| *** shaohe_feng has quit IRC | 00:00 | |
| *** jhesketh_ is now known as jhesketh | 00:02 | |
| *** alexpilotti has quit IRC | 00:02 | |
| *** boomsauce has joined #openstack-nova | 00:03 | |
| *** shaohe_feng has joined #openstack-nova | 00:03 | |
| *** thumpba has quit IRC | 00:04 | |
| *** mtanino has joined #openstack-nova | 00:08 | |
| *** moshele has joined #openstack-nova | 00:09 | |
| *** jwcroppe has joined #openstack-nova | 00:09 | |
| *** shaohe_feng has quit IRC | 00:11 | |
| *** shaohe_feng has joined #openstack-nova | 00:12 | |
| *** thumpba has joined #openstack-nova | 00:12 | |
| *** salv-orl_ has quit IRC | 00:13 | |
| *** neophy has joined #openstack-nova | 00:16 | |
| *** AmyLee7 has quit IRC | 00:18 | |
| *** mrda has joined #openstack-nova | 00:19 | |
| *** devananda has joined #openstack-nova | 00:19 | |
| *** samueldmq has joined #openstack-nova | 00:19 | |
| *** diogogmt has joined #openstack-nova | 00:19 | |
| *** dank_ has joined #openstack-nova | 00:19 | |
| *** vdrok has joined #openstack-nova | 00:20 | |
| *** patrickeast has joined #openstack-nova | 00:20 | |
| *** josecastroleon has joined #openstack-nova | 00:20 | |
| *** andrewbogott has joined #openstack-nova | 00:21 | |
| *** jamesdenton has quit IRC | 00:21 | |
| *** shaohe_feng has quit IRC | 00:21 | |
| *** ildikov has joined #openstack-nova | 00:21 | |
| *** shaohe_feng has joined #openstack-nova | 00:21 | |
| *** Jeffrey4l has joined #openstack-nova | 00:26 | |
| *** shaohe_feng has quit IRC | 00:31 | |
| *** thumpba has quit IRC | 00:31 | |
| *** shaohe_feng has joined #openstack-nova | 00:32 | |
| *** baoli has quit IRC | 00:32 | |
| *** baoli has joined #openstack-nova | 00:33 | |
| *** Sukhdev has quit IRC | 00:33 | |
| *** baoli has quit IRC | 00:33 | |
| *** baoli has joined #openstack-nova | 00:34 | |
| *** thumpba has joined #openstack-nova | 00:34 | |
| *** dimtruck is now known as zz_dimtruck | 00:34 | |
| *** pcm_ has quit IRC | 00:36 | |
| *** tbachman has quit IRC | 00:36 | |
| *** tbachman has joined #openstack-nova | 00:37 | |
| *** browne has quit IRC | 00:37 | |
| *** ducttape_ has joined #openstack-nova | 00:38 | |
| *** ducttape_ has quit IRC | 00:38 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Example verification for os-interface.inc https://review.openstack.org/315934 | 00:41 |
|---|---|---|
| *** shaohe_feng has quit IRC | 00:41 | |
| *** shaohe_feng has joined #openstack-nova | 00:42 | |
| *** hoangcx has joined #openstack-nova | 00:42 | |
| *** sheel has joined #openstack-nova | 00:43 | |
| *** pc_m has joined #openstack-nova | 00:43 | |
| *** srushti has joined #openstack-nova | 00:43 | |
| *** tpepper has quit IRC | 00:44 | |
| *** moshele has quit IRC | 00:44 | |
| *** boomsauce has quit IRC | 00:46 | |
| *** zz_dimtruck is now known as dimtruck | 00:49 | |
| *** vishwanathj has quit IRC | 00:50 | |
| *** shaohe_feng has quit IRC | 00:52 | |
| *** diogogmt has quit IRC | 00:53 | |
| *** rfolco has quit IRC | 00:54 | |
| *** Sukhdev has joined #openstack-nova | 00:54 | |
| *** shaohe_feng has joined #openstack-nova | 00:54 | |
| *** thumpba has quit IRC | 00:55 | |
| *** diogogmt has joined #openstack-nova | 00:55 | |
| *** rfolco has joined #openstack-nova | 00:56 | |
| *** jinquan has joined #openstack-nova | 00:56 | |
| *** cheneydc has joined #openstack-nova | 00:59 | |
| *** wznoinsk has joined #openstack-nova | 00:59 | |
| *** otsuka has joined #openstack-nova | 01:00 | |
| *** gongysh has quit IRC | 01:02 | |
| *** shaohe_feng has quit IRC | 01:02 | |
| *** shaohe_feng has joined #openstack-nova | 01:02 | |
| *** brad[] has quit IRC | 01:03 | |
| *** ducttape_ has joined #openstack-nova | 01:04 | |
| *** brad[] has joined #openstack-nova | 01:04 | |
| *** mtanino has quit IRC | 01:05 | |
| *** ducttape_ has quit IRC | 01:08 | |
| openstackgerrit | Giridhar Jayavelu proposed openstack/nova-specs: VMware: add SR-IOV support https://review.openstack.org/317126 | 01:09 |
| *** mvensky has joined #openstack-nova | 01:11 | |
| *** shaohe_feng has quit IRC | 01:12 | |
| *** shaohe_feng has joined #openstack-nova | 01:13 | |
| openstackgerrit | Li Yingjun proposed openstack/nova-specs: Pagination for hypervisor spec https://review.openstack.org/282209 | 01:15 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova-specs: Allow user to define the shutdown method through API when stop instances https://review.openstack.org/293790 | 01:15 |
| *** diogogmt has quit IRC | 01:18 | |
| *** shaohe_feng has quit IRC | 01:22 | |
| *** shaohe_feng has joined #openstack-nova | 01:23 | |
| *** EinstCrazy has joined #openstack-nova | 01:24 | |
| *** zhangyufei has joined #openstack-nova | 01:26 | |
| *** RuiChen has joined #openstack-nova | 01:26 | |
| *** mtanino has joined #openstack-nova | 01:30 | |
| *** jinquan has left #openstack-nova | 01:32 | |
| *** jinquan has joined #openstack-nova | 01:32 | |
| *** shaohe_feng has quit IRC | 01:33 | |
| *** shaohe_feng has joined #openstack-nova | 01:34 | |
| *** Apoorva has quit IRC | 01:34 | |
| *** mtanino has quit IRC | 01:35 | |
| openstackgerrit | Jinquan Ni proposed openstack/nova: Modify PCI without numa node info treatment https://review.openstack.org/298179 | 01:36 |
| *** jwcroppe has quit IRC | 01:36 | |
| *** cdelatte has quit IRC | 01:37 | |
| *** Sukhdev has quit IRC | 01:40 | |
| *** mvensky has quit IRC | 01:41 | |
| *** ducttape_ has joined #openstack-nova | 01:41 | |
| *** yamahata has quit IRC | 01:42 | |
| *** shaohe_feng has quit IRC | 01:43 | |
| *** shaohe_feng has joined #openstack-nova | 01:44 | |
| *** MikeG451 has quit IRC | 01:45 | |
| *** gongysh has joined #openstack-nova | 01:47 | |
| *** browne has joined #openstack-nova | 01:49 | |
| *** gongysh has quit IRC | 01:51 | |
| *** ducttape_ has quit IRC | 01:51 | |
| *** shaohe_feng has quit IRC | 01:53 | |
| *** shaohe_feng has joined #openstack-nova | 01:54 | |
| *** jwcroppe has joined #openstack-nova | 01:57 | |
| *** EinstCrazy has quit IRC | 02:00 | |
| *** EinstCrazy has joined #openstack-nova | 02:00 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: copy image id to image metadata from volume https://review.openstack.org/319905 | 02:01 |
| *** shaohe_feng has quit IRC | 02:03 | |
| *** shaohe_feng has joined #openstack-nova | 02:04 | |
| *** neophy has quit IRC | 02:07 | |
| *** yamahata has joined #openstack-nova | 02:11 | |
| *** neophy has joined #openstack-nova | 02:13 | |
| *** shaohe_feng has quit IRC | 02:14 | |
| *** unicell1 has quit IRC | 02:14 | |
| *** thorst_ has quit IRC | 02:14 | |
| *** thorst_ has joined #openstack-nova | 02:15 | |
| *** penick has quit IRC | 02:15 | |
| *** artom has quit IRC | 02:16 | |
| *** artom has joined #openstack-nova | 02:17 | |
| *** shaohe_feng has joined #openstack-nova | 02:17 | |
| *** nelsnels_ has joined #openstack-nova | 02:17 | |
| *** nelsnelson has quit IRC | 02:18 | |
| *** sacharya has joined #openstack-nova | 02:19 | |
| *** hoonetorg has quit IRC | 02:20 | |
| *** hoonetorg has joined #openstack-nova | 02:20 | |
| *** markvoelker has quit IRC | 02:21 | |
| *** ljxiash has joined #openstack-nova | 02:21 | |
| *** thorst_ has quit IRC | 02:23 | |
| *** ducttape_ has joined #openstack-nova | 02:24 | |
| *** shaohe_feng has quit IRC | 02:24 | |
| *** shaohe_feng has joined #openstack-nova | 02:25 | |
| *** browne has quit IRC | 02:30 | |
| *** ljxiash has quit IRC | 02:30 | |
| *** ljxiash has joined #openstack-nova | 02:31 | |
| *** shaohe_feng has quit IRC | 02:34 | |
| *** shaohe_feng has joined #openstack-nova | 02:34 | |
| *** ljxiash has quit IRC | 02:35 | |
| *** salv-orlando has joined #openstack-nova | 02:38 | |
| *** gjayavelu has joined #openstack-nova | 02:38 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Example verification for os-interface.inc https://review.openstack.org/315934 | 02:41 |
| *** salv-orlando has quit IRC | 02:43 | |
| *** ducttape_ has quit IRC | 02:43 | |
| *** ljxiash has joined #openstack-nova | 02:43 | |
| *** sree has joined #openstack-nova | 02:44 | |
| *** shaohe_feng has quit IRC | 02:44 | |
| *** shaohe_feng has joined #openstack-nova | 02:45 | |
| *** sree has quit IRC | 02:48 | |
| *** diogogmt has joined #openstack-nova | 02:49 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: api-ref: Example verification for os-interface.inc https://review.openstack.org/315934 | 02:51 |
| *** sudipto has joined #openstack-nova | 02:54 | |
| *** shaohe_feng has quit IRC | 02:55 | |
| *** shaohe_feng has joined #openstack-nova | 02:55 | |
| *** Sukhdev has joined #openstack-nova | 03:01 | |
| openstackgerrit | Eli Qiao proposed openstack/nova-specs: Add libvirt perf event https://review.openstack.org/319513 | 03:04 |
| *** shaohe_feng has quit IRC | 03:05 | |
| *** baoli has quit IRC | 03:05 | |
| *** raunak has joined #openstack-nova | 03:06 | |
| *** shaohe_feng has joined #openstack-nova | 03:06 | |
| *** sajeesh has quit IRC | 03:06 | |
| *** erlon_ has joined #openstack-nova | 03:08 | |
| *** rfolco has quit IRC | 03:10 | |
| *** klkumar has joined #openstack-nova | 03:10 | |
| *** shaohe_feng has quit IRC | 03:15 | |
| *** shaohe_feng has joined #openstack-nova | 03:16 | |
| *** zenoway has joined #openstack-nova | 03:20 | |
| *** thorst_ has joined #openstack-nova | 03:21 | |
| *** markvoelker has joined #openstack-nova | 03:22 | |
| *** zenoway has quit IRC | 03:24 | |
| *** shaohe_feng has quit IRC | 03:25 | |
| *** shaohe_feng has joined #openstack-nova | 03:26 | |
| *** markvoelker has quit IRC | 03:27 | |
| *** thorst_ has quit IRC | 03:28 | |
| *** sdake_ has quit IRC | 03:29 | |
| *** ccard has joined #openstack-nova | 03:33 | |
| *** ccard_ has quit IRC | 03:34 | |
| *** unicell has joined #openstack-nova | 03:35 | |
| *** shaohe_feng has quit IRC | 03:36 | |
| *** shaohe_feng has joined #openstack-nova | 03:36 | |
| *** dave-mcc_ has quit IRC | 03:38 | |
| *** xagent has joined #openstack-nova | 03:42 | |
| xagent | Hi, I have modified a nova neutronv2 API function, and I did modify the according unit test function, but tox errors out in following case. Anyone know whwat it means? | 03:44 |
| xagent | if (len(self._expected_calls_queue) == 1 and mox3.mox.ExpectedMethodCallsError: Verify: Expected methods never called: 0. <MultipleTimesGroup "default"> | 03:44 |
| *** baoli has joined #openstack-nova | 03:44 | |
| xagent | I am unsure what was not called - if I try to print methods_left, I get the following: | 03:45 |
| xagent | (Pdb) dir(self._expected_calls_queue[0]._methods_left) ['__and__', '__class__', '__cmp__', '__contains__', '__delattr__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__gt__', '__hash__', '__iand__', '__init__', '__ior__', '__isub__', '__iter__', '__ixor__', '__le__', '__len__', '__lt__', '__ne__', '__new__', '__or__', '__rand__', '__reduce__', '__reduce_ex__', '__repr__', '__ror__', '__rsub__', '__rxor__', | 03:45 |
| *** shaohe_feng has quit IRC | 03:46 | |
| *** shaohe_feng has joined #openstack-nova | 03:47 | |
| *** prateek has joined #openstack-nova | 03:49 | |
| xagent | (Pdb) left = list(self._methods_left) | 03:50 |
| xagent | (Pdb) dir(left[0]) ['AndRaise', 'AndReturn', 'GetPossibleGroup', 'InAnyOrder', 'MultipleTimes', 'WithSideEffects', '_CheckAndCreateNewGroup', '_PopNextMethod', '_VerifyMethodCall', '__call__', '__class__', '__delattr__', '__dict__', '__doc__', '__eq__', '__format__', '__getattr__', '__getattribute__', '__hash__', '__init__', '__iter__', '__module__', '__name__', '__ne__', '__new__', '__next__', '__reduce__', '__reduce_ex__', '__ | 03:50 |
| *** garyk1 has joined #openstack-nova | 03:51 | |
| *** dimtruck is now known as zz_dimtruck | 03:52 | |
| *** yamahata has quit IRC | 03:53 | |
| *** zenoway has joined #openstack-nova | 03:55 | |
| *** david-lyle_ has joined #openstack-nova | 03:55 | |
| *** shaohe_feng has quit IRC | 03:56 | |
| *** david-lyle has quit IRC | 03:57 | |
| *** shaohe_feng has joined #openstack-nova | 03:57 | |
| openstackgerrit | jingtao liang proposed openstack/nova: Normalize the options use singele quotes in nova https://review.openstack.org/320215 | 03:59 |
| *** takashin_ has joined #openstack-nova | 03:59 | |
| *** takashin has quit IRC | 03:59 | |
| *** takashin_ has left #openstack-nova | 03:59 | |
| *** takashin_ has joined #openstack-nova | 03:59 | |
| *** zenoway has quit IRC | 04:00 | |
| *** sree has joined #openstack-nova | 04:05 | |
| *** shaohe_feng has quit IRC | 04:06 | |
| *** links has joined #openstack-nova | 04:07 | |
| *** shaohe_feng has joined #openstack-nova | 04:07 | |
| *** salv-orlando has joined #openstack-nova | 04:08 | |
| *** thumpba has joined #openstack-nova | 04:10 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: libvirt: Prevent block live migration with tunnelled flag https://review.openstack.org/310707 | 04:13 |
| *** shaohe_feng has quit IRC | 04:17 | |
| *** ijw has quit IRC | 04:17 | |
| *** shaohe_feng has joined #openstack-nova | 04:17 | |
| *** salv-orlando has quit IRC | 04:17 | |
| *** sdake has joined #openstack-nova | 04:17 | |
| *** sdake has quit IRC | 04:18 | |
| *** markvoelker has joined #openstack-nova | 04:23 | |
| *** thorst_ has joined #openstack-nova | 04:26 | |
| *** shaohe_feng has quit IRC | 04:27 | |
| *** markvoelker has quit IRC | 04:28 | |
| *** shaohe_feng has joined #openstack-nova | 04:30 | |
| *** zenoway has joined #openstack-nova | 04:31 | |
| *** thorst_ has quit IRC | 04:33 | |
| *** zenoway has quit IRC | 04:35 | |
| *** xagent has quit IRC | 04:36 | |
| *** shaohe_feng has quit IRC | 04:37 | |
| *** shaohe_feng has joined #openstack-nova | 04:38 | |
| *** sudipto has quit IRC | 04:39 | |
| *** armax has quit IRC | 04:41 | |
| *** ljxiash has quit IRC | 04:42 | |
| *** otsuka has quit IRC | 04:42 | |
| *** ljxiash has joined #openstack-nova | 04:42 | |
| *** otsuka has joined #openstack-nova | 04:43 | |
| *** sridharg has joined #openstack-nova | 04:43 | |
| *** ljxiash has quit IRC | 04:47 | |
| *** shaohe_feng has quit IRC | 04:47 | |
| *** shaohe_feng has joined #openstack-nova | 04:51 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Remove deprecated configuration option network_device_mtu https://review.openstack.org/319543 | 04:55 |
| *** jaosorior has joined #openstack-nova | 04:56 | |
| *** shaohe_feng has quit IRC | 04:58 | |
| *** shaohe_feng has joined #openstack-nova | 04:58 | |
| openstackgerrit | Takashi NATSUME proposed openstack/nova: Add length check in comparing object lists https://review.openstack.org/320220 | 04:59 |
| *** flwang1 has quit IRC | 05:01 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Return instance of Guest from method write_instance_config https://review.openstack.org/265146 | 05:02 |
| *** zenoway has joined #openstack-nova | 05:06 | |
| *** shaohe_feng has quit IRC | 05:08 | |
| *** shaohe_feng has joined #openstack-nova | 05:09 | |
| *** prateek has quit IRC | 05:10 | |
| *** zenoway has quit IRC | 05:11 | |
| *** jaypipes has quit IRC | 05:13 | |
| *** dmellado has quit IRC | 05:13 | |
| *** links has quit IRC | 05:14 | |
| *** jvgrant has quit IRC | 05:15 | |
| *** adreznec has quit IRC | 05:15 | |
| *** jianghuaw has joined #openstack-nova | 05:15 | |
| *** jvgrant has joined #openstack-nova | 05:15 | |
| *** dansmith has quit IRC | 05:16 | |
| *** albertom has quit IRC | 05:16 | |
| *** baoli has quit IRC | 05:16 | |
| *** dmellado has joined #openstack-nova | 05:17 | |
| *** albertom has joined #openstack-nova | 05:17 | |
| *** links has joined #openstack-nova | 05:17 | |
| *** dansmith has joined #openstack-nova | 05:18 | |
| *** dansmith is now known as Guest71132 | 05:18 | |
| *** shaohe_feng has quit IRC | 05:18 | |
| *** prateek has joined #openstack-nova | 05:18 | |
| *** adreznec has joined #openstack-nova | 05:19 | |
| *** shaohe_feng has joined #openstack-nova | 05:19 | |
| *** sacharya_ has joined #openstack-nova | 05:21 | |
| *** laszloh has joined #openstack-nova | 05:22 | |
| *** sacharya has quit IRC | 05:23 | |
| *** links has quit IRC | 05:24 | |
| *** markvoelker has joined #openstack-nova | 05:24 | |
| *** sacharya has joined #openstack-nova | 05:26 | |
| *** sudipto has joined #openstack-nova | 05:26 | |
| *** sacharya_ has quit IRC | 05:28 | |
| *** shaohe_feng has quit IRC | 05:28 | |
| *** markvoelker has quit IRC | 05:29 | |
| *** shaohe_feng has joined #openstack-nova | 05:29 | |
| *** thorst_ has joined #openstack-nova | 05:30 | |
| *** iceyao has joined #openstack-nova | 05:34 | |
| *** thorst_ has quit IRC | 05:38 | |
| *** jaosorior has quit IRC | 05:38 | |
| *** iceyao has quit IRC | 05:39 | |
| *** shaohe_feng has quit IRC | 05:39 | |
| *** iceyao has joined #openstack-nova | 05:39 | |
| *** shaohe_feng has joined #openstack-nova | 05:39 | |
| *** jaosorior has joined #openstack-nova | 05:41 | |
| *** Sukhdev_ has joined #openstack-nova | 05:45 | |
| *** Sukhdev has quit IRC | 05:46 | |
| *** thumpba has quit IRC | 05:47 | |
| *** shaohe_feng has quit IRC | 05:49 | |
| *** shaohe_feng has joined #openstack-nova | 05:50 | |
| *** lpetrut has joined #openstack-nova | 05:53 | |
| *** erlon_ has quit IRC | 05:56 | |
| *** amotoki has joined #openstack-nova | 05:56 | |
| *** ljxiash has joined #openstack-nova | 05:56 | |
| *** shaohe_feng has quit IRC | 05:59 | |
| *** Fdaisuke has quit IRC | 06:00 | |
| *** shaohe_feng has joined #openstack-nova | 06:00 | |
| *** thorst_ has joined #openstack-nova | 06:00 | |
| *** rk4n has joined #openstack-nova | 06:00 | |
| *** rk4n has quit IRC | 06:00 | |
| *** Fdaisuke has joined #openstack-nova | 06:03 | |
| *** ijw has joined #openstack-nova | 06:03 | |
| *** links has joined #openstack-nova | 06:03 | |
| *** mjura has joined #openstack-nova | 06:07 | |
| *** gjayavelu has quit IRC | 06:08 | |
| *** rcernin has joined #openstack-nova | 06:08 | |
| *** thorst_ has quit IRC | 06:09 | |
| *** shaohe_feng has quit IRC | 06:09 | |
| *** shaohe_feng has joined #openstack-nova | 06:10 | |
| *** sdake has joined #openstack-nova | 06:11 | |
| *** Vek has quit IRC | 06:11 | |
| *** Vek has joined #openstack-nova | 06:11 | |
| *** mjura has quit IRC | 06:13 | |
| *** sridharg has quit IRC | 06:13 | |
| *** mjura has joined #openstack-nova | 06:13 | |
| *** nikhil has quit IRC | 06:14 | |
| *** vtech_ has quit IRC | 06:14 | |
| *** scottda has quit IRC | 06:14 | |
| openstackgerrit | Sampath Priyankara proposed openstack/python-novaclient: Fix nova host-evacuate for v2.14 https://review.openstack.org/320231 | 06:14 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: api-ref: parameter verification for os-hypervisors https://review.openstack.org/316681 | 06:15 |
| *** morgabra has quit IRC | 06:15 | |
| *** serverascode has quit IRC | 06:15 | |
| *** pkholkin has quit IRC | 06:15 | |
| *** junjie has quit IRC | 06:15 | |
| *** sneti_ has quit IRC | 06:15 | |
| *** DuncanT has quit IRC | 06:15 | |
| *** nikhil has joined #openstack-nova | 06:16 | |
| *** scottda has joined #openstack-nova | 06:16 | |
| *** zhenguo_ has quit IRC | 06:16 | |
| *** mrhillsman has quit IRC | 06:16 | |
| *** coreywright has quit IRC | 06:16 | |
| *** betherly has quit IRC | 06:16 | |
| *** junjie has joined #openstack-nova | 06:17 | |
| *** aunnam has quit IRC | 06:17 | |
| *** pkholkin has joined #openstack-nova | 06:18 | |
| *** zhenguo_ has joined #openstack-nova | 06:18 | |
| *** betherly has joined #openstack-nova | 06:19 | |
| *** eglynn has joined #openstack-nova | 06:19 | |
| *** DuncanT has joined #openstack-nova | 06:19 | |
| *** tziarko has quit IRC | 06:19 | |
| *** sballe_ has quit IRC | 06:19 | |
| *** shaohe_feng has quit IRC | 06:20 | |
| *** akrish9 has quit IRC | 06:20 | |
| *** isq has quit IRC | 06:20 | |
| *** shaohe_feng has joined #openstack-nova | 06:20 | |
| *** tziarko has joined #openstack-nova | 06:20 | |
| *** NightKhaos has quit IRC | 06:21 | |
| *** sballe_ has joined #openstack-nova | 06:21 | |
| *** paul-carlton2 has joined #openstack-nova | 06:21 | |
| *** fmccrthy has quit IRC | 06:21 | |
| *** macsz has quit IRC | 06:21 | |
| *** isq has joined #openstack-nova | 06:21 | |
| *** med_ has quit IRC | 06:21 | |
| *** fmccrthy has joined #openstack-nova | 06:22 | |
| *** woodster_ has quit IRC | 06:22 | |
| *** macsz has joined #openstack-nova | 06:23 | |
| *** morgabra has joined #openstack-nova | 06:23 | |
| *** aunnam has joined #openstack-nova | 06:24 | |
| *** gzhai2 has quit IRC | 06:24 | |
| *** markvoelker has joined #openstack-nova | 06:25 | |
| *** gzhai2 has joined #openstack-nova | 06:25 | |
| *** med_ has joined #openstack-nova | 06:25 | |
| *** med_ has quit IRC | 06:25 | |
| *** med_ has joined #openstack-nova | 06:25 | |
| *** sdake has quit IRC | 06:25 | |
| *** woodster_ has joined #openstack-nova | 06:26 | |
| *** moshele has joined #openstack-nova | 06:26 | |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Fix inappropriate notification send https://review.openstack.org/312887 | 06:26 |
| *** anush has joined #openstack-nova | 06:26 | |
| *** samueldmq has quit IRC | 06:27 | |
| *** samueldmq has joined #openstack-nova | 06:27 | |
| *** harlowja has quit IRC | 06:28 | |
| *** NightKhaos has joined #openstack-nova | 06:28 | |
| *** markvoelker has quit IRC | 06:29 | |
| *** shaohe_feng has quit IRC | 06:30 | |
| *** aloga has quit IRC | 06:31 | |
| *** serverascode has joined #openstack-nova | 06:32 | |
| *** shaohe_feng has joined #openstack-nova | 06:33 | |
| *** nortman has quit IRC | 06:36 | |
| *** sdake has joined #openstack-nova | 06:37 | |
| *** nortman has joined #openstack-nova | 06:37 | |
| *** salv-orlando has joined #openstack-nova | 06:37 | |
| *** rk4n has joined #openstack-nova | 06:39 | |
| *** shaohe_feng has quit IRC | 06:40 | |
| *** neophy has quit IRC | 06:40 | |
| *** mpavone has joined #openstack-nova | 06:41 | |
| openstackgerrit | Eli Qiao proposed openstack/nova-specs: Add libvirt perf event https://review.openstack.org/319513 | 06:41 |
| *** liverpooler has joined #openstack-nova | 06:41 | |
| *** shaohe_feng has joined #openstack-nova | 06:41 | |
| *** mrhillsman has joined #openstack-nova | 06:43 | |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: Revert "glance:add helper method to get client version" https://review.openstack.org/315968 | 06:43 |
| *** fragatin_ has joined #openstack-nova | 06:45 | |
| *** earlephilhower has quit IRC | 06:46 | |
| *** Sukhdev_ has quit IRC | 06:46 | |
| *** sridharg has joined #openstack-nova | 06:46 | |
| *** fragatina has quit IRC | 06:47 | |
| *** chip_ has quit IRC | 06:49 | |
| *** fragatin_ has quit IRC | 06:50 | |
| *** chip_ has joined #openstack-nova | 06:50 | |
| *** unicell has quit IRC | 06:50 | |
| *** shaohe_feng has quit IRC | 06:50 | |
| *** shaohe_feng has joined #openstack-nova | 06:51 | |
| *** unicell has joined #openstack-nova | 06:51 | |
| *** toanster has joined #openstack-nova | 06:52 | |
| *** toan has quit IRC | 06:52 | |
| *** zenoway has joined #openstack-nova | 06:53 | |
| *** sdake has quit IRC | 06:53 | |
| *** zhangyufei has quit IRC | 06:55 | |
| *** sneti_ has joined #openstack-nova | 06:55 | |
| *** zhangyufei has joined #openstack-nova | 06:57 | |
| *** fragatina has joined #openstack-nova | 06:57 | |
| *** zenoway has quit IRC | 06:58 | |
| *** lpetrut has quit IRC | 06:58 | |
| openstackgerrit | Rong Han proposed openstack/nova: libvirt: Fix the content of "disk.config" lost after migrate/resize https://review.openstack.org/288334 | 07:00 |
| *** shaohe_feng has quit IRC | 07:01 | |
| *** shaohe_feng has joined #openstack-nova | 07:01 | |
| *** fragatina has quit IRC | 07:02 | |
| *** tesseract has joined #openstack-nova | 07:04 | |
| *** jichen has joined #openstack-nova | 07:04 | |
| *** ijw has quit IRC | 07:05 | |
| *** Jeffrey4l has quit IRC | 07:06 | |
| *** thorst_ has joined #openstack-nova | 07:06 | |
| *** chip_ has quit IRC | 07:07 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: copy image id to image metadata from volume https://review.openstack.org/319905 | 07:07 |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Remove mox in tests/unit/objects/test_instance_info_cache https://review.openstack.org/307155 | 07:07 |
| *** fragatina has joined #openstack-nova | 07:07 | |
| *** chip_ has joined #openstack-nova | 07:08 | |
| *** fragatina has quit IRC | 07:08 | |
| *** tesseract has quit IRC | 07:08 | |
| *** fragatina has joined #openstack-nova | 07:09 | |
| *** yamahata has joined #openstack-nova | 07:10 | |
| *** sree_ has joined #openstack-nova | 07:10 | |
| *** rk4n has quit IRC | 07:11 | |
| *** sree_ is now known as Guest80537 | 07:11 | |
| *** aloga has joined #openstack-nova | 07:11 | |
| *** shaohe_feng has quit IRC | 07:11 | |
| *** shaohe_feng has joined #openstack-nova | 07:12 | |
| openstackgerrit | Paul Carlton proposed openstack/python-novaclient: [microversion] Bump to 2.27 https://review.openstack.org/316026 | 07:12 |
| *** sree has quit IRC | 07:13 | |
| *** sacharya has quit IRC | 07:13 | |
| *** thorst_ has quit IRC | 07:13 | |
| *** vtech has joined #openstack-nova | 07:15 | |
| *** rwsu has joined #openstack-nova | 07:19 | |
| *** shaohe_feng has quit IRC | 07:21 | |
| *** matrohon has joined #openstack-nova | 07:21 | |
| *** dedery has joined #openstack-nova | 07:21 | |
| *** shaohe_feng has joined #openstack-nova | 07:21 | |
| *** rwsu has quit IRC | 07:22 | |
| *** markvoelker has joined #openstack-nova | 07:26 | |
| *** tangchen has joined #openstack-nova | 07:27 | |
| *** nkrinner has joined #openstack-nova | 07:28 | |
| *** jaosorior is now known as jaosorior_brb | 07:28 | |
| *** jlanoux has joined #openstack-nova | 07:29 | |
| *** dedery_ has joined #openstack-nova | 07:31 | |
| *** shaohe_feng has quit IRC | 07:31 | |
| *** markvoelker has quit IRC | 07:32 | |
| *** flwang1 has joined #openstack-nova | 07:32 | |
| *** shaohe_feng has joined #openstack-nova | 07:32 | |
| *** eglynn has quit IRC | 07:33 | |
| *** dedery has quit IRC | 07:35 | |
| *** ihrachys has joined #openstack-nova | 07:37 | |
| *** rmart04 has joined #openstack-nova | 07:40 | |
| *** ecelik has joined #openstack-nova | 07:40 | |
| *** shaohe_feng has quit IRC | 07:42 | |
| *** shaohe_feng has joined #openstack-nova | 07:42 | |
| *** edand has joined #openstack-nova | 07:44 | |
| *** paul-carlton2 has quit IRC | 07:46 | |
| *** huanxie has left #openstack-nova | 07:47 | |
| *** huanxie has joined #openstack-nova | 07:47 | |
| *** woodster_ has quit IRC | 07:48 | |
| *** lpetrut has joined #openstack-nova | 07:48 | |
| *** shaohe_feng has quit IRC | 07:52 | |
| *** shaohe_feng has joined #openstack-nova | 07:53 | |
| *** markus_z has joined #openstack-nova | 07:54 | |
| *** adisky has joined #openstack-nova | 07:55 | |
| *** edand has quit IRC | 07:55 | |
| *** edand has joined #openstack-nova | 07:56 | |
| *** Fdaisuke has quit IRC | 07:57 | |
| *** Fdaisuke has joined #openstack-nova | 07:59 | |
| *** paul-carlton2 has joined #openstack-nova | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-nova | 08:00 | |
| *** fragatina has quit IRC | 08:01 | |
| *** shaohe_feng has quit IRC | 08:02 | |
| *** shaohe_feng has joined #openstack-nova | 08:03 | |
| *** vtech has quit IRC | 08:03 | |
| *** raunak has quit IRC | 08:03 | |
| *** vtech has joined #openstack-nova | 08:03 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/303345 | 08:10 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Testing latest u-c https://review.openstack.org/303345 | 08:10 |
| *** thorst_ has joined #openstack-nova | 08:11 | |
| *** lucasagomes_ is now known as lucasagomes | 08:12 | |
| *** shaohe_feng has quit IRC | 08:12 | |
| *** shaohe_feng has joined #openstack-nova | 08:13 | |
| *** sacharya has joined #openstack-nova | 08:13 | |
| *** belmoreira has joined #openstack-nova | 08:15 | |
| *** dmk0202 has joined #openstack-nova | 08:16 | |
| *** lpetrut has quit IRC | 08:17 | |
| *** sacharya has quit IRC | 08:18 | |
| *** MarcF has joined #openstack-nova | 08:18 | |
| *** thorst_ has quit IRC | 08:18 | |
| *** shaohe_feng has quit IRC | 08:23 | |
| *** shaohe_feng has joined #openstack-nova | 08:23 | |
| *** sfinucan has joined #openstack-nova | 08:25 | |
| *** salv-orlando has quit IRC | 08:25 | |
| *** afazekas_ is now known as afazekas | 08:27 | |
| *** markvoelker has joined #openstack-nova | 08:28 | |
| *** yassine has joined #openstack-nova | 08:29 | |
| *** paul-carlton2 has quit IRC | 08:29 | |
| *** yassine is now known as Guest66304 | 08:29 | |
| *** markvoelker has quit IRC | 08:32 | |
| *** shaohe_feng has quit IRC | 08:33 | |
| *** caspinol has joined #openstack-nova | 08:33 | |
| *** shaohe_feng has joined #openstack-nova | 08:34 | |
| *** efoley_ has joined #openstack-nova | 08:34 | |
| *** efoley has joined #openstack-nova | 08:34 | |
| *** Guest66304 is now known as yassou | 08:35 | |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for migration. https://review.openstack.org/241479 | 08:36 |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for evacuation. https://review.openstack.org/241478 | 08:36 |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for resize. https://review.openstack.org/241477 | 08:36 |
| openstackgerrit | Tang Chen proposed openstack/nova: Implement state machine for live-migration. https://review.openstack.org/241476 | 08:36 |
| *** efoley__ has joined #openstack-nova | 08:36 | |
| *** EinstCrazy has quit IRC | 08:37 | |
| *** roeyc has joined #openstack-nova | 08:38 | |
| *** zenoway has joined #openstack-nova | 08:39 | |
| mdbooth | lyarwood_: http://tinyurl.com/gr56e6j | 08:39 |
| *** efoley_ has quit IRC | 08:39 | |
| *** efoley has quit IRC | 08:39 | |
| mdbooth | I believe that graph is a 12 hour average, and ceph (check) seems to be on the rise | 08:39 |
| *** ig0r_ has joined #openstack-nova | 08:39 | |
| mdbooth | while the reference non-ceph job isn't | 08:40 |
| *** EinstCrazy has joined #openstack-nova | 08:40 | |
| *** efoley has joined #openstack-nova | 08:40 | |
| lyarwood_ | mdbooth: ack, I was looking at http://tinyurl.com/jmvfa2x | 08:40 |
| *** paul-carlton2 has joined #openstack-nova | 08:41 | |
| mdbooth | lyarwood_: Does that look like there was a blip around midnight, but it's gone away? | 08:42 |
| mdbooth | Doesn't seem as significant as the other graph suggests | 08:42 |
| lyarwood_ | mdbooth: yeah AFAICT it's stable now, just checked a few failures from that time and they don't appear to be ceph related anyway | 08:43 |
| mdbooth | Weird. | 08:43 |
| * mdbooth gets back to work | 08:43 | |
| *** shaohe_feng has quit IRC | 08:43 | |
| openstackgerrit | Roman Dobosz proposed openstack/nova-specs: FPGA resource use cases https://review.openstack.org/318047 | 08:44 |
| *** derekh has joined #openstack-nova | 08:44 | |
| *** shaohe_feng has joined #openstack-nova | 08:44 | |
| openstackgerrit | Zhengguang Ou proposed openstack/nova: libvirt: fix detach volume race condition https://review.openstack.org/290510 | 08:48 |
| *** jaosorior_brb is now known as jaosorior | 08:50 | |
| openstackgerrit | Jinquan Ni proposed openstack/nova: Modify PCI without numa node info treatment https://review.openstack.org/298179 | 08:51 |
| *** EinstCrazy has quit IRC | 08:51 | |
| *** gszasz has joined #openstack-nova | 08:52 | |
| *** shaohe_feng has quit IRC | 08:53 | |
| *** MarcF has joined #openstack-nova | 08:53 | |
| *** shaohe_feng has joined #openstack-nova | 08:54 | |
| *** cdent has joined #openstack-nova | 08:55 | |
| *** EinstCrazy has joined #openstack-nova | 08:57 | |
| tangchen | alex_xu: Hi Alex, would you please take a look at this BP for me ? https://blueprints.launchpad.net/nova/+spec/migration-state-machine | 08:59 |
| tangchen | alex_xu: need approve in newton. :) | 08:59 |
| *** wznoinsk has quit IRC | 08:59 | |
| *** wznoinsk has joined #openstack-nova | 08:59 | |
| *** daemontool_ has joined #openstack-nova | 09:00 | |
| *** shaohe_feng has quit IRC | 09:04 | |
| *** shaohe_feng has joined #openstack-nova | 09:05 | |
| *** lpetrut has joined #openstack-nova | 09:07 | |
| *** lpetrut1 has joined #openstack-nova | 09:12 | |
| *** caspinol has quit IRC | 09:13 | |
| *** lpetrut has quit IRC | 09:13 | |
| *** lpetrut1 is now known as lpetrut | 09:13 | |
| *** shaohe_feng has quit IRC | 09:14 | |
| *** shaohe_feng has joined #openstack-nova | 09:15 | |
| *** thorst_ has joined #openstack-nova | 09:16 | |
| *** ociuhandu has joined #openstack-nova | 09:17 | |
| *** claudiub has joined #openstack-nova | 09:18 | |
| *** sdake has joined #openstack-nova | 09:20 | |
| cdent | mdbooth: "I think this code is regularly running a pretty expensive | 09:20 |
| cdent | task looking for something which will very rarely happen, only to generate | 09:20 |
| cdent | a log message which nobody is looking for" ++ | 09:20 |
| *** ociuhandu has quit IRC | 09:20 | |
| cdent | dammit, I thought I got the linefeeds out of that. anyway: great message, thank you. | 09:21 |
| *** Dave____ is now known as Dave | 09:21 | |
| *** thorst_ has quit IRC | 09:24 | |
| *** shaohe_feng has quit IRC | 09:24 | |
| *** shaohe_feng has joined #openstack-nova | 09:25 | |
| *** ljxiash has quit IRC | 09:27 | |
| *** ljxiash has joined #openstack-nova | 09:27 | |
| *** ljxiash has quit IRC | 09:27 | |
| *** ljxiash has joined #openstack-nova | 09:27 | |
| *** zigo_ is now known as zigo | 09:28 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Remove deprecated security_group_api config option https://review.openstack.org/319537 | 09:30 |
| mdbooth | cdent: Well, if that characterisation is correct, anyway. Somebody wrote that code. Maybe I missed something. | 09:33 |
| openstackgerrit | Moshe Levi proposed openstack/nova: [WIP] OVS offload POC https://review.openstack.org/275624 | 09:34 |
| *** gmann_ has joined #openstack-nova | 09:34 | |
| *** shaohe_feng has quit IRC | 09:34 | |
| *** shaohe_feng has joined #openstack-nova | 09:35 | |
| *** haukebruno has joined #openstack-nova | 09:36 | |
| johnthetubaguy | tangchen: I wasn't sure we all agreed that was the right way to fix that problem, I guess we didn't do a spec about that one? | 09:37 |
| openstackgerrit | Anh Tran proposed openstack/nova: Remove mox stubs in api/openstack/compute/test_virtual_interfaces.py https://review.openstack.org/320325 | 09:37 |
| openstackgerrit | Laszlo Hegedus proposed openstack/nova: Add versioned exception notification https://review.openstack.org/310700 | 09:39 |
| *** gcb has quit IRC | 09:40 | |
| *** gcb has joined #openstack-nova | 09:41 | |
| cdent | mdbooth: it seems likely. There's a lot of stuff happening throughout the system which smacks of "we better puts this here just in case (and even though it would probably better handled by something external)" | 09:41 |
| *** liusheng has quit IRC | 09:41 | |
| *** dedery_ has quit IRC | 09:42 | |
| *** liusheng has joined #openstack-nova | 09:42 | |
| markus_z | nova-bugs team meeting in ~15 min in #openstack-meeting-4. New contributors are welcome as well. https://wiki.openstack.org/wiki/Meetings/Nova/BugsTeam | 09:43 |
| *** shaohe_feng has quit IRC | 09:45 | |
| *** shaohe_feng has joined #openstack-nova | 09:46 | |
| johnthetubaguy | cdent: we do have a lot of historic scope creep | 09:46 |
| johnthetubaguy | cdent: certainly worth deprecating some of that | 09:46 |
| cdent | BURN IT ALL DOWN | 09:47 |
| cdent | oh sorry, just had a moment there, better now | 09:47 |
| *** wangfeng has quit IRC | 09:47 | |
| johnthetubaguy | :) | 09:48 |
| *** dedery has joined #openstack-nova | 09:48 | |
| * johnthetubaguy hands over matches, with a release sized fuse | 09:48 | |
| tangchen | johnthetubaguy: Hi, john, in Mitaka cycle, we agreed that this doesn't need a spec. So I didn't propose one. | 09:52 |
| mdbooth | johnthetubaguy: Are you aware of any reason to keep that code, btw? | 09:53 |
| tangchen | johnthetubaguy: So OK, shall we discuss it in the nova meeting or some other time ? | 09:53 |
| *** dedery has quit IRC | 09:53 | |
| *** mvk has quit IRC | 09:53 | |
| johnthetubaguy | tangchen: yeah, I thought that was the case... I am not sure I understand what is being proposed properly from the blueprint. Its worth raising at the meeting, if you are going to attend it. | 09:53 |
| *** yassou has quit IRC | 09:54 | |
| johnthetubaguy | mdbooth: good question, do you have the link handy again? | 09:54 |
| mdbooth | cdent: Incidentally, if you're in favour of keeping it would you mind piping up on the list? Crickets would be enough to justify deleting it, I guess, but positive support would be better :) | 09:54 |
| mdbooth | johnthetubaguy: Sure, sec | 09:54 |
| johnthetubaguy | oh wait, I see you note on the ML now... | 09:55 |
| mdbooth | johnthetubaguy: http://lists.openstack.org/pipermail/openstack-dev/2016-May/095751.html | 09:55 |
| *** shaohe_feng has quit IRC | 09:55 | |
| *** sambetts|afk is now known as sambetts | 09:55 | |
| johnthetubaguy | so this is actually about some folks who were worrying about disk "drift" | 09:55 |
| johnthetubaguy | like over time its not the same bits stored on the disk as was there originally | 09:55 |
| cdent | mdbooth: I'm favour of killing it, but assumed you were primarily looking for operator/real-world feedback rather than feedback who just wants to get cruft out of code? | 09:56 |
| *** shaohe_feng has joined #openstack-nova | 09:56 | |
| tangchen | johnthetubaguy: Ok, I'd like to attend the nova meeting, but I'm not sure which one should I attend. The live migration meeting ? | 09:56 |
| mdbooth | cdent: Both would be nice :) | 09:56 |
| *** caspinol has joined #openstack-nova | 09:56 | |
| johnthetubaguy | tangchen: the main one is the default, but I think the live-migration meeting is a good idea, see what that subteam thinks | 09:56 |
| mdbooth | johnthetubaguy: Yeah, that's what I guessed. I don't think this is really achieving it, though. | 09:56 |
| cdent | mdbooth: noted, will get something on there soon | 09:56 |
| tangchen | johnthetubaguy: OK, thank you very much. :) | 09:57 |
| johnthetubaguy | mdbooth: there is the worry about bad actors as well, I guess | 09:57 |
| mdbooth | johnthetubaguy: This is possibly one I hadn't considered. What's the bad actors thing? | 09:57 |
| johnthetubaguy | mdbooth: either way, it probably makes more sense to check it on the "next" boot from the image, rather than always hammering the disk | 09:57 |
| cdent | johnthetubaguy, mdbooth wouldn't it be better to check for changes at time of use, not every periodic task? | 09:57 |
| *** roeyc has quit IRC | 09:58 | |
| cdent | or is the concern that there needs to be info as soon as it happens, for forensics? | 09:58 |
| johnthetubaguy | cdent: jinx, and yes | 09:58 |
| mdbooth | Well, we don't check other disks | 09:58 |
| johnthetubaguy | well, maybe... but that feels like over kill | 09:58 |
| mdbooth | It just seems a it random | 09:58 |
| johnthetubaguy | mdbooth: well we check them all on boot | 09:58 |
| johnthetubaguy | after download from glance | 09:58 |
| mdbooth | johnthetubaguy: I mean the actual overlay/copy | 09:58 |
| mdbooth | Disks which aren't in the image cache | 09:58 |
| mdbooth | Also ephemeral disks, swap disks | 09:59 |
| markus_z | nova-bugs team meeting in #openstack-meeting-4. New contributors are welcome as well. https://wiki.openstack.org/wiki/Meetings/Nova/BugsTeam | 09:59 |
| mdbooth | The qemu binary | 09:59 |
| mdbooth | libvirt | 09:59 |
| mdbooth | nova itself | 09:59 |
| *** iceyao_ has joined #openstack-nova | 09:59 | |
| mdbooth | I mean, there's a whole host of potentially corrupt stuff we're not checking on every boot | 09:59 |
| johnthetubaguy | so I like the idea of checking the md5 before booting, as we do that after copying it from glance | 10:00 |
| johnthetubaguy | the background always checking seems like overkill | 10:00 |
| *** iceyao_ has quit IRC | 10:00 | |
| *** iceyao__ has joined #openstack-nova | 10:00 | |
| cdent | johnthetubaguy, mdbooth maybe move this to the list so that we don't have to go over it again there? | 10:00 |
| *** iceyao__ has quit IRC | 10:00 | |
| *** iceyao_ has joined #openstack-nova | 10:01 | |
| mdbooth | johnthetubaguy: Ok, lets discuss on list. I'm not personally convinced it serves any purpose. | 10:01 |
| *** cheneydc has quit IRC | 10:01 | |
| mdbooth | cdent: jinx | 10:01 |
| mdbooth | ;) | 10:01 |
| johnthetubaguy | cdent: +1 typing something now | 10:01 |
| *** sdake has quit IRC | 10:01 | |
| cdent | I think we've used up all the available jinxes today. | 10:01 |
| * johnthetubaguy nods | 10:01 | |
| *** iceyao__ has joined #openstack-nova | 10:02 | |
| *** iceyao_ has quit IRC | 10:02 | |
| *** wangfeng has joined #openstack-nova | 10:02 | |
| *** iceyao has quit IRC | 10:02 | |
| *** hoangcx_ has joined #openstack-nova | 10:02 | |
| *** hoangcx has quit IRC | 10:03 | |
| * mdbooth notes that danpb sounded in favour of checking for bitrot, but he's not around to defend himself | 10:03 | |
| * mdbooth sticks the knife in quickly | 10:04 | |
| *** gszasz has quit IRC | 10:05 | |
| *** boomsauce has joined #openstack-nova | 10:05 | |
| *** shaohe_feng has quit IRC | 10:05 | |
| *** rk4n has joined #openstack-nova | 10:05 | |
| *** sudipto has quit IRC | 10:06 | |
| *** sudipto has joined #openstack-nova | 10:06 | |
| *** shaohe_feng has joined #openstack-nova | 10:06 | |
| johnthetubaguy | mdbooth: I would rather do that check on boot though (although that does slow down every boot) | 10:06 |
| *** ljxiash has quit IRC | 10:12 | |
| *** ljxiash has joined #openstack-nova | 10:12 | |
| *** yamahata has quit IRC | 10:13 | |
| *** sacharya has joined #openstack-nova | 10:15 | |
| *** shaohe_feng has quit IRC | 10:15 | |
| eliqiao | sudipto: hi there? | 10:16 |
| *** shaohe_feng has joined #openstack-nova | 10:16 | |
| sudipto | eliqiao, hi yeah. | 10:17 |
| *** ljxiash has quit IRC | 10:17 | |
| eliqiao | sudipto: Thanks for giving comments on spec "Add libvirt perf event", I got a question, as per your suggestion, I wonder does nova-scheduler supporting using VM's tag filter host? | 10:17 |
| eliqiao | sudipto: FYI the spec link is https://review.openstack.org/319513 | 10:18 |
| *** sacharya has quit IRC | 10:19 | |
| sudipto | eliqiao, can you explain such a tag? That would help answer i guess. | 10:19 |
| eliqiao | why you are uncomfortable with enable perf per host? | 10:19 |
| eliqiao | sudipto: from your comments: "I am somehow more comfortable enabling this as a per VM tag as opposed to a host wide tag." | 10:20 |
| sudipto | eliqiao, enabling perf on the host is OK. However, it seems like you are going to enable the perf collection on a per VM basis? | 10:20 |
| eliqiao | sudipto: yes. | 10:21 |
| *** thorst_ has joined #openstack-nova | 10:21 | |
| eliqiao | we can collect some perf event from libvirt APIs | 10:21 |
| hieulq | johnthetubaguy: could you take a look at this https://review.openstack.org/#/c/317832/ ? markus_z have confirmed with this patch-set | 10:21 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Adds Hyper-V OVS ViF driver https://review.openstack.org/140045 | 10:21 |
| sudipto | eliqiao, so if a host has 40 guests running on it. It would start gathering the same on 40 such Virtual Machines on the host right? | 10:21 |
| *** mvk has joined #openstack-nova | 10:22 | |
| eliqiao | sudipto: I don't think so, that depend on if you are calling libvirt API to gather data. | 10:22 |
| sudipto | eliqiao, and perf is something that you might need for profiling a certain virtual machine as opposed to every other VM on the host. So i recommended you to have this feature enabled via a tag on the - flavaor (extra_specs) | 10:23 |
| sudipto | eliqiao, understand. Hence i wanted to also know the implications of this feature. What happens after you have added the tag to the VM XML? | 10:23 |
| *** tbachman_ has joined #openstack-nova | 10:23 | |
| johnthetubaguy | hieulq: ah, yeah, I should check that topic again | 10:23 |
| *** tbachman has quit IRC | 10:24 | |
| *** tbachman_ is now known as tbachman | 10:24 | |
| eliqiao | sudipto: nothing, next time you calling libvirt's API getdomstats(I guess the name), you will get the perf event you configured in the VM XML. | 10:24 |
| eliqiao | sudipto: without specify perf event in XML, you won't get that event data. | 10:25 |
| *** shaohe_feng has quit IRC | 10:26 | |
| sudipto | eliqiao, ah i see. So it's like preparing the VM for perf data collection that triggering it from nova right? | 10:26 |
| sudipto | s/that/than | 10:26 |
| eliqiao | sudipto: correct. | 10:27 |
| *** shaohe_feng has joined #openstack-nova | 10:27 | |
| sudipto | eliqiao, I like your idea. However i have few more questions. | 10:28 |
| *** thorst_ has quit IRC | 10:28 | |
| *** zhangyufei has quit IRC | 10:28 | |
| eliqiao | sudipto: yes, please. | 10:29 |
| *** markvoelker has joined #openstack-nova | 10:29 | |
| *** jlanoux has quit IRC | 10:30 | |
| sudipto | eliqiao, what happens when you have tagged the VM with the cmt xml? When do you start programming in the counters for the VM? | 10:30 |
| hieulq | johnthetubaguy: thanks :-) | 10:30 |
| sudipto | eliqiao, in other words, when does perf start collecting the data against that VM? | 10:31 |
| *** jlanoux has joined #openstack-nova | 10:31 | |
| *** EinstCrazy has quit IRC | 10:32 | |
| *** EinstCrazy has joined #openstack-nova | 10:32 | |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Remove mox from api/openstack/compute/test_deferred_delete.py https://review.openstack.org/320346 | 10:32 |
| *** markvoelker has quit IRC | 10:34 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Verify requested_destination in the scheduler https://review.openstack.org/314509 | 10:34 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: API change for verifying the scheduler when live migrating https://review.openstack.org/318334 | 10:34 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: API change for verifying the scheduler when evacuating https://review.openstack.org/315572 | 10:34 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add requested_destination field to RequestSpec https://review.openstack.org/312757 | 10:34 |
| *** shaohe_feng has quit IRC | 10:36 | |
| *** EinstCrazy has quit IRC | 10:37 | |
| *** shaohe_feng has joined #openstack-nova | 10:37 | |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Added support for new block device format in vmops https://review.openstack.org/246299 | 10:38 |
| openstackgerrit | Adelina Tuvenie proposed openstack/nova: Add support for setting boot order in Hyper-V https://review.openstack.org/248029 | 10:38 |
| openstackgerrit | Tomi Juvonen proposed openstack/nova-specs: Maintenance reason to server https://review.openstack.org/310510 | 10:38 |
| *** dedery has joined #openstack-nova | 10:41 | |
| *** hoangcx_ has quit IRC | 10:41 | |
| openstackgerrit | Tomi Juvonen proposed openstack/nova-specs: Set Maintenance Reason https://review.openstack.org/320349 | 10:41 |
| eliqiao | sudipto: only when you try to getdomainstats. | 10:41 |
| *** jlanoux has quit IRC | 10:42 | |
| *** takashin_ has left #openstack-nova | 10:42 | |
| *** jlanoux has joined #openstack-nova | 10:43 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: pci: Deprecate is_new from pci requests https://review.openstack.org/310990 | 10:44 |
| openstackgerrit | Moshe Levi proposed openstack/nova: pci: Allow to update pci_requests in instance_extra https://review.openstack.org/312028 | 10:44 |
| openstackgerrit | Moshe Levi proposed openstack/nova: pci: Move PCI devices and PCI requests into migration context https://review.openstack.org/307124 | 10:44 |
| *** Jeffrey4l has joined #openstack-nova | 10:45 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Remove deprecated security_group_api config option https://review.openstack.org/319537 | 10:46 |
| *** shaohe_feng has quit IRC | 10:46 | |
| openstackgerrit | Tomi Juvonen proposed openstack/nova-specs: Maintenance Reason To Hypervisor https://review.openstack.org/320350 | 10:46 |
| *** shaohe_feng has joined #openstack-nova | 10:47 | |
| *** moshele has quit IRC | 10:48 | |
| *** adisky has quit IRC | 10:50 | |
| *** ljxiash has joined #openstack-nova | 10:50 | |
| eliqiao | sudipto: still there? Adding perf event flags to XML won't start pulling event | 10:51 |
| *** ljxiash has quit IRC | 10:52 | |
| *** ljxiash has joined #openstack-nova | 10:52 | |
| *** sudipto has quit IRC | 10:53 | |
| *** caspinol has quit IRC | 10:53 | |
| *** moshele has joined #openstack-nova | 10:53 | |
| *** shaohe_feng has quit IRC | 10:56 | |
| *** shaohe_feng has joined #openstack-nova | 10:57 | |
| *** gszasz has joined #openstack-nova | 10:58 | |
| *** sudipto has joined #openstack-nova | 11:01 | |
| sudipto | eliqiao, sorry had to switch ports. So ok, i have understood your point but my understanding is this now - after this conversation | 11:01 |
| sudipto | eliqiao, you describe the VM in nova (does it start calculating the hardware counters at this point?) - then make the getdomstats call from Ceilometer or some such metering agent (this probably means that it can now give you data based on the counters it was collecting in the past). | 11:02 |
| *** cdelatte has joined #openstack-nova | 11:03 | |
| sudipto | eliqiao, the way i see it is, while these are hardware counters - i am not sure of the greatest impact on the software stack. Let's assume, per your argument that it's not going to be much. Also from nova's perspective, there's no collection of this data involved - and hence counter overflows or periodic intervals are beyond it's control. | 11:03 |
| *** caspinol has joined #openstack-nova | 11:04 | |
| *** gszasz has quit IRC | 11:04 | |
| sudipto | eliqiao, however for elegance, why would someone want to enable these l3 misses as a generic counter for all the VMs on a host. Instead I might want to chose my VMs that i would want profiling on. Do you agree to that? | 11:04 |
| openstackgerrit | Laszlo Hegedus proposed openstack/nova: Add versioned exception notification https://review.openstack.org/310700 | 11:06 |
| *** shaohe_feng has quit IRC | 11:07 | |
| sudipto | eliqiao, i think you answered few of my queries on the review. So i will wait for others' comment on it. Thanks! | 11:07 |
| *** shaohe_feng has joined #openstack-nova | 11:08 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova-specs: Remove compute-compute communication in live-migration https://review.openstack.org/292271 | 11:08 |
| *** vtech has quit IRC | 11:11 | |
| *** roeyc has joined #openstack-nova | 11:11 | |
| *** vtech has joined #openstack-nova | 11:12 | |
| *** rtheis has joined #openstack-nova | 11:12 | |
| *** salv-orlando has joined #openstack-nova | 11:15 | |
| openstackgerrit | ChangBo Guo(gcb) proposed openstack/nova: Remove deprecated configuration option use_local https://review.openstack.org/319544 | 11:16 |
| *** brad[] has quit IRC | 11:16 | |
| eliqiao | sudipto: I just back to home now. sorry I miss some words, reading irc log.. | 11:17 |
| *** shaohe_feng has quit IRC | 11:17 | |
| *** sdake has joined #openstack-nova | 11:17 | |
| eliqiao | sudipto: so you are still have comments on why to enable all perf counter for all VMs? | 11:17 |
| *** lucasagomes is now known as lucas-hungry | 11:18 | |
| *** salv-orlando has quit IRC | 11:19 | |
| *** shaohe_feng has joined #openstack-nova | 11:19 | |
| *** roeyc has quit IRC | 11:20 | |
| *** iceyao__ has quit IRC | 11:23 | |
| *** brad[] has joined #openstack-nova | 11:25 | |
| *** otsuka has quit IRC | 11:25 | |
| *** sdague has joined #openstack-nova | 11:25 | |
| *** thorst_ has joined #openstack-nova | 11:26 | |
| *** shaohe_feng has quit IRC | 11:27 | |
| *** ociuhandu has joined #openstack-nova | 11:28 | |
| jroll | bauzas: hey, in case you missed it: https://review.openstack.org/#/c/320016/1 | 11:28 |
| *** shaohe_feng has joined #openstack-nova | 11:28 | |
| jroll | you might be interested in that | 11:28 |
| *** abalutoiu has joined #openstack-nova | 11:29 | |
| *** roeyc has joined #openstack-nova | 11:30 | |
| *** gmann_ has quit IRC | 11:30 | |
| *** markvoelker has joined #openstack-nova | 11:30 | |
| *** thorst_ has quit IRC | 11:31 | |
| *** rodrigods has quit IRC | 11:31 | |
| *** rodrigods has joined #openstack-nova | 11:31 | |
| *** tlbr_ has quit IRC | 11:31 | |
| *** caspinol has quit IRC | 11:31 | |
| bauzas | jroll: ++, I'm about to use the next week for reviewing all the specs | 11:34 |
| *** ig0r_ has quit IRC | 11:34 | |
| *** markvoelker has quit IRC | 11:35 | |
| jroll | bauzas: awesome, thanks :) | 11:35 |
| bauzas | jroll: just tbh, I'm not a nova-specs core :) | 11:35 |
| jroll | bauzas: I care more about scheduler experts' opinions at this stage, really | 11:36 |
| bauzas | :) | 11:37 |
| *** shaohe_feng has quit IRC | 11:37 | |
| *** shaohe_feng has joined #openstack-nova | 11:38 | |
| openstackgerrit | Evgeny Antyshev proposed openstack/nova: libvirt: add nova volume driver for vzstorage https://review.openstack.org/190843 | 11:39 |
| *** jaosorior has quit IRC | 11:39 | |
| openstackgerrit | Abhishek Kekane proposed openstack/nova: Use is_valid_cidr from oslo_utils https://review.openstack.org/320375 | 11:40 |
| *** jaosorior has joined #openstack-nova | 11:40 | |
| *** jaosorior has quit IRC | 11:40 | |
| *** jaosorior has joined #openstack-nova | 11:41 | |
| *** sudipto has quit IRC | 11:42 | |
| *** thorst_ has joined #openstack-nova | 11:43 | |
| *** salv-orlando has joined #openstack-nova | 11:46 | |
| *** gjayavelu has joined #openstack-nova | 11:46 | |
| *** shaohe_feng has quit IRC | 11:48 | |
| *** shaohe_feng has joined #openstack-nova | 11:48 | |
| *** sdake has quit IRC | 11:49 | |
| *** gjayavelu has quit IRC | 11:51 | |
| *** yassine has joined #openstack-nova | 11:52 | |
| *** yassine is now known as Guest9665 | 11:52 | |
| *** askb has joined #openstack-nova | 11:53 | |
| *** rfolco has joined #openstack-nova | 11:54 | |
| *** brad[] has quit IRC | 11:56 | |
| markus_z | sdague: When we talk about "type" in the api-ref, we talk about the JSON type, right? | 11:56 |
| *** ig0r_ has joined #openstack-nova | 11:56 | |
| markus_z | sdague: Which means a UUID will always be represented as a string. | 11:57 |
| openstackgerrit | Dao Cong Tien proposed openstack/nova: Remove mox from unit/api/openstack/compute/test_migrations.py https://review.openstack.org/320392 | 11:58 |
| markus_z | sdague: I'm asking because of https://bugs.launchpad.net/nova/+bug/1583701 | 11:58 |
| openstack | Launchpad bug 1583701 in OpenStack Compute (nova) "api-ref docs: use correct type for UUIDs" [Undecided,New] - Assigned to Gage Hugo (gagehugo) | 11:58 |
| *** brad[] has joined #openstack-nova | 11:58 | |
| *** shaohe_feng has quit IRC | 11:58 | |
| *** ducttape_ has joined #openstack-nova | 11:59 | |
| *** shaohe_feng has joined #openstack-nova | 11:59 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Adds RemoteFX support to the Hyper-V driver https://review.openstack.org/42529 | 12:00 |
| *** tlbr has joined #openstack-nova | 12:00 | |
| *** efoley__ has quit IRC | 12:02 | |
| *** markvoelker has joined #openstack-nova | 12:02 | |
| *** jaypipes has joined #openstack-nova | 12:03 | |
| *** efoley__ has joined #openstack-nova | 12:03 | |
| *** efoley has quit IRC | 12:03 | |
| *** efoley_ has joined #openstack-nova | 12:03 | |
| markus_z | claudiub: There are 2 new DocImpact bug reports for hyper-v: https://bugs.launchpad.net/nova/+bug/1584554 + https://bugs.launchpad.net/nova/+bug/1584199 | 12:04 |
| openstack | Launchpad bug 1584554 in OpenStack Compute (nova) " Hyper-V: Implement nova rescue" [Undecided,New] | 12:04 |
| openstack | Launchpad bug 1584199 in OpenStack Compute (nova) " HyperV: Nova serial console access support" [Undecided,New] | 12:04 |
| markus_z | claudiub: I think the one for "rescue" is already done because that patch also updating the hypervisor support matrix. | 12:05 |
| claudiub | yep | 12:06 |
| markus_z | claudiub: The "serial console" one, is that an interactive console? | 12:06 |
| claudiub | yep | 12:06 |
| markus_z | So just a check at http://docs.openstack.org/developer/nova/support-matrix.html#console_serial_interactive ? | 12:07 |
| markus_z | s/check/tick | 12:07 |
| claudiub | yeah, that should be it | 12:07 |
| markus_z | coolio, I'll close the "rescue" one and update the "serial" one, thanks! | 12:08 |
| *** shaohe_feng has quit IRC | 12:08 | |
| claudiub | sure, thank you too. :) | 12:08 |
| *** shaohe_feng has joined #openstack-nova | 12:09 | |
| *** lucas-hungry is now known as lucasagomes | 12:10 | |
| eliqiao | sudipto: around ? | 12:10 |
| *** ducttape_ has quit IRC | 12:12 | |
| *** dane-fichter has joined #openstack-nova | 12:13 | |
| dane-fichter | dane-fichter: johnthetubaguy I have a pending spec and patch up that are related to what you and mdbooth were discussing re: the image cache | 12:17 |
| dane-fichter | https://review.openstack.org/#/c/312210/ | 12:17 |
| dane-fichter | https://review.openstack.org/#/c/314222/ | 12:18 |
| *** shaohe_feng has quit IRC | 12:18 | |
| *** shaohe_feng has joined #openstack-nova | 12:19 | |
| *** ayoung has quit IRC | 12:19 | |
| *** gjayavelu has joined #openstack-nova | 12:22 | |
| *** ivase has joined #openstack-nova | 12:26 | |
| *** gjayavelu has quit IRC | 12:27 | |
| *** shaohe_feng has quit IRC | 12:29 | |
| openstackgerrit | jichenjc proposed openstack/nova: verify api-ref metadata.inc https://review.openstack.org/318600 | 12:29 |
| *** shaohe_feng has joined #openstack-nova | 12:29 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova-specs: Make checks before live-migration async https://review.openstack.org/320416 | 12:34 |
| Kevin_Zheng | johnthetubaguy: Hi could you have a check on this: https://review.openstack.org/#/c/293790/ ? | 12:34 |
| *** salv-orlando has quit IRC | 12:35 | |
| bauzas | johnthetubaguy: in case you could reapply your +2 for a very small modification ? (RequestedDestination name) | 12:36 |
| bauzas | johnthetubaguy: oops https://review.openstack.org/#/c/312757/10 | 12:37 |
| *** shaohe_feng has quit IRC | 12:39 | |
| *** shaohe_feng has joined #openstack-nova | 12:40 | |
| *** aysyd has joined #openstack-nova | 12:42 | |
| markus_z | zhenguo_: You're talking about scheduling here, right? https://bugs.launchpad.net/nova/+bug/1582663 | 12:43 |
| openstack | Launchpad bug 1582663 in OpenStack Compute (nova) "An Ironic node should not be exposed to Nova if no Ironic port is associated with it" [Undecided,New] - Assigned to Zhenguo Niu (niu-zglinux) | 12:43 |
| *** burgerk has joined #openstack-nova | 12:43 | |
| openstackgerrit | jichenjc proposed openstack/nova: api-ref verify of servers-admin-action.inc https://review.openstack.org/311727 | 12:45 |
| zhenguo_ | markus_z: yes, if a node without any port associated, we should not schedule to it | 12:45 |
| *** vladikr has joined #openstack-nova | 12:46 | |
| openstackgerrit | jichenjc proposed openstack/nova: verify api-ref os-migrations.inc https://review.openstack.org/318004 | 12:47 |
| markus_z | zhenguo_: Could you add some pointers to the bug description? | 12:47 |
| markus_z | zhenguo_: Do you want to filter those nodes out when the scheduler tries to determine a host? Or what's you plan to solve this? | 12:48 |
| zhenguo_ | markus_z: no, I can set it unavailable in ironic driver when collecting resources | 12:48 |
| *** catintheroof has joined #openstack-nova | 12:49 | |
| *** shaohe_feng has quit IRC | 12:49 | |
| *** yassine has joined #openstack-nova | 12:50 | |
| *** yassine is now known as Guest73860 | 12:50 | |
| *** shaohe_feng has joined #openstack-nova | 12:50 | |
| *** mgoddard__ has quit IRC | 12:50 | |
| *** diogogmt has quit IRC | 12:50 | |
| *** fregl_ has joined #openstack-nova | 12:51 | |
| *** mgoddard__ has joined #openstack-nova | 12:51 | |
| *** diogogmt has joined #openstack-nova | 12:52 | |
| *** Guest9665 has quit IRC | 12:52 | |
| *** rk4n has quit IRC | 12:52 | |
| markus_z | zhenguo_: So somewhere near this https://github.com/openstack/nova/blob/d2875b78b5746bfcb082a7c5385375d704518581/nova/virt/ironic/driver.py#L166-L166 ? | 12:52 |
| *** rk4n has joined #openstack-nova | 12:52 | |
| *** Guest73860 is now known as yassou | 12:53 | |
| *** akshai has joined #openstack-nova | 12:53 | |
| *** andrewbogott has quit IRC | 12:54 | |
| *** andrewbogott has joined #openstack-nova | 12:54 | |
| *** akshai_ has joined #openstack-nova | 12:54 | |
| zhenguo_ | markus_z, yeah, I plan to add the check there | 12:54 |
| *** kylek3h has joined #openstack-nova | 12:54 | |
| *** baoli has joined #openstack-nova | 12:55 | |
| markus_z | jroll: ^ That's a valid issue, isn't it? | 12:55 |
| *** sheel has quit IRC | 12:55 | |
| markus_z | jroll: context: https://bugs.launchpad.net/nova/+bug/1582663 | 12:55 |
| openstack | Launchpad bug 1582663 in OpenStack Compute (nova) "An Ironic node should not be exposed to Nova if no Ironic port is associated with it" [Undecided,Incomplete] - Assigned to Zhenguo Niu (niu-zglinux) | 12:55 |
| *** baoli_ has joined #openstack-nova | 12:56 | |
| mdbooth | dane-fichter: Will look at both, thanks | 12:56 |
| dane-fichter | mdbooth: Thanks! I think deprecation of the existing checksum mechanism could easily be added. | 12:57 |
| *** david-lyle has joined #openstack-nova | 12:57 | |
| jroll | markus_z: seems legit, though this is going to hit the ironic api like crazy | 12:57 |
| jroll | (if done how I think it will be done) | 12:57 |
| *** gjayavelu has joined #openstack-nova | 12:58 | |
| * jroll thinks maybe we shouldn't allow transition to AVAILABLE state if no ports are defined | 12:58 | |
| *** akshai has quit IRC | 12:58 | |
| openstackgerrit | Lucian Petrut proposed openstack/nova: Add Hyper-V storage QoS support https://review.openstack.org/175479 | 12:58 |
| markus_z | jroll: Ah, it's part of a periodic task, right? | 12:59 |
| *** shaohe_feng has quit IRC | 12:59 | |
| jroll | markus_z: yes, but right now we do one api call to fetch all node objects, that won't include if there are ports or not, we could either do $nodecount api calls for ports, or one huge /ports call and join in python | 13:00 |
| *** baoli has quit IRC | 13:00 | |
| markus_z | jroll: That's why it will hit the Ironic API quite often. | 13:00 |
| jroll | either way isn't great :) | 13:00 |
| jroll | markus_z: anyway, it's a valid bug, to answer your question :) | 13:00 |
| zhenguo_ | jroll: yeah, as the bug description said, at first I plan to do that in verifying process, but operators can also remove the port when it's already in available state | 13:00 |
| *** jistr is now known as jistr|mtg | 13:01 | |
| *** PaulMurray has joined #openstack-nova | 13:01 | |
| *** lucasagomes is now known as lucas-afk | 13:01 | |
| *** david-lyle_ has quit IRC | 13:01 | |
| jroll | zhenguo_: aha, that should be fine | 13:01 |
| jroll | zhenguo_: sorry, missed that verifying bit | 13:01 |
| markus_z | jroll: Yeah, I was just a little curious how Ironic works. I know next to nothing about it :) | 13:01 |
| jroll | markus_z: like most nova folks :) | 13:01 |
| *** pilgrimstack has joined #openstack-nova | 13:01 | |
| jroll | markus_z: tl;dr if there's no ports defined, we can't hook up networking | 13:02 |
| * markus_z feels safe as part of the majority :D | 13:02 | |
| *** shaohe_feng has joined #openstack-nova | 13:02 | |
| *** abalutoiu_ has joined #openstack-nova | 13:03 | |
| pilgrimstack | do you know if there is a bp or spec about extending the states of an instance to provide the information about real connectivity availability for example? | 13:03 |
| markus_z | jroll: yeah, and saying "this node is available" and then fail a provisioning on that node seems rather ugly | 13:03 |
| markus_z | cool cool, again something learned | 13:03 |
| *** pwnall has joined #openstack-nova | 13:04 | |
| *** abalutoiu has quit IRC | 13:04 | |
| *** PaulMurray has quit IRC | 13:04 | |
| pilgrimstack | the goal should be to be sure to have an ssh connection available and not targeting the boot time while ssh is not yet ready. | 13:04 |
| *** pradk has joined #openstack-nova | 13:05 | |
| *** edmondsw has joined #openstack-nova | 13:05 | |
| moshele | sean-k-mooney: hi can you join SR-IOV meeting ? | 13:06 |
| *** tbachman has quit IRC | 13:06 | |
| *** rlrossit has joined #openstack-nova | 13:07 | |
| *** diogogmt has quit IRC | 13:07 | |
| markus_z | auggy: Can you maybe host the nova-bugs-team meeting on next Tuesday, 05/31? That's the US friendly time slot and we have a local team event at this time. | 13:07 |
| pilgrimstack | ping bauzas, do you have information about it? | 13:07 |
| bauzas | pilgrimstack: mmm ? /me scrolling back | 13:08 |
| bauzas | ah | 13:08 |
| bauzas | pilgrimstack: you mean knowing whether the instance is responding to SSH ? | 13:08 |
| *** sdake has joined #openstack-nova | 13:09 | |
| bauzas | I mean, checking whether the guest OS is running good or bad ? | 13:09 |
| *** AmyLee7 has joined #openstack-nova | 13:09 | |
| sdague | pilgrimstack: the answer is no | 13:09 |
| *** abhishekk has joined #openstack-nova | 13:09 | |
| *** xyang1 has joined #openstack-nova | 13:09 | |
| bauzas | pilgrimstack: honestly, not sure it's part of the Nova mission | 13:09 |
| *** jaosorior has quit IRC | 13:10 | |
| *** shaohe_feng has quit IRC | 13:10 | |
| *** efoley has joined #openstack-nova | 13:10 | |
| *** shaohe_feng has joined #openstack-nova | 13:10 | |
| *** salv-orlando has joined #openstack-nova | 13:11 | |
| *** PaulMurray has joined #openstack-nova | 13:11 | |
| sdague | pilgrimstack: it would definitely need a spec, and it's sufficiently tricky on the edges | 13:11 |
| johnthetubaguy | +1 to what sdague just said | 13:12 |
| *** BjoernT has joined #openstack-nova | 13:12 | |
| johnthetubaguy | and honestly, +1 to bauzas too | 13:12 |
| *** efoley_ has quit IRC | 13:12 | |
| *** BjoernT is now known as Bjoern_zZzZzZzZ | 13:12 | |
| *** efoley_ has joined #openstack-nova | 13:12 | |
| *** efoley__ has quit IRC | 13:12 | |
| bauzas | pilgrimstack: the real problem is that Nova shouldn't really think about supporting this or that as supported guest OSes | 13:13 |
| bauzas | there is a Tempest (and Defcore) discussion about that IIRc | 13:13 |
| *** roeyc has quit IRC | 13:13 | |
| sdague | johnthetubaguy: sure... except, the amount of code that's written to do ping / ssh tight loops against guests in the wild world is kind of gross | 13:13 |
| *** sdake has quit IRC | 13:13 | |
| sdague | but, it is sufficiently tricky, especially when something goes wrong | 13:14 |
| bauzas | yeah | 13:14 |
| bauzas | that looks to me better to be checked by another project which could get information from the guest | 13:14 |
| sdague | bauzas: maybe, there are a lot of details you don't have access to | 13:15 |
| bauzas | or that would mean virt drivers should support an introspection state API | 13:15 |
| sdague | bauzas: it's not really introspectionm | 13:15 |
| *** ctrath has joined #openstack-nova | 13:15 | |
| sdague | because it's really "can I talk to this service from the outside world" | 13:15 |
| bauzas | sdague: I thought pilgrimstack was mentioning a guest responding to ping or ssh ? | 13:16 |
| sdague | it would more likely be some kind of structured metadata on images about services that should exist | 13:16 |
| sdague | bauzas: right, that's not guest introspection | 13:16 |
| sdague | that's network layer | 13:16 |
| *** eharney has quit IRC | 13:17 | |
| bauzas | agreed then | 13:17 |
| sdague | the virt driver wouldn't actually be involved | 13:17 |
| bauzas | it's a network-related question | 13:17 |
| bauzas | yeah, I see your point and you're right | 13:17 |
| bauzas | that looks honestly very terrible to see it in Nova | 13:17 |
| sdague | but, that also makes assurances about every part of your network config | 13:17 |
| johnthetubaguy | there is certainly a problem to be solved here that needs a solution | 13:18 |
| *** klkumar has quit IRC | 13:18 | |
| johnthetubaguy | I know we have an auto install of a monitoring agent (external and internal) that checks these kinds of things for you, but I am sure there are other ways to do this. | 13:19 |
| pilgrimstack | agree, customer needs are: I need to know when my instance/os is reachable without trying and getting timout until my service is ready | 13:20 |
| johnthetubaguy | for completeness, this is what is used: http://blueflood.io/ | 13:20 |
| *** shaohe_feng has quit IRC | 13:20 | |
| pilgrimstack | more of that, the customers need to know it from OpenStack API | 13:20 |
| *** shaohe_feng has joined #openstack-nova | 13:21 | |
| bauzas | johnthetubaguy: when I was op, we were using other ways to know the completeness of an instance creation from the guest PoV | 13:21 |
| bauzas | but all of those solutions look hardly to be implemented in the low-level that Nova is | 13:21 |
| bauzas | hard* | 13:22 |
| sdague | right, it's sufficiently tricky | 13:22 |
| johnthetubaguy | +1 on it being a tricky problem, that needs thought (i.e needs a spec) | 13:22 |
| bauzas | IIRC, Heat provides a callback mechanism for knowing whether the stack creation is ended | 13:23 |
| sdague | the place to annotate the information is clearly the image, but the thing that has to verify it seems clearly neutron, and who is responsible for keeping track of it is.... weird | 13:23 |
| *** roeyc has joined #openstack-nova | 13:23 | |
| bauzas | but that implies that the guest can reach the callback API | 13:23 |
| sdague | bauzas: does stack creation really verify that though? | 13:23 |
| johnthetubaguy | sdague: +1 it feels to me like it sits on top of multiple OpenStack APIs | 13:23 |
| bauzas | sdague: IIRC, yes | 13:23 |
| pilgrimstack | using meta-data + userdata, cloud-init get specific instruction from meta-data and PUT something on nova side | 13:23 |
| pilgrimstack | ? | 13:23 |
| bauzas | sdague: maybe at the template level tho | 13:24 |
| *** dave-mccowan has joined #openstack-nova | 13:24 | |
| pilgrimstack | juste an idea | 13:24 |
| bauzas | pilgrimstack: well, like sdague,it's a network-related problem | 13:24 |
| johnthetubaguy | so I did have a spec for allowing a guest to post server tags and metadata via the metadata service | 13:24 |
| *** Bjoern_zZzZzZzZ is now known as BjoernT | 13:24 | |
| johnthetubaguy | I do like the "i'm am done" ping from cloud-init as a callback, not so sure I like the general system | 13:25 |
| sdague | well, it also requires a very different scaling model on the metadata server | 13:25 |
| bauzas | yeah | 13:25 |
| johnthetubaguy | yeah, +1 | 13:25 |
| sdague | which is a massively untested chunk of code designed to be read only right now | 13:25 |
| johnthetubaguy | well, it does have set-ers already, for password, but its certainly not high frequency | 13:26 |
| sdague | right | 13:26 |
| *** zz_dimtruck is now known as dimtruck | 13:27 | |
| *** ayoung has joined #openstack-nova | 13:27 | |
| * johnthetubaguy wonders if monasca is hitting any of these things | 13:27 | |
| *** tbachman has joined #openstack-nova | 13:28 | |
| eliqiao | hi sdague johnthetubaguy , I got a qustion, I wonder what kinds of `sepc` can be added to flavor's extra spec filed, is there any guide? | 13:29 |
| *** READ10 has joined #openstack-nova | 13:29 | |
| *** lindycoder has joined #openstack-nova | 13:30 | |
| johnthetubaguy | eliqiao: I think jaypipes has some specs on replacing that with something that is more controlled and well defined. Its a bit messy right now | 13:30 |
| *** eharney has joined #openstack-nova | 13:30 | |
| *** shaohe_feng has quit IRC | 13:30 | |
| eliqiao | johnthetubaguy: thanks, so seems this is not good time to propose adding new extra_spec to flavor | 13:31 |
| *** abhishekk has left #openstack-nova | 13:32 | |
| *** mmandel has joined #openstack-nova | 13:32 | |
| johnthetubaguy | eliqiao: it depends how important it is, but I wouldn't recommend it | 13:32 |
| eliqiao | BTW, can you help to review https://review.openstack.org/#/c/319513/ (prevously propos to add extra_spec to flavor) | 13:32 |
| eliqiao | sdague gave some comments yesterday on use-case, and I improved them already. | 13:33 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Remove block_migration argument from live migration operations https://review.openstack.org/295316 | 13:33 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Remove block_migration from LM post operations https://review.openstack.org/295315 | 13:33 |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova: Remove block_migration from LM rollback https://review.openstack.org/292826 | 13:33 |
| *** ildikov has quit IRC | 13:34 | |
| eliqiao | Sudipta suggested to add tag to flavor's extra_spec but seems doesn't make sense now. | 13:34 |
| *** ildikov has joined #openstack-nova | 13:34 | |
| pilgrimstack | bauzas: not sure it's only a network problem, if you think about it more globally, like "I did something, now I want to update my state" from a instance point of view | 13:36 |
| pilgrimstack | ok so, I'll write a spec for the backlog, no skill to speak about implementation ;) | 13:36 |
| *** ctrath has quit IRC | 13:36 | |
| *** fesp has joined #openstack-nova | 13:36 | |
| bauzas | pilgrimstack: well, I'm just thinking that the solution is not really good for Nova because it's at least a network problem | 13:37 |
| *** tbachman_ has joined #openstack-nova | 13:38 | |
| *** efried has joined #openstack-nova | 13:38 | |
| bauzas | but I certainly acknowledge the problem which requires a solution | 13:38 |
| bauzas | just that the solution could probably involve more than nova | 13:38 |
| *** tbachman has quit IRC | 13:39 | |
| *** tbachman_ is now known as tbachman | 13:39 | |
| kashyap | BobBall: 'check-citrix-xenserver' seems to fail -- Is it just me? | 13:39 |
| BobBall | linky? | 13:39 |
| *** fesp has quit IRC | 13:40 | |
| *** shaohe_feng_ has joined #openstack-nova | 13:40 | |
| *** gjayavelu has quit IRC | 13:40 | |
| BobBall | http://ci-watch.tintri.com/project?project=nova suggests that it is just you... but before I lay blame it'd be good to have a further dig that it's not something weird about your patch :) | 13:41 |
| *** shaohe_feng has joined #openstack-nova | 13:41 | |
| *** shaohe_feng_ has quit IRC | 13:41 | |
| *** NobodyCam has joined #openstack-nova | 13:42 | |
| kashyap | BobBall: Yeah, checking the log - http://dd6b71949550285df7dc-dda4e480e005aaa13ec303551d2d8155.r49.cf1.rackcdn.com/09/319209/9/16214/run_tests.log | 13:43 |
| BobBall | stack.sh failed but it's not at all clear why... | 13:44 |
| BobBall | just says generate-subunit not found | 13:44 |
| *** iceyao has joined #openstack-nova | 13:45 | |
| *** ssurana has joined #openstack-nova | 13:45 | |
| kashyap | Yeah, looking... | 13:45 |
| *** ametts has joined #openstack-nova | 13:46 | |
| BobBall | hum de hum... very plausible that it's related to your change somehow though | 13:46 |
| kashyap | I didn't change anything w.r.t those deps | 13:47 |
| BobBall | I wonder if systemctl is installed but not active | 13:47 |
| kashyap | BobBall: If it's not active, it should resort to `service`. | 13:47 |
| BobBall | you mean sysctl itself should? or devstack? cuz devstack is only checking it exists? | 13:48 |
| BobBall | heh | 13:48 |
| *** mgoddard__ has quit IRC | 13:49 | |
| BobBall | Missing a space after True? | 13:49 |
| bauzas | pilgrimstack: sdague: johnthetubaguy: so, I remember now how Heat signals a guest OS running http://docs.openstack.org/developer/heat/template_guide/software_deployment.html#signals-and-wait-conditions | 13:49 |
| bauzas | it's what they call "wait conditions" | 13:49 |
| BobBall | I've seen lots of issues with missing spaces, but never really understood when it causes a problem and when it doesn't | 13:49 |
| kashyap | BobBall: Sigh, let me check | 13:50 |
| BobBall | kashyap: Yes - that's the problem | 13:50 |
| BobBall | kashyap: I'm surprised it works anywhere else... | 13:50 |
| kashyap | Yep, that it... | 13:50 |
| BobBall | http://paste.openstack.org/show/498614/ | 13:50 |
| bauzas | pilgrimstack: sdague: johnthetubaguy: that's basically a curl call by the guest OS to the Orchestration API | 13:50 |
| *** shaohe_feng has quit IRC | 13:51 | |
| bauzas | which is passed thru userdata | 13:51 |
| *** mrkz has joined #openstack-nova | 13:51 | |
| BobBall | Ah - looks like our CI is just so awesome that it reports very quickly. If you wait long enough, I bet Jenkins will fail the patch too :) | 13:51 |
| *** shaohe_feng has joined #openstack-nova | 13:51 | |
| *** claudiub has quit IRC | 13:51 | |
| *** klkumar has joined #openstack-nova | 13:51 | |
| *** yamahata has joined #openstack-nova | 13:52 | |
| kashyap | BobBall: Yeah, it _is_ awesome! | 13:52 |
| *** jbernard_ is now known as jbernard | 13:52 | |
| kashyap | BobBall: And, hey -- I bothered to diligently check it :P | 13:52 |
| BobBall | Yes - Thank you indeed! :) | 13:52 |
| *** burt has joined #openstack-nova | 13:53 | |
| kashyap | Likewise, for the catch. | 13:53 |
| *** coreywright has joined #openstack-nova | 13:53 | |
| *** mgoddard_ has joined #openstack-nova | 13:53 | |
| *** anteaya has joined #openstack-nova | 13:54 | |
| *** garyk1 has quit IRC | 13:54 | |
| johnthetubaguy | paul-carlton2: paul-carlton: PaulMurray: took a look at the auto complete live-migrate stuff, looks really close, I just had one thing that confused me when I was reading the spec | 13:55 |
| johnthetubaguy | it might be my inability to read, but added a comment about that | 13:55 |
| PaulMurray | thanks johnthetubaguy - paul-carlton and I had a long chat about it the other day - I think he misinterpreted your comment :) | 13:58 |
| PaulMurray | will look | 13:58 |
| openstackgerrit | Evgeny Antyshev proposed openstack/nova: libvirt: add nova volume driver for vzstorage https://review.openstack.org/190843 | 13:58 |
| PaulMurray | Live migration IRC meeting about to start on #openstack-meeting-3 PaulMurray paul-carlton2 eliqiao shaohe_feng pkoniszewski jlanoux yuntongjin mdbooth kashyap kmartin claudiub tobasco tdurakov andreas_s abhishekk diana_clarke | 13:59 |
| tdurakov | PaulMurray: hey, last week got cold so, missed the meeting, will join now | 13:59 |
| *** laszloh has quit IRC | 14:00 | |
| *** shaohe_feng has quit IRC | 14:01 | |
| *** READ10 has quit IRC | 14:01 | |
| *** shaohe_feng has joined #openstack-nova | 14:02 | |
| *** ducttape_ has joined #openstack-nova | 14:02 | |
| pkholkin | johnthetubaguy: sdague: hi folks! do you have some time to discuss this my spec? https://review.openstack.org/#/c/239663/ | 14:03 |
| *** zenoway has quit IRC | 14:03 | |
| pkholkin | I started the implementation and understood some problems/questions | 14:03 |
| *** sudipto has joined #openstack-nova | 14:04 | |
| pkholkin | mriedem: are you here, too? ^ | 14:04 |
| *** jistr|mtg is now known as jistr | 14:04 | |
| openstackgerrit | Gleb Stepanov proposed openstack/nova: Add console log offset https://review.openstack.org/292473 | 14:05 |
| kashyap | pkholkin: He's away on leave for a few days. | 14:05 |
| *** gagehugo has joined #openstack-nova | 14:05 | |
| sdague | pkholkin: sure, what's the issue? | 14:06 |
| pkholkin | kashyap: thanks! | 14:06 |
| sean-k-mooney | moshele: not sure if your still about. sorry i was not at my desk. i assume the SR-IOV meeting you wanted me to attend is over? | 14:07 |
| *** MikeG451 has joined #openstack-nova | 14:08 | |
| *** ducttape_ has quit IRC | 14:08 | |
| pkholkin | sdague: the first thing is as I see now keypairs could be got also from api db https://github.com/openstack/nova/blob/master/nova/objects/keypair.py#L198-L202 . I wanted to add to this function parameters 'limit' and 'marker' but now it should be done somehow for two dbs | 14:09 |
| *** woodster_ has joined #openstack-nova | 14:09 | |
| *** thumpba has joined #openstack-nova | 14:09 | |
| *** thumpba has quit IRC | 14:09 | |
| *** ducttape_ has joined #openstack-nova | 14:10 | |
| sdague | pkholkin: right, you will need to do that | 14:10 |
| pkholkin | e.g. limit / 2 for each, different markers | 14:10 |
| pkholkin | as I understand they should be separate | 14:11 |
| sdague | oh, because of online migrations.... | 14:11 |
| sdague | ug | 14:11 |
| *** thumpba has joined #openstack-nova | 14:11 | |
| Guest71132 | we have that for flavors.. works fine | 14:11 |
| *** Guest71132 is now known as dansmith | 14:11 | |
| *** shaohe_feng has quit IRC | 14:11 | |
| *** dansmith is now known as Guest21621 | 14:12 | |
| *** lucas-afk is now known as lucasagomes | 14:12 | |
| *** shaohe_feng has joined #openstack-nova | 14:12 | |
| pkholkin | sdague: it looks more complex than our general paging | 14:13 |
| *** jwcroppe_ has joined #openstack-nova | 14:13 | |
| Guest21621 | pkholkin: sdague: https://github.com/openstack/nova/blob/master/nova/objects/flavor.py#L651-L682 | 14:13 |
| *** penick has joined #openstack-nova | 14:13 | |
| auggy | markus_z: yes I can! | 14:13 |
| kashyap | BobBall: BTW -- the subunit thing seems to be a real bug, afazekas points out - https://bugs.launchpad.net/devstack/+bug/1547379 | 14:13 |
| openstack | Launchpad bug 1547379 in devstack "./stack.sh: line 488: generate-subunit: command not found" [Undecided,Invalid] | 14:13 |
| Guest21621 | pkholkin: also, you could just page the new location and not the old and exclude the old data if paging is requested, since that migration should be done early in a newton deploy and then it goes away | 14:13 |
| *** Guest21621 is now known as dansmith | 14:14 | |
| *** diogogmt has joined #openstack-nova | 14:14 | |
| *** 64MAAATDJ has joined #openstack-nova | 14:14 | |
| *** 64MAAATDJ has quit IRC | 14:14 | |
| BobBall | interesting | 14:14 |
| *** thumpba has quit IRC | 14:14 | |
| *** jwcroppe has quit IRC | 14:14 | |
| sdague | dansmith: ah, I get it, definitely odd on stable ordering, but I guess it's about as good as one gets | 14:14 |
| dansmith | sdague: yeah, the sorting thing is a different problem, but the paging is not hard | 14:15 |
| sdague | right | 14:15 |
| dansmith | for keypairs we could sort by owner when we do the online migrations and make the sorting more likely to be right if we cared, but .. it's a brief window | 14:16 |
| openstackgerrit | Yosef Hoffman proposed openstack/nova: Update Support Matrix https://review.openstack.org/320477 | 14:16 |
| pkholkin | dansmith: I should look into that code for details, thanks for the link | 14:16 |
| dansmith | pkholkin: np | 14:16 |
| *** takedakn has joined #openstack-nova | 14:16 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Fixed clean up process in confirm_resize() after cold migration https://review.openstack.org/320478 | 14:16 |
| kashyap | BobBall: That might not be as interesting: "that error is a false positive, it's happening because generate-subunit is called in the exit hook" [from #openstack-qa] :-) | 14:17 |
| *** sheel has joined #openstack-nova | 14:18 | |
| pkholkin | sdague: dansmith: about sorting, in sqlaclhemy.api we filter them by 'user_id' https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L3085 | 14:18 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Fixed clean up process in confirm_resize() after cold migration https://review.openstack.org/320478 | 14:18 |
| BobBall | oh haha fair enough | 14:18 |
| *** sridharg has quit IRC | 14:18 | |
| pkholkin | I don't understand for what 'order_by' is needed here | 14:18 |
| pkholkin | we can e.g. order_by them by 'name', but by 'user_id' they are already filtered | 14:19 |
| *** burt has quit IRC | 14:19 | |
| *** burt has joined #openstack-nova | 14:19 | |
| dansmith | pkholkin: I don't see order_by in that | 14:20 |
| pkholkin | in spec we said that we want to add it here | 14:20 |
| *** penick has quit IRC | 14:21 | |
| pkholkin | but now I think why ) | 14:21 |
| dansmith | unsure | 14:21 |
| *** shaohe_feng has quit IRC | 14:21 | |
| *** shaohe_feng has joined #openstack-nova | 14:22 | |
| pkholkin | maybe I mean sorting in some other place? | 14:22 |
| pkholkin | not in getting key pairs from db | 14:22 |
| pkholkin | because I'm also confused | 14:23 |
| *** diogogmt has quit IRC | 14:23 | |
| *** sacharya has joined #openstack-nova | 14:24 | |
| *** sacharya has quit IRC | 14:24 | |
| pkholkin | sdague: dansmith: ^ | 14:24 |
| pkholkin | you* mean, sorry | 14:24 |
| dansmith | pkholkin: are you referring to my "sort by user first" comment? | 14:25 |
| pkholkin | dansmith: yes | 14:26 |
| pkholkin | and for spec in general | 14:26 |
| sdague | honestly, I think we can ignore the sort by user_id bit, lets just try to create a stable order around the db key | 14:26 |
| sdague | it's a paging spec, not a sorting one | 14:26 |
| pkholkin | I have some misunderstanding now | 14:26 |
| sdague | and sorting is going to be all over the place with the 2 dbs anyway | 14:26 |
| dansmith | pkholkin: I meant in the keypair migration so that a given user is more likely to have their keypairs all unmigrated or migrated, but I was just brainstorming.. no need for you to make a change | 14:27 |
| alaski | wth http://git.openstack.org/cgit/openstack/ailuropoda/tree/README.md | 14:27 |
| *** rhagarty has joined #openstack-nova | 14:28 | |
| openstackgerrit | Pushkar Umaranikar proposed openstack/nova: Improve help text for resource tracker options https://review.openstack.org/320147 | 14:28 |
| pkholkin | sdague: dansmith: ok, I will think a little bit and try to write some implementation piece of code to look at | 14:28 |
| pkholkin | thanks for your help | 14:28 |
| *** logan- has quit IRC | 14:29 | |
| dansmith | alaski: nice | 14:29 |
| *** logan- has joined #openstack-nova | 14:30 | |
| sdague | alaski: is that their attempt for fail whale? | 14:30 |
| alaski | I guess so. The big tent is much larger than I expected | 14:31 |
| sdague | well, that's not big tent | 14:31 |
| sdague | that's just random things | 14:31 |
| sdague | sigh, I wish cgit was useful - http://git.openstack.org/cgit/openstack/ailuropoda/tree/images/v1/sad_panda01.png?id=b835ab2113da27d5c747b9e984eb74ddd19a3f8b | 14:31 |
| alaski | oh, there's no barrier to getting a repo now? | 14:32 |
| *** shaohe_feng has quit IRC | 14:32 | |
| sdague | alaski: osi license | 14:32 |
| sdague | just like stackforge | 14:32 |
| alaski | https://github.com/openstack/ailuropoda/blob/master/images/v1/sad_panda01.png | 14:32 |
| alaski | sdague: ahh | 14:32 |
| *** shaohe_feng has joined #openstack-nova | 14:32 | |
| *** sridharg has joined #openstack-nova | 14:32 | |
| *** pilgrimstack1 has joined #openstack-nova | 14:32 | |
| alaski | dansmith: I hope you're sufficiently caffeinated for a discussion. I ran into a interesting issue around deletes in cellsv2 | 14:33 |
| *** takedakn has quit IRC | 14:34 | |
| dansmith | alaski: oh boy | 14:34 |
| alaski | doffm: melwitt bauzas ^ if you're interested | 14:34 |
| alaski | so the crux of it is that right now deletes block further progress on a boot due to expected_task_state checks on instance.save | 14:34 |
| * bauzas refueling with coffee (fortunately, he can do that, not with petrol) | 14:35 | |
| alaski | but a delete of a BuildRequest does not block further progress | 14:35 |
| *** ssurana has quit IRC | 14:35 | |
| alaski | so a delete of an instance before it is scheduled does not prevent scheduling | 14:35 |
| *** rk4n_ has joined #openstack-nova | 14:35 | |
| dansmith | alaski: meaning if you're in the middle of scheduling? | 14:35 |
| alaski | yes | 14:35 |
| bauzas | interesting | 14:35 |
| alaski | so I've been considering adding a lock to the instance_mapping | 14:35 |
| dansmith | alaski: but we can avoid the build going to the cell post-scheduling or something pretty easily right? | 14:36 |
| *** rk4n has quit IRC | 14:36 | |
| alaski | dansmith: we don't have anything to do that yet, but it shouldn't be too hard to add it | 14:36 |
| dansmith | alaski: oh, you delete the buildreq once we've scheduled or something right? buildreq is short-lived? | 14:36 |
| alaski | dansmith: yes | 14:36 |
| dansmith | alaski: don't we have the reqspec too? | 14:36 |
| alaski | the flow is add build_req, map to build_req, schedule, write instance, map to instance, deleted build_req | 14:37 |
| alaski | dansmith: we do | 14:37 |
| *** askb has quit IRC | 14:37 | |
| *** ctrath has joined #openstack-nova | 14:37 | |
| dansmith | alaski: if you delete in the middle of schedule there, | 14:37 |
| *** ctrath has quit IRC | 14:37 | |
| alaski | I was thinking of catching the delete at the "map to instance" phase, but that would require cleaning up an instance that had just been written | 14:37 |
| dansmith | alaski: then when we go to map_instance we fail because the buildreq is gone right? | 14:38 |
| alaski | no, the map is independent of the buildreq | 14:38 |
| bauzas | alaski: you want to avoid a scheduling call, or you just want to prevent an exception when updating the mapping ? | 14:38 |
| *** ljjjustin has quit IRC | 14:38 | |
| *** ssurana has joined #openstack-nova | 14:38 | |
| alaski | bauzas: I want to prevent the build from proceeding after scheduling | 14:38 |
| bauzas | okay | 14:39 |
| dansmith | alaski: I think you just need to delete build_req and instance if it exists during delete, and then in write_instance make sure the buildreq is there, and in map_to_instance make sure the instance is there, no? | 14:39 |
| bauzas | I was about to say the same | 14:40 |
| *** ctrath has joined #openstack-nova | 14:40 | |
| *** jaypipes has quit IRC | 14:40 | |
| *** zzzeek has quit IRC | 14:40 | |
| bauzas | delete instance and build_req and just make sure you can't update something which is deleted | 14:40 |
| alaski | dansmith: that should work, but I think it leaves a tiny race | 14:40 |
| bauzas | we're still in the api conductor right? | 14:41 |
| *** ljjjustin has joined #openstack-nova | 14:41 | |
| alaski | because we can't write the instance in a transaction with build_req | 14:41 |
| *** abalutoiu_ has quit IRC | 14:41 | |
| alaski | bauzas: yes | 14:41 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Fixed clean up process in confirm_resize() after cold migration https://review.openstack.org/320478 | 14:42 |
| *** shaohe_feng has quit IRC | 14:42 | |
| dansmith | alaski: ah right, so the race is just that a delete could go unnoticed at some point right? | 14:42 |
| *** anteaya has quit IRC | 14:42 | |
| *** BjoernT has quit IRC | 14:42 | |
| *** david-lyle has quit IRC | 14:42 | |
| *** aysyd has quit IRC | 14:42 | |
| *** mvk has quit IRC | 14:42 | |
| *** wznoinsk has quit IRC | 14:42 | |
| *** markus_z has quit IRC | 14:42 | |
| *** josecastroleon has quit IRC | 14:42 | |
| *** ujjain has quit IRC | 14:42 | |
| *** markmcclain has quit IRC | 14:42 | |
| *** _xin has quit IRC | 14:42 | |
| *** Yingxin has quit IRC | 14:42 | |
| *** mgoddard_ has quit IRC | 14:42 | |
| *** ecelik has quit IRC | 14:42 | |
| *** nelsnels_ has quit IRC | 14:42 | |
| *** dank_ has quit IRC | 14:42 | |
| *** markmc has quit IRC | 14:42 | |
| *** pilgrimstack has quit IRC | 14:42 | |
| *** rmk has quit IRC | 14:42 | |
| bauzas | alaski: is that terrible to accept updating instance.host and just verify when mapping the instance to the cell that the row was not deleted ? | 14:42 |
| *** mmandel has quit IRC | 14:42 | |
| alaski | dansmith: yes | 14:42 |
| dansmith | alaski: we might need to check that the buildreq is still there after we create the instance before we return or someting | 14:42 |
| bauzas | f*ck, netsplit | 14:42 |
| *** akshai_ has quit IRC | 14:42 | |
| dansmith | alaski: but a single dropped delete request that isn't blocked isn't the worst thing I think | 14:43 |
| *** shaohe_feng has joined #openstack-nova | 14:43 | |
| *** pushkaru has joined #openstack-nova | 14:43 | |
| alaski | dansmith: it's not. and it's a place I wouldn't expect many deletes to happen in | 14:43 |
| alaski | a post check on buildreq, after instance write, does minimize it further I think | 14:43 |
| dansmith | alaski: you could write the instance and delete the buildreq in the same operation (separate transactions of course) and just handle the rollback if you fail to delete the buildreq | 14:44 |
| dansmith | right | 14:44 |
| *** StevenK_ is now known as StevenK | 14:44 | |
| *** _d34dh0r53_ is now known as d34dh0r53 | 14:45 | |
| *** fragatina has joined #openstack-nova | 14:45 | |
| alaski | okay. we'll I'll try going in that direction. But I'm open to further ideas | 14:45 |
| *** fragatina has quit IRC | 14:45 | |
| alaski | I was thinking of something similar with locking the instance_mapping. but checking build_req is roughly equivalent | 14:45 |
| bauzas | doffm: dansmith: FYI, I -1d https://review.openstack.org/#/c/294806/ because of a missing check, but I'm all good for the rest of the series | 14:46 |
| *** fragatina has joined #openstack-nova | 14:46 | |
| bauzas | doffm: dansmith: we could add that check later, but given it's a migration check, I thought it was better to do that in the main change | 14:46 |
| bauzas | doffm: dansmith: if you're adding it quickly, I'll +2 it | 14:46 |
| *** rcernin has quit IRC | 14:47 | |
| alaski | bauzas: I'm not sure I understood your comment earlier. when mapping to the cell the instance row should never be deleted because without the mapping the delete code will not reach that instance row | 14:47 |
| bauzas | alaski: oh right | 14:48 |
| bauzas | alaski: but deleting an instance could possibly be deleting the cell mapping afterwards, so I guess I'm silently nodding what you and dansmith proposed | 14:49 |
| lucasagomes | folks, quick q: Does nova support reloading configuration files on SIGHUP? I've tried it on devstack and it does not seem to be the case for n-api and n-cpu | 14:49 |
| *** mlavalle has joined #openstack-nova | 14:49 | |
| *** sdake has joined #openstack-nova | 14:49 | |
| mlavalle | cdent: ping | 14:49 |
| cdent | mlavalle: hello | 14:50 |
| sdague | lucasagomes: not in the general case for sure | 14:50 |
| mlavalle | cdent: hi, quick question.... are you implementing https://review.openstack.org/#/c/300176/? | 14:50 |
| dansmith | lucasagomes: oslo.config has some support for it but nova doesn't use it at all, AFAIK, and definitely may cache values causing weird behaviors if you reload | 14:50 |
| sdague | I don't know if any of the mutable annotations landed yet | 14:50 |
| *** akshai has joined #openstack-nova | 14:50 | |
| cdent | mlavalle: yes, pretty much | 14:50 |
| lucasagomes | sdague, right, what would be the case ? For context, the spec about rolling upgrades talks about SIGHUP which is currently unsupported in Ironic. So I've started looking into it | 14:51 |
| lucasagomes | any tips/guidance is appreciated | 14:51 |
| cdent | mlavalle: the relevant stack of in progress code is near https://review.openstack.org/#/c/293104/ | 14:51 |
| sdague | lucasagomes: hmmm.... I'm not sure how SIGHUP relates to rolling upgrades | 14:51 |
| lucasagomes | dansmith, cool, yeah I saw the multable parameter that can be added to the config options | 14:51 |
| dansmith | lucasagomes: we support sighup for recalculating the automatic version pin | 14:51 |
| lucasagomes | sdague, yeah, that's what I'm trying to understand too (asked in the spec) https://review.openstack.org/#/c/299245/9 | 14:51 |
| cdent | mlavalle: there's a fair bit of flailing around and back and forth between code and spec as we figure out how reality is going to interfere | 14:51 |
| dansmith | lucasagomes: but that's only in that case, not for reloading the config even for a static pin | 14:51 |
| *** jichen has quit IRC | 14:52 | |
| lucasagomes | sdague, it seems that it's a fundamental part of the current propose in Ironic, but I don't think it's fundamental as well since nova does not rely on it (that was my question there) | 14:52 |
| mlavalle | cdent: yeah, reality is tough :-) | 14:52 |
| *** shaohe_feng has quit IRC | 14:52 | |
| alaski | bauzas: cool | 14:52 |
| lucasagomes | dansmith, right, gotcha. I will investigate it further | 14:53 |
| dansmith | lucasagomes: I can try to comment on that spec if you lik | 14:53 |
| lucasagomes | dansmith, that would be great if you have some minutes to do it | 14:54 |
| lucasagomes | very appreciated | 14:54 |
| dansmith | yep | 14:54 |
| cdent | mlavalle: there's a temporary pause in progress on that stack while some database tables get moved from cell to api | 14:54 |
| mlavalle | cdent: that patchset is at the top of the stack, right? | 14:55 |
| kashyap | tdurakov: So, from the meeting channel... | 14:55 |
| *** jaypipes has joined #openstack-nova | 14:55 | |
| kashyap | tdurakov: I'll be back in an hour or so from the train. First I need to head out for a bit, I'm afraid. | 14:55 |
| *** shaohe_feng has joined #openstack-nova | 14:55 | |
| tdurakov | kashyap: ok, ping me then | 14:55 |
| cdent | mlavalle: yes, and the entire stack needs to come in under those here https://review.openstack.org/#/c/314781/ and here https://review.openstack.org/#/c/315682/ | 14:56 |
| pkholkin | dansmith: Dan, could you please look this my patch https://review.openstack.org/#/c/304180/. There are some object changes, so I think you should look as an expert there :) | 14:56 |
| shaohe_feng | johnthetubaguy: ping\ | 14:56 |
| *** links has quit IRC | 14:57 | |
| shaohe_feng | johnthetubaguy: can you have a look at this patch https://review.openstack.org/#/c/139910/20 | 14:57 |
| dansmith | pkholkin: crap on a stick! is that really as small as that patch can be? | 14:57 |
| *** mmandel has joined #openstack-nova | 14:58 | |
| *** adisky has joined #openstack-nova | 14:58 | |
| *** READ10 has joined #openstack-nova | 14:58 | |
| pkholkin | dansmith: It looks like yes) api changes need many files to add/change( | 14:59 |
| dansmith | pkholkin: :( | 14:59 |
| *** rcernin has joined #openstack-nova | 14:59 | |
| sdague | pkholkin: you should be able to do the object changes in a dedicated patch | 15:00 |
| shaohe_feng | baoli_: ping. can you help to look at this patch https://review.openstack.org/#/c/139910/20 | 15:00 |
| *** jdurgin1 has joined #openstack-nova | 15:01 | |
| *** sajeesh has joined #openstack-nova | 15:02 | |
| cdent | mlavalle: did you have some questions, or were you just getting the lay of the land? | 15:02 |
| sdague | pkholkin: especially given the object changes are wrong and break upgrades | 15:02 |
| pkholkin | do you mean that cpu models objects hierarchy? | 15:02 |
| mlavalle | cdent: getting the lay of the land. Want to follow up this. Thanks for the great response | 15:02 |
| sdague | http://logs.openstack.org/80/304180/13/check/gate-grenade-dsvm-multinode/aa3d773/logs/subnode-2/old/screen-n-cpu.txt.gz#_2016-05-20_13_38_09_708 | 15:02 |
| *** shaohe_feng has quit IRC | 15:02 | |
| cdent | mlavalle: my pleasure | 15:03 |
| sdague | pkholkin: basically do all the infrastructure on the nova objects to support the new thing | 15:03 |
| pkholkin | I want to look it too | 15:03 |
| *** shaohe_feng has joined #openstack-nova | 15:03 | |
| sdague | and make the api changes to use that new things a dedicated patch | 15:03 |
| *** zzzeek has joined #openstack-nova | 15:03 | |
| sdague | dansmith: that schedule to all nodes test wins the day of realizing that old can't talk to the cluster any more | 15:03 |
| *** RuiChen has quit IRC | 15:04 | |
| dansmith | sdague: um what? | 15:04 |
| *** jistr is now known as jistr|mtg | 15:04 | |
| sdague | https://review.openstack.org/#/c/304180 | 15:04 |
| sdague | it fails multinode grenade | 15:04 |
| * dansmith looks at the logs above | 15:04 | |
| *** mgoddard has joined #openstack-nova | 15:04 | |
| sdague | because there are objects compatibility issues | 15:04 |
| sdague | and on the new side, you can no longer land anything on old nodes | 15:05 |
| dansmith | sdague: that's just a mind-o on the sending side, but .. yeah, compat routine didn't work so fail = goodness | 15:05 |
| dansmith | sdague: I told you I'm never working on a project without a multinode grenade job right? :) | 15:06 |
| sdague | heh | 15:07 |
| *** Guest80537 has quit IRC | 15:08 | |
| *** ctrath has quit IRC | 15:09 | |
| pkholkin | sdague: I will fix that mentioned error | 15:10 |
| pkholkin | but what do you prefer to split here? | 15:10 |
| sdague | pkholkin: I still think you should split the patch | 15:10 |
| *** diogogmt has joined #openstack-nova | 15:10 | |
| sdague | the objects changes in one patch | 15:11 |
| pkholkin | there is one file https://review.openstack.org/#/c/304180/13/nova/objects/vcpu_model.py with some vcpu_model_hierarchy | 15:11 |
| *** sridharg has quit IRC | 15:11 | |
| sdague | then a follow up with the microversion bump and the changes needed for that | 15:11 |
| *** Wenzhi_ has quit IRC | 15:11 | |
| *** Kennan has quit IRC | 15:11 | |
| sdague | because the objects changes shouldn't need to touch the api layer in anything more than a very trivial way | 15:11 |
| pkholkin | ok, and what about changine compute_node object? | 15:11 |
| *** Wenzhi_ has joined #openstack-nova | 15:12 | |
| pkholkin | changing* | 15:12 |
| pkholkin | it will be in the second with all other changes except that file? | 15:12 |
| *** shaohe_feng has quit IRC | 15:13 | |
| pkholkin | or changing compute_node should be included in the first change? without microversion bumb & | 15:13 |
| pkholkin | bump* ? | 15:14 |
| *** burt has quit IRC | 15:14 | |
| *** ctrath has joined #openstack-nova | 15:14 | |
| *** Kennan has joined #openstack-nova | 15:14 | |
| *** Junhongl_ has joined #openstack-nova | 15:15 | |
| pkholkin | https://review.openstack.org/#/c/304180/13/nova/objects/compute_node.py | 15:15 |
| dansmith | pkholkin: there are lots of issues with this, I'm leaving you comments :) | 15:15 |
| pkholkin | sdague: I think we can do all object changes in one the first | 15:15 |
| pkholkin | dansmith: that is great! ) thank you ! | 15:16 |
| sdague | pkholkin: cool | 15:16 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova-specs: Add multiattach support to Nova https://review.openstack.org/304681 | 15:16 |
| pkholkin | sdague, dansmith: thanks for help, will work and come back with it) | 15:17 |
| *** ssurana has quit IRC | 15:18 | |
| *** Junhongl_ has left #openstack-nova | 15:20 | |
| *** sudipto has quit IRC | 15:21 | |
| *** jlanoux has quit IRC | 15:22 | |
| *** shaohe_feng has joined #openstack-nova | 15:23 | |
| openstackgerrit | Sean Dague proposed openstack/nova: NOT-FOR-MERGE: Dummy change to exercise live migration CI job https://review.openstack.org/319934 | 15:23 |
| *** ssurana has joined #openstack-nova | 15:24 | |
| *** burt has joined #openstack-nova | 15:24 | |
| *** mjura has quit IRC | 15:26 | |
| *** sridharg has joined #openstack-nova | 15:27 | |
| *** jaosorior has joined #openstack-nova | 15:27 | |
| ildikov | johnthetubaguy: added a few comments and uploaded a new patch set as well | 15:28 |
| ildikov | johnthetubaguy: do you know who I could ask for more details about OS assisted snapshots? | 15:28 |
| *** syjulian has joined #openstack-nova | 15:28 | |
| *** tonytan4ever has joined #openstack-nova | 15:31 | |
| *** sfinucan has quit IRC | 15:32 | |
| *** shaohe_feng has quit IRC | 15:33 | |
| *** shaohe_feng has joined #openstack-nova | 15:34 | |
| *** salv-orlando has quit IRC | 15:34 | |
| *** ishant has joined #openstack-nova | 15:34 | |
| *** cdelatte has quit IRC | 15:35 | |
| *** links has joined #openstack-nova | 15:36 | |
| *** dmk0202 has quit IRC | 15:38 | |
| *** dmk0202 has joined #openstack-nova | 15:38 | |
| *** ihrachys has quit IRC | 15:39 | |
| *** ssurana has quit IRC | 15:39 | |
| *** rk4n_ has quit IRC | 15:39 | |
| kashyap | tdurakov: Heya | 15:40 |
| *** jistr|mtg is now known as jistr | 15:40 | |
| *** claudiub|2 has joined #openstack-nova | 15:40 | |
| *** ecelik has joined #openstack-nova | 15:40 | |
| *** nelsnels_ has joined #openstack-nova | 15:40 | |
| *** dank_ has joined #openstack-nova | 15:40 | |
| *** markmc has joined #openstack-nova | 15:40 | |
| *** rmk has joined #openstack-nova | 15:40 | |
| kashyap | tdurakov: I'm here (on the bus) for 15 minutes -- then get on the train -- they available for 2hrs | 15:40 |
| *** Junhongl_ has joined #openstack-nova | 15:40 | |
| *** dedery_ has joined #openstack-nova | 15:41 | |
| *** salv-orlando has joined #openstack-nova | 15:41 | |
| *** fragatina has quit IRC | 15:43 | |
| *** shaohe_feng has quit IRC | 15:43 | |
| *** ljxiash has quit IRC | 15:44 | |
| *** Junhongl_ has quit IRC | 15:44 | |
| *** dedery has quit IRC | 15:44 | |
| *** Junhongl_ has joined #openstack-nova | 15:45 | |
| *** ctrath has quit IRC | 15:45 | |
| tdurakov | kashyap: so, I've checked again, this restart doesn't help me | 15:45 |
| *** shaohe_feng has joined #openstack-nova | 15:45 | |
| *** delattec has joined #openstack-nova | 15:45 | |
| *** sudipto_ has joined #openstack-nova | 15:45 | |
| *** anteaya has joined #openstack-nova | 15:45 | |
| *** BjoernT has joined #openstack-nova | 15:45 | |
| *** david-lyle has joined #openstack-nova | 15:45 | |
| *** aysyd has joined #openstack-nova | 15:45 | |
| *** mvk has joined #openstack-nova | 15:45 | |
| *** wznoinsk has joined #openstack-nova | 15:45 | |
| *** markus_z has joined #openstack-nova | 15:45 | |
| *** josecastroleon has joined #openstack-nova | 15:45 | |
| *** ujjain has joined #openstack-nova | 15:45 | |
| *** markmcclain has joined #openstack-nova | 15:45 | |
| *** _xin has joined #openstack-nova | 15:45 | |
| *** Yingxin has joined #openstack-nova | 15:45 | |
| *** dedery_ has quit IRC | 15:45 | |
| *** shaohe_feng has quit IRC | 15:45 | |
| *** shaohe_feng has joined #openstack-nova | 15:45 | |
| tdurakov | kashyap: any ideas? | 15:46 |
| kashyap | tdurakov: Did you check the file I sent on the meeting channel? Do you see exactly the same behavior? | 15:46 |
| *** tbachman has quit IRC | 15:47 | |
| *** Junhongl_ has quit IRC | 15:47 | |
| tdurakov | kashyap: could you explain results a bit | 15:47 |
| PaulMurray | alaski, dansmith johnthetubaguy I think we agreed in Austin to go ahead with this trivial patch, but its been lying around since: https://review.openstack.org/#/c/215483/ | 15:48 |
| PaulMurray | any chance of a review to get it out the way please? | 15:48 |
| kashyap | tdurakov: See the first 4 cases here: https://kashyapc.fedorapeople.org/behavior-of-systemctl-and-service-with-libvirt-bin-on-Xenial.txt | 15:48 |
| *** Junhongl has joined #openstack-nova | 15:48 | |
| tdurakov | kashyap: according to this, i need to do service stop /service start, right? | 15:48 |
| *** penick has joined #openstack-nova | 15:49 | |
| *** fragatina has joined #openstack-nova | 15:49 | |
| kashyap | tdurakov: It is not a prescription; it was just my results from testing. Clark said, he could get the libvirt-bin service up, and the 'gate64' model being detected once he issued an explicit restart. | 15:49 |
| *** ssurana has joined #openstack-nova | 15:49 | |
| tdurakov | kashyap: yup | 15:49 |
| kashyap | tdurakov: But yes -- according to the above what you said is what I observe. | 15:50 |
| tdurakov | the problem i faced with, even if libvirt start to see gate64 vm creation fails | 15:50 |
| *** roeyc has quit IRC | 15:51 | |
| *** belmoreira has quit IRC | 15:51 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove APIRouter of legacy v2 API code https://review.openstack.org/313902 | 15:51 |
| *** Junhongl has quit IRC | 15:51 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove legacy v2 API tests which use wsgi_app() https://review.openstack.org/313192 | 15:51 |
| kashyap | tdurakov: How are you creating 'gate64 vm'? This one -- http://git.openstack.org/cgit/openstack-infra/project-config/tree/tools/build-image.sh | 15:51 |
| *** jdurgin1 has quit IRC | 15:52 | |
| *** ssurana has quit IRC | 15:52 | |
| * kashyap back in about half-an-hour. | 15:53 | |
| *** shaohe_feng has quit IRC | 15:54 | |
| *** diogogmt has quit IRC | 15:54 | |
| *** sridharg has quit IRC | 15:54 | |
| *** shaohe_feng has joined #openstack-nova | 15:54 | |
| *** Swami has joined #openstack-nova | 15:55 | |
| *** morgabra has quit IRC | 15:55 | |
| *** morgabra has joined #openstack-nova | 15:55 | |
| mdbooth | dane-fichter: Hadn't seen https://review.openstack.org/#/c/189843/ . Nice. Any reason not to default that on? | 15:56 |
| *** paul-carlton2 has quit IRC | 15:56 | |
| *** ssurana has joined #openstack-nova | 15:56 | |
| *** delattec has quit IRC | 15:57 | |
| *** sudipto_ has quit IRC | 15:57 | |
| *** anteaya has quit IRC | 15:57 | |
| *** BjoernT has quit IRC | 15:57 | |
| *** david-lyle has quit IRC | 15:57 | |
| *** aysyd has quit IRC | 15:57 | |
| *** mvk has quit IRC | 15:57 | |
| *** wznoinsk has quit IRC | 15:57 | |
| *** markus_z has quit IRC | 15:57 | |
| *** josecastroleon has quit IRC | 15:57 | |
| *** ujjain has quit IRC | 15:57 | |
| *** markmcclain has quit IRC | 15:57 | |
| *** _xin has quit IRC | 15:57 | |
| *** Yingxin has quit IRC | 15:57 | |
| *** armax has joined #openstack-nova | 15:57 | |
| *** mgoddard_ has joined #openstack-nova | 15:58 | |
| *** diogogmt has joined #openstack-nova | 15:59 | |
| openstackgerrit | Chris Dent proposed openstack/nova: Support for both microversion headers https://review.openstack.org/300077 | 15:59 |
| cdent | sdague, johnthetubaguy: you guys +2 the spec for that ^. I reckon it is fairly close to ready. | 16:00 |
| *** jaosorior has quit IRC | 16:01 | |
| *** mgoddard has quit IRC | 16:01 | |
| *** matrohon has quit IRC | 16:02 | |
| *** fragatina has quit IRC | 16:02 | |
| *** vtech has quit IRC | 16:02 | |
| *** shaohe_feng has quit IRC | 16:04 | |
| *** shaohe_feng has joined #openstack-nova | 16:05 | |
| *** jlanoux has joined #openstack-nova | 16:06 | |
| *** delattec has joined #openstack-nova | 16:06 | |
| *** sudipto_ has joined #openstack-nova | 16:06 | |
| *** david-lyle has joined #openstack-nova | 16:06 | |
| *** aysyd has joined #openstack-nova | 16:06 | |
| *** mvk has joined #openstack-nova | 16:06 | |
| *** wznoinsk has joined #openstack-nova | 16:06 | |
| *** markus_z has joined #openstack-nova | 16:06 | |
| *** josecastroleon has joined #openstack-nova | 16:06 | |
| *** ujjain has joined #openstack-nova | 16:06 | |
| *** markmcclain has joined #openstack-nova | 16:06 | |
| *** _xin has joined #openstack-nova | 16:06 | |
| *** Yingxin has joined #openstack-nova | 16:06 | |
| *** gjayavelu has joined #openstack-nova | 16:10 | |
| johnthetubaguy | cdent: the docs appear to be "from the future" a little bit | 16:10 |
| *** tbachman has joined #openstack-nova | 16:11 | |
| johnthetubaguy | cdent: I think we need to tell people more details on when that new header is available | 16:11 |
| *** sfinucan has joined #openstack-nova | 16:11 | |
| johnthetubaguy | there are lots of deployments exposing microversions that don't support the new header | 16:11 |
| *** paul-carlton2 has joined #openstack-nova | 16:12 | |
| *** kmartin has quit IRC | 16:12 | |
| cdent | johnthetubaguy: I debated with myself on the write way to make those doc changes and eventually settled on: Explain reality as it is when this code is merged, so that people will rejigger their expectations accordingly. I figured since it was dev docs, that was the right thing to do. For the end user docs...nothing _has_ to happen yet, people can go on doing things the old way if that's what they want. | 16:12 |
| cdent | s/write way/right way/ sigh, unintentional pun | 16:13 |
| johnthetubaguy | cdent: that makes sense, I guess we just need to be specific about when the new world applies | 16:13 |
| *** raunak has joined #openstack-nova | 16:13 | |
| johnthetubaguy | cdent: anyways, just reading through, will submit the comments in a sec | 16:13 |
| cdent | specific to whom, and what are the variable on "when" | 16:13 |
| *** moshele has quit IRC | 16:13 | |
| johnthetubaguy | cdent: so we know which microversion must be present for the header to work | 16:14 |
| johnthetubaguy | if the cloud doesn't support that, then you have to send the old header, I assume | 16:14 |
| *** shaohe_feng has quit IRC | 16:14 | |
| *** gjayavelu has quit IRC | 16:14 | |
| *** lakshmiS has joined #openstack-nova | 16:15 | |
| cdent | johnthetubaguy: yeah, but if someone is reading these docs, they have the new code, yeah? | 16:15 |
| markus_z | oomichi: Is this still a thing https://bugs.launchpad.net/nova/+bug/1533487 ? The auto-expiration didn't hit because an importance is set. | 16:15 |
| openstack | Launchpad bug 1533487 in OpenStack Compute (nova) "test_create_server_with_personality fails on the gate" [Wishlist,Incomplete] | 16:15 |
| *** shaohe_feng has joined #openstack-nova | 16:15 | |
| cdent | I'm understanding the general concern (making sure people understand globally what's up with microversions), but not understanding how it fits into the context of this change. | 16:15 |
| johnthetubaguy | cdent: no, the docs are public on our website | 16:15 |
| johnthetubaguy | cdent: new api-ref is read by old and new users | 16:16 |
| *** ssurana has quit IRC | 16:16 | |
| cdent | hrmmm, and that's generically associated with a version? | 16:16 |
| johnthetubaguy | http://developer.openstack.org/api-ref/compute/ | 16:16 |
| cdent | sorry _not_ | 16:16 |
| johnthetubaguy | yeah, not | 16:16 |
| *** nkrinner is now known as nkrinner_afk | 16:16 | |
| johnthetubaguy | cdent: we only have one API, and you see the evolution of it in that one doc | 16:16 |
| cdent | for sake of understanding: what have I changed in my patch that touches the url you've just shared? | 16:17 |
| johnthetubaguy | cdent: hmm, actually none of it, I was thinking about this file: https://review.openstack.org/#/c/300077/11/doc/source/api_microversion_dev.rst | 16:18 |
| *** sacharya has joined #openstack-nova | 16:18 | |
| oomichi | markus_z: it is fine to make it incomplete at this time as the LP history. | 16:18 |
| *** rk4n has joined #openstack-nova | 16:18 | |
| hemna | johnthetubaguy, when you have a minute..... | 16:18 |
| hemna | no hurry | 16:19 |
| johnthetubaguy | cdent: those docs go there: http://docs.openstack.org/developer/nova/ | 16:19 |
| markus_z | oomichi: "incomplete" is an open state, that's why I'm asking. | 16:19 |
| *** sudipto_ has quit IRC | 16:19 | |
| cdent | johnthetubaguy: I reckon I probably do need to touch those files, but I hadn't yet, thus my confusion | 16:19 |
| *** lakshmiS has quit IRC | 16:19 | |
| johnthetubaguy | cdent: yeah, sorry, adding their location in the comments | 16:20 |
| cdent | johnthetubaguy: awesome thanks | 16:20 |
| *** gjayavelu has joined #openstack-nova | 16:20 | |
| johnthetubaguy | cdent: you could ask why we describe that in three places right now, but anyways... | 16:20 |
| oomichi | markus_z: ah, ok. I have changed it to invalid with the comment on the LP | 16:21 |
| markus_z | oomichi: thanks! | 16:21 |
| cdent | I'm (very slowly) learning to only ask questions like that maybe once a week, instead of daily. | 16:21 |
| *** ssurana has joined #openstack-nova | 16:22 | |
| *** yamahata has quit IRC | 16:22 | |
| *** derekh has quit IRC | 16:22 | |
| johnthetubaguy | cdent: oh no! thats bad | 16:23 |
| cdent | It's exhausting keeping it up every day | 16:23 |
| johnthetubaguy | true | 16:23 |
| johnthetubaguy | although its good | 16:23 |
| johnthetubaguy | hemna: hey | 16:23 |
| *** lakshmiS has joined #openstack-nova | 16:23 | |
| hemna | hey, so in my effort to remove check_attach calls during BFV, I noticed something quite odd | 16:24 |
| hemna | http://paste.openstack.org/show/498382/ | 16:24 |
| *** dmk0202 has quit IRC | 16:24 | |
| hemna | it seems that bdm.get() isn't working | 16:24 |
| hemna | in this one case | 16:24 |
| *** ig0r_ has quit IRC | 16:24 | |
| hemna | this is code that exists today that I haven't changed. | 16:24 |
| *** unicell has quit IRC | 16:24 | |
| *** shaohe_feng has quit IRC | 16:24 | |
| johnthetubaguy | hemna: oh, is that because we removed the dict compat? | 16:25 |
| *** shaohe_feng has joined #openstack-nova | 16:25 | |
| hemna | luckily that code is only for deciding what log message to output | 16:26 |
| hemna | but I noticed it was always wrong for BFV, so I put in some logs to find out what was up | 16:26 |
| hemna | and found that | 16:26 |
| johnthetubaguy | hmm, that is odd, just looking at https://github.com/openstack/nova/blob/master/nova/objects/block_device.py#L43 seems we still do the compat stuff | 16:26 |
| *** vtech has joined #openstack-nova | 16:27 | |
| hemna | I just wonder if that's causing issues elsewhere where it might actually matter | 16:27 |
| hemna | (something other than logging) | 16:27 |
| johnthetubaguy | hemna: very, very, possibly | 16:27 |
| hemna | ok, so I guess I'll file a bug then | 16:28 |
| *** catintheroof has quit IRC | 16:28 | |
| johnthetubaguy | hemna: could you reduce it to a unit test? that would be interesting | 16:28 |
| *** yassou has quit IRC | 16:29 | |
| hemna | heh nova unit tests scare me :P | 16:29 |
| hemna | I'll see what I can do | 16:29 |
| openstackgerrit | Andrey Volkov proposed openstack/nova: Fix instances with unicode symbols logging https://review.openstack.org/320564 | 16:31 |
| *** sfinucan has quit IRC | 16:32 | |
| *** mpavone has quit IRC | 16:33 | |
| *** alex_xu has quit IRC | 16:35 | |
| *** shaohe_feng has quit IRC | 16:35 | |
| *** sdake_ has joined #openstack-nova | 16:35 | |
| *** shaohe_feng has joined #openstack-nova | 16:35 | |
| *** d_code_ has quit IRC | 16:37 | |
| hemna | johnthetubaguy, so from what I can tell the nova/tests/unit/virt/test_block_device.py uses fake dicts for their bdms. | 16:38 |
| *** cheran has joined #openstack-nova | 16:38 | |
| *** sdake has quit IRC | 16:38 | |
| *** alex_xu has joined #openstack-nova | 16:38 | |
| hemna | which all seem to go through the fake_bdm_object() call...I presume that creates a real bdm from a dict ? | 16:39 |
| johnthetubaguy | hemna: I am not sure actually, we have some very bad unit tests, I have to run away in a second or two I am afraid | 16:40 |
| *** klkumar has quit IRC | 16:40 | |
| johnthetubaguy | hemna: I was thinking use create an object, fill up stuff, and see how you can get hold of it, if that makes sense? | 16:40 |
| *** rmart04 has quit IRC | 16:40 | |
| hemna | ok I'll keep hacking on this. I'll try and add a unit test in there to reproduce the same bug | 16:41 |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Automatic Live Migration Completion https://review.openstack.org/306561 | 16:41 |
| *** delatte has joined #openstack-nova | 16:43 | |
| *** edand has quit IRC | 16:43 | |
| tdurakov | kashyap: so, I'm creating gate64 as usual, by patching cpu_map with script from devstack/tools | 16:44 |
| *** Swami has quit IRC | 16:45 | |
| *** ctrath has joined #openstack-nova | 16:45 | |
| *** shaohe_feng has quit IRC | 16:45 | |
| *** mvk has quit IRC | 16:45 | |
| *** delattec has quit IRC | 16:46 | |
| kashyap | tdurakov: Hmm | 16:46 |
| tdurakov | so it works for 14.04 | 16:46 |
| *** shaohe_feng has joined #openstack-nova | 16:46 | |
| kashyap | I have an Ubuntu machine - I can fiddle around and see if I can get a working DevStack locally on Xenial with 'gate64' VM -- I use Fedora for daily-work. | 16:46 |
| tdurakov | kashyap: btw http://xsnippet.org/361722/ | 16:47 |
| tdurakov | I'm a bit confused that error message points to qemu, not libvirt | 16:47 |
| *** pilgrimstack1 has quit IRC | 16:49 | |
| *** nic has joined #openstack-nova | 16:49 | |
| *** daemontool_ has quit IRC | 16:49 | |
| *** yamahata has joined #openstack-nova | 16:49 | |
| *** daemontool_ has joined #openstack-nova | 16:49 | |
| kashyap | tdurakov: A libvirt migration dev told me: "libvirt should have translated gate64 into something different (but for some reason it didn't do it)" | 16:50 |
| kashyap | tdurakov: After looking at that error. | 16:50 |
| kashyap | And as I guessed, asked if we can reproduce it with upstream... | 16:50 |
| *** mnestratov has joined #openstack-nova | 16:51 | |
| *** harlowja has joined #openstack-nova | 16:51 | |
| tdurakov | kashyap: yup, found the same in the docs, about cpu_model translation to nearest possible | 16:52 |
| tdurakov | so, there is some kind of issue for xenial, will try to debug this either | 16:52 |
| kashyap | tdurakov: And also unfortunately nothing interesting seems to be captured from the CPU driver in the debug log | 16:53 |
| kashyap | Yep, a solid reproducer test case should tease out the problem | 16:53 |
| *** daemontool_ has quit IRC | 16:53 | |
| *** markus_z has quit IRC | 16:54 | |
| tdurakov | kashyap: hope so | 16:54 |
| *** mnestratov|2 has quit IRC | 16:54 | |
| tdurakov | kashyap: will ping you tomorrow, now need to go | 16:54 |
| rlrossit | FYI the notification meeting will start in 5 minutes | 16:55 |
| *** Apoorva has joined #openstack-nova | 16:55 | |
| rlrossit | (openstack-meeting-4) | 16:55 |
| *** tonytan4ever has quit IRC | 16:55 | |
| *** shaohe_feng has quit IRC | 16:55 | |
| kashyap | tdurakov: Sure, me too. Hope this doesn't turn into a bottom-less pit. | 16:57 |
| tdurakov | kashyap: fyi, tried to shrink amount for cpu features | 16:58 |
| tdurakov | same result | 16:58 |
| *** shaohe_feng has joined #openstack-nova | 16:58 | |
| *** kaisers has quit IRC | 17:01 | |
| *** lucasagomes is now known as lucas-afk | 17:02 | |
| *** kaisers has joined #openstack-nova | 17:02 | |
| *** prateek has quit IRC | 17:04 | |
| *** xenogear has joined #openstack-nova | 17:05 | |
| *** gjayavelu has quit IRC | 17:05 | |
| *** shaohe_feng has quit IRC | 17:05 | |
| sdague | kashyap / tdurakov / clarkb ... I think http://logs.openstack.org/34/319934/2/experimental/gate-tempest-dsvm-multinode-live-migration/b87d689/logs/screen-n-cpu.txt.gz#_2016-05-24_15_59_12_531 is not actually libvirt related | 17:06 |
| sdague | I think it's qemu | 17:06 |
| *** shaohe_feng has joined #openstack-nova | 17:06 | |
| kashyap | sdague: libvirt migration dev said this a QEMU dev: "The issue is not QEMU" | 17:07 |
| tdurakov | sdague: +1 | 17:07 |
| kashyap | sdague: Because, first that 'gate64' should not make it into QEMU command-line | 17:07 |
| kashyap | sdague: libvirt is constructing the QEMU command-line. I'll discuss on #virt OFTC in the morning with the devs to see if I can get any more useful info. | 17:07 |
| kashyap | But, as we can see here: http://logs.openstack.org/34/319934/2/experimental/gate-tempest-dsvm-multinode-live-migration/b87d689/logs/libvirt/qemu/instance-00000003.txt.gz | 17:09 |
| sdague | kashyap: ok, well then previous libvirts did this differently | 17:09 |
| tdurakov | kashyap: one more question on this, libvirt has cpu_model.xml, so, what about qemu? | 17:09 |
| kashyap | sdague: libvirt is somehow using: "-cpu gate64" into the QEMU command-line | 17:09 |
| kashyap | sdague: On a proper, working machine, | 17:09 |
| *** ducttape_ has quit IRC | 17:09 | |
| kashyap | That "-cpu" should be something else that QEMU can detect | 17:10 |
| sdague | tdurakov: for qemu it's compiled in | 17:10 |
| sdague | there is no config | 17:10 |
| kashyap | tdurakov: sdague The CPU models for QEMU can be figured by running: | 17:10 |
| kashyap | $ qemu-system-x86_64 -cpu \? | 17:10 |
| jaypipes | tdurakov: how does one get the progress/status of a live migration via Nova API? | 17:11 |
| sdague | tdurakov: http://git.qemu.org/?p=qemu.git;a=blob;f=target-i386/cpu.c;h=895a386d3b7a94e363ca1bb98821d3251e70c0e0;hb=HEAD#l694 | 17:11 |
| *** PaulMurray has quit IRC | 17:11 | |
| *** sambetts is now known as sambetts|afk | 17:11 | |
| sdague | kashyap: right, it doesn't matter how it's determined, the fact is that it's compiled ine | 17:11 |
| kashyap | tdurakov: libvirt should translate 'gate64' into one of the models from the output of `qemu-system-x86_64 -cpu \?` | 17:11 |
| kashyap | sdague: Yep | 17:11 |
| sdague | kashyap: how would it translate that? | 17:12 |
| tdurakov | kashyap, sdague, will check it on 14.04 soon | 17:12 |
| tdurakov | just for comparison | 17:12 |
| tdurakov | jaypipes: are you asking about this spec https://review.openstack.org/#/c/320416/ ? | 17:13 |
| openstackgerrit | Lucian Petrut proposed openstack/nova: Add Hyper-V storage QoS support https://review.openstack.org/175479 | 17:13 |
| kashyap | sdague: Have to look into the libvirt source to check | 17:13 |
| openstackgerrit | Paul Murray proposed openstack/nova: Add console auth object https://review.openstack.org/320063 | 17:15 |
| openstackgerrit | Paul Murray proposed openstack/nova: Add console auth tokens db api methods https://review.openstack.org/319964 | 17:15 |
| *** shaohe_feng has quit IRC | 17:16 | |
| *** shaohe_feng has joined #openstack-nova | 17:16 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Permit Live Migration of Rescued Instances https://review.openstack.org/308198 | 17:16 |
| *** moshele has joined #openstack-nova | 17:17 | |
| *** lpetrut has quit IRC | 17:17 | |
| sdague | so it looks like in libvirt 1.2 it's passing different things - http://logs.openstack.org/29/42529/24/check/gate-tempest-dsvm-multinode-full/5f504c5/logs/libvirt/qemu/instance-0000000b.txt.gz | 17:17 |
| sdague | -cpu Opteron_G1,-pse36 | 17:17 |
| *** laszloh has joined #openstack-nova | 17:17 | |
| *** jlanoux has quit IRC | 17:17 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova-specs: Make checks before live-migration async https://review.openstack.org/320416 | 17:18 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: hyperv: Adds RemoteFX release notes https://review.openstack.org/320585 | 17:18 |
| patrickeast | anyone know how to configure nova+privsep to not spam everything into warning log level? is that possible? | 17:18 |
| patrickeast | like, i want to see warnings, but not for commands that execute successfully, example: http://openstack-logs.purestorage.com/43/309443/9/check/PureISCSIDriver-tempest-dsvm-volume-multipath/d1f77c0/logs/screen-n-cpu.txt.gz?level=TRACE#_2016-05-24_15_41_09_331 | 17:18 |
| tdurakov | sdague: if it so, well, it's libvirt issue | 17:18 |
| tdurakov | libvirt somehow fails to find closest cpu model for gate64 | 17:19 |
| sdague | right it looks like old libvirt was forcing our flags definition into the qemu one | 17:20 |
| *** unicell has joined #openstack-nova | 17:20 | |
| *** unicell has quit IRC | 17:20 | |
| *** unicell has joined #openstack-nova | 17:20 | |
| *** ayoung has quit IRC | 17:22 | |
| *** ayoung has joined #openstack-nova | 17:23 | |
| kashyap | sdague: Yep, I don't know if this is reproducible upstream libvirt | 17:24 |
| kashyap | sdague: What is this older libvirt version? | 17:26 |
| * kashyap goes to look at the logs | 17:26 | |
| *** shaohe_feng has quit IRC | 17:26 | |
| sdague | 1.2.2 | 17:27 |
| openstackgerrit | Chris Dent proposed openstack/nova: Support for both microversion headers https://review.openstack.org/300077 | 17:27 |
| *** shaohe_feng has joined #openstack-nova | 17:27 | |
| cdent | johnthetubaguy: made some adjustments on the microversion stuff. not sure it's yet perfect, but I've run out of mental capacity for that sort of thing, so pushed up what I've got for further comments, notably from mr sdague as I know he's got thoughts about such things | 17:27 |
| kashyap | Thanks | 17:27 |
| *** ducttape_ has joined #openstack-nova | 17:29 | |
| *** laszloh has quit IRC | 17:31 | |
| *** cleong has joined #openstack-nova | 17:31 | |
| *** mtecer has joined #openstack-nova | 17:33 | |
| *** jwcroppe has joined #openstack-nova | 17:34 | |
| *** ayoung has quit IRC | 17:35 | |
| *** ayoung has joined #openstack-nova | 17:36 | |
| *** haukebruno has quit IRC | 17:36 | |
| dane-fichter | mdbooth: I tend to agree that we should split the verification out of the image cache. Should we just move it to the fetch() method? | 17:36 |
| *** shaohe_feng has quit IRC | 17:36 | |
| *** jwcroppe_ has quit IRC | 17:37 | |
| *** shaohe_feng has joined #openstack-nova | 17:37 | |
| *** baoli_ has quit IRC | 17:40 | |
| *** READ10 has quit IRC | 17:42 | |
| *** Jeffrey4l has quit IRC | 17:43 | |
| openstackgerrit | Paul Carlton proposed openstack/nova: Permit Live Migration of Rescued Instances https://review.openstack.org/308198 | 17:43 |
| *** paul-carlton2 has quit IRC | 17:44 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Trivial: Fixes serial console minor nits https://review.openstack.org/320599 | 17:45 |
| *** jichen has joined #openstack-nova | 17:46 | |
| *** shaohe_feng has quit IRC | 17:46 | |
| *** shaohe_feng has joined #openstack-nova | 17:47 | |
| dansmith | dane-fichter: probably a little late for mdbooth | 17:48 |
| dane-fichter | dansmith: Oh is he "across the pond"? | 17:48 |
| *** tjones has joined #openstack-nova | 17:48 | |
| *** Swami has joined #openstack-nova | 17:48 | |
| dansmith | dane-fichter: I assume you mean verification of images before boot, right? | 17:48 |
| dane-fichter | correct. | 17:48 |
| dansmith | dane-fichter: yeah, he's in the UK | 17:48 |
| dansmith | dane-fichter: so I don't understand why that's useful | 17:48 |
| dansmith | as I said on the ML | 17:48 |
| dane-fichter | IWhich aspect? The existing ImageCacheManager or my spec/patch? | 17:49 |
| dane-fichter | or the existing signature verification? | 17:49 |
| dansmith | dane-fichter: if we're not running TPM-authenticated signed code, then if you managed to muck with the image on disk before or after boot, there's really no way we're going to detect that | 17:49 |
| dansmith | and checking at boot and not after is just hoping that the attacker shows up at the right time | 17:50 |
| *** efoley has quit IRC | 17:50 | |
| dane-fichter | dansmith: Ok so for all of the image signature verification stuff, we're trusting the compute nodes | 17:50 |
| dansmith | dane-fichter: so check after download is really all we need, yes? | 17:51 |
| *** ctrath has quit IRC | 17:51 | |
| *** efoley_ has quit IRC | 17:51 | |
| dane-fichter | if you can't trust the compute nodes, you're just hosed since those run your instances | 17:51 |
| dansmith | with an optional tool to let ops scan their base images from cron if they want | 17:51 |
| dane-fichter | dansmith: well the patch I have up is meant to be optional, it's a config option | 17:51 |
| dane-fichter | dansmith: signature verification is definitely an "opt in" feature lol | 17:52 |
| dansmith | but if it's not really buying us anything, then it's really just theater and an extra config option :) | 17:52 |
| dane-fichter | I agree with you/mdbooth that the compelling case is when the image cache isn't on the same machine as the compute node | 17:53 |
| *** ctrath has joined #openstack-nova | 17:53 | |
| dansmith | dane-fichter: that case makes it even less useful I think, | 17:53 |
| dansmith | dane-fichter: because you have hundreds of instances booted from one image-cache image | 17:54 |
| dane-fichter | dansmith: how so? | 17:54 |
| dansmith | I can alter that one after those are booted and wait for them to have to page /sbin/scarything in from disk and they're hosed | 17:54 |
| *** Sukhdev has joined #openstack-nova | 17:54 | |
| *** rlrossit has quit IRC | 17:54 | |
| dansmith | the periodic scanning of that image cache definitely does not need to be done by each compute node on an interval or at boot.. it can be done constantly by a tool, or protected as write-once by the backend or something | 17:55 |
| *** Jeffrey4l has joined #openstack-nova | 17:55 | |
| *** jaugustine has joined #openstack-nova | 17:56 | |
| *** shaohe_feng has quit IRC | 17:57 | |
| dane-fichter | dansmith: I thought "done constantly by a tool" was essentially what the image cache manager was for | 17:57 |
| dane-fichter | dansmith: so as of right now we only verify image signatures on download. if we move this code up to fetch it'll check for corruption before every boot | 17:58 |
| dane-fichter | regardless of whether the image was fetched from cache of Glance | 17:58 |
| dansmith | dane-fichter: multiple computes scanning a shared image cache periodically is kindof dumb right? | 17:59 |
| dane-fichter | certainly seems like wasted effort to me haha but I didn't build that ImageCacheManager | 17:59 |
| sdague | kashyap / clarkb / tdurakov I dumped my current understanding of the situation on the mailing list | 17:59 |
| *** shaohe_feng has joined #openstack-nova | 18:00 | |
| *** ctrath has quit IRC | 18:00 | |
| *** adisky has quit IRC | 18:00 | |
| kashyap | sdague: Thanks; I'll fetch mail once I'm off the train. | 18:00 |
| dansmith | dane-fichter: so verifying them on download is really the only thing I think makes sense to me to be part of the main path | 18:00 |
| * kashyap reads the archive meanwhile | 18:00 | |
| dansmith | dane-fichter: a separate tool that could be run to validate the cache from a single spot (or on each compute node from cron) might be useful | 18:01 |
| *** sajeesh has quit IRC | 18:01 | |
| dansmith | dane-fichter: building it into the image cache manager periodic scan doesn't to me | 18:01 |
| *** rlrossit has joined #openstack-nova | 18:01 | |
| kashyap | sdague: What is the thread subject? | 18:01 |
| dane-fichter | dansmith: where would that live? how would we distribute it to operators? | 18:01 |
| dansmith | dane-fichter: if other people think it is, and it's optional, then I'm semi-meh | 18:01 |
| kashyap | sdague: Disregard me, found it | 18:02 |
| dansmith | dane-fichter: I think we're growing more needs for a nova/ops-tools directory, so I think you could make this argument there | 18:02 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: numa: Fixes NUMA topology related unit tests https://review.openstack.org/278835 | 18:02 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: libvirt: fixes python 3 related unit tests https://review.openstack.org/278834 | 18:02 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Fixes python 3 unit tests https://review.openstack.org/278810 | 18:02 |
| dane-fichter | dansmith: It will definitely be optional, everything related to signature verification is. | 18:02 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Fixes python 3 urllib quote / unquote usage https://review.openstack.org/278809 | 18:02 |
| dane-fichter | dansmith: I think we can just use the existing config flag and not add another one | 18:03 |
| dane-fichter | dansmith: so instead of "theater" + config option it'll just be "theater" | 18:03 |
| dansmith | well, I'm "meh" on adding more theater, but... whatever :) | 18:04 |
| dane-fichter | dansmith: I think pushing the verification up to fetch satisfies the requirements of the people who pay me and also gets rid of the redundant scanning dumbness | 18:05 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: compute: fixes python 3 related unit tests https://review.openstack.org/300262 | 18:05 |
| *** mgoddard_ has quit IRC | 18:06 | |
| dane-fichter | dansmith: and also makes it easier for y'all to deprecate a good portion of that ImageCacheManager code, which seems to be the plan | 18:06 |
| dansmith | dane-fichter: it doesn't because instances that are already booted are vulnerable | 18:07 |
| dansmith | dane-fichter: periodic scanning is *more* useful than on boot, IMHO | 18:07 |
| dansmith | also because every compute does it for every image in a shared cache on boot, which is just silly | 18:07 |
| *** shaohe_feng has quit IRC | 18:07 | |
| *** shaohe_feng has joined #openstack-nova | 18:08 | |
| dane-fichter | dansmith: Is the shared cache a common configuration? | 18:08 |
| dane-fichter | dansmith: or do operators configure a local image cache for each compute host? | 18:08 |
| dane-fichter | I think a decent amount of the dumbness of scanning by various compute nodes hangs on that question | 18:09 |
| *** ctrath has joined #openstack-nova | 18:09 | |
| dane-fichter | i.e. it's not so dumb if each compute host is scanning it's own cache | 18:09 |
| *** sdake_ is now known as sdake | 18:11 | |
| *** david-lyle_ has joined #openstack-nova | 18:12 | |
| *** tbachman_ has joined #openstack-nova | 18:12 | |
| *** Apoorva_ has joined #openstack-nova | 18:13 | |
| *** raunak_ has joined #openstack-nova | 18:14 | |
| *** wznoinsk_ has joined #openstack-nova | 18:14 | |
| *** josecastroleon1 has joined #openstack-nova | 18:14 | |
| *** Vek has quit IRC | 18:15 | |
| dansmith | dane-fichter: I think it's pretty dumb either way | 18:15 |
| *** aysyd has quit IRC | 18:15 | |
| dansmith | dane-fichter: because again, if I get on the host to modify an image, I'm going to first make sure nova-compute is modified so I'm not detected | 18:15 |
| *** Apoorva has quit IRC | 18:16 | |
| *** raunak has quit IRC | 18:16 | |
| *** tbachman has quit IRC | 18:16 | |
| *** david-lyle has quit IRC | 18:16 | |
| *** wznoinsk has quit IRC | 18:16 | |
| *** josecastroleon has quit IRC | 18:16 | |
| *** raunak_ is now known as raunak | 18:16 | |
| *** tbachman_ is now known as tbachman | 18:16 | |
| dane-fichter | dansmith: Still seems worthwhile to check the integrity of your cache. Doesn't it increase the level of difficulty to the attacker and increase the odds of detecting them? | 18:17 |
| *** ijw has joined #openstack-nova | 18:17 | |
| *** shaohe_feng has quit IRC | 18:17 | |
| *** ijw has quit IRC | 18:17 | |
| *** ijw has joined #openstack-nova | 18:17 | |
| dansmith | dane-fichter: I think it's amazingly trivial to get around, so no I don't think so | 18:18 |
| *** shaohe_feng has joined #openstack-nova | 18:18 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Adds RemoteFX support to the Hyper-V driver https://review.openstack.org/42529 | 18:18 |
| dansmith | dane-fichter: all I do is change the nova.conf value, restart the service and then go about my business right? if I'm root already, that's pretty trivial | 18:18 |
| dane-fichter | dansmith: assuming that the attacker has rooted the compute host is not trivial. | 18:19 |
| dane-fichter | dansmith: if you're assuming that the compute service has been rooted in your security model then any discussion of any security feature is pretty much moot | 18:20 |
| dansmith | dane-fichter: of course not, but that's the case where we have to check the integrity of our images right? | 18:20 |
| bauzas | p!spy | 18:20 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Image: add create_from_image & create_from_func https://review.openstack.org/318948 | 18:21 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Rbd: add create_from_image & create_from_func https://review.openstack.org/317694 | 18:21 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Flat: add create_from_image & create_from_func https://review.openstack.org/317701 | 18:21 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Ploop: add create_from_image & create_from_func https://review.openstack.org/317689 | 18:21 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Qcow2: add create_from_image & create_from_func https://review.openstack.org/320610 | 18:21 |
| openstackgerrit | Diana Clarke proposed openstack/nova: Lvm: add create_from_image & create_from_func https://review.openstack.org/320611 | 18:21 |
| *** salv-orlando has quit IRC | 18:21 | |
| dane-fichter | dansmith: No. the case is simply that an attacker has accessed the image cache. Maybe it's on a different machine and they compromised that, maybe there's some quirky bug that lets you replace cached images | 18:21 |
| dane-fichter | dansmith: (there was a CVE in glance last cycle that basically let you do that btw) | 18:22 |
| dansmith | dane-fichter: if they have write access to the image cache, then they can trivially root the box right? because they can construct images with write access to /etc/passwd? | 18:22 |
| *** ig0r_ has joined #openstack-nova | 18:22 | |
| *** fragatina has joined #openstack-nova | 18:22 | |
| dansmith | dane-fichter: and if they have write access to the images, they are running as the nova user, which means they can probably affect nova-compute | 18:23 |
| openstackgerrit | Jim Rollenhagen proposed openstack/nova-specs: Re-architect ironic-multiple-compute-hosts https://review.openstack.org/320016 | 18:23 |
| jroll | jaypipes: devananda: ^ friendly reminder this is a thing :) | 18:23 |
| *** thumpba has joined #openstack-nova | 18:24 | |
| dane-fichter | dansmith: I'm sorry I don't follow either claim. | 18:25 |
| *** earlephilhower has joined #openstack-nova | 18:25 | |
| *** vtech has quit IRC | 18:25 | |
| dane-fichter | I don't see how write access implies that it's easy to jump to root or that they are necessarily running as the nova user | 18:26 |
| dansmith | dane-fichter: the users that have write access to the image cache are nova and root right? | 18:26 |
| dansmith | dane-fichter: anyway, I see your narrower goal I guess.. I'm not sure I think it's overly useful, but it's also not that big of an impact if someone wants to opt-in to it | 18:26 |
| dane-fichter | I would assume so. | 18:26 |
| *** kmartin has joined #openstack-nova | 18:26 | |
| dansmith | I think it's still theaterish (maybe, midly-fictionalized biopic vs. Aliens) | 18:27 |
| *** shaohe_feng has quit IRC | 18:27 | |
| dansmith | like I said, if it's opt-in and people think it's useful, then I'm meh | 18:28 |
| *** ctrath has quit IRC | 18:28 | |
| *** shaohe_feng has joined #openstack-nova | 18:28 | |
| dane-fichter | dansmith: I guess better 'meh' then 'boo'. Thanks for taking the time to talk to me. I appreciate it. | 18:29 |
| *** rcernin has quit IRC | 18:29 | |
| dansmith | dane-fichter: meh is about the best one gets out of me, so you should be happy :P | 18:29 |
| dansmith | dane-fichter: ask cdent | 18:30 |
| dane-fichter | lol I've seen the feared dan smith -1 in action | 18:30 |
| dansmith | heh | 18:31 |
| *** thumpba has quit IRC | 18:31 | |
| *** shaohe_feng has quit IRC | 18:38 | |
| *** shaohe_feng has joined #openstack-nova | 18:39 | |
| *** clenimar has joined #openstack-nova | 18:39 | |
| *** pushkaru has quit IRC | 18:45 | |
| *** jvgrant has quit IRC | 18:45 | |
| *** d_code has joined #openstack-nova | 18:46 | |
| *** pushkaru has joined #openstack-nova | 18:46 | |
| *** iceyao has quit IRC | 18:46 | |
| *** links has quit IRC | 18:48 | |
| *** lpetrut has joined #openstack-nova | 18:48 | |
| *** shaohe_feng has quit IRC | 18:48 | |
| *** nic has quit IRC | 18:49 | |
| *** shaohe_feng has joined #openstack-nova | 18:49 | |
| *** dimtruck is now known as zz_dimtruck | 18:50 | |
| cdent | dansmith, dane-fichter ask me what? | 18:50 |
| dansmith | cdent: nothing, joke | 18:50 |
| cdent | bummer, I was hoping for something exciting | 18:51 |
| dansmith | nope | 18:51 |
| openstackgerrit | Claudiu Belu proposed openstack/nova-specs: Repropose expose-host-capabilities to Newton https://review.openstack.org/286520 | 18:51 |
| *** matrohon has joined #openstack-nova | 18:52 | |
| *** jvgrant has joined #openstack-nova | 18:54 | |
| *** markvoelker_ has joined #openstack-nova | 18:57 | |
| *** shaohe_feng has quit IRC | 18:58 | |
| *** markvoelker has quit IRC | 18:59 | |
| d_code | I’m trying to do pci-passthrough (physical function) with nova mitaka. nova-compute finds the devices that are in my pci_passthrough_whitelist, but the scheduler for pcipassthroughfilter removes all hosts for the request | 19:00 |
| *** markvoelker has joined #openstack-nova | 19:00 | |
| d_code | is there a way to find out more about what is going on? | 19:00 |
| *** shaohe_feng has joined #openstack-nova | 19:01 | |
| *** tonytan4ever has joined #openstack-nova | 19:01 | |
| *** markvoelker_ has quit IRC | 19:02 | |
| *** shaohe_feng has quit IRC | 19:08 | |
| *** shaohe_feng has joined #openstack-nova | 19:11 | |
| *** prateek has joined #openstack-nova | 19:13 | |
| *** zz_dimtruck is now known as dimtruck | 19:15 | |
| *** sheel has quit IRC | 19:15 | |
| *** mfisch has joined #openstack-nova | 19:16 | |
| *** mfisch is now known as Guest92937 | 19:16 | |
| *** zenoway has joined #openstack-nova | 19:16 | |
| melwitt | d_code: looking at nova/scheduler/filters/pci_passthrough_filter.py shows that would happen if the flavor extra_specs don't match any devices in the whitelist. you can turn on debug level logging on the scheduler to see more detailed logs about it | 19:17 |
| *** rk4n has quit IRC | 19:17 | |
| *** rk4n has joined #openstack-nova | 19:18 | |
| *** Guest92937 is now known as mfisch | 19:18 | |
| *** mfisch has quit IRC | 19:18 | |
| *** mfisch has joined #openstack-nova | 19:18 | |
| *** shaohe_feng has quit IRC | 19:19 | |
| *** ishant has quit IRC | 19:20 | |
| *** sdake has quit IRC | 19:21 | |
| *** zenoway has quit IRC | 19:21 | |
| *** sdake has joined #openstack-nova | 19:21 | |
| *** prateek_ has joined #openstack-nova | 19:22 | |
| *** shaohe_feng has joined #openstack-nova | 19:23 | |
| *** rk4n has quit IRC | 19:24 | |
| *** prateek has quit IRC | 19:25 | |
| *** zenoway has joined #openstack-nova | 19:25 | |
| edleafe | married3Months | 19:25 |
| edleafe | ugh | 19:25 |
| cdent | ugh? | 19:26 |
| rlrossit | time to change passwords! | 19:26 |
| cdent | aw, edleafe, you romantic scallywag you | 19:27 |
| *** mvk has joined #openstack-nova | 19:27 | |
| edleafe | yeah, I use easy t remember stuff for my vms | 19:27 |
| *** baoli has joined #openstack-nova | 19:27 | |
| d_code | melwitt: I turned on debug logging…but all i get is the error that “X doesn't have the required PCI devices” | 19:28 |
| *** kmartin has quit IRC | 19:29 | |
| *** shaohe_feng has quit IRC | 19:29 | |
| *** shaohe_feng has joined #openstack-nova | 19:29 | |
| *** sdake has quit IRC | 19:30 | |
| *** sdake has joined #openstack-nova | 19:31 | |
| d_code | output of `openstack flavor show` http://paste.fedoraproject.org/370492/11818114 | 19:31 |
| d_code | relevant nova.conf : http://paste.fedoraproject.org/370494/11826914 | 19:31 |
| d_code | am I using the right syntax for the pci_passthrough on the flavor? | 19:31 |
| d_code | I dug through the python…it seems that the scheduler parses nova.conf for the alias information at check time, is that correct? | 19:32 |
| *** rmart04 has joined #openstack-nova | 19:32 | |
| *** diogogmt has quit IRC | 19:32 | |
| d_code | I couldn’t find anything in the nova database that made that connection | 19:32 |
| d_code | lspci output: http://paste.fedoraproject.org/370495/18417146 | 19:33 |
| *** Jeffrey4l has quit IRC | 19:34 | |
| melwitt | I don't know that much detail about it | 19:34 |
| moshele | d_code: look at the pci_devices table in nova database | 19:34 |
| d_code | yeah, each compute node has 2 devices registered that match that whitelist :/ | 19:35 |
| *** xyang1 has quit IRC | 19:35 | |
| d_code | is it necessary to do the pci_stub thing that you have to manually do with KVM? | 19:35 |
| *** prateek_ has quit IRC | 19:36 | |
| claudiub|2 | melwitt: hiiii. can you review one small image_meta prop patch, pls? :D https://review.openstack.org/#/c/237593/ | 19:36 |
| moshele | d_code: is your pci_alias defined in nova.conf of the controller ? (nova-api uses it) | 19:37 |
| d_code | yes | 19:37 |
| *** tangchen has quit IRC | 19:38 | |
| *** tangchen has joined #openstack-nova | 19:38 | |
| d_code | I’m using kolla and since each of the servers has the nic I want to passthrough on the same pcibus, each nova service uses the same config | 19:38 |
| *** baoli has quit IRC | 19:39 | |
| *** shaohe_feng has quit IRC | 19:39 | |
| d_code | is nova-api the piece that translates pci_alias? | 19:39 |
| *** ig0r_ has quit IRC | 19:39 | |
| *** dimtruck is now known as zz_dimtruck | 19:40 | |
| *** rmart04 has quit IRC | 19:40 | |
| openstackgerrit | Giridhar Jayavelu proposed openstack/nova: VMware: set service status based on vc connection https://review.openstack.org/290315 | 19:40 |
| *** rmart04 has joined #openstack-nova | 19:41 | |
| moshele | d_code: yes | 19:41 |
| melwitt | claudiub|2: will do | 19:41 |
| d_code | moshele: k…I see it in the logs where it is parsed /sigh | 19:41 |
| d_code | can I pass vendor_id / product_id directly in the flavor or as metadata on launch ? | 19:42 |
| *** matrohon has quit IRC | 19:42 | |
| *** salv-orlando has joined #openstack-nova | 19:42 | |
| moshele | d_code: I don't think so | 19:42 |
| *** shaohe_feng has joined #openstack-nova | 19:42 | |
| d_code | yeah…I didn’t this so either.. :( | 19:43 |
| *** rmart04 has quit IRC | 19:43 | |
| *** zzelle_ has joined #openstack-nova | 19:44 | |
| *** tonytan4ever has quit IRC | 19:46 | |
| *** Jeffrey4l has joined #openstack-nova | 19:46 | |
| moshele | d_code: can you paste the pci_devices table | 19:47 |
| *** ociuhandu has quit IRC | 19:48 | |
| d_code | moshele: sure...sec | 19:48 |
| *** zz_dimtruck is now known as dimtruck | 19:49 | |
| *** ducttape_ has quit IRC | 19:49 | |
| *** shaohe_feng has quit IRC | 19:49 | |
| d_code | http://paste.fedoraproject.org/370498/64119382/ | 19:50 |
| *** shaohe_feng has joined #openstack-nova | 19:50 | |
| *** dmk0202 has joined #openstack-nova | 19:54 | |
| *** rustlebee has quit IRC | 19:54 | |
| *** rustlebee has joined #openstack-nova | 19:54 | |
| *** rustlebee is now known as russellb | 19:54 | |
| *** zenoway has quit IRC | 19:55 | |
| *** zenoway has joined #openstack-nova | 19:56 | |
| *** flwang1 has quit IRC | 19:56 | |
| *** annegentle has joined #openstack-nova | 19:56 | |
| *** shaohe_feng has quit IRC | 20:00 | |
| *** jichen has quit IRC | 20:00 | |
| *** shaohe_feng has joined #openstack-nova | 20:01 | |
| *** zenoway has quit IRC | 20:01 | |
| *** zenoway has joined #openstack-nova | 20:02 | |
| *** sdake_ has joined #openstack-nova | 20:02 | |
| *** AmyLee7 has quit IRC | 20:03 | |
| *** Apoorva_ has quit IRC | 20:04 | |
| *** ociuhandu has joined #openstack-nova | 20:05 | |
| *** sdake has quit IRC | 20:05 | |
| *** jaugustine has quit IRC | 20:05 | |
| *** baoli has joined #openstack-nova | 20:07 | |
| *** rk4n has joined #openstack-nova | 20:07 | |
| *** ducttape_ has joined #openstack-nova | 20:08 | |
| *** shaohe_feng has quit IRC | 20:10 | |
| *** tonytan4ever has joined #openstack-nova | 20:10 | |
| *** Vek has joined #openstack-nova | 20:11 | |
| *** kfarr has joined #openstack-nova | 20:12 | |
| *** shaohe_feng has joined #openstack-nova | 20:13 | |
| *** Sukhdev has quit IRC | 20:13 | |
| hoonetorg | hi | 20:14 |
| *** lpetrut has quit IRC | 20:14 | |
| hoonetorg | is volumes_dir option in nova.conf deprecated? | 20:14 |
| hoonetorg | if yes since which release? | 20:14 |
| moshele | dansmith: can you review again the pci patches here https://etherpad.openstack.org/p/newton-nova-priorities-tracking L 189, I updated them according to you comments | 20:16 |
| *** kmartin has joined #openstack-nova | 20:18 | |
| *** shaohe_feng has quit IRC | 20:20 | |
| *** omnipresent has joined #openstack-nova | 20:20 | |
| *** shaohe_feng has joined #openstack-nova | 20:21 | |
| *** ayoung has quit IRC | 20:21 | |
| *** Apoorva has joined #openstack-nova | 20:23 | |
| *** zenoway has quit IRC | 20:24 | |
| openstackgerrit | Anusha Unnam proposed openstack/nova: Improve the help text for compute timeout_opts https://review.openstack.org/317693 | 20:25 |
| *** hemanthm is now known as hemanthm|afk | 20:26 | |
| *** baoli has quit IRC | 20:29 | |
| *** baoli has joined #openstack-nova | 20:29 | |
| *** shaohe_feng has quit IRC | 20:30 | |
| *** shaohe_feng has joined #openstack-nova | 20:33 | |
| *** baoli has quit IRC | 20:34 | |
| *** tbachman_ has joined #openstack-nova | 20:34 | |
| *** baoli has joined #openstack-nova | 20:35 | |
| *** tbachman has quit IRC | 20:35 | |
| *** tbachman_ is now known as tbachman | 20:35 | |
| *** cleong has quit IRC | 20:37 | |
| *** cfriesen has joined #openstack-nova | 20:37 | |
| jaypipes | jroll: almost done with a review on that. no more than 15 mins until I push. | 20:38 |
| *** pushkaru has quit IRC | 20:40 | |
| jroll | jaypipes: you rock | 20:40 |
| *** pushkaru has joined #openstack-nova | 20:40 | |
| *** dane-fichter has quit IRC | 20:41 | |
| *** shaohe_feng has quit IRC | 20:41 | |
| *** shaohe_feng has joined #openstack-nova | 20:41 | |
| *** gyee has joined #openstack-nova | 20:45 | |
| *** rmart04 has joined #openstack-nova | 20:45 | |
| *** rmart04 has quit IRC | 20:45 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova-specs: Make checks before live-migration async https://review.openstack.org/320416 | 20:46 |
| *** nic has joined #openstack-nova | 20:46 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: WIP Create Instance from BuildRequest if not in a cell https://review.openstack.org/263927 | 20:48 |
| *** shaohe_feng has quit IRC | 20:51 | |
| jaypipes | jroll: done. | 20:51 |
| jroll | thank you sir | 20:51 |
| jaypipes | jroll: excellent work on it. you clearly understand where we're heading. | 20:51 |
| *** shaohe_feng has joined #openstack-nova | 20:51 | |
| jaypipes | jroll: +1 on it from me. | 20:51 |
| jroll | jaypipes: yay, I'm not crazy | 20:51 |
| jroll | wow | 20:51 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: objects: added 'os_secure_boot' property to ImageMetaProps object https://review.openstack.org/237593 | 20:52 |
| jroll | on patch set 1? that's allowed? | 20:52 |
| jaypipes | jroll: well, I wouldn't go that far ;P | 20:52 |
| jroll | :) | 20:52 |
| jaypipes | jroll: :) technically it's an amendment to an existing spec patch, so that doesn't count ;) | 20:52 |
| jroll | heh | 20:52 |
| *** eharney has quit IRC | 20:53 | |
| *** ducttape_ has quit IRC | 20:53 | |
| *** dimtruck is now known as zz_dimtruck | 20:54 | |
| *** fragatina has quit IRC | 20:57 | |
| d_code | SamYaple: ping | 20:57 |
| dansmith | jaypipes: wait wht? | 20:58 |
| dansmith | jaypipes: jroll isn't crazy? | 20:58 |
| *** ivase has quit IRC | 20:58 | |
| jroll | oh he didn't say that, I did | 20:58 |
| jroll | clearly wrong | 20:58 |
| dansmith | jroll: he seemed to agree, which is why I was looking for clarification | 20:58 |
| *** ivase has joined #openstack-nova | 20:58 | |
| jroll | sounds like he disagreed to me | 20:59 |
| *** shaohe_feng has quit IRC | 21:01 | |
| *** shaohe_feng has joined #openstack-nova | 21:02 | |
| *** rcernin has joined #openstack-nova | 21:02 | |
| *** ivase has quit IRC | 21:03 | |
| *** rfolco has quit IRC | 21:03 | |
| *** tbachman_ has joined #openstack-nova | 21:04 | |
| *** Sukhdev has joined #openstack-nova | 21:04 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove service crud from test_resource_tracker https://review.openstack.org/319699 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove StatsDicTestCase from test_resource_tracker https://review.openstack.org/319696 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove conductor from test_resource_tracker https://review.openstack.org/319697 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: rt-unit: Replace hard-coded strings with constants https://review.openstack.org/319695 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove useless test of incorrect stats value https://review.openstack.org/319693 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove RT duplicate unit test for PCI stats https://review.openstack.org/319692 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove more duplicate RT unit tests https://review.openstack.org/319691 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Removes test_claim_saves_numa_topology() https://review.openstack.org/319690 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove duplicate unit test in test_tracker https://review.openstack.org/319723 | 21:04 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Remove delete stubs in test_resource_tracker https://review.openstack.org/319722 | 21:04 |
| jaypipes | alaski, dansmith: ^^ just rebased and fixed merge conflict...' | 21:05 |
| *** tbachman has quit IRC | 21:05 | |
| *** tbachman_ is now known as tbachman | 21:05 | |
| *** ayoung has joined #openstack-nova | 21:05 | |
| *** jaypipes has quit IRC | 21:05 | |
| *** jdurgin has quit IRC | 21:07 | |
| cfriesen | Is anyone looking at https://review.openstack.org/#/c/270482/ and https://review.openstack.org/#/c/200870/? They seem to be fixing a valid disk accounting issue (https://bugs.launchpad.net/nova/+bug/1469179) but don't seem to be getting a lot of attention. | 21:07 |
| openstack | Launchpad bug 1469179 in OpenStack Compute (nova) "instance.root_gb should be 0 for volume-backed instances" [Medium,In progress] - Assigned to Ankit Agrawal (ankitagrawal) | 21:07 |
| *** jdurgin has joined #openstack-nova | 21:08 | |
| *** shaohe_feng has quit IRC | 21:11 | |
| *** smatzek has joined #openstack-nova | 21:12 | |
| *** cdent has quit IRC | 21:12 | |
| *** shaohe_feng has joined #openstack-nova | 21:12 | |
| dansmith | alaski: I got em | 21:13 |
| alaski | dansmith: cool | 21:14 |
| mikal | Morning | 21:15 |
| SamYaple | d_code: pong... | 21:16 |
| d_code | SamYaple: we discussed a few weeks ago in #kolla about pci passthrough.. | 21:16 |
| SamYaple | probably been a few months. I havent worked on #kolla for a while | 21:17 |
| SamYaple | but what was your question? | 21:17 |
| d_code | I used kolla to deploy mitaka and have followed a number of conflicting wiki articles on how to configure it, but the pcipassthroughfilter keeps telling me there are no instances available | 21:17 |
| d_code | yeah…it was probably early march, I guess | 21:17 |
| d_code | have you used pci passthrough of a full pci device in Mitaka? | 21:18 |
| d_code | seems SR-IOV VFs are much more popular | 21:18 |
| SamYaple | hmmm off the top of my head i think you may not be passing the /dev:/dev into the container | 21:18 |
| d_code | so, it’s in the nova database… nova-compute sees the correct device and registers it…so the whitelist is working | 21:18 |
| *** haplo37_ has joined #openstack-nova | 21:19 | |
| d_code | it seems the issue is that my pci_alias is not matching the db entry | 21:19 |
| d_code | and even with debug = True, I can’t get any additional resolution on that | 21:19 |
| *** kfarr has quit IRC | 21:19 | |
| *** pushkaru has quit IRC | 21:19 | |
| *** pushkaru has joined #openstack-nova | 21:20 | |
| *** fragatina has joined #openstack-nova | 21:20 | |
| SamYaple | the compute node has the pcipassthroughfilter error d_code ? or is that only on the sceudler | 21:20 |
| d_code | http://paste.fedoraproject.org/370494/11826914/ | 21:20 |
| *** vishwanathj has joined #openstack-nova | 21:20 | |
| d_code | I’ve got the output from the scheduler…sec | 21:21 |
| d_code | http://paste.fedoraproject.org/370492/11818114/ <— flavor info | 21:21 |
| *** zz_dimtruck is now known as dimtruck | 21:21 | |
| *** AmyLee7 has joined #openstack-nova | 21:21 | |
| d_code | scheduler output in logs: http://paste.fedoraproject.org/370482/14641159/ | 21:22 |
| *** shaohe_feng has quit IRC | 21:22 | |
| *** shaohe_feng has joined #openstack-nova | 21:22 | |
| *** ducttape_ has joined #openstack-nova | 21:22 | |
| *** rlrossit has quit IRC | 21:23 | |
| *** sfinucan has joined #openstack-nova | 21:23 | |
| SamYaple | d_code: isnt the whitelist supposed to be an actual list? will it work when passed a dict rather than a list of dicts | 21:23 |
| d_code | yes, apparently it went back and forth back in Grizzly…but now it supports both | 21:24 |
| SamYaple | ok | 21:24 |
| *** haplo37_ has quit IRC | 21:24 | |
| *** yamahata has quit IRC | 21:24 | |
| *** gagehugo has quit IRC | 21:25 | |
| *** baoli has quit IRC | 21:25 | |
| d_code | debug output from scheduler : http://paste.fedoraproject.org/370519/41251431/ | 21:26 |
| *** baoli has joined #openstack-nova | 21:26 | |
| SamYaple | d_code: looks good from here. what are the logs on the actual compute node when this tries to get scheduled and fails? | 21:26 |
| *** dimtruck is now known as zz_dimtruck | 21:29 | |
| *** pwnall has quit IRC | 21:29 | |
| d_code | afaik, nova-compute log doesn’t reflect anything w/ this particular instance… it just does it’s occasional check as normal | 21:31 |
| *** lakshmiS has quit IRC | 21:32 | |
| *** shaohe_feng has quit IRC | 21:32 | |
| *** dmk0202 has quit IRC | 21:32 | |
| *** sfinucan has quit IRC | 21:32 | |
| SamYaple | ah yea. its failing before it schedules to the compute node it appears. so this is an issue with the stats reported back from the compute node. either its not reporting back the pci devices or you aren't sepcifying a device that exists on the server | 21:33 |
| *** flwang1 has joined #openstack-nova | 21:33 | |
| SamYaple | can you `lspci -k` on one of the computes for me | 21:33 |
| *** shaohe_feng has joined #openstack-nova | 21:33 | |
| *** ducttape_ has quit IRC | 21:33 | |
| d_code | http://paste.fedoraproject.org/370522/25615146 shows the pci device being used | 21:33 |
| d_code | yeah...sec | 21:33 |
| d_code | lspci -k -> http://paste.fedoraproject.org/370523/25651146 | 21:34 |
| *** annegentle has quit IRC | 21:34 | |
| *** sdake has joined #openstack-nova | 21:34 | |
| d_code | 07:00 is the card I want | 21:35 |
| *** ametts has quit IRC | 21:35 | |
| d_code | lspci -k -nn (w/ vendor/device id’s) http://paste.fedoraproject.org/370524/25716146 | 21:35 |
| d_code | does the device have to get added to pci_stub like when you manually do it w/ KVM? | 21:36 |
| d_code | the pci_stub kernel module | 21:36 |
| *** diogogmt has joined #openstack-nova | 21:36 | |
| SamYaple | d_code: is the goal here to use sr-iov with neutron? because thats a bit different a setup | 21:37 |
| *** thorst_ has quit IRC | 21:37 | |
| d_code | nope…just want good ol’ pci passthrough of the full device | 21:37 |
| *** dmk0202 has joined #openstack-nova | 21:37 | |
| SamYaple | ok fair enough | 21:38 |
| *** thorst_ has joined #openstack-nova | 21:38 | |
| d_code | I know there’s SR-IOV PF…but I don’t know if that works the same | 21:38 |
| d_code | but..really I just want to dedicate hardware to a couple VMs | 21:38 |
| *** syjulian has quit IRC | 21:38 | |
| *** sdake_ has quit IRC | 21:38 | |
| SamYaple | hmmm looking over some other configs i think you might need to wrap both sides of the address is astrisk | 21:40 |
| d_code | wait…I just noticed… w/ lspci -k -nn, the vendor ID / device ID for the subsystem is different | 21:40 |
| openstackgerrit | Claudiu Belu proposed openstack/nova-specs: Adds Nova discoverable policy CLI spec https://review.openstack.org/289405 | 21:40 |
| d_code | is that what I should be targeting? | 21:40 |
| SamYaple | "address": "*:07:00.*" | 21:40 |
| d_code | 103c:17d3 (for HP 560SFP+ adapter) versus 8086:10fb for Intel 82599ES | 21:41 |
| SamYaple | you know d_code ive never done pci passtrough of a nic with sr-iov without goign through neutron. i really don't know the answer here | 21:41 |
| *** lindycoder has quit IRC | 21:41 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: Save migration context in the context of an Instance.save() only https://review.openstack.org/320677 | 21:42 |
| openstackgerrit | Moshe Levi proposed openstack/nova: intance numa topoligy Save intance numa topoligy in the context of an Instance.save() only https://review.openstack.org/320678 | 21:42 |
| openstackgerrit | Moshe Levi proposed openstack/nova: One db call to save instance extra field https://review.openstack.org/320679 | 21:42 |
| d_code | heh…alright man…I appreciate your help | 21:42 |
| *** rtheis has quit IRC | 21:42 | |
| *** shaohe_feng has quit IRC | 21:42 | |
| d_code | I truly thought this was a little more understood…I’ve run into a couple people that have done GPUs, but everything else seems to look right | 21:42 |
| SamYaple | i have passed through gpus and entire nics though using the setup you have | 21:43 |
| *** thorst_ has quit IRC | 21:43 | |
| *** shaohe_feng has joined #openstack-nova | 21:43 | |
| d_code | oh weird | 21:43 |
| d_code | so, I want the entire nic… have you done it with mitaka ? | 21:43 |
| *** rcernin has quit IRC | 21:43 | |
| *** thorst_ has joined #openstack-nova | 21:44 | |
| openstackgerrit | Sarafraj Singh proposed openstack/nova: Deprecate barbican options https://review.openstack.org/317684 | 21:45 |
| SamYaple | i have not. but my configs do have me wrapping the address with two * *, i can suggest trying that | 21:45 |
| SamYaple | ill play with my gpu stuff when i get home and see if i can see what might be different with your environment | 21:45 |
| *** penick_ has joined #openstack-nova | 21:46 | |
| *** penick has quit IRC | 21:46 | |
| *** penick_ is now known as penick | 21:46 | |
| *** zzelle_ has quit IRC | 21:47 | |
| d_code | I wrapped the address w/ astierks on both ends… I’m also gonna dry the HP device IDs to see if that makes a difference | 21:47 |
| d_code | s/dry/try/ | 21:48 |
| *** thorst_ has quit IRC | 21:48 | |
| *** akshai has quit IRC | 21:48 | |
| *** thorst_ has joined #openstack-nova | 21:50 | |
| *** Sukhdev has quit IRC | 21:50 | |
| *** annegentle has joined #openstack-nova | 21:50 | |
| *** burgerk has quit IRC | 21:52 | |
| *** shaohe_feng has quit IRC | 21:52 | |
| *** Sukhdev has joined #openstack-nova | 21:53 | |
| *** ducttape_ has joined #openstack-nova | 21:53 | |
| *** shaohe_feng has joined #openstack-nova | 21:53 | |
| *** thorst_ has quit IRC | 21:54 | |
| *** burt has quit IRC | 21:56 | |
| *** baoli has quit IRC | 21:59 | |
| *** baoli has joined #openstack-nova | 21:59 | |
| *** penick has joined #openstack-nova | 22:01 | |
| *** penick has quit IRC | 22:02 | |
| openstackgerrit | Moshe Levi proposed openstack/nova: Save migration context only in the context of an Instance.save() https://review.openstack.org/320677 | 22:03 |
| openstackgerrit | Moshe Levi proposed openstack/nova: One db call to save instance extra field https://review.openstack.org/320679 | 22:03 |
| openstackgerrit | Moshe Levi proposed openstack/nova: Save intance numa topoligy only in the context of an Instance.save() https://review.openstack.org/320678 | 22:03 |
| *** shaohe_feng has quit IRC | 22:03 | |
| *** shaohe_feng has joined #openstack-nova | 22:03 | |
| *** mmandel has quit IRC | 22:06 | |
| *** penick has joined #openstack-nova | 22:07 | |
| *** moshele has quit IRC | 22:07 | |
| *** dmk0202 has quit IRC | 22:09 | |
| *** darrenc is now known as darrenc_afk | 22:09 | |
| *** sdake has quit IRC | 22:09 | |
| *** pushkaru has quit IRC | 22:09 | |
| *** albertom has quit IRC | 22:10 | |
| *** tjones has left #openstack-nova | 22:11 | |
| *** pushkaru has joined #openstack-nova | 22:13 | |
| *** AmyLee7 has quit IRC | 22:13 | |
| *** shaohe_feng has quit IRC | 22:13 | |
| *** AmyLee7 has joined #openstack-nova | 22:14 | |
| oomichi | sdague: johnthetubaguy: alex_xu: can you take a look at https://review.openstack.org/#/c/313192/ ? | 22:14 |
| *** shaohe_feng has joined #openstack-nova | 22:14 | |
| *** thorst_ has joined #openstack-nova | 22:15 | |
| *** darrenc_afk is now known as darrenc | 22:16 | |
| *** AmyLee7 has quit IRC | 22:17 | |
| *** jamesdenton has joined #openstack-nova | 22:17 | |
| *** sdake has joined #openstack-nova | 22:19 | |
| *** thorst_ has quit IRC | 22:19 | |
| *** albertom has joined #openstack-nova | 22:19 | |
| *** claudiub has joined #openstack-nova | 22:22 | |
| *** shaohe_feng has quit IRC | 22:23 | |
| *** albertom has quit IRC | 22:24 | |
| *** claudiub|2 has quit IRC | 22:26 | |
| *** shaohe_feng has joined #openstack-nova | 22:26 | |
| *** annegentle has quit IRC | 22:29 | |
| *** albertom has joined #openstack-nova | 22:31 | |
| *** thumpba has joined #openstack-nova | 22:31 | |
| *** shaohe_feng has quit IRC | 22:33 | |
| *** shaohe_feng has joined #openstack-nova | 22:34 | |
| *** jwcroppe_ has joined #openstack-nova | 22:35 | |
| *** jwcroppe has quit IRC | 22:36 | |
| *** claudiub|2 has joined #openstack-nova | 22:36 | |
| *** albertom has quit IRC | 22:40 | |
| *** claudiub has quit IRC | 22:40 | |
| *** gjayavelu has joined #openstack-nova | 22:41 | |
| *** rk4n has quit IRC | 22:42 | |
| *** Sukhdev has quit IRC | 22:43 | |
| *** shaohe_feng has quit IRC | 22:44 | |
| *** shaohe_feng has joined #openstack-nova | 22:45 | |
| *** ducttape_ has quit IRC | 22:46 | |
| *** mlavalle has left #openstack-nova | 22:47 | |
| *** edmondsw has quit IRC | 22:50 | |
| *** pushkaru has quit IRC | 22:53 | |
| *** shaohe_feng has quit IRC | 22:54 | |
| *** shaohe_feng has joined #openstack-nova | 22:55 | |
| *** nic has quit IRC | 22:55 | |
| *** baoli has quit IRC | 22:55 | |
| *** mrkz has quit IRC | 22:57 | |
| *** ducttape_ has joined #openstack-nova | 22:57 | |
| *** ducttape_ has quit IRC | 22:58 | |
| *** yamahata has joined #openstack-nova | 22:58 | |
| *** openstackgerrit has quit IRC | 23:03 | |
| *** vishwanathj has quit IRC | 23:03 | |
| *** openstackgerrit has joined #openstack-nova | 23:03 | |
| *** shaohe_feng has quit IRC | 23:04 | |
| *** shaohe_feng has joined #openstack-nova | 23:07 | |
| *** sdague has quit IRC | 23:09 | |
| *** markvoelker has quit IRC | 23:11 | |
| *** david-lyle_ is now known as david-lyle | 23:12 | |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: WIP: limits.inc example verification https://review.openstack.org/315714 | 23:14 |
| *** shaohe_feng has quit IRC | 23:14 | |
| *** shaohe_feng has joined #openstack-nova | 23:18 | |
| openstackgerrit | Augustina Ragwitz proposed openstack/nova: limits.inc example verification https://review.openstack.org/315714 | 23:18 |
| *** rbridgeman has joined #openstack-nova | 23:19 | |
| *** Sukhdev has joined #openstack-nova | 23:21 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Remove legacy v2 API code completely https://review.openstack.org/311653 | 23:21 |
| *** thumpba has quit IRC | 23:24 | |
| *** shaohe_feng has quit IRC | 23:25 | |
| *** shaohe_feng has joined #openstack-nova | 23:26 | |
| *** Swami has quit IRC | 23:31 | |
| *** albertom has joined #openstack-nova | 23:31 | |
| *** jamesdenton has quit IRC | 23:33 | |
| *** annegentle has joined #openstack-nova | 23:35 | |
| *** shaohe_feng has quit IRC | 23:35 | |
| *** shaohe_feng has joined #openstack-nova | 23:36 | |
| *** Sukhdev has quit IRC | 23:39 | |
| *** omnipresent has quit IRC | 23:41 | |
| *** zenoway has joined #openstack-nova | 23:41 | |
| *** thorst_ has joined #openstack-nova | 23:45 | |
| *** shaohe_feng has quit IRC | 23:45 | |
| *** zenoway has quit IRC | 23:46 | |
| *** shaohe_feng has joined #openstack-nova | 23:46 | |
| *** thorst__ has joined #openstack-nova | 23:47 | |
| *** yamahata has quit IRC | 23:48 | |
| *** rbridgeman has quit IRC | 23:48 | |
| *** Sukhdev has joined #openstack-nova | 23:49 | |
| *** thorst_ has quit IRC | 23:52 | |
| *** sdake has quit IRC | 23:55 | |
| *** thorst__ is now known as thorst_ | 23:55 | |
| *** shaohe_feng has quit IRC | 23:55 | |
| *** shaohe_feng has joined #openstack-nova | 23:56 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
