| *** achanda has joined #openstack-nova | 00:07 | |
| *** achanda has quit IRC | 00:12 | |
| *** jcru has joined #openstack-nova | 00:25 | |
| *** aix has quit IRC | 00:26 | |
| *** jcru has quit IRC | 00:28 | |
| *** jcru has joined #openstack-nova | 00:28 | |
| *** jcru has quit IRC | 00:29 | |
| *** aix has joined #openstack-nova | 00:30 | |
| *** shaohe_feng1 has quit IRC | 00:31 | |
| *** annashen has joined #openstack-nova | 00:36 | |
| *** otter768 has quit IRC | 00:37 | |
| *** matrohon has quit IRC | 00:37 | |
| *** mjturek1 has quit IRC | 00:37 | |
| *** promulo__ has quit IRC | 00:37 | |
| *** annashen has quit IRC | 00:41 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Pass project_id when create networks by os-tenant-networks https://review.openstack.org/164549 | 00:45 |
|---|---|---|
| *** mwagner_lap has quit IRC | 00:48 | |
| tonyb | When is the pre-release string freeze? | 00:49 |
| *** vmtrooper has joined #openstack-nova | 00:53 | |
| *** iamjarvo has joined #openstack-nova | 00:53 | |
| *** eliqiao has joined #openstack-nova | 00:54 | |
| openstackgerrit | shihanzhang proposed openstack/nova: Fix AggregateCoreFilter return incorrect value https://review.openstack.org/164128 | 00:57 |
| *** vmtrooper has quit IRC | 00:58 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Clean up Fake_Url for unit test of flavor_access https://review.openstack.org/162869 | 01:00 |
| *** mwagner_lap has joined #openstack-nova | 01:00 | |
| *** Longgeek has joined #openstack-nova | 01:08 | |
| *** liusheng has quit IRC | 01:13 | |
| *** Longgeek has quit IRC | 01:15 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove the elevated context when get network https://review.openstack.org/164551 | 01:15 |
| *** alexpilotti has quit IRC | 01:19 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Cleanup exception handler when attaching interface https://review.openstack.org/131026 | 01:21 |
| *** shaohe_feng1 has joined #openstack-nova | 01:21 | |
| *** salv-orlando has quit IRC | 01:22 | |
| *** futurama008 has joined #openstack-nova | 01:23 | |
| *** mtanino has joined #openstack-nova | 01:25 | |
| *** futurama008 has quit IRC | 01:30 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Handle RESIZE_PREP status when nova compute do init_instance https://review.openstack.org/132827 | 01:40 |
| *** baoli has quit IRC | 01:48 | |
| *** ijw_ has quit IRC | 01:48 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove inner multiple version method support from devref and unitests https://review.openstack.org/164555 | 01:49 |
| *** otter768 has joined #openstack-nova | 01:49 | |
| *** park has joined #openstack-nova | 01:51 | |
| *** annashen has joined #openstack-nova | 01:54 | |
| *** Marga_ has joined #openstack-nova | 01:57 | |
| *** annashen has quit IRC | 01:58 | |
| *** annashen has joined #openstack-nova | 01:59 | |
| *** baoli has joined #openstack-nova | 02:07 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Cleanup exception handler when attaching interface https://review.openstack.org/131026 | 02:09 |
| *** dims_ has quit IRC | 02:13 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for network_disassociate https://review.openstack.org/161624 | 02:16 |
| *** popw has joined #openstack-nova | 02:22 | |
| *** salv-orlando has joined #openstack-nova | 02:22 | |
| *** erkules_ has joined #openstack-nova | 02:23 | |
| *** salv-orlando has quit IRC | 02:25 | |
| *** erkules has quit IRC | 02:26 | |
| *** salv-orlando has joined #openstack-nova | 02:29 | |
| *** salv-orlando has quit IRC | 02:38 | |
| *** vmtrooper has joined #openstack-nova | 02:41 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Handle exception when doing detach_interface https://review.openstack.org/130151 | 02:47 |
| *** vmtrooper has quit IRC | 02:47 | |
| *** takedakn has joined #openstack-nova | 02:49 | |
| *** eliqiao has quit IRC | 02:49 | |
| *** erkules_ is now known as erkules | 02:54 | |
| *** erkules has quit IRC | 02:55 | |
| *** erkules has joined #openstack-nova | 02:55 | |
| *** browne has joined #openstack-nova | 02:56 | |
| *** takedakn has quit IRC | 02:57 | |
| *** coolsvap|afk is now known as coolsvap | 02:58 | |
| *** nelsnelson has joined #openstack-nova | 02:59 | |
| *** annashen has quit IRC | 03:05 | |
| *** Longgeek has joined #openstack-nova | 03:11 | |
| *** iamjarvo has quit IRC | 03:12 | |
| *** salv-orlando has joined #openstack-nova | 03:12 | |
| *** Longgeek has quit IRC | 03:13 | |
| *** Longgeek has joined #openstack-nova | 03:14 | |
| *** dimsum__ has joined #openstack-nova | 03:14 | |
| *** salv-orlando has quit IRC | 03:16 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Pass project_id when create networks by os-tenant-networks https://review.openstack.org/164549 | 03:19 |
| openstackgerrit | Alex Xu proposed openstack/nova: Deassociate before delete network in os-tenant-networks delete method https://review.openstack.org/164563 | 03:19 |
| *** Longgeek_ has joined #openstack-nova | 03:19 | |
| *** dimsum__ has quit IRC | 03:19 | |
| *** Longgeek has quit IRC | 03:20 | |
| *** ijw has joined #openstack-nova | 03:23 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for pci https://review.openstack.org/150355 | 03:25 |
| *** salv-orlando has joined #openstack-nova | 03:26 | |
| *** salv-orlando has quit IRC | 03:29 | |
| *** eliqiao has joined #openstack-nova | 03:30 | |
| *** salv-orlando has joined #openstack-nova | 03:32 | |
| *** salv-orlando has quit IRC | 03:32 | |
| *** takedakn has joined #openstack-nova | 03:35 | |
| *** salv-orlando has joined #openstack-nova | 03:36 | |
| *** salv-orlando has quit IRC | 03:38 | |
| *** salv-orlando has joined #openstack-nova | 03:41 | |
| *** Longgeek has joined #openstack-nova | 03:41 | |
| *** Longgeek_ has quit IRC | 03:44 | |
| *** salv-orlando has quit IRC | 03:45 | |
| *** deepthi has joined #openstack-nova | 03:46 | |
| *** iamjarvo has joined #openstack-nova | 03:46 | |
| *** iamjarvo has joined #openstack-nova | 03:47 | |
| *** takedakn has quit IRC | 03:48 | |
| *** haomaiwang has joined #openstack-nova | 03:49 | |
| *** salv-orlando has joined #openstack-nova | 03:50 | |
| *** salv-orlando has quit IRC | 03:57 | |
| *** iamjarvo has quit IRC | 04:03 | |
| *** eliqiao has quit IRC | 04:09 | |
| *** shyama has joined #openstack-nova | 04:09 | |
| *** achanda has joined #openstack-nova | 04:10 | |
| *** Longgeek has quit IRC | 04:11 | |
| *** Longgeek has joined #openstack-nova | 04:11 | |
| *** salv-orlando has joined #openstack-nova | 04:19 | |
| *** salv-orlando has quit IRC | 04:24 | |
| *** otter768 has quit IRC | 04:27 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Handle RESIZE_PREP status when nova compute do init_instance https://review.openstack.org/132827 | 04:27 |
| *** salv-orlando has joined #openstack-nova | 04:27 | |
| *** salv-orlando has quit IRC | 04:28 | |
| *** vmtrooper has joined #openstack-nova | 04:30 | |
| *** vmtrooper has quit IRC | 04:36 | |
| *** bkopilov has quit IRC | 04:39 | |
| *** gokrokve has joined #openstack-nova | 04:40 | |
| *** mtanino is now known as mtanino_away | 04:40 | |
| *** hanzhang has quit IRC | 04:41 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Libvirt: preallocate_images CONFIG can be arbitrary characters to enable preallocate images https://review.openstack.org/160105 | 04:42 |
| *** hanzhang has joined #openstack-nova | 04:42 | |
| *** gokrokve has quit IRC | 04:45 | |
| *** gokrokve has joined #openstack-nova | 04:46 | |
| *** annashen has joined #openstack-nova | 04:47 | |
| *** changbl has quit IRC | 04:48 | |
| *** baoli has quit IRC | 04:50 | |
| *** gokrokve has quit IRC | 04:50 | |
| *** eliqiao has joined #openstack-nova | 04:55 | |
| *** bkopilov has joined #openstack-nova | 04:55 | |
| *** annashen has quit IRC | 05:03 | |
| *** armax has joined #openstack-nova | 05:05 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Move policy enforcement into REST API layer for v2.1 api volume_attachment https://review.openstack.org/159376 | 05:11 |
| openstackgerrit | Alex Xu proposed openstack/nova: Rename and move the v2.1 api policy into separated files https://review.openstack.org/150267 | 05:11 |
| *** aswadr has joined #openstack-nova | 05:15 | |
| *** gokrokve has joined #openstack-nova | 05:16 | |
| *** gokrokve has quit IRC | 05:18 | |
| *** gokrokve has joined #openstack-nova | 05:18 | |
| *** gokrokve has quit IRC | 05:23 | |
| openstackgerrit | Lan Qi Song proposed openstack/nova: Vmware:Fix how VolumeOps is getting host information https://review.openstack.org/135182 | 05:29 |
| *** ijw has quit IRC | 05:31 | |
| *** Nikolay_St has quit IRC | 05:31 | |
| *** dsanders has quit IRC | 05:33 | |
| *** dsanders has joined #openstack-nova | 05:33 | |
| *** eliqiao has quit IRC | 05:34 | |
| *** garyk has quit IRC | 05:37 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for network_associate https://review.openstack.org/161234 | 05:37 |
| openstackgerrit | Alex Xu proposed openstack/nova: Remove db layer hard-code permission checks for network_disassociate https://review.openstack.org/161624 | 05:39 |
| openstackgerrit | Alex Xu proposed openstack/nova: Deassociate before delete network in os-tenant-networks delete method https://review.openstack.org/164563 | 05:39 |
| openstackgerrit | Alex Xu proposed openstack/nova: Pass project_id when create networks by os-tenant-networks https://review.openstack.org/164549 | 05:39 |
| *** ijw has joined #openstack-nova | 05:41 | |
| *** hblixt has joined #openstack-nova | 05:42 | |
| *** sluo_wfh has joined #openstack-nova | 05:46 | |
| *** eliqiao has joined #openstack-nova | 05:46 | |
| *** armax has quit IRC | 05:51 | |
| *** ijw has quit IRC | 05:51 | |
| *** ijw has joined #openstack-nova | 05:52 | |
| *** hblixt has quit IRC | 05:53 | |
| *** sluo_wfh has quit IRC | 05:55 | |
| *** changbl has joined #openstack-nova | 05:58 | |
| *** lpetrut has joined #openstack-nova | 06:01 | |
| *** ferest has joined #openstack-nova | 06:02 | |
| *** dboik_ has quit IRC | 06:05 | |
| *** ferest has quit IRC | 06:07 | |
| *** sluo_wfh has joined #openstack-nova | 06:08 | |
| *** nelsnelson has quit IRC | 06:09 | |
| *** achanda has quit IRC | 06:15 | |
| *** irenab has quit IRC | 06:16 | |
| *** salv-orlando has joined #openstack-nova | 06:17 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/162030 | 06:17 |
| *** gokrokve has joined #openstack-nova | 06:18 | |
| *** irenab has joined #openstack-nova | 06:18 | |
| *** vmtrooper has joined #openstack-nova | 06:19 | |
| *** amotoki has joined #openstack-nova | 06:21 | |
| *** gokrokve has quit IRC | 06:23 | |
| *** salv-orlando has quit IRC | 06:23 | |
| *** afazekas has joined #openstack-nova | 06:23 | |
| *** vmtrooper has quit IRC | 06:24 | |
| *** pradipta has joined #openstack-nova | 06:25 | |
| *** salv-orlando has joined #openstack-nova | 06:34 | |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: scheduler:refactor group info in filter_properties https://review.openstack.org/148275 | 06:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: api: add soft-affinity policies for server groups https://review.openstack.org/147520 | 06:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: Add functional tests to soft affinity capabilities https://review.openstack.org/147521 | 06:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: scheduler: add num of instance per host in a group https://review.openstack.org/148277 | 06:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: object:add count_members_by_hosts to InstanceGroup https://review.openstack.org/147517 | 06:35 |
| openstackgerrit | Ildiko Vancsa proposed openstack/nova: scheduler: add soft-(anti-)affinity weighers https://review.openstack.org/147516 | 06:35 |
| *** _gryf|afk has quit IRC | 06:41 | |
| *** achanda has joined #openstack-nova | 06:42 | |
| *** nkrinner has joined #openstack-nova | 06:42 | |
| *** amotoki has quit IRC | 06:43 | |
| *** amotoki has joined #openstack-nova | 06:44 | |
| *** amotoki has quit IRC | 06:48 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Move unlock_override policy enforcement into V2.1 REST API layer https://review.openstack.org/162168 | 06:48 |
| *** _gryf has joined #openstack-nova | 06:49 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: DO NOT MERGE - Merge V2 and V2.1 functional tests https://review.openstack.org/162016 | 06:49 |
| *** eliqiao has quit IRC | 06:49 | |
| *** salv-orlando has quit IRC | 06:50 | |
| *** Longgeek_ has joined #openstack-nova | 06:52 | |
| *** Longgeek has quit IRC | 06:52 | |
| *** salv-orlando has joined #openstack-nova | 06:57 | |
| *** eliqiao has joined #openstack-nova | 06:57 | |
| *** vivekd has joined #openstack-nova | 07:06 | |
| *** garyk has joined #openstack-nova | 07:08 | |
| *** achanda has quit IRC | 07:12 | |
| *** pcaruana has quit IRC | 07:14 | |
| *** gokrokve has joined #openstack-nova | 07:18 | |
| *** gokrokve has quit IRC | 07:20 | |
| *** gokrokve has joined #openstack-nova | 07:20 | |
| *** oro has joined #openstack-nova | 07:21 | |
| *** oro_ has joined #openstack-nova | 07:21 | |
| *** browne has quit IRC | 07:22 | |
| *** Marga_ has quit IRC | 07:23 | |
| *** sahid has joined #openstack-nova | 07:24 | |
| *** gokrokve has quit IRC | 07:24 | |
| *** kaisers has joined #openstack-nova | 07:24 | |
| *** TobiasE has joined #openstack-nova | 07:26 | |
| *** ijw has quit IRC | 07:28 | |
| *** mtanino_away has quit IRC | 07:28 | |
| *** Nikolay_St has joined #openstack-nova | 07:28 | |
| *** romainh has joined #openstack-nova | 07:36 | |
| *** andreykurilin_ has joined #openstack-nova | 07:37 | |
| *** salv-orlando has quit IRC | 07:37 | |
| *** mpavone has joined #openstack-nova | 07:38 | |
| hanzhang | hi all! who can help me review this patch:https://review.openstack.org/#/c/164128/ | 07:41 |
| *** rwsu has joined #openstack-nova | 07:43 | |
| *** amotoki has joined #openstack-nova | 07:44 | |
| *** liusheng has joined #openstack-nova | 07:44 | |
| *** Marga_ has joined #openstack-nova | 07:45 | |
| *** ajayaa has joined #openstack-nova | 07:48 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for fixed_ip_associate_* https://review.openstack.org/150687 | 07:49 |
| *** dsanders has quit IRC | 07:55 | |
| *** andreykurilin_ has quit IRC | 07:56 | |
| *** dsanders has joined #openstack-nova | 07:59 | |
| *** Marga_ has quit IRC | 08:01 | |
| *** vivekd has quit IRC | 08:04 | |
| *** vivekd has joined #openstack-nova | 08:04 | |
| *** oro_ has quit IRC | 08:08 | |
| *** vmtrooper has joined #openstack-nova | 08:08 | |
| *** oro has quit IRC | 08:08 | |
| *** markus_z has joined #openstack-nova | 08:08 | |
| *** moshele has joined #openstack-nova | 08:11 | |
| *** vmtrooper has quit IRC | 08:13 | |
| *** dulek has joined #openstack-nova | 08:16 | |
| *** gokrokve has joined #openstack-nova | 08:18 | |
| *** park has quit IRC | 08:21 | |
| *** dsanders has quit IRC | 08:21 | |
| *** pradipta has quit IRC | 08:22 | |
| *** gokrokve has quit IRC | 08:23 | |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova: VMware: Replace get_dynamic_property with get_object_property https://review.openstack.org/164220 | 08:23 |
| *** eliqiao has quit IRC | 08:24 | |
| *** eliqiao has joined #openstack-nova | 08:28 | |
| *** lpetrut has quit IRC | 08:28 | |
| *** gszasz has joined #openstack-nova | 08:29 | |
| *** jcoufal has joined #openstack-nova | 08:33 | |
| *** ndipanov has joined #openstack-nova | 08:33 | |
| gmann | lifeless: ping | 08:33 |
| openstackgerrit | sahid proposed openstack/nova: compute: make test rpcapi more flexible https://review.openstack.org/164177 | 08:37 |
| *** salv-orlando has joined #openstack-nova | 08:38 | |
| *** nellysmitt has joined #openstack-nova | 08:40 | |
| *** jlanoux has joined #openstack-nova | 08:41 | |
| *** jang has quit IRC | 08:41 | |
| *** jang has joined #openstack-nova | 08:42 | |
| *** ildikov has quit IRC | 08:47 | |
| *** salv-orl_ has joined #openstack-nova | 08:50 | |
| openstackgerrit | Eli Qiao proposed openstack/nova-specs: Judge the service state when perform a deleting on it https://review.openstack.org/163274 | 08:51 |
| *** salv-orlando has quit IRC | 08:51 | |
| *** oro_ has joined #openstack-nova | 08:54 | |
| *** oro has joined #openstack-nova | 08:55 | |
| openstackgerrit | wuhao proposed openstack/nova: Add previous page links for Nova GET servers API https://review.openstack.org/164169 | 08:55 |
| *** lpetrut has joined #openstack-nova | 08:56 | |
| *** lpetrut has quit IRC | 08:57 | |
| *** jcoufal has quit IRC | 08:57 | |
| *** salv-orl_ has quit IRC | 08:57 | |
| *** apuimedo has joined #openstack-nova | 09:01 | |
| *** salv-orlando has joined #openstack-nova | 09:02 | |
| *** mpovolny has joined #openstack-nova | 09:02 | |
| *** salv-orlando has quit IRC | 09:05 | |
| *** eglynn-office has joined #openstack-nova | 09:06 | |
| *** eglynn-officeafk is now known as eglynn | 09:06 | |
| *** eglynn-office has quit IRC | 09:07 | |
| *** eglynn-office has joined #openstack-nova | 09:07 | |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova: VMware: Replace get_dynamic_properties with get_object_properties_dict https://review.openstack.org/164604 | 09:09 |
| *** jistr has joined #openstack-nova | 09:11 | |
| *** derekh has joined #openstack-nova | 09:14 | |
| *** lucasagomes has joined #openstack-nova | 09:17 | |
| *** gokrokve has joined #openstack-nova | 09:18 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: use oslo.vmware methods for handling tokens https://review.openstack.org/125073 | 09:19 |
| openstackgerrit | Lan Qi Song proposed openstack/nova: Vmware:Fix how VolumeOps is getting host information https://review.openstack.org/135182 | 09:22 |
| *** salv-orlando has joined #openstack-nova | 09:22 | |
| *** gokrokve has quit IRC | 09:22 | |
| *** jang has quit IRC | 09:24 | |
| *** salv-orlando has quit IRC | 09:25 | |
| *** jang has joined #openstack-nova | 09:26 | |
| *** vivekd_ has joined #openstack-nova | 09:30 | |
| *** vivekd has quit IRC | 09:31 | |
| openstackgerrit | garyk proposed openstack/nova: DB: add support for volume_type_id by BDMS https://review.openstack.org/139992 | 09:32 |
| openstackgerrit | garyk proposed openstack/nova: Object: add volume_type_id to block_device objects https://review.openstack.org/139994 | 09:33 |
| *** salv-orlando has joined #openstack-nova | 09:34 | |
| *** vivekd_ has quit IRC | 09:36 | |
| *** salv-orlando has quit IRC | 09:37 | |
| *** salv-orlando has joined #openstack-nova | 09:38 | |
| *** dheeraj-gupta-4 has joined #openstack-nova | 09:38 | |
| *** salv-orlando has quit IRC | 09:43 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 09:43 | |
| *** salv-orlando has joined #openstack-nova | 09:44 | |
| *** dimsum__ has joined #openstack-nova | 09:46 | |
| *** jang has quit IRC | 09:46 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: Variable 'name' already declared in 'for' loop https://review.openstack.org/164631 | 09:48 |
| *** pixelb has joined #openstack-nova | 09:49 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: enable configuring of addition information for virtual machine https://review.openstack.org/140669 | 09:50 |
| *** vivekd has joined #openstack-nova | 09:52 | |
| *** kashyapc_ has joined #openstack-nova | 09:53 | |
| *** jang1 has joined #openstack-nova | 09:54 | |
| *** salv-orlando has quit IRC | 09:56 | |
| *** vmtrooper has joined #openstack-nova | 09:57 | |
| *** salv-orl_ has joined #openstack-nova | 09:58 | |
| *** sabari has quit IRC | 10:00 | |
| *** moshele has quit IRC | 10:00 | |
| *** sabari has joined #openstack-nova | 10:01 | |
| *** vmtrooper has quit IRC | 10:02 | |
| *** asettle has joined #openstack-nova | 10:04 | |
| *** kashyapc_ has quit IRC | 10:04 | |
| *** kashyapc has joined #openstack-nova | 10:05 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: Unused e variable in exception https://review.openstack.org/164635 | 10:05 |
| *** Maike has joined #openstack-nova | 10:06 | |
| *** salv-orl_ has quit IRC | 10:08 | |
| *** gszasz has quit IRC | 10:08 | |
| *** asettle has quit IRC | 10:08 | |
| *** gszasz has joined #openstack-nova | 10:09 | |
| *** salv-orlando has joined #openstack-nova | 10:13 | |
| *** moshele has joined #openstack-nova | 10:13 | |
| *** salv-orlando has quit IRC | 10:17 | |
| *** gokrokve has joined #openstack-nova | 10:18 | |
| *** coolsvap is now known as coolsvap|afk | 10:19 | |
| *** jang1 has quit IRC | 10:20 | |
| *** eliqiao has quit IRC | 10:20 | |
| *** salv-orlando has joined #openstack-nova | 10:22 | |
| *** jang1 has joined #openstack-nova | 10:22 | |
| *** gokrokve has quit IRC | 10:23 | |
| *** salv-orlando has quit IRC | 10:29 | |
| *** oro_ has quit IRC | 10:29 | |
| *** oro has quit IRC | 10:30 | |
| *** ildikov has joined #openstack-nova | 10:31 | |
| *** salv-orlando has joined #openstack-nova | 10:33 | |
| *** salv-orlando has quit IRC | 10:36 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for floating_ips_bulk https://review.openstack.org/150704 | 10:36 |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Add floating_ips_bulk policy enforcement test case for v2.1 REST API layer https://review.openstack.org/160120 | 10:36 |
| *** jamie_h has joined #openstack-nova | 10:45 | |
| *** popw has quit IRC | 10:46 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: DO NOT MERGE - Merge V2 and V2.1 functional tests https://review.openstack.org/162016 | 10:47 |
| *** salv-orlando has joined #openstack-nova | 10:51 | |
| *** exploreshaifali has joined #openstack-nova | 10:51 | |
| *** shaohe_feng1 has quit IRC | 10:51 | |
| *** ZZelle has joined #openstack-nova | 10:55 | |
| *** salv-orlando has quit IRC | 10:56 | |
| *** kashyapc has quit IRC | 11:02 | |
| *** dims_ has joined #openstack-nova | 11:02 | |
| *** jang1 has quit IRC | 11:04 | |
| *** dimsum__ has quit IRC | 11:04 | |
| *** jang1 has joined #openstack-nova | 11:05 | |
| *** matrohon has joined #openstack-nova | 11:06 | |
| *** haomaiwang has quit IRC | 11:06 | |
| *** dims_ has quit IRC | 11:07 | |
| *** gszasz has quit IRC | 11:09 | |
| *** dimsum__ has joined #openstack-nova | 11:13 | |
| *** gszasz has joined #openstack-nova | 11:13 | |
| *** gszasz has quit IRC | 11:13 | |
| *** takedakn has joined #openstack-nova | 11:14 | |
| *** exploreshaifali has quit IRC | 11:14 | |
| *** gszasz has joined #openstack-nova | 11:16 | |
| *** takedakn has quit IRC | 11:18 | |
| *** gokrokve has joined #openstack-nova | 11:18 | |
| *** takedakn has joined #openstack-nova | 11:18 | |
| openstackgerrit | sahid proposed openstack/nova: compute: introduce binary constants https://review.openstack.org/157363 | 11:20 |
| *** aix has quit IRC | 11:21 | |
| *** VW_ has joined #openstack-nova | 11:22 | |
| *** gokrokve has quit IRC | 11:23 | |
| *** eliqiao has joined #openstack-nova | 11:23 | |
| *** ajayaa has quit IRC | 11:24 | |
| hanzhang | who can help to review this patch:https://review.openstack.org/#/c/164128/ | 11:30 |
| *** IanGovett has joined #openstack-nova | 11:31 | |
| alex_xu | good morning everyone, there are fun policy patch ready to merge and review https://etherpad.openstack.org/p/apipolicycheck Thanks in advance for anyone want to enjoy review them! | 11:34 |
| alex_xu | mikal: johnthetubaguy sdague , the last two v2.1 policy patches are ready to review https://review.openstack.org/159376 https://review.openstack.org/150267 , then we can focus on db cleanup, thanks in advance! | 11:37 |
| *** ajayaa has joined #openstack-nova | 11:37 | |
| *** rushiagr_away is now known as rushiagr | 11:39 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: DO NOT MERGE - Merge V2 and V2.1 functional tests https://review.openstack.org/162016 | 11:43 |
| johnthetubaguy | alex_xu: does 150267 need devstack changes to make the test pass? | 11:43 |
| *** vmtrooper has joined #openstack-nova | 11:46 | |
| *** mwagner_lap has quit IRC | 11:46 | |
| sdague | alex_xu: can you set a Depends-On to show it tested with the devstack change? | 11:50 |
| *** jaypipes has joined #openstack-nova | 11:51 | |
| sdague | or, I guess I can do that | 11:51 |
| *** vmtrooper has quit IRC | 11:51 | |
| openstackgerrit | Sean Dague proposed openstack/nova: Rename and move the v2.1 api policy into separated files https://review.openstack.org/150267 | 11:51 |
| johnthetubaguy | sdague: thanks, that makes sense | 11:52 |
| sdague | alex_xu: updated with depends on so that it will test against the devstack change | 11:52 |
| dimsum__ | nova cores - fresh set of quick hit bugs, lots of new ones - https://etherpad.openstack.org/p/kilo-nova-priorities-tracking | 11:53 |
| sdague | I finally figured out why we get all these no network tracebacks in the api samples tests | 11:53 |
| sdague | we're actually using the database, and never preallocate any networks | 11:54 |
| gmann | sdague: can you please have a look into - https://review.openstack.org/#/c/162016/ | 11:55 |
| *** rushiagr is now known as rushiagr_away | 11:55 | |
| gmann | sdague: doing test merge with testscenario but looks like scenario generated tests gets hang - http://logs.openstack.org/16/162016/4/check/gate-nova-tox-functional/cdcd378/console.html | 11:55 |
| *** mwagner_lap has joined #openstack-nova | 11:56 | |
| gmann | sdague: any idea m missing anything here | 11:56 |
| sdague | ok, it won't be until later this week, because I'm trying to get this scheduler thing sorted right now with some tests. | 11:56 |
| *** VW_ has quit IRC | 11:56 | |
| sdague | gmann: so, lets assume some of the test cleanup can happen post freeze | 11:56 |
| *** deepthi has quit IRC | 11:57 | |
| sdague | the real question I think needs to be addressed before freeze is what's our paste.ini going to look like | 11:57 |
| sdague | are we going to replace the v2 endpoint with v2.1 ? | 11:57 |
| sdague | can we drop v3 entirely | 11:57 |
| gmann | sdague: yes, running same tests cases with changing /v2 with v2.1 | 11:58 |
| sdague | right, but the paste.ini is what people use to deploy as well | 11:58 |
| gmann | sdague: as v3 tests are as per extension and in good shape , we can drop v2 tests and share v3(v2.1) for v2 and v2.1 | 11:59 |
| gmann | sdague: that will in tmp file through fixture not in original one | 12:00 |
| *** TobiasE1 has joined #openstack-nova | 12:00 | |
| sdague | right, so lets ignore the tests issue for right now | 12:01 |
| sdague | I actually want to understand the contents of paste.ini on release | 12:01 |
| *** rushiagr_away is now known as rushiagr | 12:01 | |
| *** jcru has joined #openstack-nova | 12:01 | |
| *** jcru has quit IRC | 12:01 | |
| sdague | and what we are recommending people do to change their paste.ini in production | 12:01 |
| *** TobiasE has quit IRC | 12:01 | |
| *** jcru has joined #openstack-nova | 12:02 | |
| *** gszasz has quit IRC | 12:02 | |
| gmann | sdague: ohk, sounds good. I will hold up tests merge thing till release | 12:03 |
| *** henrique_ has joined #openstack-nova | 12:04 | |
| *** dprince has joined #openstack-nova | 12:07 | |
| alex_xu | johnthetubaguy: sdague thanks, sorry for just saw your message | 12:07 |
| *** salv-orlando has joined #openstack-nova | 12:07 | |
| gmann | sdague: so changing /v2 endpoint with v2.1 can cause backward compatibility for /v2 users due to strong input validation | 12:07 |
| alex_xu | how that patch can pass test~ | 12:08 |
| openstackgerrit | garyk proposed openstack/nova: DB: add support for volume_type_id by BDMS https://review.openstack.org/139992 | 12:08 |
| sdague | gmann: well for existing users, they are going to have to make a change manually | 12:09 |
| openstackgerrit | garyk proposed openstack/nova: Object: add volume_type_id to block_device objects https://review.openstack.org/139994 | 12:09 |
| sdague | johnthetubaguy: you have any thoughts on this one? | 12:10 |
| *** aysyd has joined #openstack-nova | 12:10 | |
| gmann | sdague: ohk | 12:10 |
| * alex_xu go home now, will be back after one and half hour | 12:12 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Removed unused variables, methods and classes https://review.openstack.org/164666 | 12:12 |
| eliqiao | alex_xu: bye | 12:12 |
| *** ildikov has quit IRC | 12:13 | |
| *** salv-orlando has quit IRC | 12:14 | |
| garyk | sdague: can you please look at https://review.openstack.org/162652. i needed to rebase after yo approved | 12:14 |
| johnthetubaguy | sdague: good question, I was just thinking about that… I am trying to remember whats in the keystone catalog | 12:15 |
| *** gszasz has joined #openstack-nova | 12:15 | |
| sdague | garyk: approved | 12:16 |
| sdague | johnthetubaguy: well the keystone catalog is whatever you put there | 12:16 |
| garyk | sdague: gracias | 12:16 |
| *** ildikov has joined #openstack-nova | 12:16 | |
| johnthetubaguy | sdague: I was just checking, the v2 bit is included in the catalog | 12:17 |
| sdague | right | 12:17 |
| *** ankit_ag has joined #openstack-nova | 12:18 | |
| *** gokrokve has joined #openstack-nova | 12:18 | |
| *** baoli has joined #openstack-nova | 12:18 | |
| sdague | so, honestly, it seems like the right thing to do is to set v2 as v2.1 in our paste.ini and leave a commented out section for how to restore old v2 as the v2 end point | 12:18 |
| sdague | existing deploys will stay on v2 until they update their paste.ini | 12:19 |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Remove cell api overrides for force-delete https://review.openstack.org/135202 | 12:19 |
| johnthetubaguy | sdague: I didn't think we were ready for that though? as we are not sure all SDKs will work with that new endpoint? | 12:19 |
| sdague | well, v2 is hard frozen at this point now, right? | 12:20 |
| johnthetubaguy | sdague: I have a slight preference for defining a specific special name for v2.1, and add that into the keystone catalog in a def core checked place, and go down that line | 12:20 |
| johnthetubaguy | sdague: agreed its hard frozen, but I don't like us breaking so many clients "out of the box", all be it, badly behaved clients | 12:21 |
| sdague | the whole point of v2.1 was that it was largely a seemless roll forward | 12:21 |
| *** salv-orlando has joined #openstack-nova | 12:21 | |
| *** raildo has joined #openstack-nova | 12:21 | |
| sdague | johnthetubaguy: ok, except I'm not sure I understand what the break is | 12:22 |
| *** salv-orlando has quit IRC | 12:22 | |
| johnthetubaguy | sdague: I maybe wrongly thought we would have both running in parallel, while the major SDKs are caught up | 12:22 |
| johnthetubaguy | sdague: the break is the extra validation right? | 12:22 |
| sdague | I expect that if we don't put v2.1 on v2, no one is going to clean up anything | 12:22 |
| sdague | except, existing deploys won't automatically flip | 12:22 |
| *** salv-orlando has joined #openstack-nova | 12:22 | |
| *** rajesht has joined #openstack-nova | 12:23 | |
| sdague | because no one is crazy enough to CD their paste.ini file | 12:23 |
| *** gokrokve has quit IRC | 12:23 | |
| sdague | so they get release notes about how to move onto v2.1 | 12:23 |
| johnthetubaguy | sdague: true, I was thinking that we do that in liberty, switch the default, after a release of testing, but maybe thats not forcing the issue quick enough… | 12:23 |
| sdague | new deploys start in v2.1 | 12:23 |
| johnthetubaguy | ah, thats a good point... | 12:23 |
| sdague | and we document how to easily use old v2 if you want instead | 12:24 |
| sdague | but it seems like we should default to moving forward. | 12:24 |
| *** dkliban_afk is now known as dkliban | 12:24 | |
| johnthetubaguy | sdague: if we knew all major SDKs worked with the new version, I would be cool with that, but I am not convinced we know that yet | 12:25 |
| johnthetubaguy | sdague: but I am torn, I like the idea of forcing the issue, I just wonder if we should wait six months to get there, maybe thats dumb | 12:26 |
| sdague | ok, so let's define what criteria you think is needed there | 12:26 |
| *** rfolco has joined #openstack-nova | 12:26 | |
| sdague | well if we wait six months that means when we flip to v2.1 you just consumed 40 new microversions, not 4 | 12:26 |
| *** claudiub has joined #openstack-nova | 12:27 | |
| sdague | also, if we don't default to v2.1, ec2 out of tree doesn't work | 12:27 |
| sdague | because that needs v2.1 | 12:27 |
| sdague | v2.3 actualy | 12:27 |
| *** alexpilotti has joined #openstack-nova | 12:27 | |
| johnthetubaguy | sdague: I was assuming we made both available by default, making the new local the new name, and the old one will eventually redirect/forward to the new code, once people are happy | 12:27 |
| sdague | ok, this seems sufficiently nuanced to need an ML thread to get us an answer | 12:27 |
| sdague | johnthetubaguy: yeh, I think the issue is a bunch of that is out of our control. Nova doesn't manage it's catalog entries. | 12:28 |
| johnthetubaguy | sdague: yeah, thats true, although I was hoping def core would give us a define name that has to be present, and we make that point to v2.1 from the start? | 12:29 |
| sdague | defcore is still working on havana | 12:29 |
| sdague | so... they really won't be helpful in forward looking statements here I don't think | 12:29 |
| johnthetubaguy | sdague: I am over thinking this, I think the key question is, do we want new deployes using v2.1 by default | 12:30 |
| johnthetubaguy | I worry about broken SDKs because there is actually an API change their | 12:30 |
| johnthetubaguy | sdague: we did talk about a soft mode, where validation error for extra attributes would just log errors, during the transition, but I am not sure we ever implemented that | 12:31 |
| *** dimsum__ has quit IRC | 12:32 | |
| *** dimsum__ has joined #openstack-nova | 12:32 | |
| sdague | the reality is, the validation mostly just means you don't have corrupt data in your db which fails later | 12:32 |
| johnthetubaguy | sdague: maybe that would be enough, its the error out on extra params for requests without micro versions specified that worry me | 12:32 |
| johnthetubaguy | sdague: I am only worried about the stuff we never looked at really | 12:33 |
| sdague | because what was happening otherwise is db types were causing exceptions to bubble up when we tried to insert | 12:33 |
| johnthetubaguy | sdague: right, so I want that validation, thats fine, its the extra param we used to just ignore that worry we me most, and only requests that have no micro version specified | 12:33 |
| *** lucasagomes is now known as lucas-hungry | 12:34 | |
| sdague | are we doing that on all requests, or just micro version ones? | 12:34 |
| johnthetubaguy | some old SDKs made otherwise valid requests, that worked, no db corruption, but I think v2.1 would reject some of their requests, as they have extra params that are ignored | 12:34 |
| johnthetubaguy | sdague: so that might be my missunderstanding, I thought it was all requests, as we delete the old validation code inside the body of the code? | 12:35 |
| eliqiao | hi all, how to let devstack to git pull all project in /opt/stack ? | 12:35 |
| *** dave-mccowan has quit IRC | 12:36 | |
| * johnthetubaguy goes to get some lunch before a meeting in a bit… | 12:36 | |
| gmann | sdague: i think its for all request as we have additionalProperties False in schema files | 12:37 |
| sdague | johnthetubaguy: do we have a list of SDKs that someone could go off and investigate that would make you comfortable? | 12:37 |
| *** dheeraj-gupta-4 has left #openstack-nova | 12:37 | |
| *** subscope has joined #openstack-nova | 12:37 | |
| johnthetubaguy | sdague: in a way I do, but its not the best list, its very selfish, I am thinking about the ones we at rackspace recommend here: https://developer.rackspace.com/sdks/ | 12:38 |
| sdague | hey, it's a list :) | 12:38 |
| sdague | I think oomichi did some validation on a bunch of these for a similar tempest question in the past | 12:39 |
| *** jang1 has quit IRC | 12:39 | |
| sdague | I wonder if he'd do that here | 12:39 |
| *** jang has joined #openstack-nova | 12:39 | |
| johnthetubaguy | OK, cool, might be worth skipping pyrax on that list, but the others seems sound | 12:40 |
| edleafe | johnthetubaguy: why skip pyrax? | 12:40 |
| johnthetubaguy | sdague: although, it might just be that single extra param relax that we need, just strip them from the incoming request instead of erroring out on them. | 12:41 |
| *** vladikr has joined #openstack-nova | 12:41 | |
| *** ajayaa has quit IRC | 12:41 | |
| johnthetubaguy | edleafe: its more we should test python-novaclient instead, which I hope is OK | 12:41 |
| sdague | right, that | 12:41 |
| sdague | we should be testing the openstack codepaths in these libs, not the rax specific ones | 12:42 |
| johnthetubaguy | edleafe: pyrax should be a rackspace problem, and I think it just uses novaclient underneath anyways, unless my memory is playing tricks on me | 12:42 |
| johnthetubaguy | sdague: +1 | 12:42 |
| sdague | johnthetubaguy: I don't think it does, I think mordred was ragey about that at one point | 12:42 |
| edleafe | johnthetubaguy: yes, it uses novaclient | 12:42 |
| edleafe | johnthetubaguy: but it isn't a rax-only sdk | 12:42 |
| mordred | aroo? | 12:42 |
| mordred | yes | 12:42 |
| mordred | what? | 12:42 |
| mordred | pyrax should be tested for whatever issue you're discussing | 12:43 |
| mordred | because it's a clear and valid consumer of python-novaclient | 12:43 |
| *** jang has quit IRC | 12:43 | |
| mordred | and might find library use cases that someone didn't consider | 12:44 |
| *** jang has joined #openstack-nova | 12:44 | |
| *** VW_ has joined #openstack-nova | 12:45 | |
| *** lpetrut has joined #openstack-nova | 12:47 | |
| *** sdake has joined #openstack-nova | 12:47 | |
| * mordred goes to read the scrollback | 12:47 | |
| mordred | sdague: the v2.1 thing above - is there anything a novaclient consumer needs to know? or if I specify v2 it'll happily give me v2.1 from the server? | 12:49 |
| *** openstackgerrit has quit IRC | 12:50 | |
| *** openstackgerrit has joined #openstack-nova | 12:50 | |
| mordred | sdague: because I'm passing '2' as the first parameter to the Client constructor right now | 12:50 |
| sdague | that's part of this discussion, because currently it's a different paste endpoint | 12:50 |
| *** adalbas has joined #openstack-nova | 12:51 | |
| *** fifieldt has joined #openstack-nova | 12:51 | |
| sdague | so you would need to specify something different | 12:52 |
| mordred | how is it discoverable? | 12:52 |
| sdague | if you GET / you'll get both versions listed with endpoints | 12:53 |
| mordred | will those both be listed in keystone? | 12:54 |
| mordred | (I am a consumer of novaclient, I don't GET / ever - I use the library) | 12:54 |
| sdague | registering in keystone is beyond scope for nova, I can tell you what devstack does | 12:56 |
| sdague | which is 2 endpoints | 12:56 |
| *** oro has joined #openstack-nova | 12:56 | |
| mordred | btw - not to derail this - the thing I got ragey about re: pyrax was that I Was trying to look in the heat source code to figure out how heat handled something on rackspace since I knew that rax was running heat, and I found that rather than encode any re-usable logic I could learn from they'd just made a rax driver which used pyrax instead | 12:56 |
| mordred | sdague: ok - so, let me try to rephrase ... | 12:56 |
| *** oro_ has joined #openstack-nova | 12:56 | |
| mordred | sdague: as a person who has 3 different clouds that I have scripts running against, how do I know how to hand the right thing to Client() for each of them - and how do I detect if one of them makes this deployment change? | 12:57 |
| *** wuhg has joined #openstack-nova | 12:57 | |
| *** dboik has joined #openstack-nova | 12:57 | |
| mordred | will '2' as the opaque magic parameter I'm supposed to pass keep working? | 12:57 |
| * mordred is happy to change the magic parameter passed to novaclient - just wants to know which parameter he can pass that says "just get me something that works | 13:00 | |
| sdague | mordred: honestly, it's a good question. Which I'm trying to sift at the moment. The problem is there are multiple stages between nova upstream code, and what a deployment looks like. | 13:01 |
| sdague | where's the shade code that calls novaclient? | 13:01 |
| sdague | I'd like to figure out if that opaque '2' can be removed entirely | 13:01 |
| edleafe | sdague: pyrax imports v2 of novaclient: from novaclient.v2 import client as _cs_client | 13:03 |
| edleafe | sdague: so as long as novaclient handles that correctly, it should be fine | 13:04 |
| edleafe | sdague: i.e., if novaclient maps v2 to v2.1 | 13:04 |
| *** ijw has joined #openstack-nova | 13:06 | |
| *** dboik has quit IRC | 13:09 | |
| *** wverdugo500 has left #openstack-nova | 13:09 | |
| sdague | yeh, honestly, I don't know. The nova client code sometimes uses the service catalog, and sometimes seems to bypass it. Looking now | 13:09 |
| *** dave-mccowan has joined #openstack-nova | 13:10 | |
| *** aix has joined #openstack-nova | 13:10 | |
| mordred | sdague: I _used_ to use import novaclient.v2 like pyrax does - then I was told that was wrong and that I should use the unversioned one | 13:11 |
| mordred | sdague: http://git.openstack.org/cgit/openstack-infra/shade/tree/shade/__init__.py#n221 | 13:11 |
| mordred | sdague: which, unless a cloud is explictly configured differently, will wind up with the vaule here: http://git.openstack.org/cgit/stackforge/os-client-config/tree/os_client_config/defaults.py#n17 | 13:12 |
| mordred | http://git.openstack.org/cgit/openstack-infra/shade/tree/shade/__init__.py#n29 import statement, fwiw | 13:12 |
| sdague | mordred: yeh, so, in my reading the magic '2' stays | 13:13 |
| sdague | honestly, everything maps to the magic '2' now | 13:13 |
| sdague | https://github.com/openstack/python-novaclient/blob/master/novaclient/client.py#L704-L707 | 13:14 |
| sdague | which is a whole other wonkiness that we should figure out if we can unwind | 13:15 |
| *** kaufer has joined #openstack-nova | 13:15 | |
| mordred | sdague: AWESOME. I would also love to unwind that | 13:16 |
| mordred | sdague: and I'd honestly love if we could just make that a keyword param that defaults to '2' | 13:16 |
| sdague | yeh, so for compat reasons I think we need to be trickier | 13:16 |
| mordred | so that a consumer of the library who does not know what value to put in can just have it default to the magic value | 13:16 |
| sdague | but, regardless, lets put that off until post release, because there are bigger fish to fry atm | 13:17 |
| mordred | if it's a keyword param but is the first param in the list, positional '2' passed in from old code should still continue to work as normal - but new code could just remove it as an arg | 13:17 |
| mordred | sdague: ++ | 13:17 |
| *** shaohe_feng1 has joined #openstack-nova | 13:17 | |
| sdague | yeh, but I'd actually like to deprecate the usage to the caller, and have them know they can drop the version | 13:17 |
| sdague | so drop the arg, inspect args, if value 0 is 2 emit deprecation warning | 13:18 |
| *** gokrokve has joined #openstack-nova | 13:18 | |
| *** TobiasE has joined #openstack-nova | 13:18 | |
| sdague | and give it 6 months to cycle out | 13:19 |
| *** TobiasE1 has quit IRC | 13:19 | |
| sdague | because I don't think we're ever going to have multiple backends here in this code, we're unwinding all that | 13:20 |
| mordred | ++ | 13:20 |
| mordred | I support your end goal | 13:20 |
| *** gokrokve has quit IRC | 13:22 | |
| *** IanGovett has quit IRC | 13:24 | |
| sdague | hmmm.... man jogo made a hell of a fake compute driver which supports 800 GB of mem and 1000 vcpu :) | 13:25 |
| *** amotoki has quit IRC | 13:26 | |
| *** claudiub_ has joined #openstack-nova | 13:29 | |
| *** sandywalsh has joined #openstack-nova | 13:31 | |
| *** claudiub has quit IRC | 13:32 | |
| *** IanGovett has joined #openstack-nova | 13:32 | |
| dansmith | meh, I can trivially make one that supports more | 13:33 |
| dimsum__ | sdague: haha | 13:33 |
| *** dimsum__ is now known as dims | 13:33 | |
| sdague | yeh, well, I was trying to build a test case for scheduling hitting limits | 13:34 |
| sdague | and was wondering why 20 medium instances didn't put me over the edge | 13:34 |
| sdague | now i know | 13:34 |
| dims | sdague: jogo's amazon-killer! | 13:34 |
| *** lucas-hungry is now known as lucasagomes | 13:35 | |
| *** vmtrooper has joined #openstack-nova | 13:35 | |
| sdague | well, I actually overran my quotas first (which are apparently enabled during tests) | 13:35 |
| *** shyama has quit IRC | 13:36 | |
| *** mriedem has joined #openstack-nova | 13:37 | |
| dansmith | yep, that's the first thin you have to fix when you use the fake driver | 13:37 |
| dansmith | I always do "a lot of nines" and then have to add more | 13:37 |
| sdague | why not just -1 | 13:38 |
| *** dboik has joined #openstack-nova | 13:39 | |
| *** dboik has quit IRC | 13:39 | |
| *** dboik has joined #openstack-nova | 13:40 | |
| obondarev | jaypipes: ping | 13:40 |
| *** vmtrooper has quit IRC | 13:40 | |
| *** exploreshaifali has joined #openstack-nova | 13:41 | |
| jaypipes | obondarev: on phone :) can you do in 20 minutes? | 13:41 |
| obondarev | jaypipes: sure, thanks | 13:41 |
| *** sahid has quit IRC | 13:44 | |
| *** sahid has joined #openstack-nova | 13:44 | |
| dansmith | sdague: will all the quota values take a -1? I thought there was some that wouldn't so I always just blow it out | 13:44 |
| mriedem | -1 would just mean unlimited | 13:44 |
| mriedem | s/would/should/ | 13:44 |
| *** rajesht has quit IRC | 13:44 | |
| *** pixelb has quit IRC | 13:45 | |
| mriedem | dansmith: sdague: https://github.com/openstack-dev/devstack/blob/master/lib/nova_plugins/hypervisor-fake#L38 | 13:46 |
| dansmith | well okay then | 13:46 |
| *** gmann_ has joined #openstack-nova | 13:46 | |
| mriedem | not sure why that script just doesn't set the fake quota driver | 13:46 |
| *** ajayaa has joined #openstack-nova | 13:47 | |
| ankit_ag | sdague: Could you please check my reply on https://review.openstack.org/#/c/145738/4/nova/compute/api.py | 13:47 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova-specs: libvirt: virtio-net multiqueue https://review.openstack.org/128825 | 13:48 |
| PaulMurray | dansmith, I have a quick objects question - we talked briefly about object equivalence at the mid-cycle meetup | 13:48 |
| dansmith | yup | 13:48 |
| PaulMurray | dansmith, in https://review.openstack.org/#/c/160942/6/nova/compute/resource_tracker.py I do a comparison of an old compute_node and a new changed one | 13:49 |
| PaulMurray | dansmith, ndipanov reviewed it and suggested putting the comparison in the objects code | 13:49 |
| PaulMurray | dansmith, I did it by getting a primitive for each, removing the changes part and then doing a comparson of the primitives | 13:50 |
| PaulMurray | dansmith, its pretty generic, could go in base - what do you think? | 13:50 |
| PaulMurray | dansmith, as a special method - not __equals__ | 13:50 |
| dansmith | first off, which line is the compare? | 13:50 |
| dansmith | oh, I see | 13:51 |
| dansmith | you're calling it resources | 13:51 |
| *** burt has joined #openstack-nova | 13:51 | |
| PaulMurray | tha's it - around 580 | 13:51 |
| dansmith | so a utility method that takes a list of keys to ignore, serializes two objects, strips them and compares? | 13:51 |
| *** mjturek1 has joined #openstack-nova | 13:51 | |
| PaulMurray | yes, basically - didn't have any to ignore, but can do that | 13:52 |
| PaulMurray | could also decide if want to strip the changes | 13:52 |
| ndipanov | dansmith, that really has no business living in the RT | 13:52 |
| dansmith | we discussed not implementing __eq__ on the base object, but if you do it on compute_node that'd be fine, right? | 13:52 |
| *** BobBall_AWOL is now known as BobBall | 13:52 | |
| *** gmann_ has quit IRC | 13:53 | |
| ndipanov | dansmith, PaulMurray I was thinking more along the lines of compute_node.compare_with_dict(values) | 13:53 |
| PaulMurray | ndipanov, yes, makes sense - didn't want to do an __eq__ | 13:53 |
| ndipanov | and copy_resource would be from_resource_dict or something | 13:54 |
| *** ankit_ag has quit IRC | 13:54 | |
| PaulMurray | dansmith, I'll go with putting it on the ComputeNode then - just wondered if you thought it might be useful on the base NovaObject | 13:55 |
| PaulMurray | ndipanov, ack | 13:55 |
| dansmith | PaulMurray: I don't think it needs to be on the base object, because it could just be a utility function and our base object is already super busy | 13:55 |
| *** gokrokve has joined #openstack-nova | 13:56 | |
| dansmith | PaulMurray: but if you create the utillity method and then use it from a method on compute_node, that seems fine | 13:56 |
| *** thomasem has joined #openstack-nova | 13:56 | |
| *** thomasem has quit IRC | 13:58 | |
| *** thomasem has joined #openstack-nova | 13:58 | |
| PaulMurray | dansmith, not sure what you mean for the utility method - where would it go? I was going to add it to the ComputeNode object | 13:59 |
| dansmith | PaulMurray: we have other utility methods in the base module | 13:59 |
| PaulMurray | dansmith, I'll take a look | 13:59 |
| PaulMurray | dansmith, thanks | 13:59 |
| *** samueldmq has quit IRC | 14:00 | |
| *** gokrokve_ has joined #openstack-nova | 14:00 | |
| *** ildikov has quit IRC | 14:03 | |
| *** gokrokve has quit IRC | 14:04 | |
| *** whenry has joined #openstack-nova | 14:04 | |
| garyk | is zuul alive? | 14:05 |
| mriedem | they are talking about it in -infra | 14:05 |
| obondarev | jaypipes: replied to your comments on https://review.openstack.org/#/c/157755/ will you please take a look when you have time? | 14:06 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VolumeOps unit tests https://review.openstack.org/139797 | 14:06 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds MigrationOps unit tests https://review.openstack.org/139796 | 14:07 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: removes test_hypervapi tests https://review.openstack.org/139798 | 14:07 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VMOps unit tests (part 2) https://review.openstack.org/138934 | 14:07 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests https://review.openstack.org/148980 | 14:07 |
| *** takedakn has quit IRC | 14:09 | |
| *** nelsnelson has joined #openstack-nova | 14:09 | |
| *** nelsnelson has quit IRC | 14:11 | |
| tdurakov | jaypipes: here is the patch we just discussed: https://review.openstack.org/#/c/164263/ Could you checked it?:) | 14:13 |
| *** markmcclain has left #openstack-nova | 14:15 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for network_associate https://review.openstack.org/161234 | 14:17 |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: add neutron api NotImplemented test cases for Netowrk V2.1 https://review.openstack.org/162564 | 14:17 |
| *** Nikolay_St has quit IRC | 14:17 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for network_associate https://review.openstack.org/161234 | 14:19 |
| *** jecarey has joined #openstack-nova | 14:21 | |
| *** nelsnelson has joined #openstack-nova | 14:22 | |
| *** bkopilov has quit IRC | 14:22 | |
| *** krykowski has joined #openstack-nova | 14:23 | |
| *** thangp has joined #openstack-nova | 14:26 | |
| *** garyk has quit IRC | 14:26 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: DO NOT MERGE: enable libvirt live snapshot (testing) https://review.openstack.org/147332 | 14:26 |
| *** amotoki has joined #openstack-nova | 14:26 | |
| *** garyk has joined #openstack-nova | 14:30 | |
| *** otter768 has joined #openstack-nova | 14:32 | |
| *** amotoki has quit IRC | 14:32 | |
| *** bkopilov has joined #openstack-nova | 14:33 | |
| *** zz_jgrimm- is now known as jgrimm | 14:35 | |
| *** otter768 has quit IRC | 14:36 | |
| sdague | johnthetubaguy: http://lists.openstack.org/pipermail/openstack-dev/2015-March/059160.html - hopefully the beginning of a conversation there | 14:38 |
| *** Marga_ has joined #openstack-nova | 14:40 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for network_create_safe https://review.openstack.org/161626 | 14:40 |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Pass project_id when create networks by os-tenant-networks https://review.openstack.org/164549 | 14:40 |
| *** Marga_ has quit IRC | 14:41 | |
| *** Marga_ has joined #openstack-nova | 14:41 | |
| openstackgerrit | wuhao proposed openstack/nova: Add previous page links for Nova GET servers API https://review.openstack.org/164169 | 14:42 |
| *** iamjarvo has joined #openstack-nova | 14:49 | |
| *** hemnafk is now known as hemna | 14:49 | |
| *** Maike has quit IRC | 14:50 | |
| *** VW_ has quit IRC | 14:52 | |
| *** krykowski has quit IRC | 14:55 | |
| *** adalbas has quit IRC | 14:56 | |
| *** browne has joined #openstack-nova | 14:56 | |
| *** Nikolay_St has joined #openstack-nova | 14:56 | |
| *** neelashah has joined #openstack-nova | 14:57 | |
| *** VW_ has joined #openstack-nova | 15:00 | |
| *** zzzeek has joined #openstack-nova | 15:01 | |
| *** mpovolny has quit IRC | 15:01 | |
| *** rushiagr is now known as rushiagr_away | 15:02 | |
| *** nelsnelson has quit IRC | 15:03 | |
| *** Maike has joined #openstack-nova | 15:03 | |
| *** Maike has quit IRC | 15:03 | |
| alex_xu | hi, dear cores, there are three patches need another +2 https://review.openstack.org/150352 https://review.openstack.org/150353 https://review.openstack.org/150354 thanks for any review! | 15:03 |
| *** Maike has joined #openstack-nova | 15:04 | |
| alex_xu | johnthetubaguy: sdague https://review.openstack.org/#/c/150267/ passed the test :) | 15:04 |
| *** garyk has quit IRC | 15:05 | |
| *** armax has joined #openstack-nova | 15:06 | |
| alex_xu | oops, there are -1 on the devstack patch, johnthetubaguy sdague forget it, looks like need address devstack patch comment | 15:07 |
| *** garyk has joined #openstack-nova | 15:07 | |
| *** shaohe_feng2 has joined #openstack-nova | 15:07 | |
| *** tonytan4ever has joined #openstack-nova | 15:08 | |
| *** shaohe_feng1 has quit IRC | 15:10 | |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for security_group_default_rule_destroy https://review.openstack.org/160257 | 15:11 |
| openstackgerrit | ShaoHe Feng proposed openstack/nova: Remove db layer hard-code permission checks for security_group_default_rule_create https://review.openstack.org/150718 | 15:11 |
| *** aswadr has quit IRC | 15:14 | |
| *** mpovolny has joined #openstack-nova | 15:15 | |
| *** david-lyle_afk is now known as david-lyle | 15:15 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Handle exception when doing detach_interface https://review.openstack.org/130151 | 15:16 |
| openstackgerrit | Steven Kaufer proposed openstack/nova-specs: Nova Server Count API Parameter https://review.openstack.org/134279 | 15:17 |
| *** achanda has joined #openstack-nova | 15:17 | |
| *** dsanders has joined #openstack-nova | 15:18 | |
| dansmith | edleafe: jaypipes: left a question/concern on the base scheduler instance update logic patch | 15:19 |
| dansmith | the faster you tell me I'm stupid for being concerned, the quicker we can move on :P | 15:19 |
| edleafe | dansmith: you're stupid | 15:20 |
| edleafe | dansmith: oh, wait, you wanted me to read your comment first... | 15:20 |
| edleafe | :) | 15:20 |
| dansmith | edleafe: optional :) | 15:20 |
| *** thedodd has joined #openstack-nova | 15:21 | |
| *** tsekiyama has joined #openstack-nova | 15:22 | |
| *** nelsnelson has joined #openstack-nova | 15:23 | |
| edleafe | dansmith: agree with the high load on scheduler startup. Async sounds like a good approach | 15:23 |
| *** vmtrooper has joined #openstack-nova | 15:24 | |
| dansmith | okay | 15:24 |
| edleafe | dansmith: the orginall proposal was to do a series of queries by host, but that was deemed to require too many DB calls | 15:24 |
| edleafe | dansmith: it was recommended to do it all in one call, and then sort it in the scheduler | 15:24 |
| dansmith | edleafe: so are you proposing doing the same call, but async completion? | 15:25 |
| dansmith | hmm | 15:25 |
| edleafe | dansmith: I'm not really sure | 15:25 |
| edleafe | I thought the series of calls would have been better | 15:25 |
| dansmith | was that discussed in the comments here somewhere? | 15:25 |
| dansmith | yeah, I would think it would be a lot better | 15:25 |
| edleafe | dansmith: IIRC it was during the spec review | 15:25 |
| *** achanda has quit IRC | 15:25 | |
| dansmith | the problem with the monolithic call is that the result has to fit entirely into memory twice | 15:25 |
| dansmith | not to mention being sent over the wire from the database | 15:26 |
| dansmith | I'm sure jaypipes has an opinion here | 15:26 |
| *** Nikolay_St has quit IRC | 15:27 | |
| *** dsanders has quit IRC | 15:27 | |
| *** eliqiao has quit IRC | 15:27 | |
| edleafe | dansmith: I'm sure he has several. :) | 15:27 |
| dansmith | hopefully non-conflicting | 15:27 |
| edleafe | dansmith: no promises | 15:27 |
| dansmith | edleafe: what was the name of the spec for this? | 15:28 |
| dansmith | isolate-scheduler-db was a parent one, right? didn't we do another one just for this specific thing? | 15:28 |
| edleafe | dansmith: in the meantime I'll play around with the async idea. Is there anything similar in the codebase I can steal from, um, model it on? | 15:28 |
| *** mjturek1 has quit IRC | 15:28 | |
| dansmith | ah, isolate-scheduler-db-filters | 15:28 |
| edleafe | dansmith: spec: https://review.openstack.org/#/c/138444 | 15:29 |
| dansmith | edleafe: well, just utils.spawn_n() | 15:29 |
| *** vmtrooper has quit IRC | 15:29 | |
| edleafe | dansmith: ok, that's what I was thinkning | 15:29 |
| edleafe | thinking, even | 15:29 |
| *** mjturek1 has joined #openstack-nova | 15:29 | |
| dansmith | edleafe: should be all you need to spawn a thread, and then either do the monolithic thing, or the series of queries so that you're updating self._instance_info as you go | 15:29 |
| openstackgerrit | Chris Friesen proposed openstack/nova: Validate server group anti-affinity policy https://review.openstack.org/162746 | 15:29 |
| *** rushiagr_away is now known as rushiagr | 15:30 | |
| edleafe | dansmith: regarding the second comment about requerying - currently each of the filters that use instance info do that on every request | 15:30 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Log exception from deallocate_port_for_instance for triage https://review.openstack.org/164738 | 15:31 |
| edleafe | dansmith: so this isn't a radical departure, and might be more efficient overall | 15:31 |
| dansmith | edleafe: okay | 15:33 |
| dansmith | edleafe: well, I clicked through a bunch of reviews on the spec and didn't find any arguments about multiple calls to build the list before I got sick of looking | 15:33 |
| edleafe | dansmith: yeah, it might have been over IRC | 15:33 |
| dansmith | edleafe: especially if you yield in your async loop, I think that it's an overall win | 15:34 |
| dansmith | if you think about it, | 15:34 |
| dansmith | if you spawn a thread, and then it does a mysql call that takes two minutes to complete, | 15:34 |
| dansmith | then you're blocking the whole scheduler for the whole time anyway, | 15:34 |
| dansmith | and it will potentially even do things like time out the service check interval before it finishes, etc | 15:34 |
| dansmith | if you split into multiple calls then at least you have the potential to yield some execution time to other things in between calls | 15:35 |
| *** krykowski has joined #openstack-nova | 15:36 | |
| *** Longgeek_ has quit IRC | 15:36 | |
| dansmith | edleafe: you might also be able to use InstanceList.get_by_filters() to query for ten compute nodes at a time, making your batches larger and call count smaller | 15:36 |
| dansmith | filters={'host': ['host1', 'host2', ... ] | 15:36 |
| edleafe | dansmith: ok, good idea | 15:37 |
| *** romainh has quit IRC | 15:38 | |
| edleafe | dansmith: I don't see the comments, but the spec changed to an all-at-once approch between PS 15 and 16. https://review.openstack.org/#/c/138444/15..16/specs/kilo/approved/isolate-scheduler-db-filters.rst,cm | 15:39 |
| dansmith | edleafe: okay, well, I'm not going to go digging through IRC logs and timestamps to find it :D | 15:40 |
| dansmith | edleafe: let's see what jaypipes says when he pops up | 15:40 |
| edleafe | dansmith: me neither | 15:40 |
| *** garyk1 has joined #openstack-nova | 15:44 | |
| *** garyk has quit IRC | 15:45 | |
| *** iamjarvo has quit IRC | 15:46 | |
| *** ZZelle_ has joined #openstack-nova | 15:46 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VolumeOps unit tests https://review.openstack.org/139797 | 15:47 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds MigrationOps unit tests https://review.openstack.org/139796 | 15:47 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: removes test_hypervapi tests https://review.openstack.org/139798 | 15:47 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VMOps unit tests (part 2) https://review.openstack.org/138934 | 15:47 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests https://review.openstack.org/148980 | 15:47 |
| bauzas | edleafe: dansmith: mmm good point, I also missed that performance-impacting change in the spec | 15:47 |
| *** achanda has joined #openstack-nova | 15:48 | |
| *** wuhg has quit IRC | 15:49 | |
| *** vivekd has quit IRC | 15:49 | |
| openstackgerrit | Julien Danjou proposed openstack/nova: Stop using exception.message https://review.openstack.org/164752 | 15:52 |
| openstackgerrit | Julien Danjou proposed openstack/nova: simple_tenant_usage: levera timeutils, drop strtime() https://review.openstack.org/164753 | 15:52 |
| *** exploreshaifali has quit IRC | 15:52 | |
| *** rushiagr is now known as rushiagr_away | 15:53 | |
| *** gokrokve_ has quit IRC | 15:53 | |
| openstackgerrit | Julien Danjou proposed openstack/nova: simple_tenant_usage: levera timeutils, drop strtime() https://review.openstack.org/164753 | 15:53 |
| *** itzikb_ has joined #openstack-nova | 15:54 | |
| itzikb_ | baoli: ping | 15:54 |
| *** dave-mccowan has quit IRC | 15:56 | |
| *** gszasz_ has joined #openstack-nova | 15:57 | |
| *** gokrokve has joined #openstack-nova | 15:57 | |
| *** dsanders has joined #openstack-nova | 15:58 | |
| *** gszasz has quit IRC | 15:59 | |
| *** mpovolny has quit IRC | 16:00 | |
| *** nkrinner has quit IRC | 16:01 | |
| *** marun has quit IRC | 16:01 | |
| *** gszasz_ has quit IRC | 16:01 | |
| *** rushiagr_away is now known as rushiagr | 16:02 | |
| *** apuimedo has quit IRC | 16:03 | |
| garyk1 | sdague: can you please see https://review.openstack.org/#/c/135625/ - the code has an exception and does not have a MS +1 | 16:03 |
| *** achanda has quit IRC | 16:04 | |
| sdague | garyk1: why is MS relevant? | 16:05 |
| sdague | also, is the vmware ci dead? | 16:05 |
| sdague | it hasn't voted on that patch since Jan | 16:06 |
| dansmith | sdague: MS == minesweeper | 16:06 |
| *** marun has joined #openstack-nova | 16:06 | |
| garyk1 | sdague: it has voted and is voting. it it a little deeep in the queue | 16:06 |
| garyk1 | sdague: when i tested this code it broke and threw an exception and did not run an instance. that is my main concern | 16:07 |
| garyk1 | i need to check why it fails MS - i think the same reason too | 16:07 |
| *** ijw has quit IRC | 16:07 | |
| *** Maike has quit IRC | 16:07 | |
| jaypipes | dansmith: what's up? | 16:07 |
| *** rfolco has quit IRC | 16:08 | |
| garyk1 | sdague: this is the last +1 from MS. our cloud has a few problems at the moment | 16:08 |
| *** ZZelle_ has quit IRC | 16:08 | |
| dansmith | jaypipes: my comment here: https://review.openstack.org/#/c/160511/14 | 16:08 |
| *** rfolco has joined #openstack-nova | 16:08 | |
| dansmith | jaypipes: and the backlog above between edleafe and I | 16:08 |
| *** gszasz has joined #openstack-nova | 16:09 | |
| *** dulek has quit IRC | 16:11 | |
| sdague | oh | 16:11 |
| *** tjones1 has joined #openstack-nova | 16:11 | |
| sdague | can we call it vmware ci instead | 16:11 |
| *** openstackgerrit has quit IRC | 16:11 | |
| edleafe | dansmith: re-worked _init_instance_info, to be called with utils.spawn_n: http://paste.openstack.org/show/192666/ | 16:11 |
| sdague | because MS is very confusing in a context where microsoft also votes | 16:11 |
| garyk1 | sdague: i can speak with tracy and ryn and see if we can change/update | 16:12 |
| *** openstackgerrit has joined #openstack-nova | 16:12 | |
| edleafe | dansmith: is that close to what you had in mind? | 16:12 |
| *** iamjarvo has joined #openstack-nova | 16:12 | |
| openstackgerrit | wuhao proposed openstack/nova: Add previous page links for Nova GET servers API https://review.openstack.org/164169 | 16:12 |
| garyk1 | sdague: it is actually called "VMware NSX CI" | 16:12 |
| dansmith | edleafe: well, you're still doing the get_all at the top of that | 16:13 |
| sdague | garyk1: ok, that's where I was getting confused by your comments | 16:13 |
| edleafe | dansmith: doh! forgot to delete that line | 16:13 |
| edleafe | dansmith: pretend it isn't there :) | 16:13 |
| garyk1 | sdague: sorry, next time i will write VMWare CI | 16:13 |
| sdague | also, because it doesn't provide info in the standard format, it doesn't pop up into the CI block at the top | 16:13 |
| tjones1 | @sdague it used to be called minesweeper - it's now VMware NSX CI | 16:14 |
| sdague | that would be good to address | 16:14 |
| dansmith | edleafe: also, I think it's ugly to modify the list inside the loop. why not just use range and step? | 16:14 |
| jaypipes | edleafe: why the need to greenthread.sleep(0) in there? this code should only be run on startup of the scheduler service daemons, and we don't want to accept connections until all the instance infos have been populated... | 16:14 |
| *** marun_ has joined #openstack-nova | 16:14 | |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: libvirt: make fakelibvirt more customizable https://review.openstack.org/155003 | 16:15 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: introduce a NUMAServersTest class https://review.openstack.org/155006 | 16:15 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: enhance functional tests primitives https://review.openstack.org/155004 | 16:15 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova: tests: add a ServersTestBase class https://review.openstack.org/155005 | 16:15 |
| dansmith | jaypipes: because async, that's what I was looking for comments on | 16:15 |
| *** marun has quit IRC | 16:15 | |
| dansmith | jaypipes: I think that at the point it's being run in the existing patch, RPC has already been setup and the service record has been made, right? | 16:15 |
| jaypipes | dansmith: I don't know why we'd want to do asnc there... we want to block there until all the instance infos are loaded, no? | 16:16 |
| *** amotoki has joined #openstack-nova | 16:16 | |
| *** itzikb_ has quit IRC | 16:16 | |
| dansmith | jaypipes: if that takes ten minutes, do you really want the scheduler hamstrung until then? | 16:16 |
| *** aix has quit IRC | 16:16 | |
| dansmith | although, if this is for anti-affinity, maybe that would cause us to make bad decisions...? | 16:16 |
| *** claudiub_ has quit IRC | 16:16 | |
| jaypipes | dansmith: if it's not, then if the scheduler takes requests, it's not going to be able to make a corrrect decision. | 16:16 |
| *** mpavone has quit IRC | 16:17 | |
| dansmith | ugh | 16:17 |
| jaypipes | dansmith: right? | 16:17 |
| edleafe | jaypipes: it won't use this info at all until updates are flowing | 16:17 |
| dansmith | only for affinity-related things, but yeah, I guess | 16:17 |
| edleafe | so it will use the old queries to get instance info until then | 16:17 |
| dansmith | edleafe: oh, right because unless updated=True, it will fall back? | 16:17 |
| edleafe | dansmith: yep | 16:17 |
| jaypipes | edleafe: I see... | 16:17 |
| dansmith | right, okay | 16:17 |
| dansmith | blocking the scheduler for a long time at startup seems, like, really bad | 16:18 |
| jaypipes | seems quite a complicated design, and complicated < simple IMO :) | 16:18 |
| *** browne has quit IRC | 16:18 | |
| jaypipes | edleafe: have you given any thought to how you will test this complicated interaction? | 16:18 |
| dansmith | jaypipes: for a large cloud, this startup could seriously take many minutes, right? | 16:19 |
| jaypipes | dansmith: yes. | 16:19 |
| edleafe | jaypipes: I've only tested on my devstack | 16:19 |
| edleafe | jaypipes: I don't have access to much else | 16:19 |
| dansmith | jaypipes: so scheduler crashes and startup means we slam the database for 10+ minutes and can't schedule anything | 16:19 |
| dansmith | that seems rage-inducing to me | 16:20 |
| *** rmoe has quit IRC | 16:20 | |
| jaypipes | dansmith: which is why the end-goal was to support sharding of the scheduler, instead of having each scheduler do all things for all nodes. | 16:20 |
| *** marun_ has quit IRC | 16:20 | |
| *** amotoki has quit IRC | 16:21 | |
| dansmith | yeah, understand, but making it super bad for another 6*N months seems worse than what we have now, IMHO | 16:21 |
| jaypipes | dansmith: I agree with you about rage-inducing. but at the same time, I find it more rage-inducing to not be able to debug what precisely is going on inside a daemon due to complex design vs. dealing with the rage of waiting for startup to finish. | 16:21 |
| openstackgerrit | Chris Friesen proposed openstack/nova: Validate server group anti-affinity policy https://review.openstack.org/162746 | 16:21 |
| dansmith | it doesn't seem very much more complicated to me, but maybe I'm missing something | 16:22 |
| dansmith | and, it's a single-threaded thing that has to run at large scale, so it can't be too simple :) | 16:22 |
| openstackgerrit | Tobias Engelbert proposed openstack/nova: Allow multiple volume attachments https://review.openstack.org/153033 | 16:22 |
| openstackgerrit | Chris Friesen proposed openstack/nova: Fix race in server group policy validation https://review.openstack.org/164762 | 16:22 |
| jaypipes | dansmith: the whole "which state am I in" thing with the updated=True flag seems complicated to me, but I suppose I can go along with this design if it's well tested. thus my question to edleafe ;) | 16:23 |
| dansmith | well, I was taking updated=True as part of the existing proposal, so making that loop async doesn't really make it (significantly) more complicated (than the current proposal), IMHO | 16:23 |
| jaypipes | ok. | 16:24 |
| *** ijw has joined #openstack-nova | 16:24 | |
| dansmith | I guess I thought you had +2d this in the past, so you were on board with that part of the proposal, | 16:25 |
| dansmith | but sounds like maybe that was a bad assumption | 16:25 |
| dansmith | (because it doesn't look like you had) | 16:25 |
| jaypipes | I +2d an earlier version of this that did everything on startup to get the list of instances. Or perhaps I didn't understand the earlier draft. | 16:25 |
| dansmith | okay | 16:26 |
| edleafe | jaypipes: you might be thinking of the spec | 16:26 |
| jaypipes | dansmith: actually, I only ever -1'd this particular patch. | 16:26 |
| dansmith | yeah | 16:26 |
| dansmith | current patch has updated=True | 16:26 |
| edleafe | jaypipes: I don't see any +2 on this PS | 16:26 |
| jaypipes | dansmith: because I don't like adding yet another conf option for this. | 16:26 |
| dansmith | jaypipes: yeah | 16:26 |
| jaypipes | dansmith: since whether or not to do the instance updates can be determined by looking at whether certain filters are in the pipeline. | 16:27 |
| sdague | so where in the fake virt driver should I actually check allocations and throw an error that we've overrun? spawn? | 16:27 |
| *** TobiasE has quit IRC | 16:27 | |
| *** tjones1 has quit IRC | 16:27 | |
| sdague | I'd like to actually make it be able to run out of resources for testing purposes. | 16:27 |
| edleafe | jaypipes: that's not the only determinant | 16:27 |
| *** garyk1 has quit IRC | 16:27 | |
| jaypipes | sdague: that's not the virt driver :) that's the resource tracker in nova/compute/resource_tracker.py and nova/compute/claim.py | 16:27 |
| edleafe | jaypipes: some deployments may see to much memory usage with this design, and may want to shut it off | 16:27 |
| sdague | jaypipes: well, it's not that either | 16:28 |
| dansmith | personally I don't like looking at the filter config in the compute node, but I seem to be in the minority there :) | 16:28 |
| sdague | as that happy lets me overrun by at least an order of magnitude | 16:28 |
| edleafe | dansmith: I also don't like hard-coding filter names | 16:28 |
| *** garyk has joined #openstack-nova | 16:28 | |
| jaypipes | edleafe, dansmith: meh, ok :) | 16:28 |
| *** iamjarvo has quit IRC | 16:29 | |
| *** marun_ has joined #openstack-nova | 16:29 | |
| jaypipes | sdague: so you are asking how to change the virt driver code to just not have as many resources? | 16:29 |
| *** dsanders has quit IRC | 16:29 | |
| sdague | jaypipes: or is there something else the fake driver needs to do to get it to enforce | 16:29 |
| dansmith | sdague: it needs to report an exhaustible amount of resources ;) | 16:29 |
| sdague | http://paste.openstack.org/show/192669/ | 16:30 |
| jaypipes | sdague: the way to do that would be to simply reduce the total amount of resources the fake virt driver exposes, no? | 16:30 |
| *** dboik_ has joined #openstack-nova | 16:30 | |
| *** nellysmitt has quit IRC | 16:30 | |
| sdague | jaypipes: probably, I'm asking for where exactly does that expose from so I can fix the fake virt to do that correctly | 16:30 |
| jaypipes | sdague: ah, I see now. | 16:30 |
| jaypipes | sdague: one minute, lemme grab a link. | 16:31 |
| *** baoli has quit IRC | 16:31 | |
| sdague | jaypipes: thanks, appreciate | 16:31 |
| kashyap | Ugh, pastebins in source code -- https://github.com/openstack/nova/blob/master/nova/virt/libvirt/utils.py#L197 | 16:31 |
| *** salv-orlando has quit IRC | 16:31 | |
| *** iamjarvo has joined #openstack-nova | 16:32 | |
| kashyap | Now, I can't see the 'rationale' that's pointed to in the pastebin, it's expired. Using pastebins in bugs alone is bad, as code comments. . . | 16:32 |
| sdague | kashyap: you could poke harlowja to figure out the crux of the issue | 16:32 |
| sdague | kashyap: agree | 16:32 |
| bauzas | jaypipes: I'm on a call right now, but I saw something you said which hitted me | 16:32 |
| kashyap | sdague, I was tracking down root cause of a bug and came down to that. | 16:32 |
| *** baoli_ has joined #openstack-nova | 16:32 | |
| *** otter768 has joined #openstack-nova | 16:32 | |
| *** jaypipes has left #openstack-nova | 16:32 | |
| bauzas | (17:20:14) jaypipes: dansmith: which is why the end-goal was to support sharding of the scheduler, instead of having each scheduler do all things for all nodes. | 16:33 |
| *** jaypipes has joined #openstack-nova | 16:33 | |
| bauzas | jaypipes: I just wanted to make sure that you know one paper that I have in mind | 16:33 |
| *** dboik has quit IRC | 16:33 | |
| bauzas | jaypipes: http://research.google.com/pubs/pub41684.html and http://eurosys2013.tudos.org/wp-content/uploads/2013/paper/Schwarzkopf.pdf | 16:33 |
| *** gokrokve has quit IRC | 16:34 | |
| *** marun_ has quit IRC | 16:35 | |
| bauzas | jaypipes: so you'll understand that saying that sharding is the a solution is fine... until a certain point | 16:35 |
| *** gokrokve has joined #openstack-nova | 16:36 | |
| *** tjones1 has joined #openstack-nova | 16:36 | |
| *** Nic has joined #openstack-nova | 16:36 | |
| *** iamjarvo has quit IRC | 16:36 | |
| *** sandywalsh has quit IRC | 16:36 | |
| raorn | greetings | 16:36 |
| raorn | can you, please, enlight me on availability zones usage? | 16:37 |
| *** otter768 has quit IRC | 16:37 | |
| *** krykowski has quit IRC | 16:37 | |
| jaypipes | sdague: :( seems the fake virt driver is specifically designed to not run out of resources.. | 16:37 |
| jaypipes | sdague: and is hard-coded all over the place to prevernt that. | 16:37 |
| sdague | jaypipes: ok, what interface would I need to expose? | 16:37 |
| jaypipes | raorn: this channel is for Nova developer discussions. For usage questions, please use the mailing list or the #openstack IRC channel. Please see /topic. Thanks! | 16:38 |
| jaypipes | sdague: well, https://github.com/openstack/nova/blob/master/nova/virt/fake.py#L389 | 16:38 |
| jaypipes | sdague: just always reports the same things.. | 16:38 |
| jaypipes | sdague: which is this: https://github.com/openstack/nova/blob/master/nova/virt/fake.py#L96-L100 | 16:39 |
| *** rmoe has joined #openstack-nova | 16:39 | |
| sdague | ok, so if I make that be correct, we're good? | 16:39 |
| sdague | or are there any other interfaces needed? | 16:39 |
| jaypipes | sdague: :( I'm not sure. | 16:39 |
| bauzas | raorn: http://docs.openstack.org/openstack-ops/content/scaling.html#segregate_cloud can be a first point of reading | 16:39 |
| sdague | jaypipes: well, that's a starting point | 16:39 |
| sdague | going to get some lunch, then get hacking on this | 16:40 |
| jaypipes | sdague: It woudl seem to me that the resource tracker uses the DB values of the compute_nodes table as its source of truth, not the hypervisor, thus the reason that those links, while hard-coded, don't seem to be considered by the resource tracker. | 16:40 |
| sdague | jaypipes: are there other interfaces you expect from claims here? | 16:40 |
| *** flwang has joined #openstack-nova | 16:41 | |
| flwang | mriedem: ping | 16:41 |
| sdague | jaypipes: so the fact that spawn doesn't persist to the db might be the bigger issue? | 16:41 |
| jaypipes | sdague: well, each virt driver's spawn() method isn't responsible for interacting with the DB.. | 16:42 |
| *** gszasz has quit IRC | 16:42 | |
| sdague | ok | 16:42 |
| raorn | bauzas: thanks! | 16:42 |
| jaypipes | sdague: the compute manager does that, not the virt driver. | 16:42 |
| jaypipes | sdague: still digging here... | 16:42 |
| sdague | ok, so that content should be there | 16:42 |
| sdague | which, actually, it is | 16:42 |
| dansmith | well, spawn will do some things with the database | 16:43 |
| sdague | because I can GET /servers and get everything | 16:43 |
| openstackgerrit | Vladik Romanovsky proposed openstack/nova-specs: libvirt: virtio-net multiqueue https://review.openstack.org/128825 | 16:43 |
| *** sbfox has joined #openstack-nova | 16:43 | |
| *** oro has quit IRC | 16:43 | |
| sdague | dansmith: inside the driver itself, or above it at the compute manager? | 16:43 |
| dansmith | sdague: in a real driver, inside spawn | 16:43 |
| *** oro_ has quit IRC | 16:44 | |
| *** exploreshaifali has joined #openstack-nova | 16:44 | |
| dansmith | sdague: as it updates things like progress, potentially some state, network info, etc | 16:44 |
| dansmith | sdague: none of it is required for claims and quotas though, AFAIK | 16:44 |
| sdague | ok, I'll give making get_resources not completely rediculous a try | 16:45 |
| *** kaufer has quit IRC | 16:45 | |
| dansmith | sdague: well, it's not ridiculous, it's there like that to serve a purpose of course :) | 16:45 |
| jaypipes | sdague: danpb was working on that as well... might want to email him about it. | 16:45 |
| sdague | jaypipes: any idea if there are patches up? | 16:46 |
| openstackgerrit | Josh Gachnang proposed openstack/nova: Adjust resource tracker for new Ironic states https://review.openstack.org/164313 | 16:47 |
| jaypipes | sdague: https://review.openstack.org/#/q/status:merged+project:openstack/nova+branch:master+topic:bp/resource-objects,n,z | 16:47 |
| *** mwagner_lap has quit IRC | 16:47 | |
| jaypipes | sdague: that's all I've got. he would be the person to chat with though :) | 16:47 |
| sdague | jaypipes: ok, well I'm on that code | 16:47 |
| *** neelashah has quit IRC | 16:47 | |
| sdague | https://review.openstack.org/#/q/file:%255Enova/virt/fake.py+status:open,n,z doesn't look promissing | 16:47 |
| sdague | ok, anyway, this is reasonable breadcrumbs, I can go poke | 16:48 |
| *** marun_ has joined #openstack-nova | 16:48 | |
| *** dsanders has joined #openstack-nova | 16:49 | |
| openstackgerrit | Pavel Kholkin proposed openstack/python-novaclient: Enable pep8 H404 rule https://review.openstack.org/164778 | 16:50 |
| *** pixelb has joined #openstack-nova | 16:52 | |
| *** markus_z has quit IRC | 16:52 | |
| *** dsanders has quit IRC | 16:53 | |
| openstackgerrit | Jennifer Mulsow proposed openstack/nova: Edit ServerGroup inval frmt msg to be more verbose https://review.openstack.org/164308 | 16:53 |
| *** tonytan4ever has quit IRC | 16:54 | |
| *** openstackgerrit has quit IRC | 16:54 | |
| *** browne has joined #openstack-nova | 16:54 | |
| *** openstackgerrit has joined #openstack-nova | 16:54 | |
| raorn | I think there's a bug with moving HV's between aggregates and migrating instances... | 16:57 |
| *** kaufer has joined #openstack-nova | 16:57 | |
| *** nellysmitt has joined #openstack-nova | 16:58 | |
| *** mtanino has joined #openstack-nova | 16:59 | |
| *** neelashah has joined #openstack-nova | 17:00 | |
| *** yamahata has quit IRC | 17:00 | |
| *** neelashah1 has joined #openstack-nova | 17:00 | |
| *** annashen has joined #openstack-nova | 17:00 | |
| openstackgerrit | Josh Gachnang proposed openstack/nova: Add support for cleaning in Ironic driver https://review.openstack.org/161474 | 17:00 |
| *** iamjarvo has joined #openstack-nova | 17:01 | |
| *** Marga_ has quit IRC | 17:01 | |
| *** iamjarvo has quit IRC | 17:01 | |
| *** henrique_ has quit IRC | 17:02 | |
| *** htruta has quit IRC | 17:02 | |
| *** iamjarvo has joined #openstack-nova | 17:02 | |
| *** iamjarvo has quit IRC | 17:02 | |
| *** matrohon has quit IRC | 17:02 | |
| *** htruta has joined #openstack-nova | 17:03 | |
| *** iamjarvo has joined #openstack-nova | 17:03 | |
| *** neelashah has quit IRC | 17:04 | |
| *** iamjarvo has quit IRC | 17:04 | |
| *** jlanoux has quit IRC | 17:05 | |
| *** iamjarvo has joined #openstack-nova | 17:05 | |
| *** marun_ has quit IRC | 17:05 | |
| flwang | jaypipes: ping | 17:06 |
| *** dboik_ has quit IRC | 17:06 | |
| *** salv-orlando has joined #openstack-nova | 17:07 | |
| *** baoli_ has quit IRC | 17:08 | |
| jaypipes | flwang: pong | 17:09 |
| *** marun_ has joined #openstack-nova | 17:09 | |
| *** dboik has joined #openstack-nova | 17:09 | |
| flwang | jaypipes: could u pls help review https://review.openstack.org/#/c/164494/ ? | 17:09 |
| flwang | it exists since Icehouse | 17:09 |
| jaypipes | flwang: will do my best. | 17:10 |
| flwang | jaypipes: awesome, thanks a lot | 17:10 |
| jaypipes | np | 17:10 |
| *** baoli has joined #openstack-nova | 17:11 | |
| flwang | it totally breaks the nova backup function | 17:11 |
| *** gokrokve has quit IRC | 17:11 | |
| *** zzzeek has quit IRC | 17:12 | |
| *** vmtrooper has joined #openstack-nova | 17:13 | |
| *** Nic has quit IRC | 17:13 | |
| jaypipes | flwang: I'm not a fan of the nova backup function to begin with... | 17:13 |
| *** garyk has quit IRC | 17:13 | |
| flwang | jaypipes: lol, i agree, it should be handled by a customized script from the user instead of nova :) | 17:14 |
| jaypipes | flwang: ++. regardless, looking at yoru patch and the bug now. | 17:14 |
| flwang | jaypipes: cool, cheers | 17:14 |
| *** garyk has joined #openstack-nova | 17:15 | |
| *** patrickeast has joined #openstack-nova | 17:15 | |
| *** marun_ has quit IRC | 17:15 | |
| flwang | the interesting thing is there is user want to use it, at least in our cloud | 17:15 |
| *** cbader has joined #openstack-nova | 17:16 | |
| *** cbader02 has joined #openstack-nova | 17:16 | |
| *** garyk has quit IRC | 17:16 | |
| *** ijw has quit IRC | 17:16 | |
| *** romainh has joined #openstack-nova | 17:17 | |
| *** vmtrooper has quit IRC | 17:18 | |
| *** erlon has joined #openstack-nova | 17:19 | |
| *** cbader02 has quit IRC | 17:21 | |
| jaypipes | flwang: reviewed. basically, that logic you put into the two REST API controller plugins needs to be moved directly into the nova/compute/api:backup() method. | 17:23 |
| dansmith | jaypipes: said the same thing a minute earlier | 17:23 |
| jaypipes | dansmith: yes, I see that now. | 17:23 |
| jaypipes | dansmith: sorry. | 17:23 |
| dansmith | jaypipes: no, feel free to agree with me anytime | 17:23 |
| dansmith | :) | 17:23 |
| jaypipes | :) | 17:24 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add CellMapping object https://review.openstack.org/161867 | 17:26 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Test fixture for the api database https://review.openstack.org/164794 | 17:26 |
| flwang | thanks guys, i know the code works, but i just really need some comments where to place it :) | 17:26 |
| flwang | however | 17:26 |
| *** tonytan4ever has joined #openstack-nova | 17:26 | |
| flwang | jaypipes: dansmith: see this https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/servers.py#L1059 | 17:27 |
| *** harlowja has joined #openstack-nova | 17:28 | |
| dansmith | we told you where to stick it :) | 17:28 |
| flwang | yep, i know | 17:28 |
| dansmith | and yeah, that could use some refactoring too, would be awesome if you fixed it up too | 17:28 |
| *** dave-mccowan has joined #openstack-nova | 17:28 | |
| flwang | dansmith: ok, make more sense for me, thanks for the clarification | 17:29 |
| *** ijw has joined #openstack-nova | 17:29 | |
| flwang | dansmith: thanks a lot | 17:29 |
| dansmith | np | 17:29 |
| *** Marga_ has joined #openstack-nova | 17:30 | |
| jaypipes | flwang: yes, here is where to add that logic that switches to use snapshot_volume_backed() in case of a volume-backed instance: https://github.com/openstack/nova/blob/master/nova/compute/api.py#L2109 | 17:30 |
| *** ijw has quit IRC | 17:31 | |
| *** ijw has joined #openstack-nova | 17:31 | |
| flwang | jaypipes: yep, i know :) https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/servers.py#L1059 gives me some wrong hint :) | 17:31 |
| *** moshele has quit IRC | 17:32 | |
| *** ijw has quit IRC | 17:33 | |
| *** ijw has joined #openstack-nova | 17:34 | |
| *** VW_ has quit IRC | 17:35 | |
| *** pixelb has quit IRC | 17:36 | |
| *** dprince has quit IRC | 17:36 | |
| *** rushiagr is now known as rushiagr_away | 17:36 | |
| *** Nic has joined #openstack-nova | 17:37 | |
| *** marun has joined #openstack-nova | 17:37 | |
| *** saikrishna has joined #openstack-nova | 17:38 | |
| *** Sukhdev has joined #openstack-nova | 17:38 | |
| *** yamahata has joined #openstack-nova | 17:38 | |
| *** gokrokve has joined #openstack-nova | 17:39 | |
| *** marun_ has joined #openstack-nova | 17:40 | |
| *** dsanders has joined #openstack-nova | 17:41 | |
| *** nelsnelson has quit IRC | 17:42 | |
| *** marun has quit IRC | 17:42 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: Don't fetch kernel/ramdisk files if you already have them. https://review.openstack.org/163661 | 17:43 |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: libvirt: Don't fetch kernel/ramdisk files if you already have them. https://review.openstack.org/163661 | 17:44 |
| *** afazekas has quit IRC | 17:44 | |
| sdague | jaypipes: so, where are host limits actually populated? | 17:48 |
| jaypipes | sdague: in the scheduler's filters :( | 17:49 |
| sdague | well, at some point that data has to come from the compute node, right? | 17:49 |
| sdague | that's what I'm having a hard time back tracking | 17:50 |
| *** achanda has joined #openstack-nova | 17:50 | |
| *** dprince has joined #openstack-nova | 17:50 | |
| jaypipes | sdague: nope. :( the scheduler gets the cpu_allocation_ratio and ram_allocation_ratio for a compute host by checking the *scheduler's* nova.conf option values. :( | 17:50 |
| *** ybathia has joined #openstack-nova | 17:50 | |
| *** jistr has quit IRC | 17:51 | |
| sdague | for ratios, fine | 17:51 |
| jaypipes | sdague: or, if the compute host is in an aggregate, and the aggregate has an override, it uses that. | 17:51 |
| sdague | but at some point you have to actually know how much ram is on a node, right? | 17:51 |
| jaypipes | sdague: that comes from the compute_nodes table, which the scheduler re-reads from the conductor on every scheduler select_destinations() method. | 17:52 |
| openstackgerrit | Pavel Kholkin proposed openstack/python-novaclient: Enable pep8 H405 rule https://review.openstack.org/164803 | 17:52 |
| sdague | ok, so that's the part that I need to make sure actually gets populated | 17:52 |
| sdague | it wasn't being obvious to me | 17:52 |
| jaypipes | sdague: https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L483 | 17:52 |
| *** thomasem has quit IRC | 17:52 | |
| jaypipes | sdague: the list of ALL compute nodes is grabbed from the conductor here: https://github.com/openstack/nova/blob/master/nova/scheduler/host_manager.py#L468 | 17:53 |
| openstackgerrit | Pavel Kholkin proposed openstack/python-novaclient: Enable pep8 H404 rule https://review.openstack.org/164778 | 17:53 |
| jaypipes | sdague: and that get_all_host_states() method is called once at the start of every single call to select_destinations() :( | 17:53 |
| sdague | ok, I apparently blew right pass that get_all when I was jumping through the call stack | 17:54 |
| sdague | jaypipes: thanks, that's helpful | 17:55 |
| jaypipes | sdague: no worries. it's ugly ass code. | 17:55 |
| *** achanda_ has joined #openstack-nova | 17:55 | |
| openstackgerrit | Bart Wensley proposed openstack/nova: Add numa topology to migration object https://review.openstack.org/163440 | 17:56 |
| *** sahid has quit IRC | 17:56 | |
| sdague | yeh, going up through the call stack through about 10 layers where everyone is passing limits got me lost :) | 17:57 |
| *** ajayaa has quit IRC | 17:57 | |
| *** romainh has quit IRC | 17:57 | |
| *** salv-orl_ has joined #openstack-nova | 17:57 | |
| *** adalbas has joined #openstack-nova | 17:58 | |
| openstackgerrit | Chris St. Pierre proposed openstack/nova: Allow blacklisting os-api extensions https://review.openstack.org/160986 | 17:58 |
| *** achanda has quit IRC | 17:58 | |
| *** Nikolay_St has joined #openstack-nova | 17:59 | |
| *** marun_ has quit IRC | 17:59 | |
| *** yamahata has quit IRC | 18:00 | |
| *** yamahata has joined #openstack-nova | 18:00 | |
| *** lucasagomes is now known as lucas-dinner | 18:00 | |
| openstackgerrit | Pavel Kholkin proposed openstack/python-novaclient: Enable pep8 H405 rule https://review.openstack.org/164803 | 18:00 |
| *** baoli has quit IRC | 18:00 | |
| *** mwagner_lap has joined #openstack-nova | 18:01 | |
| *** baoli_ has joined #openstack-nova | 18:01 | |
| *** salv-orlando has quit IRC | 18:01 | |
| *** thomasem has joined #openstack-nova | 18:01 | |
| *** garyk has joined #openstack-nova | 18:02 | |
| *** jdurgin has joined #openstack-nova | 18:02 | |
| garyk | sdague: thanks | 18:02 |
| *** marun_ has joined #openstack-nova | 18:06 | |
| *** nelsnelson has joined #openstack-nova | 18:07 | |
| *** iamjarvo has quit IRC | 18:08 | |
| *** marun has joined #openstack-nova | 18:09 | |
| *** mpavone has joined #openstack-nova | 18:11 | |
| *** marun_ has quit IRC | 18:11 | |
| *** derekh has quit IRC | 18:12 | |
| *** marun_ has joined #openstack-nova | 18:13 | |
| *** marun has quit IRC | 18:13 | |
| *** oro_ has joined #openstack-nova | 18:14 | |
| *** welldannit has joined #openstack-nova | 18:16 | |
| *** thomasem has quit IRC | 18:16 | |
| *** IanGovett has quit IRC | 18:16 | |
| *** oro has joined #openstack-nova | 18:17 | |
| *** rushiagr_away is now known as rushiagr | 18:17 | |
| *** IanGovett has joined #openstack-nova | 18:17 | |
| *** IanGovett has quit IRC | 18:18 | |
| *** iamjarvo has joined #openstack-nova | 18:18 | |
| *** thomasem has joined #openstack-nova | 18:18 | |
| *** paul-- has quit IRC | 18:20 | |
| *** marun_ has quit IRC | 18:20 | |
| openstackgerrit | David J Hu proposed openstack/nova-specs: Nova admin role https://review.openstack.org/164813 | 18:20 |
| *** marun_ has joined #openstack-nova | 18:21 | |
| *** packet has joined #openstack-nova | 18:21 | |
| *** melwitt has joined #openstack-nova | 18:22 | |
| *** saikrishna has quit IRC | 18:25 | |
| *** jecarey has quit IRC | 18:25 | |
| *** tjones1 has quit IRC | 18:26 | |
| *** flwang has quit IRC | 18:26 | |
| *** tjones1 has joined #openstack-nova | 18:26 | |
| *** bwensley has joined #openstack-nova | 18:27 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: use instance UUID as instance name https://review.openstack.org/160796 | 18:28 |
| openstackgerrit | garyk proposed openstack/nova: VMware: update test_vm_util to use instance object https://review.openstack.org/160273 | 18:28 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 18:29 | |
| *** paul-- has joined #openstack-nova | 18:29 | |
| jogo | looking for some reviews for https://review.openstack.org/#/c/122347 | 18:30 |
| jogo | fixes a serious quotas bug | 18:30 |
| jogo | and its a non trivial fix | 18:31 |
| *** melwitt_ has joined #openstack-nova | 18:31 | |
| *** IanGovett has joined #openstack-nova | 18:32 | |
| *** otter768 has joined #openstack-nova | 18:33 | |
| *** Maike has joined #openstack-nova | 18:34 | |
| dansmith | jogo: man, I don't even understand comstud's comment there | 18:34 |
| dansmith | if the instance is in deleting, we need to create some reservations? | 18:34 |
| jogo | dansmith: quota reservation yes | 18:34 |
| jogo | dansmith: a negative reservation | 18:34 |
| jogo | to decrement the quotas | 18:35 |
| *** melwitt has quit IRC | 18:35 | |
| *** melwitt_ is now known as melwitt | 18:35 | |
| *** dprince has quit IRC | 18:35 | |
| *** thomasem has quit IRC | 18:36 | |
| *** thomasem has joined #openstack-nova | 18:36 | |
| openstackgerrit | garyk proposed openstack/nova: Virt: update fake driver to use UUID as lookup key https://review.openstack.org/160164 | 18:37 |
| *** otter768 has quit IRC | 18:38 | |
| dansmith | jogo: okay, I thought delete would do the decrement | 18:38 |
| *** welldannit has quit IRC | 18:38 | |
| jogo | dansmith: the compute node does taht | 18:39 |
| openstackgerrit | garyk proposed openstack/nova: VMware: save instance object creation in test_vmops https://review.openstack.org/162456 | 18:39 |
| jogo | dansmith: n-api creates the negative quota, and the quotas.commit() happens on the compute node | 18:40 |
| openstackgerrit | garyk proposed openstack/nova: VMware: fill in instance metadata when launching instances https://review.openstack.org/141028 | 18:40 |
| jogo | but if the compute node goes down before it can finish deleting the quota is lost | 18:41 |
| dansmith | okay | 18:41 |
| dansmith | I'm pretty quotas-ignorant | 18:41 |
| *** thomasem has quit IRC | 18:41 | |
| *** emagana has joined #openstack-nova | 18:41 | |
| jogo | we can't decrement the quota until it deletes successfully | 18:42 |
| openstackgerrit | garyk proposed openstack/nova: VMware: fill in instance metadata when resizing instances https://review.openstack.org/144436 | 18:42 |
| jogo | since error state still counts as a quota usage | 18:42 |
| dansmith | sure | 18:42 |
| bwensley | Is there a way to re-open a bug that was set to Fix Released, if the fix didn't fix the problem described in the bug? | 18:43 |
| *** annashen has quit IRC | 18:44 | |
| *** dprince has joined #openstack-nova | 18:44 | |
| jogo | bwensley: yes | 18:45 |
| *** annashen has joined #openstack-nova | 18:45 | |
| bwensley | How would I do that? I don't seem to have the ability to change the status for the bug (1267862). | 18:46 |
| *** moshele has joined #openstack-nova | 18:47 | |
| *** jecarey has joined #openstack-nova | 18:47 | |
| jogo | bwensley: join the nova bug triage team | 18:47 |
| jogo | its an open group | 18:47 |
| jogo | https://launchpad.net/~nova-bugs | 18:47 |
| *** mikeit has joined #openstack-nova | 18:48 | |
| *** dsanders1 has joined #openstack-nova | 18:51 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Implement instance update logic in Scheduler https://review.openstack.org/160511 | 18:51 |
| *** dsanders has quit IRC | 18:51 | |
| edleafe | dansmith: jaypipes: ^^ | 18:51 |
| *** thomasem has joined #openstack-nova | 18:52 | |
| *** VW_ has joined #openstack-nova | 18:52 | |
| *** eglynn is now known as eglynn-officeafk | 18:53 | |
| *** mwagner_lap has quit IRC | 18:53 | |
| *** oro_ has quit IRC | 18:54 | |
| *** annashen has quit IRC | 18:55 | |
| *** ijw_ has joined #openstack-nova | 18:56 | |
| jogo | dansmith: the quotas log is all kinds of crazy | 18:56 |
| *** eglynn-office has quit IRC | 18:56 | |
| dansmith | jogo: s/ log// | 18:56 |
| jogo | dansmith: thanks | 18:57 |
| *** Marga_ has quit IRC | 18:57 | |
| dansmith | np :) | 18:57 |
| *** afazekas has joined #openstack-nova | 18:57 | |
| *** Marga_ has joined #openstack-nova | 18:57 | |
| *** baoli_ has quit IRC | 18:58 | |
| *** annashen has joined #openstack-nova | 18:58 | |
| *** ijw has quit IRC | 18:59 | |
| *** welldannit has joined #openstack-nova | 18:59 | |
| *** Nikolay_St has quit IRC | 18:59 | |
| *** baoli has joined #openstack-nova | 19:00 | |
| *** Nikolay_St has joined #openstack-nova | 19:00 | |
| *** mikeit has quit IRC | 19:01 | |
| *** vmtrooper has joined #openstack-nova | 19:02 | |
| *** marun_ has quit IRC | 19:02 | |
| *** mpavone has quit IRC | 19:02 | |
| *** matrohon has joined #openstack-nova | 19:03 | |
| *** Marga_ has quit IRC | 19:03 | |
| *** Marga_ has joined #openstack-nova | 19:03 | |
| lifeless | gmann: pong | 19:03 |
| *** baoli has quit IRC | 19:04 | |
| *** vmtrooper has quit IRC | 19:06 | |
| *** lpetrut has quit IRC | 19:08 | |
| *** iamjarvo has quit IRC | 19:08 | |
| *** jecarey has quit IRC | 19:08 | |
| *** eglynn-office has joined #openstack-nova | 19:10 | |
| *** annashen has quit IRC | 19:11 | |
| *** oro_ has joined #openstack-nova | 19:11 | |
| *** annashen has joined #openstack-nova | 19:11 | |
| *** dboik has quit IRC | 19:12 | |
| *** unicell has joined #openstack-nova | 19:12 | |
| *** dboik has joined #openstack-nova | 19:13 | |
| *** afazekas has quit IRC | 19:13 | |
| *** neelashah1 has quit IRC | 19:13 | |
| dansmith | edleafe: teensynit | 19:14 |
| dansmith | edleafe: and some calls for logging I think, to help address jaypipes' complexity concern(s) | 19:14 |
| dansmith | edleafe: but I think it's much better than what you had, IMHO | 19:14 |
| edleafe | dansmith: ok, thanks | 19:14 |
| edleafe | I'll add some logging to that method | 19:15 |
| *** jecarey has joined #openstack-nova | 19:18 | |
| *** neelashah has joined #openstack-nova | 19:18 | |
| ybathia | what settings are required in nova.conf with recent devstack to get notifications on rabbitmq? | 19:20 |
| *** afazekas has joined #openstack-nova | 19:21 | |
| *** Marga_ has quit IRC | 19:21 | |
| *** Marga_ has joined #openstack-nova | 19:22 | |
| *** eglynn-office has quit IRC | 19:24 | |
| *** annashen has quit IRC | 19:25 | |
| *** Marga_ has quit IRC | 19:25 | |
| *** Marga_ has joined #openstack-nova | 19:26 | |
| *** samueldmq has joined #openstack-nova | 19:26 | |
| *** annashen has joined #openstack-nova | 19:28 | |
| *** afazekas has quit IRC | 19:30 | |
| *** iamjarvo has joined #openstack-nova | 19:30 | |
| *** annashen has quit IRC | 19:31 | |
| *** annashen has joined #openstack-nova | 19:32 | |
| *** annashen has quit IRC | 19:37 | |
| *** annashen has joined #openstack-nova | 19:37 | |
| *** packet has quit IRC | 19:37 | |
| *** Sukhdev has quit IRC | 19:38 | |
| *** mriedem has quit IRC | 19:39 | |
| *** rushiagr is now known as rushiagr_away | 19:42 | |
| *** packet has joined #openstack-nova | 19:43 | |
| *** tjones1 has quit IRC | 19:43 | |
| *** eglynn-office has joined #openstack-nova | 19:45 | |
| sdague | lifeless: I expect gmann is asleep at this point | 19:46 |
| openstackgerrit | melanie witt proposed openstack/python-novaclient: Ensure the use of volume endpoint in volumes apis https://review.openstack.org/164321 | 19:46 |
| *** aysyd has quit IRC | 19:46 | |
| lifeless | sdague: ah well :) | 19:46 |
| *** mriedem has joined #openstack-nova | 19:47 | |
| *** moshele has quit IRC | 19:47 | |
| *** moshele has joined #openstack-nova | 19:47 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add InstanceMapping object https://review.openstack.org/161904 | 19:49 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Objects: use setattr rather than dict syntax in remotable https://review.openstack.org/163974 | 19:49 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add support for multiple database engines https://review.openstack.org/159266 | 19:49 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add get_api_session to db api https://review.openstack.org/161866 | 19:49 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Test fixture for the api database https://review.openstack.org/164794 | 19:49 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Use the proper database engine for nova-manage https://review.openstack.org/161833 | 19:49 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Split NovaPersistentObject mixin https://review.openstack.org/163605 | 19:49 |
| alaski | hmm, that shouldn't have happened | 19:49 |
| *** aysyd has joined #openstack-nova | 19:50 | |
| *** alexpilotti has quit IRC | 19:50 | |
| *** sdake has quit IRC | 19:50 | |
| alaski | oh, I guess I haven't encountered that rebase behavior in git-review before | 19:51 |
| dansmith | alaski: you can put the obj_as_admin on the base class if you want.. I made it separate so it didn't get swept up in the export, but that's done now | 19:51 |
| dansmith | alaski: so that will trim that patch down a little further | 19:51 |
| *** sdake has joined #openstack-nova | 19:51 | |
| alaski | dansmith: ok | 19:51 |
| *** xyang1 has joined #openstack-nova | 19:53 | |
| *** amotoki has joined #openstack-nova | 19:53 | |
| *** nellysmitt has quit IRC | 19:54 | |
| *** salv-orl_ has quit IRC | 19:55 | |
| *** afazekas has joined #openstack-nova | 19:57 | |
| *** amotoki has quit IRC | 19:58 | |
| *** r-daneel has joined #openstack-nova | 19:58 | |
| *** thomasem has quit IRC | 20:02 | |
| jaypipes | edleafe: https://review.openstack.org/#/c/160511/ reviewed. | 20:02 |
| edleafe | jaypipes: thanks - digging into it now | 20:03 |
| *** flwang has joined #openstack-nova | 20:03 | |
| *** rushiagr_away is now known as rushiagr | 20:07 | |
| *** thomasem has joined #openstack-nova | 20:10 | |
| *** thomasem has quit IRC | 20:11 | |
| *** pixelb has joined #openstack-nova | 20:12 | |
| *** romainh has joined #openstack-nova | 20:12 | |
| *** tjones1 has joined #openstack-nova | 20:14 | |
| *** afazekas has quit IRC | 20:15 | |
| *** thomasem has joined #openstack-nova | 20:15 | |
| *** ndipanov has quit IRC | 20:16 | |
| *** marun_ has joined #openstack-nova | 20:16 | |
| *** afazekas has joined #openstack-nova | 20:16 | |
| edleafe | jaypipes: I thought it mock.patch on the setUp() method was considered bad form | 20:17 |
| edleafe | s/thought it/thought | 20:17 |
| jaypipes | edleafe: I wasn't aware of that... | 20:17 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Objects: use setattr rather than dict syntax in remotable https://review.openstack.org/163974 | 20:18 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add CellMapping object https://review.openstack.org/161867 | 20:18 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Test fixture for the api database https://review.openstack.org/164794 | 20:18 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Split out NovaTimestampObject https://review.openstack.org/163605 | 20:18 |
| edleafe | jaypipes: bauzas had told me that. That's why he put the aggs mock inside the method | 20:18 |
| jaypipes | edleafe: well, I don't agree with bauzas :) | 20:18 |
| edleafe | jaypipes: I'll bet that's a first! :) | 20:18 |
| jaypipes | ha! | 20:19 |
| edleafe | jaypipes: ok, I'll move both our mocks to be decorators | 20:19 |
| *** alexpilotti has joined #openstack-nova | 20:19 | |
| dansmith | I don't like mocks in setUp() either, FWIW :) | 20:19 |
| sdague | garyk / tjones1 - so what's the vmware ci voting story - https://review.openstack.org/#/c/135625 still seems to not have a vote | 20:20 |
| edleafe | dansmith: as decorators, or anywhere? | 20:20 |
| dansmith | edleafe: I don't like setting them in setUp() and expecting to clean them up later | 20:20 |
| edleafe | dansmith: the decorator and context manager forms clean themselves up | 20:21 |
| dansmith | edleafe: right | 20:21 |
| dansmith | edleafe: is the question about decorating setUp()? | 20:22 |
| *** moshele has quit IRC | 20:22 | |
| edleafe | dansmith: yes | 20:22 |
| edleafe | dansmith: vs. a nested context manager in the setUp code | 20:22 |
| dansmith | edleafe: yeah, that's fine with me, although it can lead to some hard-to-pin-down behavior | 20:22 |
| dansmith | edleafe: agreed +1 for decorator in that case | 20:22 |
| *** dprince has quit IRC | 20:22 | |
| edleafe | dansmith: ok, decorator it is... | 20:23 |
| *** marun_ is now known as marun | 20:23 | |
| mriedem | sdague: vmware ci has been gone since at least last week | 20:23 |
| *** afazekas has quit IRC | 20:23 | |
| tjones1 | @sdague - triggering a vmware ci run on it | 20:24 |
| mriedem | tjones1: there are others | 20:24 |
| sdague | tjones1: ok, why isn't it processing the queue? | 20:24 |
| tjones1 | mriedem: its up now | 20:24 |
| tjones1 | hold on | 20:24 |
| tjones1 | has a bunch in the queue | 20:24 |
| *** lpetrut has joined #openstack-nova | 20:25 | |
| *** gokrokve has quit IRC | 20:27 | |
| *** gokrokve has joined #openstack-nova | 20:27 | |
| *** rushiagr is now known as rushiagr_away | 20:27 | |
| openstackgerrit | Romain Hardouin proposed openstack/nova: heal_instance_info_cache_interval help clearer https://review.openstack.org/163897 | 20:28 |
| sdague | tjones1: also, where are you folks tracking bugs on it. Because it's not using standard format hideci doesn't bring up the results | 20:29 |
| tjones1 | sdague: our ci owner is out sick today so i can't ask him. where should we be tracking them ? infra? | 20:30 |
| tjones1 | if you write a bug i will assign it to him and ping him when he gets back | 20:30 |
| sdague | I have no idea, I was mostly asking where it's todo list was managed | 20:31 |
| *** igordcard_ has joined #openstack-nova | 20:32 | |
| tjones1 | if you want to drop me an email telling me what's wrong i can write a bug and assign it to him. or will what you have written above be clear enough for a ci maintainer to know what you mean :-) | 20:32 |
| *** otter768 has joined #openstack-nova | 20:34 | |
| *** salv-orlando has joined #openstack-nova | 20:35 | |
| *** afazekas has joined #openstack-nova | 20:38 | |
| sdague | tjones1: I don't know, how about we pretend it is, and if not ask him to get on irc once he's feeling better | 20:38 |
| tjones1 | ok i just sent it to him | 20:38 |
| *** dkliban is now known as dkliban_brb | 20:39 | |
| *** otter768 has quit IRC | 20:39 | |
| *** dboik has quit IRC | 20:40 | |
| *** annashen has quit IRC | 20:40 | |
| *** dboik has joined #openstack-nova | 20:41 | |
| mriedem | tjones1: sdague: docs http://ci.openstack.org/third_party.html#requirements point to https://wiki.openstack.org/wiki/ThirdPartySystems which points to https://wiki.openstack.org/wiki/ThirdPartySystems/VMware_CI which doesn't point to https://wiki.openstack.org/wiki/NovaVMware/Minesweeper | 20:41 |
| mriedem | the old wiki, https://wiki.openstack.org/wiki/NovaVMware/Minesweeper, had a section for current status/issues https://wiki.openstack.org/wiki/NovaVMware/Minesweeper/Status | 20:42 |
| mriedem | would be nice if the latter were linked from the former, more centralized, wiki for all third party ci | 20:42 |
| *** bwensley has quit IRC | 20:43 | |
| mriedem | i could summon she who knows the rules | 20:43 |
| *** thedodd has quit IRC | 20:44 | |
| *** jcru has quit IRC | 20:44 | |
| tjones1 | mriedem: i'll update it | 20:44 |
| *** jcru has joined #openstack-nova | 20:46 | |
| sdague | jaypipes: so... this is curious. How often should get_available_resource be called? | 20:47 |
| *** afazekas has quit IRC | 20:47 | |
| mriedem | all the time | 20:47 |
| mriedem | or every minute | 20:48 |
| jaypipes | sdague: it is called every periodic interval (sync_compute_nodes() in the RT) | 20:48 |
| jaypipes | sdague: as mriedem says, I think the default is 60 seconds. | 20:48 |
| mriedem | isn't it also done on every claim? | 20:48 |
| jaypipes | no, I don't think so, but lemme check. | 20:48 |
| *** wverdugo500 has joined #openstack-nova | 20:49 | |
| sdague | yeh, it only seems to be called on startup here | 20:49 |
| mriedem | we hit a lock timeout with that when the periodic task was running and an api operation was starved waiting for the lock | 20:49 |
| *** wverdugo500 has left #openstack-nova | 20:49 | |
| mriedem | b/c lvm hung for 2 minutes | 20:49 |
| *** afazekas has joined #openstack-nova | 20:49 | |
| sdague | so that's part of my overrunning the resource tracker, is it never polls for new data | 20:49 |
| *** packet has quit IRC | 20:50 | |
| *** lucas-dinner has quit IRC | 20:50 | |
| *** vmtrooper has joined #openstack-nova | 20:50 | |
| *** andreykurilin_ has joined #openstack-nova | 20:50 | |
| *** annashen has joined #openstack-nova | 20:51 | |
| jaypipes | sdague: it's called on nova-compute startup, and it's called after live migration is done | 20:54 |
| bauzas | jaypipes: evening | 20:55 |
| *** thedodd has joined #openstack-nova | 20:55 | |
| *** vmtrooper has quit IRC | 20:55 | |
| bauzas | sdague: nope, get_avail_resources() is only done by each 60sec and clean the claims | 20:56 |
| *** jaypipes has quit IRC | 20:56 | |
| bauzas | jaypipes: edleafe: so just to be clear, I dislike mocking methods in setUp() | 20:56 |
| bauzas | ergh, jay left | 20:57 |
| *** jaypipes has joined #openstack-nova | 20:57 | |
| bauzas | jaypipes: edleafe: so as I said, I dislike mocking in setUp() | 20:57 |
| sdague | bauzas: ok, so that means we've got an overrun race right? And we're just relying on compute drivers to barf if they can't schedule | 20:57 |
| jaypipes | bauzas: why? | 20:57 |
| bauzas | jaypipes: because it's not clear that methods are mocked for each test | 20:58 |
| bauzas | jaypipes: so, as I said, I did a context manager mocking only for initiating HostManager because that's ok just for it | 20:58 |
| jaypipes | bauzas: and it's clearer to have with mock.patch() in the setUp()? :( | 20:58 |
| *** harlowja is now known as harlowja_away | 20:58 | |
| bauzas | jaypipes: only for a ctx manager | 20:59 |
| sdague | bauzas: given our inheritance depth of all these tests, it seems like a silly thing to complain about. The code is in the same file at least | 20:59 |
| jaypipes | bauzas: why do you think that applying a mock.patch decorator around setUp() would make the mock last for longer than just the setUp() method? | 20:59 |
| bauzas | sdague: I don't have all the convo, I was just replying on your question about this method | 21:00 |
| bauzas | :) | 21:00 |
| jaypipes | bauzas: I think you may be confusing using the decorators on the class itself instead of the setUp() method? | 21:00 |
| dansmith | bauzas: are you thinking of decorating the class? | 21:00 |
| jaypipes | dansmith: jinx. | 21:00 |
| dansmith | jaypipes: I owe you a coke or something now, huh? | 21:00 |
| *** packet has joined #openstack-nova | 21:00 | |
| bauzas | jaypipes: ok, let's be clear, I'm only +1 if within a setUp() method, someone is doing : | 21:00 |
| dansmith | based on standard high school jinx rules of course | 21:00 |
| bauzas | with mock.patch(): do_sth() | 21:00 |
| jaypipes | bauzas: why? | 21:01 |
| *** Sukhdev has joined #openstack-nova | 21:01 | |
| bauzas | jaypipes: because it's clear that the fake is just done for an item | 21:01 |
| bauzas | jaypipes: so it's creating a Fake object | 21:01 |
| bauzas | jaypipes: using this Fake object is fine by me | 21:02 |
| bauzas | jaypipes: but when writing | 21:02 |
| bauzas | def setUp(): a = mock.patch(); a.start() | 21:03 |
| *** ijw_ has quit IRC | 21:03 | |
| bauzas | it then creates a mocked method for each test | 21:03 |
| bauzas | which is not clear | 21:03 |
| edleafe | bauzas: but we're talking about using the decorator mock for setUp | 21:03 |
| bauzas | edleafe: that's even worst IMHO | 21:03 |
| edleafe | bauzas: why? | 21:04 |
| dansmith | disagree :) | 21:04 |
| dansmith | using a decorator or a context manager means it's gone by the time setUp() is complete and before the test is run | 21:04 |
| *** ijw has joined #openstack-nova | 21:04 | |
| dansmith | a = mock.patch(); a.start() should almost never be used, IMHO | 21:04 |
| jaypipes | bauzas: what dansmith said. | 21:04 |
| bauzas | dansmith: ah sorry, I readed too fast | 21:05 |
| bauzas | dansmith: I'm -1 on mocking against a class | 21:05 |
| bauzas | dansmith: but I'm fine by using a decorator for setUp because that's the same as for a context manager | 21:06 |
| dansmith | mocking a class or decorating one? | 21:06 |
| *** oro_ has quit IRC | 21:06 | |
| dansmith | okay | 21:06 |
| dansmith | I think I'm confused about what we're arguing about then :) | 21:06 |
| *** ijw has quit IRC | 21:06 | |
| *** ijw has joined #openstack-nova | 21:06 | |
| bauzas | dansmith: yeah my point was to say : *only mock something if the mock is cleaned by end of setUp() | 21:06 |
| *** htruta has quit IRC | 21:06 | |
| *** iamjarvo has quit IRC | 21:06 | |
| *** oro has quit IRC | 21:07 | |
| dansmith | which a decorator on setUp() will do | 21:07 |
| dansmith | sounds like maybe there was confusion over "decorating setup" and "decorating the class" | 21:07 |
| bauzas | dansmith: erm, chatty typpy convo ? | 21:07 |
| bauzas | okay guys, I wanted to shoot something I saw too many times, that's it | 21:07 |
| *** jgrimm is now known as zz_jgrimm | 21:07 | |
| bauzas | which was basically : *don't mock methods in setUp that will still be mocked once the test begins* | 21:08 |
| bauzas | the rest is matter of jumping into a convo without getting the whole context, my bad | 21:08 |
| *** amotoki has joined #openstack-nova | 21:09 | |
| *** andreykurilin_ has quit IRC | 21:09 | |
| edleafe | bauzas: the context was that jaypipes dinged me in a review for nesting context manager mocks in setUp | 21:10 |
| edleafe | bauzas: the aggs and instances stuff | 21:10 |
| jaypipes | DING! | 21:10 |
| edleafe | bauzas: so I'm moving them both to decorators | 21:10 |
| bauzas | edleafe: oic | 21:10 |
| bauzas | edleafe: well, I'm usually fine by using with ctx_mgr() as a, ctx_mgr2() as b: | 21:11 |
| *** neelashah has quit IRC | 21:11 | |
| bauzas | edleafe: but using decorators is fine too | 21:11 |
| bauzas | edleafe: my point was that you were nesting context managers in a way which was okay, but not great | 21:11 |
| edleafe | bauzas: understood | 21:12 |
| edleafe | bauzas: but earlier you had told me not to decorate setUp, but to use context manager inside the method | 21:12 |
| bauzas | edleafe: I also preferred to use a context manager in my own series, because it was clear that the mock was only on instanciating HostManager | 21:12 |
| *** annashen has quit IRC | 21:12 | |
| *** tellesnobrega has quit IRC | 21:13 | |
| bauzas | edleafe: decorating is good, but could be not so clear that it would be for HostManager | 21:13 |
| bauzas | edleafe: hence the ctx manager | 21:13 |
| *** thangp has quit IRC | 21:16 | |
| *** afazekas has quit IRC | 21:16 | |
| *** browne has quit IRC | 21:16 | |
| *** tellesnobrega has joined #openstack-nova | 21:18 | |
| *** thomasem has quit IRC | 21:19 | |
| *** iamjarvo has joined #openstack-nova | 21:20 | |
| *** iamjarvo has quit IRC | 21:21 | |
| *** iamjarvo has joined #openstack-nova | 21:21 | |
| openstackgerrit | Chris Friesen proposed openstack/nova: Validate server group anti-affinity policy https://review.openstack.org/162746 | 21:22 |
| *** browne has joined #openstack-nova | 21:22 | |
| openstackgerrit | Chris Friesen proposed openstack/nova: Fix race in server group policy validation https://review.openstack.org/164762 | 21:23 |
| *** lpetrut has quit IRC | 21:23 | |
| *** browne has quit IRC | 21:23 | |
| *** afazekas has joined #openstack-nova | 21:23 | |
| *** asettle has joined #openstack-nova | 21:24 | |
| *** packet has quit IRC | 21:27 | |
| *** oro has joined #openstack-nova | 21:27 | |
| *** dkliban_brb is now known as dkliban | 21:28 | |
| *** browne has joined #openstack-nova | 21:28 | |
| *** oro_ has joined #openstack-nova | 21:28 | |
| *** afazekas has quit IRC | 21:29 | |
| *** annashen has joined #openstack-nova | 21:30 | |
| edleafe | jaypipes: got a second for your comments on test_ironic_host_manager and the self.context changes? | 21:31 |
| *** hemna has quit IRC | 21:32 | |
| *** mriedem is now known as mriedem_away | 21:33 | |
| *** jcru has quit IRC | 21:33 | |
| jaypipes | edleafe: for you? of course :) | 21:33 |
| edleafe | jaypipes: aw, gee... :) | 21:34 |
| *** iamjarvo has quit IRC | 21:35 | |
| edleafe | jaypipes: for some of the tests I needed to add self.context (an actual context object, not a string like many of the tests used). | 21:35 |
| edleafe | jaypipes: I replaced the rest of the usages to make all the tests consistent. | 21:35 |
| edleafe | jaypipes: would you rather I leave some in one state, and the others as they were | 21:35 |
| edleafe | jaypipes: or make them all work the same? | 21:36 |
| jaypipes | edleafe: one minute... | 21:36 |
| *** jamie_h has quit IRC | 21:37 | |
| *** aysyd has quit IRC | 21:38 | |
| *** aysyd has joined #openstack-nova | 21:40 | |
| *** irenab has quit IRC | 21:41 | |
| *** sseverson has joined #openstack-nova | 21:42 | |
| *** ijw_ has joined #openstack-nova | 21:42 | |
| *** hemna has joined #openstack-nova | 21:43 | |
| *** browne_ has joined #openstack-nova | 21:44 | |
| *** browne_ has quit IRC | 21:44 | |
| *** browne1 has joined #openstack-nova | 21:44 | |
| edleafe | jaypipes: never mind - I figure out a work-around | 21:44 |
| *** browne1 has quit IRC | 21:44 | |
| edleafe | jaypipes: I'll leave the ugly code untouched :) | 21:45 |
| *** ijw has quit IRC | 21:45 | |
| *** eglynn-office has quit IRC | 21:45 | |
| jaypipes | edleafe: ok :) | 21:46 |
| *** hemna has quit IRC | 21:47 | |
| *** browne_ has joined #openstack-nova | 21:47 | |
| *** browne has quit IRC | 21:47 | |
| *** zzzeek has joined #openstack-nova | 21:50 | |
| *** hemna has joined #openstack-nova | 21:52 | |
| dims | quick poll for nova cores - is the quick hit still useful? or has it outlived its utiliity? or too stale every day? https://etherpad.openstack.org/p/kilo-nova-priorities-tracking | 21:53 |
| *** matrohon has quit IRC | 21:54 | |
| dansmith | dims: last I looked, we were scraping the barrel for "trivial things" | 21:55 |
| dansmith | I think we cleared out a lot of baggage, | 21:55 |
| dansmith | and then the bar for trivial started moving | 21:55 |
| dansmith | I go through there expecting to spend ten minutes on all of them, the first few requiring actual thought, and close them | 21:55 |
| dansmith | aside from trivialness, there are also things hitting the list that aren't bugs | 21:56 |
| dansmith | maybe trivial, but not really bugs, which is annoying | 21:56 |
| dims | dansmith: thanks for the feedback :) will try to line up really easy ones at the top | 21:57 |
| dims | at the very least :) | 21:57 |
| *** adalbas has quit IRC | 21:59 | |
| *** aysyd has quit IRC | 21:59 | |
| bauzas | dims: eh I will get time by this week for helping this out \o/ | 22:00 |
| *** aysyd has joined #openstack-nova | 22:03 | |
| *** irenab has joined #openstack-nova | 22:04 | |
| *** burt has quit IRC | 22:04 | |
| *** Maike has quit IRC | 22:05 | |
| *** pixelb has quit IRC | 22:09 | |
| *** iamjarvo has joined #openstack-nova | 22:10 | |
| *** aysyd has quit IRC | 22:11 | |
| *** harlowja_away is now known as harlowja | 22:11 | |
| jamielennox | hey all - i'm still looking for a last +2 on https://review.openstack.org/#/c/136931/ changing nova->neutron auth process | 22:12 |
| *** welldannit has quit IRC | 22:13 | |
| *** ijw_ has quit IRC | 22:13 | |
| *** annashen has quit IRC | 22:14 | |
| *** annashen has joined #openstack-nova | 22:15 | |
| *** browne_ has quit IRC | 22:16 | |
| *** browne has joined #openstack-nova | 22:18 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Implement instance update logic in Scheduler https://review.openstack.org/160511 | 22:20 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add the instance update calls from Compute https://review.openstack.org/160513 | 22:20 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Modify filters to get instance info from HostState https://review.openstack.org/160512 | 22:20 |
| openstackgerrit | Ed Leafe proposed openstack/nova: Add the RPC calls for instance updates. https://review.openstack.org/160507 | 22:20 |
| edleafe | dansmith: jaypipes: have at it!! ^^ | 22:20 |
| *** dave-mccowan has quit IRC | 22:24 | |
| *** dboik_ has joined #openstack-nova | 22:30 | |
| *** dboik has quit IRC | 22:33 | |
| *** nelsnelson has quit IRC | 22:34 | |
| *** dboik_ has quit IRC | 22:34 | |
| *** ijw has joined #openstack-nova | 22:34 | |
| *** otter768 has joined #openstack-nova | 22:35 | |
| *** romainh has left #openstack-nova | 22:36 | |
| *** VW_ has quit IRC | 22:37 | |
| *** salv-orl_ has joined #openstack-nova | 22:37 | |
| *** nelsnelson has joined #openstack-nova | 22:38 | |
| *** salv-orlando has quit IRC | 22:39 | |
| *** vmtrooper has joined #openstack-nova | 22:39 | |
| *** otter768 has quit IRC | 22:40 | |
| *** dims has quit IRC | 22:43 | |
| *** nelsnels_ has joined #openstack-nova | 22:44 | |
| *** vmtrooper has quit IRC | 22:45 | |
| *** nelsnelson has quit IRC | 22:45 | |
| *** dims has joined #openstack-nova | 22:46 | |
| JoshNang | hey, i'd be very grateful if anyone had time to review two fairly small ironic driver patches. one fixes a fairly big bug in the resource tracker for ironic with our new state machine (https://review.openstack.org/#/c/164313/) which we discovered while reviewing code for another spec/patch (https://review.openstack.org/#/c/161474/) | 22:47 |
| *** dims has quit IRC | 22:47 | |
| *** dims has joined #openstack-nova | 22:47 | |
| *** dave-mccowan has joined #openstack-nova | 22:50 | |
| *** tsekiyam_ has joined #openstack-nova | 22:52 | |
| *** sdake has quit IRC | 22:55 | |
| *** tsekiyama has quit IRC | 22:55 | |
| *** kaufer has quit IRC | 22:56 | |
| *** tsekiyam_ has quit IRC | 22:57 | |
| openstackgerrit | Praveen Yalagandula proposed openstack/nova: neutron port security-group improperly updated on nova interface-attach https://review.openstack.org/144398 | 23:00 |
| *** amotoki has quit IRC | 23:01 | |
| *** Sukhdev has quit IRC | 23:01 | |
| *** patrickeast_ has joined #openstack-nova | 23:03 | |
| *** patrickeast has quit IRC | 23:04 | |
| *** patrickeast_ is now known as patrickeast | 23:04 | |
| alexpilotti | cyeoh: hello | 23:05 |
| *** jecarey has quit IRC | 23:05 | |
| cyeoh | alexpilotti: hi | 23:05 |
| alexpilotti | cyeoh: me and claudiub are wondering about the X509 support in python-novaclient | 23:06 |
| alexpilotti | cyeoh: https://review.openstack.org/#/c/136458/ | 23:06 |
| *** claudiub has joined #openstack-nova | 23:06 | |
| alexpilotti | cyeoh: btw thanks a lot with all the help on the nova side of this BP :-) | 23:06 |
| claudiub | cyeoh: hello :) | 23:06 |
| alexpilotti | cyeoh: the question here is how to handle microversions | 23:06 |
| *** iamjarvo has quit IRC | 23:07 | |
| cyeoh | claudiub: hi!. Yea we need to get suppport for it merged | 23:07 |
| claudiub | cyeoh: how to handle the microversions on the novaclient side.. from what i saw, the novaclient does not include the microversion request in the header | 23:07 |
| cyeoh | so off the top of my head there is a novaclient patch we sort of need in first. | 23:07 |
| cyeoh | let me just look for the patch | 23:08 |
| *** mwagner_lap has joined #openstack-nova | 23:08 | |
| cyeoh | last time I looked it wasn't quite ready yet | 23:08 |
| claudiub | cyeoh: yeah... i have a commit on it, and i was working on some functional tests for it, but I could not test the v2.2 microversion | 23:08 |
| *** dave-mccowan has quit IRC | 23:09 | |
| cyeoh | claudiub, alexpilotti: https://review.openstack.org/#/c/152569/ | 23:09 |
| devananda | seconding JoshNang's request above -- those two patches would help us a lot | 23:10 |
| cyeoh | this should allow you to send the microversion header information to nova, but I dont think its quite right yet and i haven't been able to get back to it really in the last week or so) | 23:10 |
| devananda | both patches are confined to our driver code | 23:11 |
| cyeoh | claudiub: so even though its not right yet, I am curious as to see if there is enough there to be able to specify the microversion number to access the keypairs nova side? | 23:11 |
| alexpilotti | cyeoh: tx, so we could rebase on it. Do you think it’ll make it in time for Kilo? | 23:12 |
| cyeoh | I need to do a proper review of 152569 and should hopefully be able to get to that in a couple of hours | 23:12 |
| *** hemna is now known as hemnafk | 23:13 | |
| alexpilotti | cyeoh: yeiii tx | 23:13 |
| claudiub | cyeoh: i see. So, I will take a look at it in the morning, and review it. Then rebase my commit on top of it. Also, I am uncertain on how the microversion request will work on novaclient. For example, let's say that the novaclient version is 2.5.0... That means that the 2.5.0 version will make include the microversion request for v2.2? | 23:14 |
| cyeoh | alexpilotti, I'm honestly not sure, might have to just get tnough in for sending the version info. | 23:14 |
| *** thedodd has quit IRC | 23:14 | |
| claudiub | cyeoh: and then, 2.5.1 will include a microversion request for v2.3? | 23:14 |
| cyeoh | yes if you request 2.5 you will get assuming your implementation doesn't have a max_version (pretty sure it doesn't) will also be included | 23:15 |
| cyeoh | there is no 2.5.1 - we only have X.Y | 23:15 |
| claudiub | it was just just an example. :) | 23:16 |
| *** oro_ has quit IRC | 23:17 | |
| cyeoh | yea, so as long the version requested lies with the range specified for the api_version decorator then it will get executed | 23:17 |
| *** lnr has joined #openstack-nova | 23:17 | |
| *** sbfox has quit IRC | 23:17 | |
| *** oro_ has joined #openstack-nova | 23:17 | |
| *** oro has quit IRC | 23:17 | |
| claudiub | but still, how are any future microversion requests added to the novaclient request? (v2.3, v2.4, v2.5...etc)? | 23:18 |
| *** dims has quit IRC | 23:18 | |
| cyeoh | so I just need to finish looking at this current review then I'll get back on novaclientand see what minimal stuff we can get through | 23:18 |
| *** dims has joined #openstack-nova | 23:18 | |
| *** r-daneel has quit IRC | 23:19 | |
| claudiub | cyeoh: sure, if there is anything I can help you with, let me know. Also, this should also be done in horizon. Is there anything that does this, for the time being? | 23:19 |
| cyeoh | claudiub: we still need to work out what sort extra bits we need to add. For example,if its just an extra parameter and novaclient always prints all parameters returned, then theoreticaly nothing? | 23:19 |
| *** Roamer` has quit IRC | 23:20 | |
| *** andreykurilin_ has joined #openstack-nova | 23:20 | |
| cyeoh | claudiub, yea I don't know mich about the horizon side :-( | 23:20 |
| claudiub | cyeoh: I see. I'll try and talk to the horizon folks about this and see how to implement the microversions there. :) | 23:22 |
| *** salv-orl_ has quit IRC | 23:23 | |
| cyeoh | claudiub: thx! | 23:23 |
| *** sdake has joined #openstack-nova | 23:24 | |
| *** baoli has joined #openstack-nova | 23:24 | |
| *** ZZelle has quit IRC | 23:25 | |
| *** lnr has left #openstack-nova | 23:26 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VolumeOps unit tests https://review.openstack.org/139797 | 23:29 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds MigrationOps unit tests https://review.openstack.org/139796 | 23:29 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: removes test_hypervapi tests https://review.openstack.org/139798 | 23:29 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds VMOps unit tests (part 2) https://review.openstack.org/138934 | 23:29 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: Hyper-V: Adds HyperVDriver unit tests https://review.openstack.org/148980 | 23:29 |
| *** salv-orlando has joined #openstack-nova | 23:31 | |
| *** oro has joined #openstack-nova | 23:34 | |
| *** salv-orlando has quit IRC | 23:35 | |
| *** promulo has joined #openstack-nova | 23:40 | |
| *** jaypipes has quit IRC | 23:41 | |
| *** r-daneel has joined #openstack-nova | 23:42 | |
| *** oro_ has quit IRC | 23:42 | |
| *** oro has quit IRC | 23:43 | |
| *** sdake__ has joined #openstack-nova | 23:45 | |
| *** r-daneel has quit IRC | 23:47 | |
| *** andreykurilin_ has quit IRC | 23:48 | |
| *** dboik has joined #openstack-nova | 23:48 | |
| *** shaohe_feng2 has quit IRC | 23:48 | |
| *** gokrokve has quit IRC | 23:49 | |
| *** sdake has quit IRC | 23:49 | |
| *** dboik_ has joined #openstack-nova | 23:49 | |
| *** dboik has quit IRC | 23:53 | |
| *** alexpilotti has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!