr has joined #openstack-nova | 14:29 | |
| mdbooth | Any chance we can just get it done and fix it later? | 14:29 |
| *** erbadr is now known as Jinxing | 14:29 | |
| mdbooth | How bad are the negatives? | 14:29 |
| *** Jinxing has quit IRC | 14:30 | |
| *** Jinxing has joined #openstack-nova | 14:30 | |
| mdbooth | Incidentally, octopus merge ftw | 14:30 |
| sfinucan | This one's the real issue https://review.openstack.org/#/c/403425 | 14:31 |
| sfinucan | fwict, it could potentially bork "rbd", whatever that is | 14:31 |
| *** asettle has quit IRC | 14:31 | |
| sfinucan | mdbooth: That's actually something you might be able to help with though, now that I think of it | 14:31 |
| *** liverpooler has quit IRC | 14:32 | |
| *** asettle has joined #openstack-nova | 14:32 | |
| *** liverpooler has joined #openstack-nova | 14:32 | |
| sfinucan | the other one https://review.openstack.org/#/c/403419/ looks like a simple enough change, but if I do it then I can't +2 | 14:32 |
| *** tbachman has joined #openstack-nova | 14:33 | |
| *** roeyc has quit IRC | 14:34 | |
| *** ducttape_ has quit IRC | 14:34 | |
| *** abalutoiu_ has quit IRC | 14:34 | |
| mdbooth | sfinucan: I might be able to check that out | 14:35 |
| openstackgerrit | Hironori Shiina proposed openstack/nova: ironic: Add soft power off support to ironic driver. https://review.openstack.org/407977 | 14:35 |
| *** smatzek has quit IRC | 14:35 | |
| mdbooth | sfinucan: Yeah, hadn't appreciated your -1 was against non-test code | 14:35 |
| mdbooth | I'll have a look at that | 14:35 |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: fix formatting in base https://review.openstack.org/404939 | 14:36 |
| jaypipes | mdbooth, sfinucan: added an alternative solution on 403425. | 14:36 |
| *** mvk has joined #openstack-nova | 14:36 | |
| *** hshiina has quit IRC | 14:36 | |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: Remove check_attach https://review.openstack.org/335358 | 14:37 |
| sfinucan | jaypipes: Yup, that looks much saner to me | 14:38 |
| * sfinucan will be sooo much happier when Python 3 takes over the world and solves this stuff for us | 14:38 | |
| * sfinucan doubts he'll live to see that day | 14:38 | |
| ildikov | johnthetubaguy: Hi. Do you happen to have any pointers for me for a test failure with this exception: 'This test uses methods that set internal oslo_db'? :) | 14:38 |
| jaypipes | sfinucan: :) | 14:39 |
| mdbooth | jaypipes sfinucan: I'll check it out when I've finished this rebase. | 14:39 |
| mriedem | ildikov: it means you're usinga NoDbTestCase but have code in the tests hitting the db | 14:39 |
| mriedem | so you're missing some mocks | 14:39 |
| ildikov | mriedem: hmm, interesting | 14:40 |
| johnthetubaguy | ildikov: yeah, +1 mriedem missing a mock by the sound of it | 14:40 |
| ildikov | I think I haven't touched those tests that report this, but I'll dig deeper then | 14:40 |
| johnthetubaguy | you maybe made something stop failing earlier on? worth a dig | 14:41 |
| * ildikov admits I modified only existing test cases and not even those are the ones that are failing... :S | 14:41 | |
| mdbooth | sfinucan: I'd be so much happier if the python community conceded python3 was a huge mistake and concentrated on fixing python2. | 14:41 |
| ildikov | johnthetubaguy: no idea yet :) | 14:41 |
| *** marst has joined #openstack-nova | 14:41 | |
| jaypipes | ildikov: if you change the test case class base from NoDBTestCase to TestCase and re-run, if it succeeds, the issue is what mriedem said. | 14:42 |
| mriedem | ildikov: doesn't the failure have a stacktrace going back to what unmocked code was triggering the failure? | 14:42 |
| ildikov | johnthetubaguy: fixed your comments on the patch at least and added a few answers to those, where I don't think we will need unreserve | 14:42 |
| mriedem | ildikov: i'm pretty sure it can be walked back via the stacktrace | 14:43 |
| *** gouthamr has joined #openstack-nova | 14:43 | |
| *** andymaier has joined #openstack-nova | 14:44 | |
| *** moshele has quit IRC | 14:44 | |
| ildikov | mriedem: I will look into that deeper, just wasn't 100% sure what I'm looking for :) | 14:44 |
| ildikov | mriedem: I think I have an idea now where the problem might be | 14:45 |
| ildikov | mriedem: tnx! | 14:45 |
| *** moshele has joined #openstack-nova | 14:45 | |
| *** dinobot_ has quit IRC | 14:45 | |
| *** thorst_ has quit IRC | 14:46 | |
| *** jbernard has quit IRC | 14:46 | |
| *** jbernard has joined #openstack-nova | 14:46 | |
| *** gcb has joined #openstack-nova | 14:47 | |
| *** tblakes has joined #openstack-nova | 14:47 | |
| *** kaisers_ has quit IRC | 14:47 | |
| mdbooth | Incidentally, I have a bugfix change which is feeling a bit sad: https://review.openstack.org/#/c/383859/ | 14:48 |
| *** burt has joined #openstack-nova | 14:49 | |
| *** tongli has quit IRC | 14:49 | |
| mdbooth | It's passed all the CI we can think of, but I suspect the issue it's in an unloved area of code which nobody's that familiar with | 14:49 |
| mriedem | i loved it over once | 14:50 |
| mriedem | it didn't love me back | 14:50 |
| mdbooth | mriedem: It told me it thinks very highly of you | 14:50 |
| mriedem | ooo, sounds like middle school romance | 14:50 |
| mdbooth | lol | 14:51 |
| mriedem | tell your patch that i definitely like it, like, like like | 14:51 |
| jaypipes | mdbooth, mriedem: your eyeballs on 409943 would be useful... thx in advance. | 14:51 |
| *** mtanino has joined #openstack-nova | 14:52 | |
| *** thorst_ has joined #openstack-nova | 14:52 | |
| *** yufei has quit IRC | 14:53 | |
| jaypipes | mriedem: you gonna ask it to slow dance? | 14:54 |
| mriedem | jaypipes: gross no, | 14:54 |
| mriedem | i'm going to hang out with my buddies against the wall | 14:54 |
| mriedem | and just watch | 14:54 |
| *** sreeram_v has quit IRC | 14:55 | |
| jaypipes | mriedem: totes. | 14:55 |
| sdague | mriedem / jaypipes I just sent a summary of the server param filtering discussion from the api meeting this morning to the ML | 14:55 |
| mriedem | thanks | 14:56 |
| sdague | it would be great if you could look at that, and chime in with anything we missed | 14:56 |
| sdague | because it's largely the summary plan forward | 14:56 |
| mriedem | i'm distracted with this instance usage audit log thing that was brought up in the ML | 14:56 |
| jaypipes | sdague: cheers, will do. | 14:56 |
| *** links has quit IRC | 14:57 | |
| *** mnasiadka has quit IRC | 14:58 | |
| *** Swami has joined #openstack-nova | 14:59 | |
| *** abalutoiu_ has joined #openstack-nova | 15:00 | |
| mriedem | wtf, how does this work? http://developer.openstack.org/api-ref/compute/?expanded=list-server-usage-audits-for-an-tenant-detail | 15:01 |
| mriedem | it's an admin-only api | 15:01 |
| mriedem | and it's meant to get usage audit for a tenant, but doesn't take a tenant on the request | 15:01 |
| *** jlk has quit IRC | 15:01 | |
| *** yufei has joined #openstack-nova | 15:02 | |
| *** smatzek has joined #openstack-nova | 15:02 | |
| *** mvk has quit IRC | 15:02 | |
| *** jlk has joined #openstack-nova | 15:03 | |
| *** jlk has quit IRC | 15:03 | |
| *** jlk has joined #openstack-nova | 15:03 | |
| mriedem | i guess it's not per-tenant like the docs say, | 15:04 |
| mriedem | it's just, get me all data from all instances and all computes in the cloud | 15:04 |
| mdbooth | jaypipes: Can we send dicts over rpc? | 15:05 |
| jaypipes | mdbooth: you mean do o.vo objects support dict fields? | 15:06 |
| mdbooth | jaypipes: I just think it's kinda weird to call instance_fault_get_by_instance_uuids | 15:06 |
| mdbooth | which returns a dict | 15:06 |
| mdbooth | Turn it into a list | 15:06 |
| jaypipes | mdbooth: no disagreement with you there :) | 15:07 |
| mdbooth | Only for the caller to turn it back into an identically-keyed dict | 15:07 |
| sdague | mriedem: which ML post is this? | 15:08 |
| mriedem | sdague: http://lists.openstack.org/pipermail/openstack-dev/2016-December/108887.html | 15:09 |
| * mriedem goes to a meeting | 15:09 | |
| sdague | yeh the simple tenant usage API is an oddball | 15:09 |
| mriedem | sdague: this isn't that | 15:10 |
| sdague | oh... ? | 15:10 |
| mriedem | this is an even less used oddball | 15:10 |
| mriedem | yeah there is a periodic task on the computes which generates audit task logs, | 15:10 |
| mriedem | which this api pulls that data from for all instances on all hosts | 15:10 |
| mriedem | so it doesn't have anything to do with tenants, so the api-ref is wrong | 15:11 |
| *** psachin has quit IRC | 15:11 | |
| mriedem | this code is all sorts of borked | 15:12 |
| mriedem | e.g. _get_audit_task_logs in the API controller has begin/end logic which is dead code | 15:12 |
| mriedem | ok meeting for real this time | 15:13 |
| *** gouthamr has quit IRC | 15:13 | |
| *** esberglu has joined #openstack-nova | 15:14 | |
| *** gcb has quit IRC | 15:14 | |
| *** gcb has joined #openstack-nova | 15:14 | |
| *** whenry has quit IRC | 15:15 | |
| *** gouthamr has joined #openstack-nova | 15:15 | |
| *** whenry has joined #openstack-nova | 15:16 | |
| *** mvk has joined #openstack-nova | 15:16 | |
| *** ducttape_ has joined #openstack-nova | 15:16 | |
| *** bvanhav_ is now known as bvanhav | 15:17 | |
| *** gcb has quit IRC | 15:20 | |
| sdague | mriedem: when you get back, gibi +1ed - https://review.openstack.org/#/c/409951/ | 15:23 |
| sfinucan | alex_xu: out of curiosity, how does the validation of parameters for the API work? | 15:26 |
| *** cardeois has left #openstack-nova | 15:26 | |
| sfinucan | I'm looking at this patch https://review.openstack.org/#/c/403536/6/nova/tests/unit/api/openstack/compute/test_disk_config.py | 15:26 |
| openstackgerrit | Guillaume Espanel proposed openstack/nova: Catch neutronclient.NotFound on floating deletion https://review.openstack.org/410815 | 15:27 |
| *** whenry has quit IRC | 15:27 | |
| *** roeyc has joined #openstack-nova | 15:27 | |
| sfinucan | I think the issue is that there's a 'u' prefix leaking into the error message, but I can't figure out where those error messages are built | 15:27 |
| *** nklenke has joined #openstack-nova | 15:27 | |
| *** gszasz_ has quit IRC | 15:28 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: WIP: Add new configuration option live_migration_scheme https://review.openstack.org/410817 | 15:29 |
| *** baoli has quit IRC | 15:29 | |
| *** esmiurium_ has quit IRC | 15:30 | |
| *** raunak has joined #openstack-nova | 15:30 | |
| pkoniszewski | mriedem: hey, don't know whether you saw my message, i think i found the root cause why LM grenade job is failing - https://bugs.launchpad.net/nova/+bug/1649586/comments/7 | 15:31 |
| openstack | Launchpad bug 1649586 in OpenStack Compute (nova) ""HTTP exception thrown: Cannot 'os-migrateLive' instance while it is in task_state migrating" in gate-grenade-dsvm-neutron-multinode-live-migration-nv" [Medium,Confirmed] | 15:31 |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: enable console log for virtuozzo https://review.openstack.org/397166 | 15:31 |
| *** mlavalle has joined #openstack-nova | 15:32 | |
| *** sridharg has quit IRC | 15:33 | |
| *** sridharg has joined #openstack-nova | 15:34 | |
| mdbooth | jaypipes: I may have an alternative suggestion on that query, btw | 15:34 |
| mdbooth | Just testing it | 15:35 |
| *** eharney has joined #openstack-nova | 15:35 | |
| *** zz_dimtruck is now known as dimtruck | 15:37 | |
| *** fragatina has joined #openstack-nova | 15:37 | |
| *** fragatina has quit IRC | 15:37 | |
| *** fragatina has joined #openstack-nova | 15:38 | |
| *** siva_krish has quit IRC | 15:39 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Mark live_migration_downtime_steps as deprecated for removal https://review.openstack.org/408002 | 15:39 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Change exponential function to linear https://review.openstack.org/407518 | 15:39 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Move migration_downtime_steps to libvirt/migration https://review.openstack.org/407517 | 15:39 |
| *** kfarr has joined #openstack-nova | 15:39 | |
| *** sridharg has quit IRC | 15:39 | |
| *** siva_krish has joined #openstack-nova | 15:40 | |
| *** eharney has quit IRC | 15:40 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Add CLI to show instance usage audit logs https://review.openstack.org/409413 | 15:40 |
| mdbooth | jaypipes: How about this instead: http://paste.openstack.org/show/592359/ | 15:41 |
| mdbooth | Advantage is it returns the same base model for both queries, so less chance of them getting out of sync | 15:41 |
| *** annegentle has joined #openstack-nova | 15:41 | |
| mdbooth | It's also slightly shorter | 15:41 |
| mdbooth | It trusts that the query optimiser will be sensible with the subquery, which I'm pretty sure is the case | 15:42 |
| *** thorst_ has quit IRC | 15:44 | |
| *** NobodyCam is now known as WeAreAllMiLan | 15:46 | |
| *** dave-mccowan has quit IRC | 15:46 | |
| *** siva_krish has quit IRC | 15:48 | |
| *** WeAreAllMiLan is now known as NobodyCam | 15:48 | |
| *** lpetrut1 has joined #openstack-nova | 15:49 | |
| *** Sukhdev has joined #openstack-nova | 15:50 | |
| *** lpetrut has quit IRC | 15:51 | |
| *** lpetrut1 is now known as lpetrut | 15:51 | |
| *** ralonsoh_ has joined #openstack-nova | 15:51 | |
| *** ralonsoh has quit IRC | 15:51 | |
| *** annegentle has quit IRC | 15:52 | |
| *** claudiub has quit IRC | 15:52 | |
| *** diogogmt has quit IRC | 15:53 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Add limit and offset to server-groups list https://review.openstack.org/409282 | 15:53 |
| *** eharney has joined #openstack-nova | 15:53 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: Added CLI args for nova-policy check command https://review.openstack.org/382915 | 15:53 |
| *** yufei has left #openstack-nova | 15:54 | |
| rpodolyaka | mdbooth: I'd actually suggest to *not* return a model in the first place | 15:54 |
| rpodolyaka | as we basically throw it away the moment we get a model instance | 15:54 |
| rpodolyaka | and convert to a dict | 15:54 |
| rpodolyaka | performance-wise it's more work for no good reason | 15:55 |
| mdbooth | rpodolyaka: That's fine too. I guess my main concern was that there's no good reason for both sides of the branch to be so different when they're returning the same data. | 15:55 |
| *** abalutoiu_ has quit IRC | 15:55 | |
| rpodolyaka | yeah, that makes sense | 15:55 |
| jaypipes | mdbooth: nice suggestion, yep. | 15:56 |
| jaypipes | mdbooth: I probably would have done it that way originally had I known about the subquery() thing that rpodolyaka taught me :) | 15:56 |
| jaypipes | mdbooth: if you or rpodolyaka want to push a rev on that, cool with me :) | 15:56 |
| jaypipes | (hint hint) | 15:56 |
| rpodolyaka | jaypipes: on it! | 15:57 |
| jaypipes | :) yay! | 15:57 |
| *** ayoung has joined #openstack-nova | 15:57 | |
| *** _ducttape_ has joined #openstack-nova | 15:57 | |
| *** fragatina has quit IRC | 15:58 | |
| *** hieulq__ has joined #openstack-nova | 15:59 | |
| *** tongli has joined #openstack-nova | 16:00 | |
| *** ducttape_ has quit IRC | 16:00 | |
| jbernard | ftersin: http://logs.openstack.org/50/281550/7/check/gate-tempest-dsvm-full-devstack-plugin-ceph-ubuntu-xenial/54ce5c8/logs/screen-c-vol.txt.gz?level=INFO#_2016-12-14_13_53_01_188 | 16:00 |
| jbernard | ftersin: it appears to be hitting there | 16:00 |
| *** annegentle has joined #openstack-nova | 16:00 | |
| *** cleong has joined #openstack-nova | 16:01 | |
| jbernard | ftersin: i should add logging in the cleanup routine as well | 16:01 |
| *** gouthamr has quit IRC | 16:01 | |
| dansmith | vladikr these are different failures from the last time I pinged you right? http://logs.openstack.org/79/319379/48/check/gate-nova-tox-db-functional-ubuntu-xenial/9f160ae/testr_results.html.gz | 16:01 |
| *** thorst_ has joined #openstack-nova | 16:02 | |
| dansmith | vladikr: these show the numa or pci filters removing candidate hosts | 16:02 |
| mriedem | sdague: got it | 16:03 |
| *** gszasz has joined #openstack-nova | 16:03 | |
| ftersin | jbernard: but this happened for some other test, which did not fail | 16:03 |
| vladikr | dansmith, looking | 16:03 |
| *** gouthamr has joined #openstack-nova | 16:04 | |
| dansmith | vladikr: this set is complex and I've been chasing my tail, so I'm having a hard time remembering exactly what the other failure looked like | 16:04 |
| *** sacharya has joined #openstack-nova | 16:04 | |
| mriedem | pkoniszewski: ok, so rabbit is the problem.... | 16:04 |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.volume_detach notification https://review.openstack.org/408676 | 16:04 |
| mriedem | pkoniszewski: it seems odd that it has that high of a failure rate though | 16:04 |
| jbernard | ftersin: the prenense of that log message should mean that a failure was avoided | 16:04 |
| mriedem | i.e. if rabbit were just randomly failing, i would expect this to be more rare | 16:05 |
| jbernard | ftersin: else it would have returned ImageBusy | 16:05 |
| jbernard | ftersin: no? | 16:05 |
| *** gouthamr has quit IRC | 16:06 | |
| vladikr | dansmith, yes, I think these are different. the previous were only related to sriov tests and were silently failing on 'nova valid host found' | 16:07 |
| dansmith | vladikr: these are also "no valid host" right? or you mean failing but not because of the scheduler filter? | 16:07 |
| vladikr | dansmith, yes, sorry, it was because the the instance were actually started , while shouldn't be | 16:08 |
| *** sacharya has quit IRC | 16:08 | |
| *** siva_krish has joined #openstack-nova | 16:10 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.volume_attach notification https://review.openstack.org/401992 | 16:10 |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.volume_detach notification https://review.openstack.org/408676 | 16:10 |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Add more field's in InstancePayload https://review.openstack.org/407128 | 16:10 |
| ftersin | jbernard: look at ci on https://review.openstack.org/#/c/213433/ | 16:10 |
| *** Jinxing has quit IRC | 16:10 | |
| ftersin | jbernard: it failed on that test | 16:11 |
| *** jvgrant has quit IRC | 16:11 | |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: Remove check_attach https://review.openstack.org/335358 | 16:11 |
| ftersin | jbernard: and we see appropriate message in logs http://logs.openstack.org/33/213433/10/check/gate-tempest-dsvm-full-devstack-plugin-ceph-ubuntu-xenial/1d1b9dc/logs/screen-c-vol.txt.gz#_2016-12-13_16_40_13_637 | 16:11 |
| *** jvgrant has joined #openstack-nova | 16:11 | |
| *** mingyu has joined #openstack-nova | 16:12 | |
| dansmith | vladikr: so this is failing to boot something we expect should be bootable right? | 16:12 |
| ftersin | jbernard: but we also see the same message for VolumesV1SnapshotTest http://logs.openstack.org/33/213433/10/check/gate-tempest-dsvm-full-devstack-plugin-ceph-ubuntu-xenial/1d1b9dc/logs/screen-c-vol.txt.gz#_2016-12-13_16_42_46_707 | 16:13 |
| dansmith | vladikr: I'm not sure what changed that made this break again | 16:13 |
| mriedem | pkoniszewski: is it possible that we're running this live migration test after we've upgraded and rabbit isn't fully up yet? | 16:13 |
| mriedem | but we don't actually take rabbit down during grenade do we? | 16:13 |
| ftersin | jbernard: which passed though | 16:13 |
| *** rmart04 has quit IRC | 16:13 | |
| jbernard | ftersin: we're not seeing the postpone message | 16:13 |
| jbernard | ftersin: if we had (if the patch was applied), then it would not have failed | 16:13 |
| *** rcernin has quit IRC | 16:14 | |
| jbernard | ftersin: oh wait, you're saying the test passed anyway? | 16:14 |
| ftersin | jbernard: so i suppose that VolumesV1SnapshotTest allows some kind of errors | 16:14 |
| dansmith | vladikr: in the case of test_create_server_with_numa_fails, isn't it trying to create a server that will fail? | 16:14 |
| ftersin | jbernard: yep, i think we must wait for this case hitting exactly by 'ebs' test | 16:15 |
| jbernard | ftersin: or merge it and cross fingers ;) | 16:15 |
| *** moshele has quit IRC | 16:15 | |
| vladikr | dansmith, looking at test_pci_sriov_servers test, it seems that the instance that we have just created is not in a list of existing instances | 16:16 |
| * sfinucan just discovered you can run the same tests under multiple environments with tox using 'tox -e py34,py37 <classpath to tests...>' | 16:16 | |
| dansmith | vladikr: yeah, I'll check on that one, but what about the numa one? | 16:16 |
| vladikr | dansmith, looking | 16:17 |
| *** dave-mccowan has joined #openstack-nova | 16:17 | |
| ftersin | jbernard: heh, you know, if shit can happen, it happens | 16:17 |
| dansmith | vladikr: oh, I think I get it.. | 16:18 |
| dansmith | vladikr: these are building instances expected to go to ERROR state but it doesn't see them there | 16:18 |
| dansmith | vladikr: which is similar to the thing last time I think.. so I must have regressed something | 16:18 |
| *** jaosorior has quit IRC | 16:20 | |
| vladikr | dansmith, yea, test_create_server_with_numa_fails expects the created instance to end up in error state | 16:20 |
| *** slaweq_ has joined #openstack-nova | 16:20 | |
| *** jaosorior has joined #openstack-nova | 16:20 | |
| openstackgerrit | Sujitha proposed openstack/nova: Allow finding instance with invalid regex name https://review.openstack.org/392305 | 16:20 |
| *** mjura has quit IRC | 16:23 | |
| sfinucan | What are folk's thoughts on https://review.openstack.org/#/c/408561/3/tox.ini ? | 16:23 |
| *** bvanhav_ has joined #openstack-nova | 16:24 | |
| vladikr | dansmith, but I'm not sure it's similar to earlier errors, because previously the instance was created and could spawn , instead of getting into an error state | 16:24 |
| sfinucan | Namely, is there any value to running functional tests under both Python 2.7 and Python 3.5 | 16:24 |
| *** bvanhav has quit IRC | 16:24 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: api-ref: Fix description of os-instance-usage-audit-log https://review.openstack.org/410843 | 16:24 |
| mriedem | sdague: that can help some of the confusion on the audit log stuff ^ | 16:24 |
| *** andymaier has quit IRC | 16:24 | |
| *** nicolasbock has quit IRC | 16:25 | |
| vladikr | dansmith, here the problem is that we are creating an instance with created_server = self.api.post_server(post) | 16:25 |
| vladikr | but then it's not in the found_server = self.api.get_server(created_server_id) | 16:25 |
| dansmith | vladikr: right, it's similar in that the instance is created somewhere but compute api isn't showing it in the list, Imean | 16:25 |
| dansmith | vladikr: in this case, it's in cell0 but we're not showing it | 16:26 |
| vladikr | dansmith, ok, yea | 16:26 |
| dansmith | vladikr: dammit, was hoping it was your fault instead of mine :) | 16:26 |
| dansmith | well, the day is still young | 16:26 |
| vladikr | dansmith, I was expecting it to be mine :) | 16:27 |
| vladikr | ;) | 16:27 |
| dansmith | vladikr: I always assume.. :D | 16:27 |
| dansmith | you know, in 'murica, we're innocent until proven guilty | 16:27 |
| *** burgerk has joined #openstack-nova | 16:27 | |
| *** tojuvone has quit IRC | 16:27 | |
| *** Guest52040 has joined #openstack-nova | 16:28 | |
| *** hieulq__ has quit IRC | 16:28 | |
| vladikr | dansmith, :) | 16:29 |
| *** hieulq__ has joined #openstack-nova | 16:29 | |
| *** cmartin has joined #openstack-nova | 16:30 | |
| *** raunak has quit IRC | 16:31 | |
| *** kaisers_ has joined #openstack-nova | 16:31 | |
| *** Oku_OS is now known as Oku_OS-away | 16:32 | |
| *** raunak has joined #openstack-nova | 16:32 | |
| *** mvk has quit IRC | 16:32 | |
| *** _ducttape_ has quit IRC | 16:32 | |
| diana_clarke | hmmm... is anyone looking into the releasenotes job failures? | 16:33 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Remove dead begin/end code from InstanceUsageAuditLogController https://review.openstack.org/410847 | 16:33 |
| *** ducttape_ has joined #openstack-nova | 16:33 | |
| mriedem | diana_clarke: link? | 16:33 |
| diana_clarke | one sec, let me grab one | 16:33 |
| *** dikonoor has quit IRC | 16:34 | |
| mriedem | diana_clarke: looks like https://review.openstack.org/#/c/410839/ is related | 16:35 |
| diana_clarke | mriedem: http://logs.openstack.org/93/386093/20/gate/gate-nova-releasenotes/6e53933/console.html#_2016-12-14_16_19_32_721037 | 16:35 |
| *** nicolasbock has joined #openstack-nova | 16:35 | |
| mriedem | they are talking about it in openstack-release | 16:35 |
| mriedem | yeah stable/liberty was deleted | 16:35 |
| *** andreas_s has quit IRC | 16:35 | |
| diana_clarke | mriedem: ah, okay. thanks! | 16:36 |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.volume_detach notification https://review.openstack.org/408676 | 16:36 |
| *** claudiub has joined #openstack-nova | 16:38 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Use liberty-eol tag for liberty release notes https://review.openstack.org/410851 | 16:39 |
| mriedem | diana_clarke: let's watch this ^ | 16:39 |
| * diana_clarke watching | 16:40 | |
| *** diogogmt has joined #openstack-nova | 16:41 | |
| rpodolyaka | mdbooth: ah, sorry, I missed this at first. In the snippet you provided - http://paste.openstack.org/show/592359/ - line 24 | 16:42 |
| rpodolyaka | it's actually something jaypipes wanted to avoid - a dependent subquery | 16:42 |
| rpodolyaka | the idea was to use a derived table | 16:42 |
| rpodolyaka | mdbooth: here is what I get when using your snippet - http://paste.openstack.org/show/592368/ | 16:43 |
| rpodolyaka | vs refined version of the current patch set - http://paste.openstack.org/show/592369/ | 16:43 |
| mdbooth | rpodolyaka: I was kinda assuming the query optimiser would work out they're the same thing | 16:44 |
| mdbooth | You know... magic | 16:44 |
| rpodolyaka | :) | 16:44 |
| rpodolyaka | I'm afraid it won't | 16:44 |
| mdbooth | rpodolyaka: So, is the performance of the dependent subquery poor? | 16:46 |
| *** rcernin has joined #openstack-nova | 16:46 | |
| rpodolyaka | mdbooth: I'll let jaypipes comment, but my understanding is that you essentially have a nested loop | 16:46 |
| rpodolyaka | for each row in the original select | 16:46 |
| sdague | mriedem: +@ | 16:47 |
| jaypipes | rpodolyaka: right, which is what I wanted to avoid :) | 16:47 |
| *** tongli has quit IRC | 16:48 | |
| *** jaosorior has quit IRC | 16:48 | |
| *** jaosorior has joined #openstack-nova | 16:48 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Remove dead begin/end code from InstanceUsageAuditLogController https://review.openstack.org/410847 | 16:48 |
| mdbooth | rpodolyaka: 2 questions | 16:48 |
| mdbooth | Is it possible to add a join in the if block? | 16:49 |
| *** diogogmt has quit IRC | 16:49 | |
| *** roeyc has quit IRC | 16:49 | |
| rpodolyaka | yes | 16:49 |
| mdbooth | Does it affect performance if the joined query doesn't filter by instance_uuid? | 16:49 |
| * mdbooth would really hope the answer to the second is no | 16:49 | |
| *** diogogmt has joined #openstack-nova | 16:49 | |
| mdbooth | Seems redundant (and potentially error prone) to do that filter twice | 16:50 |
| *** raunak has quit IRC | 16:51 | |
| mdbooth | rpodolyaka: Anyway, I may be in over-analysing territory here :) I just wanted a single query which we conditionally modified. | 16:52 |
| rpodolyaka | mdbooth: yeah, I see your point, but I'm not sure it would filter first and only then perform a join | 16:53 |
| rpodolyaka | here's what I have now | 16:53 |
| rpodolyaka | http://paste.openstack.org/show/592370/ | 16:53 |
| rpodolyaka | and the SQL it generates - http://paste.openstack.org/show/592369/ | 16:53 |
| mdbooth | rpodolyaka: How about if you make the else block unconditional, and remove the same filtering from the joined query? | 16:55 |
| mdbooth | rpodolyaka: I'd hope that would have the same plan | 16:56 |
| rpodolyaka | let me give it a try | 16:56 |
| *** fragatina has joined #openstack-nova | 16:57 | |
| mdbooth | sfinucan jaypipes: In https://review.openstack.org/#/c/403425/5/nova/virt/libvirt/storage/rbd_utils.py I'm looking for something which converts its input to something for which 'isinstance(foo, str)' returns true in both python2 and python3 | 16:59 |
| mdbooth | Whilst remaining meaningful | 17:00 |
| *** diogogmt has quit IRC | 17:00 | |
| sfinucan | mdbooth: six.binary_type looks like the one https://pythonhosted.org/six/#constants | 17:01 |
| *** kaisers_ has quit IRC | 17:02 | |
| mdbooth | sfinucan: How would that be used in this case? | 17:02 |
| *** kaisers_ has joined #openstack-nova | 17:02 | |
| sfinucan | mdbooth: Pretty much the same way, iirc | 17:03 |
| mdbooth | We need a convert function | 17:03 |
| *** Zer0Byte__ has joined #openstack-nova | 17:03 | |
| mdbooth | same way as what? | 17:03 |
| sfinucan | So 'isinstance(foo, six.binary_type)' | 17:03 |
| mdbooth | We can't change that, it's not ours | 17:03 |
| mdbooth | The test is isinstance(foo, str) | 17:03 |
| openstackgerrit | Guillaume Espanel proposed openstack/nova: Reduce time and memory usage of tenant usage. https://review.openstack.org/343734 | 17:03 |
| sfinucan | Oh, I've just seen your comments | 17:04 |
| *** lpetrut has quit IRC | 17:05 | |
| *** sacharya has joined #openstack-nova | 17:05 | |
| *** mingyu has quit IRC | 17:05 | |
| *** hieulq__ has quit IRC | 17:05 | |
| *** slaweq_ has quit IRC | 17:06 | |
| *** Kevin_Zheng has quit IRC | 17:07 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Remove internal attribute access from shell https://review.openstack.org/410867 | 17:07 |
| *** kaisers_ has quit IRC | 17:07 | |
| sfinucan | mdbooth: Hmm, so that code will accept unicode and reject bytestrings in Python 3, while doing the opposite in Python 2 | 17:07 |
| sfinucan | ...which leads me to believe that that code is not Python 3 compatible yet :( | 17:07 |
| mdbooth | sfinucan: That was the result of my testing | 17:07 |
| mdbooth | \o/ | 17:07 |
| sfinucan | Yup | 17:08 |
| mdbooth | That's awesome, because it means the whole effort is pointless :) | 17:08 |
| sfinucan | Exactly | 17:08 |
| sfinucan | So a valid fix for that is to disable all those tests in Python 3 | 17:08 |
| sfinucan | ...and the feature with it | 17:08 |
| sfinucan | IMO, anyway | 17:08 |
| mdbooth | So, 60% of deployments can't use python3... | 17:09 |
| rpodolyaka | mdbooth: so it gives http://paste.openstack.org/show/592377/ but I'm not entirely sure how to interpret it :P | 17:09 |
| sfinucan | That sounds about right | 17:09 |
| sfinucan | Until someone goes and hardens the lib to support Python 3 | 17:09 |
| sfinucan | oh, hold up | 17:10 |
| sfinucan | "This branch is 26473 commits behind ceph:master." | 17:10 |
| sfinucan | that's not the main repo | 17:10 |
| mdbooth | rpodolyaka: I'm going to butt out :) | 17:10 |
| mdbooth | rpodolyaka: I don't know the answer either. | 17:10 |
| * mdbooth would prefer to maximise common code, but not if the resulting performance means it doesn't meet the purpose of the change. | 17:11 | |
| *** nklenke has quit IRC | 17:12 | |
| mdbooth | So, if you're going to duplicate the filter in both branches, for eg, you should probably add a comment against both filters which points out you need to change the other one. | 17:12 |
| mriedem | sdague: jaypipes: dansmith: would you like to unblock all nova changes on master? https://review.openstack.org/#/c/410851/ | 17:12 |
| *** tpatzig_ has joined #openstack-nova | 17:13 | |
| sfinucan | mdbooth: I'm not fluent in pyx (whatever that is) but it /looks/ like it supports Python 3 https://github.com/ceph/ceph/blob/master/src/pybind/rados/rados.pyx#L31-L36 | 17:13 |
| rpodolyaka | mdbooth: ok, will do | 17:13 |
| *** ihrachys has quit IRC | 17:14 | |
| *** nklenke has joined #openstack-nova | 17:14 | |
| *** tpatzig_ has quit IRC | 17:15 | |
| sdague | mriedem: +A | 17:15 |
| sdague | are we getting that promoted? | 17:15 |
| mriedem | sdague: no as it only breaks nova | 17:15 |
| sdague | or does that only impact changes with renos? | 17:15 |
| mriedem | it breaks all nova changes | 17:16 |
| mriedem | but that nova reno job only runs on nova changes | 17:16 |
| mriedem | promos are for cross-project | 17:16 |
| sdague | mriedem: well if it is reseting the gate for everyone it can count | 17:16 |
| sdague | but, that's fine | 17:16 |
| *** diogogmt has joined #openstack-nova | 17:17 | |
| *** tpatzig_ has joined #openstack-nova | 17:17 | |
| openstackgerrit | Gábor Antal proposed openstack/nova: Transform instance.reboot notifications https://review.openstack.org/382959 | 17:18 |
| *** fragatina has quit IRC | 17:19 | |
| *** tpatzig_ has quit IRC | 17:19 | |
| *** tojuvone has joined #openstack-nova | 17:20 | |
| *** bdeschenes has joined #openstack-nova | 17:21 | |
| mriedem | sdague: did you run your -1 abandon script on novaclient too? | 17:24 |
| mriedem | if not, that would be good | 17:24 |
| *** hongbin has quit IRC | 17:25 | |
| *** fragatina has joined #openstack-nova | 17:25 | |
| mriedem | BobBall: is anyone working on fixing the vif plugging timeout failures in the citrix xenserver ci? | 17:26 |
| mriedem | this thing: | 17:26 |
| mriedem | 2016-12-14 04:52:58.046 8342 WARNING nova.virt.xenapi.vmops [req-c913d37c-b9a2-44c9-94ce-b6117fee3c03 tempest-ServersTestManualDisk-2008587945 tempest-ServersTestManualDisk-2008587945] [instance: 980c5069-222f-40fb-afc7-d354c921088a] Timeout waiting for vif plugging callback | 17:26 |
| *** rpodolyaka is now known as rpodolyaka_afk | 17:27 | |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: Only return latest instance fault for instances https://review.openstack.org/409943 | 17:27 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1640993 | 17:27 |
| openstack | Launchpad bug 1640993 in OpenStack Compute (nova) "xenserver hits vif plugging timeout with neutron CI job" [Medium,Confirmed] | 17:27 |
| mriedem | looks like https://review.openstack.org/#/c/390931/ | 17:28 |
| sdague | mriedem: I did not | 17:28 |
| sdague | I can do that now | 17:28 |
| mriedem | BobBall: would it make sense to make vif plugging still wait but not be fatal until this is more stable? | 17:28 |
| *** lucasagomes is now known as lucas-afk | 17:30 | |
| sdague | mriedem: oh, actually, it does novaclient | 17:30 |
| sdague | so yes | 17:30 |
| mriedem | sdague: ok | 17:30 |
| mriedem | thorst_: are you going to be around tomorrow for the nova team meeting? | 17:32 |
| mriedem | i'd like to get an update in there on the powervm third party ci | 17:32 |
| pkoniszewski | mriedem: i don't think that rabbit is not up yet, as the rabbit issue can be observed only on subnode-2 | 17:33 |
| pkoniszewski | mriedem: i haven't found any traces on controller+compute node | 17:33 |
| *** bdeschenes has quit IRC | 17:34 | |
| *** bvanhav__ has joined #openstack-nova | 17:35 | |
| *** bvanhav__ is now known as bvanhav | 17:35 | |
| *** bvanhav_ has quit IRC | 17:36 | |
| *** baoli has joined #openstack-nova | 17:36 | |
| *** mkoderer has quit IRC | 17:36 | |
| pkoniszewski | mriedem: im not very surprised about high failure ratio there, there's the blocking RPC call to subnode-2, that's enough to fail | 17:36 |
| mriedem | pkoniszewski: but then why wouldn't we have a high failure rate in the live-migration specific multinode (non-grenade) job too? | 17:37 |
| pkoniszewski | mriedem: and that's a good question, so far i've observed this issue only in grenade jobs, rabbit is not causing any troubles in non-grenade jobs | 17:39 |
| mriedem | the main differences would be the subnode is newton, | 17:39 |
| mriedem | and with grenade we run other smoke tests, whereas with the live migration specific job we just run live migration tests | 17:40 |
| mriedem | so the grenade job will have a bit more test load | 17:40 |
| mriedem | but maybe the n-1 subnode is part of the problem? | 17:40 |
| bauzas | mriedem: quick q, when are you planning to cut o-2 ? | 17:40 |
| bauzas | thrusday EOB ? | 17:41 |
| mriedem | bauzas: probably end of day today for me | 17:41 |
| thorst_ | mriedem: I can be. Will ask adreznec to be around too | 17:41 |
| bauzas | mriedem: okay, I haven't looked at the relnotes | 17:41 |
| pkoniszewski | e.g. q-agt is tracing in all grenade multinode jobs on subnode-2 | 17:41 |
| pkoniszewski | this does not happen in any non-grenade multinode job | 17:41 |
| bauzas | mriedem: that's a milestone, so that's fine to have some issues with relnotes but I want to make sure we think about those at least for RC1 and the release itself | 17:41 |
| bauzas | mriedem: oh snap, just saw your email (facepalm) | 17:42 |
| *** yamahata_ has quit IRC | 17:42 | |
| pkoniszewski | neutron multinode greande job - http://logs.openstack.org/18/407518/3/check/gate-grenade-dsvm-neutron-multinode-ubuntu-xenial/2b0384e/logs/subnode-2/old/screen-q-agt.txt.gz?level=TRACE | 17:42 |
| pkoniszewski | a lot of traces | 17:42 |
| pkoniszewski | tempest (non-grenade) multinode job - http://logs.openstack.org/18/407518/3/check/gate-tempest-dsvm-neutron-multinode-full-ubuntu-xenial-nv/c83dd1a/logs/subnode-2/screen-q-agt.txt.gz?level=TRACE | 17:43 |
| pkoniszewski | no traces | 17:43 |
| mriedem | bauzas: so one thing about that, i was talking to dansmith and melwitt the other night about docs for cells v2, and we have https://review.openstack.org/#/c/267153/ but it needs updating, or we might just want to start from scratch on that, sort of like the placement docs | 17:43 |
| pkoniszewski | no idea | 17:43 |
| mriedem | bauzas: we have some things in the release notes for newton about upgrading to cells v2, but it's kind of scatter brain, | 17:44 |
| mriedem | bauzas: so now that cellsv2 is required in ocata, i want to get a doc by rc1 where we can point to that in the preface part of the ocata renos | 17:44 |
| bauzas | mriedem: yeah I had the same problem with communicating about the RP being a thing in Newton | 17:44 |
| mriedem | it's less of an issue until the thing is actually required | 17:44 |
| bauzas | mriedem: I think the Newton prelude can help, but we unfortunately merged it in a point release, not 14.0.0 | 17:45 |
| bauzas | mriedem: sure, but the idea behind having it optional and then mandatory is that deployers have one cycle for preparing themselves | 17:45 |
| bauzas | if those people are waiting their CI to break, then it's a problem | 17:46 |
| mriedem | sure, but i also think most deployments aren't doing CD | 17:46 |
| bauzas | having it communicated properly in a release note helps us explaining it's not us badly communicating, it's them not paying attention to what we say :) | 17:46 |
| mriedem | so they should be reading release notes before prepping to pull in a new release of code | 17:46 |
| mriedem | anywho | 17:47 |
| mriedem | we're in agreement either way | 17:47 |
| bauzas | :) | 17:47 |
| bauzas | anywho, I'll review that patch | 17:47 |
| *** asettle has quit IRC | 17:48 | |
| bauzas | I mean, I'll redo that patch | 17:48 |
| mriedem | bauzas: also https://review.openstack.org/#/c/410009/ | 17:48 |
| bauzas | mriedem: that ^ seems interesting to have for ocata-2 | 17:48 |
| *** thorst_ has quit IRC | 17:48 | |
| bauzas | people don't really CD, agreed, but they look at the milestones | 17:49 |
| *** diogogmt has quit IRC | 17:49 | |
| *** unicell has quit IRC | 17:49 | |
| *** thorst_ has joined #openstack-nova | 17:49 | |
| *** ociuhandu has quit IRC | 17:50 | |
| *** kaisers_ has joined #openstack-nova | 17:52 | |
| *** jaosorior has quit IRC | 17:54 | |
| mriedem | melwitt: on https://review.openstack.org/#/c/410009/ | 17:55 |
| mriedem | melwitt: i think we should probably be explicit about the expected cell0 database name | 17:55 |
| mriedem | in grenade/devstack it's nova_api_cell0 | 17:55 |
| melwitt | mriedem: do we prescribe a name for the databases or do operators name them whatever they want? (nova_api_cell0 vs cell0) | 17:55 |
| mriedem | but our release notes just say the cell0 database | 17:55 |
| melwitt | is what I was about to ask | 17:55 |
| mriedem | so https://github.com/openstack/nova/blob/bce8e4b9e378f1bf0d4595c7c3f8b26d86e6527b/nova/cmd/manage.py#L1270 | 17:56 |
| mriedem | by default we use the name of the api db and then append _cell0 to it | 17:56 |
| mriedem | and that's how simple_cell_setup works | 17:56 |
| melwitt | ah, I see. I didn't notice that | 17:56 |
| mriedem | so you could name your nova api db nova-squish-squash and then our reno would be wrong, | 17:56 |
| melwitt | okay, cool. I'll change it. I was just curious about the names because I hadn't thought about it before | 17:56 |
| mriedem | but i'd tend to think we should follow what devstack/grenade use | 17:57 |
| melwitt | makes sense | 17:57 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: placement: adds REST API for nested providers https://review.openstack.org/384807 | 17:57 |
| openstackgerrit | Roman Dobosz proposed openstack/nova: placement: update client to set parent provider https://review.openstack.org/385693 | 17:57 |
| *** ihrachys has joined #openstack-nova | 17:57 | |
| *** Jeffrey4l has quit IRC | 17:58 | |
| *** thorst_ has quit IRC | 17:58 | |
| *** Sukhdev has quit IRC | 17:59 | |
| *** markus_z has quit IRC | 17:59 | |
| *** thorst_ has joined #openstack-nova | 18:00 | |
| *** nic has joined #openstack-nova | 18:01 | |
| *** asettle has joined #openstack-nova | 18:04 | |
| *** ralonsoh_ has quit IRC | 18:04 | |
| *** thorst_ has quit IRC | 18:04 | |
| *** thorst has joined #openstack-nova | 18:06 | |
| *** derekh has quit IRC | 18:09 | |
| *** sudipto has joined #openstack-nova | 18:09 | |
| *** sudipto_ has joined #openstack-nova | 18:09 | |
| *** ihrachys has quit IRC | 18:10 | |
| *** nic has quit IRC | 18:11 | |
| *** thorst has quit IRC | 18:11 | |
| *** ihrachys has joined #openstack-nova | 18:11 | |
| *** raunak has joined #openstack-nova | 18:11 | |
| *** thorst has joined #openstack-nova | 18:12 | |
| *** raunak has quit IRC | 18:13 | |
| *** nic has joined #openstack-nova | 18:14 | |
| *** raunak has joined #openstack-nova | 18:15 | |
| *** tojuvone has quit IRC | 18:16 | |
| *** thorst has quit IRC | 18:17 | |
| *** asettle has quit IRC | 18:17 | |
| *** riccardo_ has joined #openstack-nova | 18:18 | |
| mriedem | johnthetubaguy: if you're still around https://review.openstack.org/#/c/404456/ | 18:19 |
| *** Guest89456 has quit IRC | 18:21 | |
| *** lpetrut has joined #openstack-nova | 18:21 | |
| *** david-lyle has quit IRC | 18:21 | |
| johnthetubaguy | mriedem: dang, I meant to hit that before I was away, looks good | 18:21 |
| *** yamahata_ has joined #openstack-nova | 18:22 | |
| mriedem | thanks | 18:22 |
| *** thorst has joined #openstack-nova | 18:23 | |
| *** riccardo_ has quit IRC | 18:23 | |
| mriedem | jaypipes: https://review.openstack.org/#/c/398469/9/nova/compute/resource_tracker.py@202 | 18:25 |
| mriedem | jaypipes: doesn't vcenter also have some node sillies too? | 18:26 |
| *** unicell has joined #openstack-nova | 18:28 | |
| mriedem | jaypipes: or maybe that's dead since liberty with https://review.openstack.org/#/c/103916/ | 18:29 |
| *** markvoelker has quit IRC | 18:29 | |
| *** tjones has joined #openstack-nova | 18:29 | |
| *** nklenke has quit IRC | 18:30 | |
| *** thorst has quit IRC | 18:30 | |
| *** david-lyle has joined #openstack-nova | 18:30 | |
| mriedem | tjones: is vmware ci totally dead? | 18:32 |
| tjones | right now everything is dead | 18:33 |
| tjones | not just external but interal | 18:33 |
| mriedem | as in.... | 18:33 |
| tjones | we are wroking it | 18:33 |
| mriedem | bodies in the hallway? | 18:33 |
| mriedem | ok | 18:33 |
| tjones | lol - ok not that bad but all of our CI is down so we are running around like mad | 18:33 |
| tjones | will update you shortly | 18:34 |
| mriedem | thanks | 18:34 |
| tjones | sure sorry (again and again) | 18:35 |
| *** markvoelker has joined #openstack-nova | 18:35 | |
| *** sudipto_ has quit IRC | 18:35 | |
| *** sudipto has quit IRC | 18:35 | |
| *** thorst has joined #openstack-nova | 18:36 | |
| *** thorst has quit IRC | 18:36 | |
| *** nklenke has joined #openstack-nova | 18:37 | |
| jaypipes | mriedem: yeah, I don't believe the vCenter driver relies on the Nova scheduler for picking a node within the vCenter cluster. | 18:37 |
| *** thorst has joined #openstack-nova | 18:37 | |
| jaypipes | mriedem: but I could be wrong :) | 18:37 |
| mriedem | +2 either way | 18:38 |
| tjones | no it does not. | 18:38 |
| *** ducttape_ has quit IRC | 18:39 | |
| *** lpetrut has quit IRC | 18:39 | |
| openstackgerrit | int32bit proposed openstack/python-novaclient: Add CLI to show instance usage audit logs https://review.openstack.org/409413 | 18:40 |
| openstackgerrit | Sujitha proposed openstack/nova: Change host for osapi and metadata services in db. https://review.openstack.org/384724 | 18:42 |
| *** ijw has joined #openstack-nova | 18:45 | |
| mriedem | jaypipes: looks like you might have some unnecessary churn in here https://review.openstack.org/#/c/398470/9 | 18:47 |
| *** Guest52040 has quit IRC | 18:48 | |
| *** dharinic is now known as dharinic|lunch | 18:49 | |
| *** ijw has quit IRC | 18:50 | |
| *** ducttape_ has joined #openstack-nova | 18:50 | |
| *** lpetrut has joined #openstack-nova | 18:51 | |
| *** ijw has joined #openstack-nova | 18:56 | |
| *** ijw has quit IRC | 18:57 | |
| *** ijw has joined #openstack-nova | 18:58 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: libvirt: Remove slowpath listing of instances https://review.openstack.org/410287 | 18:58 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: libvirt: Flatten 'get_domain' function https://review.openstack.org/290706 | 18:58 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: libvirt: Stop misusing NovaException https://review.openstack.org/410288 | 18:58 |
| openstackgerrit | Stephen Finucane proposed openstack/nova: fakelibvirt: Remove unused functions https://review.openstack.org/410921 | 18:58 |
| *** ihrachys has quit IRC | 19:00 | |
| *** kaisers_ has quit IRC | 19:02 | |
| *** lamt has joined #openstack-nova | 19:03 | |
| *** thorst has quit IRC | 19:03 | |
| *** kaisers_ has joined #openstack-nova | 19:04 | |
| jaypipes | mriedem: answered. | 19:04 |
| *** ijw has quit IRC | 19:04 | |
| *** hongbin has joined #openstack-nova | 19:05 | |
| *** kaisers__ has joined #openstack-nova | 19:05 | |
| *** kaisers_ has quit IRC | 19:05 | |
| *** ihrachys has joined #openstack-nova | 19:06 | |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: expose virtual interfaces with vlans to metadata https://review.openstack.org/405954 | 19:06 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: neutron: update vlan in virtual_interface object https://review.openstack.org/405953 | 19:06 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Parse os attributes of LibvirtConfigGuest from xml https://review.openstack.org/410665 | 19:07 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Parse filesystem elements of guest config https://review.openstack.org/410666 | 19:07 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Use config types to parse XML for instance disks https://review.openstack.org/410667 | 19:07 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Simplify internal usage of get_instance_disk_info https://review.openstack.org/334901 | 19:07 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Remove unused disk_info parameter https://review.openstack.org/382025 | 19:07 |
| *** ihrachys has quit IRC | 19:09 | |
| *** jgriffith is now known as jgriffith_AutoAw | 19:10 | |
| *** liverpooler has quit IRC | 19:10 | |
| *** amoralej is now known as amoralej|off | 19:11 | |
| openstackgerrit | Lee Yarwood proposed openstack/nova: libvirt: Detach volumes from a domain before detaching any encryptors https://review.openstack.org/398479 | 19:11 |
| *** liverpooler has joined #openstack-nova | 19:12 | |
| openstackgerrit | Lee Yarwood proposed openstack/nova: libvirt: Reconnect volumes and encryptors during a hard reboot https://review.openstack.org/400384 | 19:14 |
| *** marst has quit IRC | 19:16 | |
| *** marst has joined #openstack-nova | 19:16 | |
| *** nklenke has quit IRC | 19:16 | |
| *** thorst has joined #openstack-nova | 19:16 | |
| *** nklenke has joined #openstack-nova | 19:17 | |
| *** thorst has quit IRC | 19:17 | |
| *** Sukhdev has joined #openstack-nova | 19:20 | |
| *** siva_krish has quit IRC | 19:22 | |
| *** dharinic|lunch is now known as dharinic | 19:23 | |
| openstackgerrit | Stephen Finucane proposed openstack/nova: libvirt: Stop misusing NovaException https://review.openstack.org/410288 | 19:25 |
| *** siva_krish has joined #openstack-nova | 19:25 | |
| *** thorst has joined #openstack-nova | 19:26 | |
| *** thorst has quit IRC | 19:26 | |
| *** thorst has joined #openstack-nova | 19:28 | |
| *** lpetrut has quit IRC | 19:28 | |
| *** thorst has quit IRC | 19:31 | |
| *** roeyc has joined #openstack-nova | 19:32 | |
| *** Swami has quit IRC | 19:33 | |
| *** Swami has joined #openstack-nova | 19:34 | |
| *** nic has quit IRC | 19:35 | |
| *** fragatina has quit IRC | 19:35 | |
| *** fragatina has joined #openstack-nova | 19:36 | |
| *** nic has joined #openstack-nova | 19:37 | |
| *** fragatina has quit IRC | 19:38 | |
| *** thorst has joined #openstack-nova | 19:38 | |
| mriedem | jaypipes: when you say self.compute_node do you mean self.nodename? | 19:39 |
| mriedem | i'm also not sure how that applies to _update_usage_from_instance which isn't using self.nodename today | 19:40 |
| mriedem | and isn't using the nodename parameter you added | 19:40 |
| mriedem | ah but here https://review.openstack.org/#/c/398473/12/nova/compute/resource_tracker.py@897 | 19:40 |
| jaypipes | mriedem: no, I mean self.compute_node | 19:43 |
| *** sdague has quit IRC | 19:43 | |
| jaypipes | mriedem: so, a few patches later, we go from the resource tracker having a *single* ComputeNode object in memory to instead having a map of COmputeNode objects, keyed by hypervisor_hostname (nodename) | 19:44 |
| mriedem | yar | 19:44 |
| jaypipes | mriedem: as mentioned in the comment on there, I didn't want to make the patch huge :( | 19:44 |
| jaypipes | mriedem: because of test changes, trying to do that all in a single patch ballooned shit to >1K LOCC | 19:45 |
| nicolasbock | Hi guys, I am trying to use the nova-docker driver | 19:45 |
| nicolasbock | I set compute_driver = novadocker.virt.docker.DockerDriver | 19:45 |
| jaypipes | mriedem: and made it very difficult to track (even more difficult than it already is!) | 19:45 |
| mriedem | jaypipes: yeah i get it | 19:45 |
| mriedem | needed the context of the follow on patches | 19:45 |
| *** kaisers__ has quit IRC | 19:45 | |
| nicolasbock | but because of nova/virt/driver.py:1614 | 19:45 |
| nicolasbock | 'nova.virt.%s' % compute_driver, | 19:46 |
| nicolasbock | oslo tries to load nova.virt.novadocker.virt.docker | 19:46 |
| *** sdague has joined #openstack-nova | 19:46 | |
| mriedem | nicolasbock: this channel doesn't support nova-docker, you should probably ask in the #nova-docker channel | 19:46 |
| *** nklenke has quit IRC | 19:46 | |
| nicolasbock | mriedem, thanks, I didn't realize there is a separate channel for that | 19:46 |
| mriedem | nova-docker is also basically no longer maintained afaik | 19:46 |
| *** nklenke has joined #openstack-nova | 19:49 | |
| openstackgerrit | Dan Smith proposed openstack/nova: Make servers api view load instance fault from proper cell https://review.openstack.org/407238 | 19:50 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add schedule_and_build_instances conductor method https://review.openstack.org/367557 | 19:50 |
| openstackgerrit | Dan Smith proposed openstack/nova: Multicell support for instance listing https://review.openstack.org/396775 | 19:50 |
| openstackgerrit | Dan Smith proposed openstack/nova: Fix non-parameterized service id in hypervisors sample tests https://review.openstack.org/410471 | 19:50 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make the SingleCellSimple fixture a little more comprehensive https://review.openstack.org/410472 | 19:50 |
| openstackgerrit | Dan Smith proposed openstack/nova: WIP: Make api_samples tests use simple cell environment https://review.openstack.org/410473 | 19:50 |
| openstackgerrit | Dan Smith proposed openstack/nova: WIP Move instance creation to conductor https://review.openstack.org/319379 | 19:50 |
| *** liangy has quit IRC | 19:52 | |
| *** raunak has quit IRC | 19:52 | |
| openstackgerrit | Maciej Szankin proposed openstack/nova: conf: move few console opts to xenserver group https://review.openstack.org/389647 | 19:53 |
| *** liangy has joined #openstack-nova | 19:55 | |
| *** raunak has joined #openstack-nova | 19:57 | |
| * jaypipes digs in to review dansmith's series. | 19:57 | |
| *** raunak has quit IRC | 19:58 | |
| *** raunak has joined #openstack-nova | 19:58 | |
| *** ayoung has quit IRC | 19:59 | |
| mriedem | jaypipes: ok, meeting time, left a couple more comments but need to come back to that one | 19:59 |
| *** Swami_ has joined #openstack-nova | 20:00 | |
| jaypipes | mriedem: no worries, thanks for the reviews. | 20:00 |
| *** Swami has quit IRC | 20:00 | |
| *** jgriffith_AutoAw is now known as jgriffith | 20:02 | |
| *** siva_krish has quit IRC | 20:07 | |
| *** gszasz has quit IRC | 20:07 | |
| *** siva_krish has joined #openstack-nova | 20:10 | |
| *** ociuhandu has joined #openstack-nova | 20:15 | |
| bauzas | dansmith: planning to run the cells meeting today? | 20:16 |
| dansmith | yeah | 20:17 |
| bauzas | dansmith: if so, just highlight me by that time, please :) | 20:17 |
| dansmith | bauzas: I don't do that.. get a calendar :) | 20:17 |
| melwitt | dansmith likes to use large ping lists for meetings | 20:17 |
| bauzas | heh okay | 20:17 |
| *** moshele has joined #openstack-nova | 20:18 | |
| *** asettle has joined #openstack-nova | 20:18 | |
| *** asettle has quit IRC | 20:22 | |
| *** ekuris has quit IRC | 20:24 | |
| *** stvnoyes has quit IRC | 20:24 | |
| *** stvnoyes has joined #openstack-nova | 20:24 | |
| *** liangy has quit IRC | 20:30 | |
| *** Swami_ has quit IRC | 20:30 | |
| openstackgerrit | melanie witt proposed openstack/nova: WIP Make Quotas object favor the API database https://review.openstack.org/410945 | 20:32 |
| openstackgerrit | melanie witt proposed openstack/nova: WIP Add online migration to move quotas to API database https://review.openstack.org/410946 | 20:32 |
| *** gouthamr has joined #openstack-nova | 20:33 | |
| *** pcaruana has quit IRC | 20:33 | |
| *** ociuhandu has quit IRC | 20:35 | |
| *** moshele has quit IRC | 20:36 | |
| *** ociuhandu has joined #openstack-nova | 20:37 | |
| *** bvanhav has quit IRC | 20:37 | |
| *** StevenK has quit IRC | 20:38 | |
| *** roeyc has quit IRC | 20:38 | |
| *** StevenK has joined #openstack-nova | 20:38 | |
| *** ociuhandu has quit IRC | 20:41 | |
| *** rtheis has quit IRC | 20:42 | |
| jaypipes | gibi: mind reading the comments between dansmith and melwitt about notifications impact on line 836 here? https://review.openstack.org/#/c/367557/34/nova/conductor/manager.py | 20:45 |
| *** ayoung has joined #openstack-nova | 20:46 | |
| *** nklenke has quit IRC | 20:46 | |
| *** tlian has quit IRC | 20:46 | |
| *** diogogmt has joined #openstack-nova | 20:46 | |
| *** mvk has joined #openstack-nova | 20:46 | |
| *** ayoung has quit IRC | 20:49 | |
| *** thorst has quit IRC | 20:51 | |
| *** ociuhandu has joined #openstack-nova | 20:52 | |
| *** jgriffith is now known as jgriffith_AutoAw | 20:53 | |
| *** ociuhandu has quit IRC | 20:56 | |
| *** dtp has joined #openstack-nova | 20:58 | |
| *** _ducttape_ has joined #openstack-nova | 20:58 | |
| *** tbachman has quit IRC | 21:01 | |
| *** ducttape_ has quit IRC | 21:02 | |
| *** lamt has quit IRC | 21:02 | |
| mriedem | omfg it's started | 21:02 |
| *** thorst has joined #openstack-nova | 21:06 | |
| jaypipes | mriedem: what? | 21:08 |
| mriedem | cells meeting bro! | 21:09 |
| *** _ducttape_ has quit IRC | 21:09 | |
| mriedem | i wasn't even notified personally | 21:09 |
| *** ducttape_ has joined #openstack-nova | 21:10 | |
| *** Swami has joined #openstack-nova | 21:15 | |
| *** jgriffith_AutoAw is now known as jgriffith | 21:18 | |
| *** r-daneel has joined #openstack-nova | 21:20 | |
| *** asettle has joined #openstack-nova | 21:21 | |
| *** smatzek has quit IRC | 21:22 | |
| openstackgerrit | Sujitha proposed openstack/nova: Change os-server-tags default policy https://review.openstack.org/396420 | 21:22 |
| *** tbachman has joined #openstack-nova | 21:23 | |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Parse os attributes of LibvirtConfigGuest from xml https://review.openstack.org/410665 | 21:24 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Parse filesystem elements of guest config https://review.openstack.org/410666 | 21:24 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Use config types to parse XML for instance disks https://review.openstack.org/410667 | 21:24 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Simplify internal usage of get_instance_disk_info https://review.openstack.org/334901 | 21:24 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: libvirt: Remove unused disk_info parameter https://review.openstack.org/382025 | 21:24 |
| *** catintheroof has joined #openstack-nova | 21:24 | |
| *** asettle has quit IRC | 21:25 | |
| *** erlon has quit IRC | 21:30 | |
| *** liverpooler has quit IRC | 21:30 | |
| *** nherciu_ has quit IRC | 21:33 | |
| *** liverpooler has joined #openstack-nova | 21:35 | |
| *** slaweq_ has joined #openstack-nova | 21:38 | |
| *** jgriffith is now known as jgriffith_AutoAw | 21:40 | |
| *** liverpooler has quit IRC | 21:40 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: [WIP] Add service_token for nova-cinder interaction https://review.openstack.org/397399 | 21:40 |
| *** jgriffith_AutoAw is now known as jgriffith | 21:45 | |
| mriedem | hmm, so we deprecated the quota_driver config option in newton, we could constrain that to just noop or db | 21:46 |
| mriedem | so we can't do classloading anymore | 21:46 |
| melwitt | mriedem: that's not what quota classes are though, right? or are you talking about something else | 21:47 |
| mriedem | melwitt: i'm just pulling threads | 21:47 |
| melwitt | ah | 21:47 |
| mriedem | for out of tree quota things | 21:47 |
| melwitt | gotcha | 21:47 |
| mriedem | it's also weird that these drivers don't extend an ABC | 21:48 |
| dtp | sdague are you here? | 21:50 |
| mriedem | melwitt: so as much as i can tell from the rest api, there isn't a way to associate a tenant with a different quota class | 21:51 |
| mriedem | Vek: are you aware of anything in the upstream compute API that allows you to associate a quota class with a tenant? | 21:52 |
| melwitt | that must be how it doesn't work then. odd that isn't possible | 21:52 |
| mriedem | melwitt: well it might be possible in a rax api | 21:53 |
| mriedem | but not in nova | 21:53 |
| mriedem | from what i can tell | 21:53 |
| mriedem | johnthetubaguy isn't around anymore probably, is there anyone else here that works at rackspace? | 21:54 |
| mriedem | mikal: ? | 21:54 |
| *** thorst has quit IRC | 21:56 | |
| *** shaner has joined #openstack-nova | 21:57 | |
| *** nherciu_ has joined #openstack-nova | 21:57 | |
| *** thorst_ has joined #openstack-nova | 21:59 | |
| dtp | melwitt dansmith all the various console types will need instance_uuid added to url, right? | 22:00 |
| melwitt | mriedem: looks like the association is made through the RequestContext.quota_class. so now to see if anything ever sets that. probably not in nova but somehow rax could have a paste middleware that sets it or something? | 22:01 |
| mriedem | melwitt: that's what turnstile did | 22:01 |
| mriedem | which was the middleware thing | 22:01 |
| *** tjones has quit IRC | 22:02 | |
| mriedem | https://lists.launchpad.net/openstack/msg12200.html | 22:02 |
| melwitt | dtp: yes. each console type will need to be able to contact the cell where the instance is, and we lookup the cell via instance uuid (or host) | 22:02 |
| melwitt | mriedem: okay, I think I get it now | 22:03 |
| mriedem | melwitt: ok, so i'm going to start trying to unwind some of this and remove it | 22:03 |
| *** slaweq_ has quit IRC | 22:04 | |
| mriedem | melwitt: i think for now though we'd be safe to not migrate quota_classes to the api db | 22:04 |
| melwitt | mriedem: sweet. I shall rebase my patches on yours then | 22:04 |
| mriedem | melwitt: well, it's going to be awhile probably | 22:05 |
| mriedem | and this is going to be a gut job | 22:05 |
| mriedem | i think i'll start by making quota_driver config a choices opt, unrelated but it's something i'd like to not forget about | 22:05 |
| *** slaweq_ has joined #openstack-nova | 22:05 | |
| melwitt | mriedem: okay. I'll work on updating to not move quota classes at all. need to also figure out what's needed, if anything, to replace what the default quota class does | 22:06 |
| mriedem | i believe the default quota class just reads from config if you haven't changed it in the api | 22:07 |
| mriedem | i.e. with a fresh install you can do 'nova quota-class-show default' and it will give you things, based on the config | 22:07 |
| mriedem | but nova.quota_classes table is empty in the db | 22:07 |
| mriedem | if you do 'nova quota-class-update --instances 1 default' | 22:07 |
| melwitt | right. I didn't realize it was default quota class doing that | 22:07 |
| mriedem | then you get a 'default' quota_class in the db with just that hard-limit set | 22:07 |
| mriedem | for instances | 22:07 |
| melwitt | ah, okay | 22:08 |
| *** cleong has quit IRC | 22:08 | |
| mriedem | yup, and if you change the conf default for instances, you'll still get whatever is in the db | 22:08 |
| melwitt | oh, actually I was thinking of the default quota limits | 22:09 |
| mriedem | yeah i guess it's really the limit | 22:09 |
| mriedem | mysql> select * from nova.quota_classes; | 22:09 |
| mriedem | +---------------------+------------+------------+----+-----------------+-----------+------------+---------+ | 22:09 |
| mriedem | | created_at | updated_at | deleted_at | id | class_name | resource | hard_limit | deleted | | 22:09 |
| mriedem | +---------------------+------------+------------+----+-----------------+-----------+------------+---------+ | 22:09 |
| mriedem | | 2016-12-14 21:36:24 | NULL | NULL | 1 | single-instance | instances | 1 | 0 | | 22:09 |
| mriedem | | 2016-12-14 22:08:13 | NULL | NULL | 2 | default | instances | 20 | 0 | | 22:09 |
| mriedem | +---------------------+------------+------------+----+-----------------+-----------+------------+---------+ | 22:09 |
| mriedem | if there isn't a resource with a hard limit for a given quota class, the quota engine code fills that in based on the config | 22:10 |
| mriedem | and given you can't assign a non-default quota class to a tenant, you'll only ever get the default, from config or db if you override that via api | 22:10 |
| mriedem | it's super useful o-) | 22:10 |
| *** ayoung has joined #openstack-nova | 22:11 | |
| melwitt | right. I'm thinking I'll need to change the behavior to update the quotas and project_user_quotas tables instead | 22:11 |
| melwitt | for rest api quota updates | 22:11 |
| melwitt | and migrate people's default quota class entries to those tables | 22:13 |
| melwitt | I'll do some testing with devstack to see what all gets updated by the api and needs to be migrated/translated over | 22:14 |
| melwitt | 'nova quota-update' I mean | 22:15 |
| *** lamt has joined #openstack-nova | 22:16 | |
| bauzas | dansmith: I'm fine with https://review.openstack.org/#/c/404279/3 after reviewing a 3rd time, do you basically agree ? | 22:17 |
| melwitt | from the code it looks like updating limits has nothing to do with the quota classes | 22:17 |
| mriedem | melwitt: as in os-quota-sets:update? | 22:17 |
| dansmith | bauzas: the only reason I didn't +2 before was to avoid clashing with your -1, but I think I've said several times that I am, so.. YES :) | 22:18 |
| bauzas | okay :) | 22:18 |
| *** tbachman has quit IRC | 22:18 | |
| melwitt | mriedem: yeah. that's the api for changing quota limits. the challenge though will be if someone updated their default quota class and that somehow overrides what's in the normal quota limits | 22:18 |
| *** tbachman has joined #openstack-nova | 22:19 | |
| mriedem | oh...so if we don't migrate the quota_classes table, we lose that | 22:19 |
| *** catintheroof has quit IRC | 22:19 | |
| mriedem | because you'd have entries in there for the 'default' quota class | 22:19 |
| mriedem | so basically we have an entire table just for a single class of quota, which is 'default' | 22:19 |
| melwitt | yeah. that's why I was thinking, if we want to drop quota_classes table, we need to be able to take what's in the default quota class and update the quotas table with it, during the online migration | 22:20 |
| mriedem | hmm, so we probably do need the quota_classes table, at least until we have a replacement for that logic, | 22:20 |
| melwitt | maybe. I'm not 100% clear on how quotas and quota_classes tables work together yet | 22:21 |
| mriedem | which i'd think is just setting the hard_limit in project_user_quotas | 22:21 |
| mriedem | yeah this is gd terrible | 22:22 |
| melwitt | yeah, that's what I'm thinking is it would just be copying those limits over from one table to the others | 22:22 |
| melwitt | yeah | 22:22 |
| *** cmartin has quit IRC | 22:25 | |
| *** annegentle has quit IRC | 22:25 | |
| mriedem | melwitt: ok yeah so i did 'nova quota-update --instances -1 <admin tenant id>' and then i get an entry in nova.quotas table for that resource and tenant | 22:25 |
| mriedem | mysql> select * from nova.quotas; | 22:25 |
| mriedem | +----+---------------------+------------+------------+----------------------------------+-----------+------------+---------+ | 22:25 |
| mriedem | | id | created_at | updated_at | deleted_at | project_id | resource | hard_limit | deleted | | 22:25 |
| mriedem | +----+---------------------+------------+------------+----------------------------------+-----------+------------+---------+ | 22:25 |
| mriedem | | 1 | 2016-12-14 22:24:00 | NULL | NULL | cc495759b32b49a7b7dfd321f22cc680 | instances | -1 | 0 | | 22:25 |
| mriedem | +----+---------------------+------------+------------+----------------------------------+-----------+------------+---------+ | 22:25 |
| mriedem | 1 row in set (0.01 sec) | 22:25 |
| mriedem | otherwise they come from nova.quota_classes for the default quota i think, | 22:26 |
| mriedem | and if that's not set, then config | 22:26 |
| melwitt | mriedem: yeah, that's the thing I'm aware of. the question is, if you change the default quota class, does it change your quota? | 22:26 |
| mriedem | let me see | 22:26 |
| *** annegentle has joined #openstack-nova | 22:26 | |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Add more field's in InstancePayload https://review.openstack.org/407128 | 22:26 |
| melwitt | I have a bad feeling it's like you say, first the quotas table is used, then falls back onto default quota class, then falls back onto config | 22:27 |
| mriedem | melwitt: changed default instances quota to 50, but for this tenant i still get back -1 | 22:27 |
| mriedem | i changed --cores default to 40 and i get that back for the tenant | 22:27 |
| mriedem | but for this tenant i don't have a cores limit set in nova.quotas | 22:28 |
| mriedem | so it's coming from nova.quota_classes | 22:28 |
| *** liangy has joined #openstack-nova | 22:28 | |
| melwitt | yep, so you're right. it's a cascade | 22:28 |
| mriedem | mysql> select * from nova.quota_classes where class_name='default'; | 22:28 |
| mriedem | +---------------------+---------------------+------------+----+------------+-----------+------------+---------+ | 22:28 |
| mriedem | | created_at | updated_at | deleted_at | id | class_name | resource | hard_limit | deleted | | 22:28 |
| mriedem | +---------------------+---------------------+------------+----+------------+-----------+------------+---------+ | 22:28 |
| mriedem | | 2016-12-14 22:08:13 | 2016-12-14 22:26:38 | NULL | 2 | default | instances | 50 | 0 | | 22:28 |
| mriedem | | 2016-12-14 22:27:35 | NULL | NULL | 3 | default | cores | 40 | 0 | | 22:28 |
| mriedem | +---------------------+---------------------+------------+----+------------+-----------+------------+---------+ | 22:28 |
| mriedem | and if i change default ram quota in config, it'll come from there | 22:28 |
| *** annegentle has quit IRC | 22:29 | |
| mriedem | b/c i don't have ram set in nova.quotas per tenant, or nova.quota_classes per 'default' | 22:29 |
| melwitt | now I'm thinking about new tenants, where do they pick up their starting quota from. probably quota_classes | 22:29 |
| mriedem | probably | 22:29 |
| mriedem | i think default quota_classes are global | 22:29 |
| melwitt | so that might break us. I don't think we can replicate that otherwise | 22:30 |
| melwitt | that is, we might have to have the quota_classes table with one thing in it, the defaults | 22:30 |
| *** tblakes has quit IRC | 22:34 | |
| melwitt | and if we remove quota classes, that will take away the ability for people to update the default class, if they were doing things that way | 22:34 |
| melwitt | we don't have api ref docs for the quota class api | 22:35 |
| mriedem | melwitt: yeah that's another bug | 22:35 |
| mriedem | which started some of this | 22:35 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1602400 | 22:36 |
| openstack | Launchpad bug 1602400 in OpenStack Compute (nova) "os-quota-class-sets APIs are undocumented" [Low,Confirmed] | 22:36 |
| *** browne has joined #openstack-nova | 22:37 | |
| melwitt | mriedem: if we deprecate quota classes we'll need in the relnotes to say something like "copy any custom default quota class values to your nova.conf" | 22:37 |
| mriedem | melwitt: we could drop the quota_classes table if we just read default quotas from the config, but that breaks the api | 22:37 |
| mriedem | right | 22:37 |
| melwitt | yeah | 22:38 |
| mriedem | we could deprecate the quota classes API, but then we'd still have to support it for the older microversions, which rely on the table | 22:38 |
| melwitt | I don't think we can break existing users that have done this until at least next cycle right | 22:38 |
| mriedem | i don't really see how we get around the api compat issue | 22:39 |
| *** jgriffith is now known as jgriffith_AutoAw | 22:39 | |
| mriedem | w/o the table | 22:39 |
| melwitt | yeah, I wasn't thinking of the compat part. we have the same problem with nova-network though, right? | 22:39 |
| mriedem | well, nova-network is special | 22:39 |
| melwitt | when we remove it, no more older microversions will work | 22:39 |
| mriedem | like, short bus special | 22:39 |
| melwitt | lol | 22:39 |
| melwitt | okay | 22:39 |
| mriedem | yeah i think we've just admitted that we're breaking those apis when nova-network is gone | 22:39 |
| mriedem | and it sucks but it's a thing | 22:40 |
| melwitt | yeah. either way, for now we have to have the table. even if we could break compat someday, we wouldn't be able to in ocata | 22:41 |
| *** pradk has quit IRC | 22:42 | |
| *** ayoung is now known as ayoung_dadmode | 22:43 | |
| *** baoli has quit IRC | 22:44 | |
| *** gouthamr has quit IRC | 22:45 | |
| *** marst has quit IRC | 22:45 | |
| *** esberglu has quit IRC | 22:47 | |
| *** Swami has quit IRC | 22:50 | |
| *** jgriffith_AutoAw is now known as jgriffith | 22:51 | |
| openstackgerrit | melanie witt proposed openstack/nova: WIP Make Quotas object favor the API database https://review.openstack.org/410945 | 22:51 |
| openstackgerrit | melanie witt proposed openstack/nova: WIP Add online migration to move quotas to API database https://review.openstack.org/410946 | 22:51 |
| *** ducttape_ has quit IRC | 22:51 | |
| *** slaweq_ has quit IRC | 22:54 | |
| *** burgerk has quit IRC | 22:55 | |
| *** dave-mccowan has quit IRC | 22:56 | |
| *** gouthamr has joined #openstack-nova | 22:57 | |
| *** vladikr has quit IRC | 22:57 | |
| *** eharney has quit IRC | 23:00 | |
| *** esberglu has joined #openstack-nova | 23:01 | |
| *** avolkov` has quit IRC | 23:02 | |
| *** jheroux has quit IRC | 23:02 | |
| *** thorst_ has quit IRC | 23:04 | |
| *** rcernin has quit IRC | 23:05 | |
| melwitt | mriedem: what do you think of this xenapi config change? it looks fine to me but I'd like your opinion before I +2 it https://review.openstack.org/#/c/299092/ | 23:06 |
| *** esberglu has quit IRC | 23:06 | |
| mriedem | hmmm | 23:08 |
| mriedem | knee jerk reaction was that's not following a deprecation type policy, but it's not really possible to straddle that with a deprecation since we're not depracating the option | 23:08 |
| mriedem | and you can't check the value in the code and say 'if xapi1: log warning' because xapi1 might be legit | 23:09 |
| dansmith | mriedem: I think you mean "deprecating" | 23:09 |
| mriedem | yes i do | 23:09 |
| mriedem | sorry | 23:09 |
| melwitt | yeah | 23:09 |
| mriedem | dansmith: irregardless of how i spell depracating | 23:09 |
| dansmith | heh | 23:09 |
| mriedem | melwitt: looking at their CI http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/92/299092/19/check/dsvm-tempest-neutron-network/afa7936/logs/etc/nova/nova.conf.txt.gz | 23:10 |
| mriedem | they use xapi4 | 23:10 |
| mriedem | i don't really know why xapi1 is almost always wrong though | 23:10 |
| mriedem | is that like, if you're using xen from 1998? | 23:10 |
| dansmith | more like 2008 | 23:11 |
| melwitt | hah yeah. I guess I assumed it must be more harm than good to have the default based on how the whole subteam was reviewing it | 23:11 |
| mriedem | i posted a question in there | 23:12 |
| *** burt has quit IRC | 23:12 | |
| mriedem | they should have probably provided more background in the commit message about why it's almost never correct | 23:12 |
| dansmith | well, 4 is clearly 3 better than 1 | 23:12 |
| dansmith | I mean, come on | 23:12 |
| mriedem | wrong | 23:13 |
| mriedem | pumpkinhead 4 is clearly not better than 1 | 23:13 |
| mriedem | or tremors 4 | 23:13 |
| mriedem | i mean, c'mon | 23:13 |
| mriedem | if it doesn't have kevin bacon in it, i don't watch it | 23:13 |
| nic | ^ a sound strategy | 23:14 |
| melwitt | do you have an alert for bacon? | 23:14 |
| diana_clarke | lol | 23:14 |
| *** mtanino has quit IRC | 23:15 | |
| *** thorst_ has joined #openstack-nova | 23:15 | |
| *** mtanino has joined #openstack-nova | 23:15 | |
| *** dave-mccowan has joined #openstack-nova | 23:15 | |
| *** jschwarz has quit IRC | 23:16 | |
| *** mlavalle has quit IRC | 23:16 | |
| *** nic has quit IRC | 23:16 | |
| *** migi has quit IRC | 23:17 | |
| mriedem | i imagine nic has left to relive some classic kevin bacon silver screen moments | 23:17 |
| *** migi has joined #openstack-nova | 23:18 | |
| dansmith | mriedem: and if not, we probably don't want to know | 23:19 |
| mriedem | https://www.youtube.com/watch?v=j8XGmZ8HDIU | 23:19 |
| mriedem | +1 on the vaseline lense, -1 on the smoking | 23:19 |
| mriedem | i guess he is the bad boy in that movie | 23:19 |
| *** nic has joined #openstack-nova | 23:20 | |
| *** yamahata_ has quit IRC | 23:21 | |
| *** jschwarz has joined #openstack-nova | 23:21 | |
| *** asettle has joined #openstack-nova | 23:22 | |
| *** ijw has joined #openstack-nova | 23:22 | |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Disassociate floating ip when fixed ip is removed https://review.openstack.org/350299 | 23:24 |
| *** ducttape_ has joined #openstack-nova | 23:24 | |
| *** dave-mccowan has quit IRC | 23:27 | |
| tbachman | mriedem: I love how he’s also apparrently an expert gymnast | 23:27 |
| *** asettle has quit IRC | 23:27 | |
| mriedem | apparently? | 23:27 |
| mriedem | he just is | 23:27 |
| tbachman | lol | 23:27 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Restrict CONF.quota_driver to DB and noop quota drivers https://review.openstack.org/410996 | 23:30 |
| *** sdague has quit IRC | 23:31 | |
| melwitt | dansmith: so you think we should just have obj_target_cell here even though it doesn't do anything yet other than make the functional tests work? or would you be cool with rebasing on top of a change that adds mq switching back to target_cell? | 23:36 |
| *** thorst_ has quit IRC | 23:36 | |
| dansmith | melwitt: I think we should start targeting cells when we know they need targeted and follow up with mq switching in target_cell() after to start cutting over yeah | 23:37 |
| mriedem | melwitt: what am i missing here? https://review.openstack.org/#/c/410009/3/releasenotes/notes/cell-id-db-sync-nova-manage-8504b54dd115a2e9.yaml | 23:37 |
| mriedem | i'm a bit lost on what you and bauzas are saying | 23:37 |
| mriedem | i did want to get this into the o-2 milestone https://review.openstack.org/#/c/410009/3/releasenotes/notes/ocata-requires-cellv2-96bd243be874d77f.yaml | 23:37 |
| melwitt | mriedem: I was trying to say to him that no matter what, operators have to create a new database. the only choice is what they name it | 23:41 |
| dansmith | melwitt: you know, regarding the notification question... | 23:42 |
| dansmith | melwitt: since we still call the old path for cellsv1, it might be weird to have a different notification depending on the config right? | 23:42 |
| mriedem | melwitt: but the naming convention of $<api-db-name>_cell0 is baked into the simple_cell_setup flow | 23:42 |
| mriedem | so that's my point | 23:42 |
| dansmith | I guess I don't really now how much notification consumers look at that field | 23:42 |
| melwitt | mriedem: yeah. can I mention simple_cell_setup in the notes or does that rewrite history? it would be nice to say when the name matters | 23:43 |
| mriedem | melwitt: https://review.openstack.org/#/c/410009/3/releasenotes/notes/ocata-requires-cellv2-96bd243be874d77f.yaml hasn't been released yet | 23:43 |
| mriedem | it's new for o-2 | 23:44 |
| mriedem | so i think we have some freedom there | 23:44 |
| melwitt | dansmith: yeah, I was agreeing that it sounds like we wouldn't want to change the name as this stuff is supposed to be transparent to the end user | 23:44 |
| mriedem | i wanted to make sure ^ specifically is clear before we release, | 23:44 |
| melwitt | I just wasn't thinking about it that way at first | 23:44 |
| mriedem | because o-2 is what makes cellsv2 required | 23:44 |
| dansmith | melwitt: yeah, I just had that thought as another potential thing | 23:44 |
| melwitt | oh, for cells v1 too. yeah | 23:44 |
| melwitt | mriedem: okay. yeah | 23:45 |
| melwitt | that one should definitely use the name that simple_cell_setup will use | 23:46 |
| melwitt | let me update my patch | 23:46 |
| mriedem | i'm fine with fudging the rest | 23:46 |
| *** baoli has joined #openstack-nova | 23:46 | |
| mriedem | the other notes i mean | 23:46 |
| melwitt | mriedem: but it's not okay to introduce the word "simple_cell_setup" to notes that don't already mention it, right? because the command may not have existed when the note was first published? | 23:48 |
| mriedem | i think those notes are all newton or newer, and simple_cell_setup existed in newton | 23:49 |
| mriedem | if it didn't our from-newton script in grenade wouldn't work | 23:49 |
| melwitt | okay | 23:49 |
| mriedem | man, going through our release notes for o-2 | 23:51 |
| mriedem | i'm not sure who is reviewing these | 23:51 |
| mriedem | i know it's not dan because these wouldn't slide | 23:51 |
| melwitt | I hope it wasn't me for the bad ones | 23:51 |
| mriedem | nope | 23:51 |
| mriedem | well, you might have approved them | 23:51 |
| mriedem | i didn't check that... | 23:51 |
| melwitt | yeah, that's what I meant | 23:52 |
| melwitt | my point on the create database instruction is that even if they provide a database connection and map cell0 manually, they still have to have created a database for cell0. cell0 doesn't use the API DB schema | 23:54 |
| melwitt | so I didn't think I can take out the words "create a database" | 23:54 |
| mriedem | yeah, if they map cell0 manually though, they aren't using simple_cell_setup probably | 23:54 |
| mriedem | right we need the 'create a database' in there | 23:55 |
| melwitt | yeah. I was just saying I think the only thing to argue about is whether we call out the name. the db connection thing to me is irrelevant | 23:55 |
| mriedem | IRregardless of what they do, they have to create a new db for cell0 | 23:55 |
| mriedem | what they name it depends on which commands that run | 23:55 |
| melwitt | yeah | 23:55 |
| melwitt | I'm editing these now, hopefully to something really good | 23:55 |
| mriedem | but if you create the cell0 db as just 'cell0', the mapping is going to be wrong when using simple_cell_setup, right? | 23:55 |
| mriedem | because the db connection for the cell0 mapping is going to be looking for nova_api_cell0 | 23:56 |
| melwitt | yeah because you can't do anything to tell simple_cell_setup a name | 23:56 |
| mriedem | right, so i think if someone read that as literally a new db named 'cell0' they are going to get into trouble | 23:56 |
| mriedem | when using simple_cell_setup | 23:56 |
| mriedem | then they'll come here and complain | 23:56 |
| *** Swami has joined #openstack-nova | 23:56 | |
| mriedem | and i'll say, this is all melwitt's fault | 23:56 |
| melwitt | yes. I'm adding that specific stuff for the simple_cell_setup. I just hope this isn't going to be too confusing | 23:57 |
| melwitt | hah | 23:57 |
| mriedem | it might be :) | 23:57 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Fix some release notes in preparation for the o-2 beta release https://review.openstack.org/411005 | 23:57 |
| mriedem | bauzas: dansmith: ^ | 23:57 |
| bauzas | I love people highlighting me just 2 secs before going to bed :) | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!