| *** EinstCrazy has quit IRC | 00:03 | |
| *** zhangjn has quit IRC | 00:04 | |
| *** baoli has joined #openstack-nova | 00:11 | |
| *** davideagnello has quit IRC | 00:12 | |
| *** davideagnello has joined #openstack-nova | 00:12 | |
| *** mdrabe has joined #openstack-nova | 00:15 | |
| *** baoli has quit IRC | 00:16 | |
| *** otter768 has joined #openstack-nova | 00:20 | |
| *** acc_ has quit IRC | 00:20 | |
| *** salv-orlando has joined #openstack-nova | 00:21 | |
| *** otter768 has quit IRC | 00:24 | |
| *** tbowman94 has joined #openstack-nova | 00:26 | |
| *** salv-orl_ has joined #openstack-nova | 00:26 | |
| *** yuner027 has joined #openstack-nova | 00:26 | |
| *** salv-orlando has quit IRC | 00:27 | |
| *** pixelb has joined #openstack-nova | 00:28 | |
| *** browne has joined #openstack-nova | 00:29 | |
| *** baoli has joined #openstack-nova | 00:29 | |
| *** baoli has quit IRC | 00:29 | |
| *** baoli has joined #openstack-nova | 00:33 | |
| *** su_zhang has quit IRC | 00:40 | |
| *** dimsum__ has joined #openstack-nova | 00:45 | |
| *** diegows has quit IRC | 00:46 | |
| *** dimsum__ has quit IRC | 00:47 | |
| *** tbowman94 has quit IRC | 00:50 | |
| *** achanda has quit IRC | 00:55 | |
| *** zhangjn has joined #openstack-nova | 00:55 | |
| *** EinstCrazy has joined #openstack-nova | 00:57 | |
| *** zhangjn has quit IRC | 00:57 | |
| *** zhangjn has joined #openstack-nova | 00:58 | |
| *** davideagnello has quit IRC | 00:58 | |
| *** vilobhmm_11 has quit IRC | 00:59 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Add test of claim context manager abort https://review.openstack.org/213591 | 00:59 |
|---|---|---|
| *** salv-orl_ has quit IRC | 01:00 | |
| *** rotbart has joined #openstack-nova | 01:02 | |
| *** piyanai has quit IRC | 01:05 | |
| *** rotbeard has quit IRC | 01:06 | |
| *** jwcroppe has joined #openstack-nova | 01:09 | |
| *** otter768 has joined #openstack-nova | 01:14 | |
| *** derekh has quit IRC | 01:14 | |
| *** otter768 has quit IRC | 01:14 | |
| *** otter768 has joined #openstack-nova | 01:15 | |
| *** wanghua has joined #openstack-nova | 01:16 | |
| *** stevemar has joined #openstack-nova | 01:16 | |
| *** baoli has quit IRC | 01:17 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Add test of claim context manager abort https://review.openstack.org/213591 | 01:27 |
| openstackgerrit | Jay Pipes proposed openstack/nova: Move resize claim abort test to test_tracker https://review.openstack.org/213592 | 01:44 |
| *** dimsum__ has joined #openstack-nova | 01:45 | |
| *** armax has joined #openstack-nova | 01:46 | |
| *** baoli has joined #openstack-nova | 01:49 | |
| openstackgerrit | Jay Pipes proposed openstack/nova: Move resize claim abort test to test_tracker https://review.openstack.org/213592 | 01:50 |
| *** pixelb has quit IRC | 01:53 | |
| *** yamahata has joined #openstack-nova | 01:53 | |
| *** jwcroppe has quit IRC | 01:56 | |
| *** shaohe_robot has joined #openstack-nova | 01:57 | |
| *** rotbeard has joined #openstack-nova | 01:59 | |
| *** dimsum__ has quit IRC | 02:00 | |
| *** rotbart has quit IRC | 02:03 | |
| *** su_zhang has joined #openstack-nova | 02:06 | |
| *** haomaiwa_ has joined #openstack-nova | 02:07 | |
| *** EinstCrazy has quit IRC | 02:09 | |
| *** shaohe_robot1 has joined #openstack-nova | 02:11 | |
| *** shaohe_robot2 has joined #openstack-nova | 02:12 | |
| *** shaohe_robot has quit IRC | 02:13 | |
| *** shaohe_robot1 has quit IRC | 02:16 | |
| *** armax has quit IRC | 02:16 | |
| *** Daisy has joined #openstack-nova | 02:18 | |
| *** haomaiwa_ has quit IRC | 02:18 | |
| *** EinstCrazy has joined #openstack-nova | 02:19 | |
| *** Daisy has quit IRC | 02:23 | |
| *** baoli has quit IRC | 02:28 | |
| *** EinstCrazy has quit IRC | 02:34 | |
| *** mragupat has joined #openstack-nova | 02:34 | |
| *** mragupat_ has joined #openstack-nova | 02:35 | |
| *** mragupat has quit IRC | 02:39 | |
| *** asettle is now known as asettle-afk | 02:44 | |
| *** shyu has joined #openstack-nova | 02:48 | |
| *** haomaiwa_ has joined #openstack-nova | 02:48 | |
| *** rprakash has joined #openstack-nova | 02:54 | |
| rprakash | #info mesaage for alex_xu can you include discussion on cloudlet API this week Tuesday Nova API meeting? | 02:56 |
| rprakash | #undo | 02:56 |
| *** mragupat_ has quit IRC | 02:56 | |
| *** davideagnello has joined #openstack-nova | 02:59 | |
| *** salv-orlando has joined #openstack-nova | 03:00 | |
| *** haomaiwa_ has quit IRC | 03:01 | |
| *** dimsum__ has joined #openstack-nova | 03:01 | |
| *** haomaiwang has joined #openstack-nova | 03:01 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Give instance default hostname if hostname is empty https://review.openstack.org/224438 | 03:01 |
| *** asettle-afk is now known as asettle | 03:02 | |
| *** davideagnello has quit IRC | 03:04 | |
| *** salv-orlando has quit IRC | 03:04 | |
| *** su_zhang has quit IRC | 03:05 | |
| *** rprakash has quit IRC | 03:05 | |
| *** salv-orlando has joined #openstack-nova | 03:06 | |
| *** dimsum__ has quit IRC | 03:08 | |
| *** otter768 has quit IRC | 03:20 | |
| *** EinstCrazy has joined #openstack-nova | 03:22 | |
| *** cfriesen has joined #openstack-nova | 03:23 | |
| *** ildikov has joined #openstack-nova | 03:28 | |
| *** shuquan has joined #openstack-nova | 03:28 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 03:41 |
| *** salv-orlando has quit IRC | 03:43 | |
| *** acc_ has joined #openstack-nova | 03:45 | |
| *** yuner027 has quit IRC | 03:48 | |
| *** yamahata has quit IRC | 03:50 | |
| *** baoli has joined #openstack-nova | 03:59 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-nova | 04:01 | |
| openstackgerrit | Jay Pipes proposed openstack/nova-specs: WIP: Add concept of resource providers https://review.openstack.org/225546 | 04:02 |
| *** baoli has quit IRC | 04:03 | |
| *** mdrabe has quit IRC | 04:05 | |
| *** EinstCrazy has quit IRC | 04:06 | |
| *** su_zhang has joined #openstack-nova | 04:11 | |
| *** davideagnello has joined #openstack-nova | 04:12 | |
| *** acc_ has quit IRC | 04:19 | |
| *** fawadkhaliq has joined #openstack-nova | 04:20 | |
| *** yuner027 has joined #openstack-nova | 04:23 | |
| *** VikasC has joined #openstack-nova | 04:27 | |
| *** achanda has joined #openstack-nova | 04:34 | |
| alex_xu | rpodolyaka: yea, sure | 04:37 |
| *** vilobhmm_11 has joined #openstack-nova | 04:39 | |
| *** Sree has joined #openstack-nova | 04:51 | |
| *** yamahata has joined #openstack-nova | 04:55 | |
| *** rotbeard has quit IRC | 05:00 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-nova | 05:01 | |
| *** deepthi has joined #openstack-nova | 05:04 | |
| *** davideagnello has quit IRC | 05:05 | |
| *** dimsum__ has joined #openstack-nova | 05:05 | |
| *** browne has quit IRC | 05:07 | |
| openstackgerrit | Sean McCully proposed openstack/nova: Creates disk.fallocate_test https://review.openstack.org/188249 | 05:09 |
| *** dimsum__ has quit IRC | 05:11 | |
| *** browne has joined #openstack-nova | 05:11 | |
| *** moshele has joined #openstack-nova | 05:19 | |
| *** ankit_ag has joined #openstack-nova | 05:20 | |
| *** ssurana has joined #openstack-nova | 05:21 | |
| *** otter768 has joined #openstack-nova | 05:21 | |
| *** otter768 has quit IRC | 05:26 | |
| *** deepakcs has joined #openstack-nova | 05:28 | |
| *** deepthi has quit IRC | 05:29 | |
| *** deepthi has joined #openstack-nova | 05:30 | |
| *** wanghua has quit IRC | 05:31 | |
| *** wanghua has joined #openstack-nova | 05:32 | |
| *** salv-orlando has joined #openstack-nova | 05:39 | |
| *** ssurana has quit IRC | 05:40 | |
| *** wanghua has quit IRC | 05:43 | |
| *** salv-orlando has quit IRC | 05:44 | |
| *** ssurana has joined #openstack-nova | 05:45 | |
| *** salv-orlando has joined #openstack-nova | 05:45 | |
| *** wanghua has joined #openstack-nova | 05:45 | |
| *** shuquan has quit IRC | 05:54 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-nova | 06:01 | |
| *** stevemar has quit IRC | 06:05 | |
| *** e0ne has joined #openstack-nova | 06:10 | |
| *** haomaiwang has quit IRC | 06:12 | |
| *** lsmola has joined #openstack-nova | 06:12 | |
| *** e0ne has quit IRC | 06:14 | |
| *** rajesht has joined #openstack-nova | 06:14 | |
| *** fawadkhaliq has quit IRC | 06:14 | |
| *** fawadkhaliq has joined #openstack-nova | 06:15 | |
| *** Marga_ has quit IRC | 06:18 | |
| *** fawadkhaliq has quit IRC | 06:19 | |
| *** Marga_ has joined #openstack-nova | 06:19 | |
| *** Marga_ has quit IRC | 06:21 | |
| *** atuvenie has joined #openstack-nova | 06:22 | |
| *** rdopiera has joined #openstack-nova | 06:23 | |
| *** park_jhei has joined #openstack-nova | 06:24 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/225397 | 06:25 |
| *** salv-orlando has quit IRC | 06:28 | |
| *** rdopiera has quit IRC | 06:30 | |
| *** rdopiera has joined #openstack-nova | 06:30 | |
| *** ssurana has quit IRC | 06:32 | |
| *** scheuran has joined #openstack-nova | 06:32 | |
| *** tutj7 has joined #openstack-nova | 06:34 | |
| *** sudipto has joined #openstack-nova | 06:39 | |
| *** sahid has joined #openstack-nova | 06:46 | |
| *** su_zhang has quit IRC | 06:48 | |
| *** salv-orlando has joined #openstack-nova | 06:50 | |
| *** ildikov has quit IRC | 06:54 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Execute _init_instance() concurrently https://review.openstack.org/223572 | 06:55 |
| openstackgerrit | Sean McCully proposed openstack/nova: Creates disk.fallocate_test https://review.openstack.org/188249 | 06:55 |
| *** shihanzhang has quit IRC | 06:56 | |
| *** vlaza has joined #openstack-nova | 06:58 | |
| *** nkrinner has joined #openstack-nova | 06:59 | |
| *** shihanzhang has joined #openstack-nova | 07:00 | |
| *** asettle is now known as asettle-GONE | 07:04 | |
| *** asettle-GONE is now known as asettle-afk | 07:04 | |
| *** e0ne has joined #openstack-nova | 07:06 | |
| *** apuimedo_ has joined #openstack-nova | 07:06 | |
| *** davideagnello has joined #openstack-nova | 07:06 | |
| *** fawadkhaliq has joined #openstack-nova | 07:08 | |
| *** dimsum__ has joined #openstack-nova | 07:08 | |
| *** haomaiwang has joined #openstack-nova | 07:09 | |
| *** jamielennox is now known as jamielennox|away | 07:09 | |
| *** davideagnello has quit IRC | 07:11 | |
| *** hanzhang has joined #openstack-nova | 07:12 | |
| *** dimsum__ has quit IRC | 07:13 | |
| *** shihanzhang has quit IRC | 07:15 | |
| *** cfriesen has quit IRC | 07:15 | |
| *** jichen has joined #openstack-nova | 07:19 | |
| *** matrohon has joined #openstack-nova | 07:19 | |
| *** otter768 has joined #openstack-nova | 07:22 | |
| *** otter768 has quit IRC | 07:27 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 07:28 |
| openstackgerrit | Rajesh Tailor proposed openstack/nova: Fix order of arguments in assertEqual https://review.openstack.org/221271 | 07:29 |
| *** achanda has quit IRC | 07:38 | |
| *** romainh has joined #openstack-nova | 07:39 | |
| *** achanda has joined #openstack-nova | 07:39 | |
| *** ndipanov has joined #openstack-nova | 07:40 | |
| *** ildikov has joined #openstack-nova | 07:40 | |
| *** baoli has joined #openstack-nova | 07:41 | |
| *** jlanoux has joined #openstack-nova | 07:43 | |
| *** vilobhmm_11 has quit IRC | 07:43 | |
| *** browne has quit IRC | 07:44 | |
| *** baoli has quit IRC | 07:46 | |
| *** lpetrut has joined #openstack-nova | 07:46 | |
| *** lpetrut1 has joined #openstack-nova | 07:49 | |
| *** lpetrut has quit IRC | 07:50 | |
| *** lpetrut1 is now known as lpetrut | 07:50 | |
| *** Marga_ has joined #openstack-nova | 07:52 | |
| *** MarcF has joined #openstack-nova | 07:53 | |
| *** vilobhmm_11 has joined #openstack-nova | 07:53 | |
| *** shihanzhang has joined #openstack-nova | 07:55 | |
| *** Marga_ has quit IRC | 07:57 | |
| *** hanzhang has quit IRC | 07:57 | |
| *** dguitarbite has quit IRC | 07:58 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-nova | 08:01 | |
| *** salv-orlando has quit IRC | 08:03 | |
| *** lucasagomes has joined #openstack-nova | 08:03 | |
| *** dguitarbite has joined #openstack-nova | 08:04 | |
| *** acc_ has joined #openstack-nova | 08:05 | |
| *** yuner027 has quit IRC | 08:05 | |
| *** su_zhang has joined #openstack-nova | 08:06 | |
| *** tdurakov has joined #openstack-nova | 08:06 | |
| *** su_zhang has quit IRC | 08:10 | |
| *** pradipta has joined #openstack-nova | 08:11 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 08:13 |
| *** danpb has joined #openstack-nova | 08:13 | |
| *** yassine_ has joined #openstack-nova | 08:17 | |
| *** apuimedo has quit IRC | 08:18 | |
| *** yamahata has quit IRC | 08:19 | |
| *** achanda has quit IRC | 08:20 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Give instance default hostname if hostname is empty https://review.openstack.org/224438 | 08:20 |
| *** haomaiwang has quit IRC | 08:21 | |
| *** apuimedo_ has quit IRC | 08:22 | |
| *** apuimedo has joined #openstack-nova | 08:22 | |
| *** acc_ has quit IRC | 08:24 | |
| openstackgerrit | Marian Horban proposed openstack/nova: Refactoring of HTTP exception message https://review.openstack.org/223496 | 08:24 |
| *** acc_ has joined #openstack-nova | 08:24 | |
| *** fawadkhaliq has quit IRC | 08:26 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add checks for volume status when rebuilding https://review.openstack.org/217567 | 08:26 |
| *** Sree has quit IRC | 08:27 | |
| *** jlanoux has quit IRC | 08:27 | |
| *** markus_z has joined #openstack-nova | 08:27 | |
| *** jlanoux has joined #openstack-nova | 08:28 | |
| kashyap | PaulMurray: Hi, when you're around, just curious - what do you mean by "TLC" in your email? -- "live migration could do with a little TLC". Please spell out three-letter acronyms in their first occurance. :-) | 08:30 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: WIP: Use legacy engineFacade from oslo_db.enginefacade https://review.openstack.org/221310 | 08:31 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: WIP: use oslo_db.enginefacade https://review.openstack.org/225065 | 08:32 |
| *** atuvenie_ has joined #openstack-nova | 08:32 | |
| *** vilobhmm_11 has quit IRC | 08:35 | |
| *** atuvenie has quit IRC | 08:36 | |
| bauzas | good morning Nova | 08:36 |
| *** baoli has joined #openstack-nova | 08:42 | |
| *** alexpilotti has joined #openstack-nova | 08:43 | |
| *** e0ne has quit IRC | 08:46 | |
| *** Sree has joined #openstack-nova | 08:47 | |
| *** baoli has quit IRC | 08:47 | |
| openstackgerrit | garyk proposed openstack/nova: Revert "libvirt-vif: Allow to configure a script on bridge interface" https://review.openstack.org/225585 | 08:51 |
| *** garyk has joined #openstack-nova | 08:52 | |
| garyk | johnthetubaguy: danpb: please see https://review.openstack.org/225585. This is a major blocker for the NSX neutron plugin. I know I gave a +1 but only checkd that with the ML2 plugin and not ours. | 08:53 |
| *** boris-42 has quit IRC | 08:53 | |
| *** dougwig has quit IRC | 08:53 | |
| *** patrickeast has quit IRC | 08:53 | |
| *** rex_lee_ has quit IRC | 08:53 | |
| *** Marga_ has joined #openstack-nova | 08:53 | |
| garyk | this is a regression and i think that we need to deal with the issue better. | 08:54 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: verify cpu bw policy capability for host https://review.openstack.org/224726 | 08:54 |
| *** aix has quit IRC | 08:54 | |
| johnthetubaguy | garyk: OK, thats interesting, does the NSX plugin supply a script argument or something? | 08:55 |
| openstackgerrit | huan proposed openstack/nova: Do not rely on XenAPI to guess device names https://review.openstack.org/213112 | 08:56 |
| garyk | johnthetubaguy: no, it does not provide scrips. | 08:56 |
| garyk | that is the issue, i think that the change specifally expects scripts to be supplied | 08:56 |
| johnthetubaguy | garyk: so how does that patch break things? | 08:56 |
| *** alex_klimov has joined #openstack-nova | 08:57 | |
| danpb | garyk: huh how is a libvirt patch breaking vmware nsx ?!?!? | 08:57 |
| johnthetubaguy | garyk: honestly, it reads the opposite to me? | 08:57 |
| garyk | johnthetubaguy: danpb: can you guys please look at the trace of the bug. | 08:58 |
| garyk | danpb: in the past the code did not expect a script and now it does | 08:58 |
| *** Marga_ has quit IRC | 08:58 | |
| garyk | danpb: this is the exception we get: unsupported configuration: scripts are not supported on interfaces of type bridge | 08:59 |
| *** salv-orlando has joined #openstack-nova | 08:59 | |
| garyk | that is from libvirt | 08:59 |
| garyk | so could it be the libvirt version? | 08:59 |
| *** derekh has joined #openstack-nova | 08:59 | |
| garyk | the plugin does not provide a script | 09:00 |
| *** park_jhei has quit IRC | 09:00 | |
| johnthetubaguy | garyk: so I guess its maybe this bit? https://review.openstack.org/#/c/225585/1/nova/virt/libvirt/designer.py,cm still quite confused | 09:00 |
| garyk | johnthetubaguy: i tried setting that too and it still happened | 09:01 |
| danpb | originally the designer.py would set script="" but this was a no op | 09:01 |
| danpb | becuase the format_dom() method ignored the script attribute entirely | 09:01 |
| garyk | johnthetubaguy: danpb: please note that this is not on the NSx related plugin but all neutron plugins that use the type "bridge" | 09:01 |
| danpb | in the code we now honour the script attribute | 09:01 |
| johnthetubaguy | oh, so script == "" and not None? | 09:01 |
| danpb | so have to get rid of the bogus script="" which should never have been there | 09:01 |
| johnthetubaguy | thinking about line 1178 here: https://review.openstack.org/#/c/225585/1/nova/virt/libvirt/config.py,cm | 09:02 |
| garyk | my guessing is that if self.script is not None is the problem. but then everyone should break | 09:03 |
| garyk | i can debug this further, but i think that we should revert in the mean time | 09:03 |
| danpb | it would be helpful to see the logs of the CI system | 09:03 |
| salv-orlando | johnthetubaguy: I'm not sure if garyk mentioned that this is not only affecting the NSX plugin, but all the plugins that do direct OVS plugging | 09:03 |
| danpb | the traceback isn't really enough | 09:03 |
| johnthetubaguy | salv-orlando: that totally makes sense | 09:04 |
| *** Sree has quit IRC | 09:04 | |
| johnthetubaguy | salv-orlando: I guess we don't have any of those on our check queue | 09:04 |
| danpb | it seems like the actual problem is that set_vif_host_backend_ovs_config() still has the conf.script="" | 09:04 |
| salv-orlando | johnthetubaguy: nah all the neutron jobs do hybrid vif plugging | 09:04 |
| garyk | danpb: ok, so that should be set as None? | 09:05 |
| danpb | ie the original patch was incomplete | 09:05 |
| salv-orlando | garyk: probably. Worth testing. | 09:05 |
| danpb | we shouldjust delete that line as we did for the set_vif_host_backend_bridge_config() method | 09:05 |
| *** josecastroleon has joined #openstack-nova | 09:06 | |
| garyk | danpb: ok, i'll do that and repost after testing | 09:06 |
| danpb | ok | 09:06 |
| danpb | if it still fails please attach a n-cpu.log file to the bug | 09:06 |
| garyk | danpb: sue, will do | 09:07 |
| *** davideagnello has joined #openstack-nova | 09:07 | |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Split functional tests for legacy(V2.1) and microversions https://review.openstack.org/224594 | 09:07 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Specify api_version for functional tests https://review.openstack.org/224593 | 09:07 |
| johnthetubaguy | danpb: garyk: so I have made this block RC1, let me know what works best here | 09:07 |
| garyk | johnthetubaguy: ok, will something in a few minutes | 09:07 |
| johnthetubaguy | :) | 09:08 |
| *** acc_ has quit IRC | 09:10 | |
| *** davideagnello has quit IRC | 09:11 | |
| *** asettle-afk has quit IRC | 09:12 | |
| ndipanov | johnthetubaguy, when do we want to have the RC1 out - this week? | 09:12 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: rebuild: RPC sends additional args and claims are done https://review.openstack.org/200485 | 09:12 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: claims: move a debug msg to a warn on missing migration https://review.openstack.org/224013 | 09:12 |
| *** park_jhei has joined #openstack-nova | 09:13 | |
| *** yuner027 has joined #openstack-nova | 09:16 | |
| *** dougwig has joined #openstack-nova | 09:16 | |
| *** eglynn has joined #openstack-nova | 09:18 | |
| johnthetubaguy | ndipanov: good question, in theory we tag RC1 tomorrow | 09:19 |
| johnthetubaguy | ndipanov: depends how many bad bugs we spot in between now and then really | 09:19 |
| openstackgerrit | Man Li Qi proposed openstack/nova: (WIP)Filter instance query result for DB who doesn't support regex https://review.openstack.org/130174 | 09:19 |
| ndipanov | johnthetubaguy, any chance we could get this https://review.openstack.org/200485 on the list? | 09:20 |
| *** haomaiwang has joined #openstack-nova | 09:20 | |
| *** claudiub has joined #openstack-nova | 09:21 | |
| johnthetubaguy | ndipanov: yeah, I was wondering about adding that one, mostly due to being unable to backport it | 09:21 |
| *** haomaiwang has quit IRC | 09:21 | |
| *** haomaiwa_ has joined #openstack-nova | 09:21 | |
| *** ociuhandu has quit IRC | 09:21 | |
| johnthetubaguy | ndipanov: so I have added it so it gets more attention, but may have to remove it, hope that works | 09:21 |
| ndipanov | johnthetubaguy, thanks - let's see | 09:22 |
| *** ociuhandu has joined #openstack-nova | 09:22 | |
| *** ociuhandu has quit IRC | 09:22 | |
| *** otter768 has joined #openstack-nova | 09:23 | |
| *** haomaiwa_ has quit IRC | 09:23 | |
| *** haomaiwa_ has joined #openstack-nova | 09:23 | |
| *** sayali has quit IRC | 09:24 | |
| *** boris-42 has joined #openstack-nova | 09:24 | |
| *** aix has joined #openstack-nova | 09:25 | |
| *** sayali has joined #openstack-nova | 09:26 | |
| *** claudiub has quit IRC | 09:26 | |
| *** markvoelker has quit IRC | 09:27 | |
| ndipanov | johnthetubaguy, where is the list btw? | 09:27 |
| *** otter768 has quit IRC | 09:27 | |
| johnthetubaguy | ndipanov: for the blockers, its technically just the milestone list: https://launchpad.net/nova/+milestone/liberty-rc1 | 09:28 |
| johnthetubaguy | ndipanov: for the possibles we have the tag: https://bugs.launchpad.net/nova/+bugs?field.tag=liberty-rc-potential | 09:28 |
| openstackgerrit | garyk proposed openstack/nova: libvirt: fix direct OVS plugging https://review.openstack.org/225585 | 09:29 |
| garyk | johnthetubaguy: danpb: ^ (in parallel I am adding specific designer tests). hopefully we can do that in a separate patch set. | 09:29 |
| ndipanov | johnthetubaguy, thanks | 09:29 |
| garyk | johnthetubaguy: danpb: i have tested and that solves the issue | 09:29 |
| *** Marga_ has joined #openstack-nova | 09:30 | |
| danpb | ok | 09:30 |
| johnthetubaguy | garyk: danpb: my previous comment was about line 1178 here: https://review.openstack.org/#/c/199092/3/nova/virt/libvirt/config.py,cm | 09:31 |
| johnthetubaguy | garyk: danpb: would it be better to check for None and "" ? i.e. just if self.script? | 09:31 |
| johnthetubaguy | anyways, if the other fixes our immediate issue, that seems like a good quick fix | 09:32 |
| danpb | it is valid to set the script to "" for some hypervisors, in order to disable the default script | 09:32 |
| danpb | though we don't need it for openstack, i think it is better to keep it as is | 09:32 |
| johnthetubaguy | danpb: ah, OK, that totally answers my question then | 09:32 |
| *** tojuvone has joined #openstack-nova | 09:34 | |
| *** Marga_ has quit IRC | 09:35 | |
| johnthetubaguy | salv-orlando: garyk: it would be great if you can confirm if/when this fixes all the current CI things that have gone red on your side: https://review.openstack.org/#/c/225585/2 | 09:35 |
| *** tutj7 has quit IRC | 09:36 | |
| johnthetubaguy | garyk: thank you for jumping on that for us! | 09:36 |
| *** park_jhei has quit IRC | 09:36 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Give instance default hostname if hostname is empty https://review.openstack.org/224438 | 09:38 |
| *** eglynn has quit IRC | 09:40 | |
| *** eglynn has joined #openstack-nova | 09:40 | |
| *** kmARC_ has joined #openstack-nova | 09:40 | |
| garyk | johnthetubaguy: at the moment we have issues with our CI. The nova voting one is using Nova net. We are in the process of moving it to Neutron but that is goig to take a while. | 09:41 |
| *** IanGovett has joined #openstack-nova | 09:42 | |
| *** rex_lee_ has joined #openstack-nova | 09:42 | |
| *** baoli has joined #openstack-nova | 09:43 | |
| *** shyu has quit IRC | 09:44 | |
| johnthetubaguy | garyk: OK, understood, CI take a heap of work to get right, they are tricky things | 09:45 |
| *** dimsum__ has joined #openstack-nova | 09:47 | |
| *** baoli has quit IRC | 09:48 | |
| *** Marga_ has joined #openstack-nova | 09:52 | |
| *** patrickeast has joined #openstack-nova | 09:53 | |
| *** Marga_ has quit IRC | 09:54 | |
| *** Marga__ has joined #openstack-nova | 09:54 | |
| *** ociuhandu has joined #openstack-nova | 09:56 | |
| *** salv-orlando has quit IRC | 09:57 | |
| *** e0ne has joined #openstack-nova | 09:58 | |
| *** Marga__ has quit IRC | 09:59 | |
| *** haomaiwa_ has quit IRC | 10:01 | |
| *** yassine_ has quit IRC | 10:01 | |
| *** haomaiwa_ has joined #openstack-nova | 10:01 | |
| *** fawadkhaliq has joined #openstack-nova | 10:02 | |
| *** eglynn_ has joined #openstack-nova | 10:02 | |
| *** kiran-r has joined #openstack-nova | 10:03 | |
| *** eglynn has quit IRC | 10:04 | |
| *** zhangjn has quit IRC | 10:10 | |
| *** tutj7 has joined #openstack-nova | 10:14 | |
| johnthetubaguy | alex_xu: ndipanov: it seems worth a chat about this patch: https://review.openstack.org/#/c/200485/23 | 10:16 |
| johnthetubaguy | I am just reading through that again myself | 10:17 |
| alex_xu | johnthetubaguy: yea | 10:17 |
| *** zhangjn has joined #openstack-nova | 10:17 | |
| alex_xu | johnthetubaguy: ndipanov we won't drop claim when we booting and resize instance https://github.com/openstack/nova/blob/master/nova/compute/claims.py#L51 | 10:18 |
| ndipanov | alex_xu, we do | 10:18 |
| alex_xu | when action succss | 10:18 |
| ndipanov | we do | 10:18 |
| alex_xu | ndipanov: where, correct me if I'm misunderstand again | 10:19 |
| *** zhangjn has quit IRC | 10:19 | |
| ndipanov | alex_xu, rebuild has a confirm-revert model | 10:19 |
| ndipanov | er migratte | 10:19 |
| ndipanov | wich rebuild doesn;t | 10:19 |
| ndipanov | it happens in the confirm path of migrate | 10:20 |
| ndipanov | alex_xu, go look for it there | 10:20 |
| alex_xu | ndipanov: but we haven't rebuild confrim | 10:20 |
| alex_xu | ndipanov: and booting instance won't drop the claim also | 10:21 |
| alex_xu | ndipanov: if we waiting for 'update_available_resource', that maybe a long time | 10:21 |
| alex_xu | so there maybe happen a lot of race | 10:21 |
| alex_xu | also the last thing, we have option to disable 'update_available_resource' https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L192 | 10:22 |
| ndipanov | alex_xu, booting instance does not create a migration record | 10:22 |
| ndipanov | so it does not have to drop it | 10:22 |
| alex_xu | ndipanov: ok, if the rebuild success, why we drop it? | 10:23 |
| ndipanov | because as I explained on the previous patch - from that point the resources will be counted using the instance record | 10:23 |
| openstackgerrit | Ritesh proposed openstack/python-novaclient: Modify nova help list message for --tenant https://review.openstack.org/225612 | 10:23 |
| ndipanov | drop | 10:24 |
| alex_xu | ndipanov: ok, that happened after 60 secs? | 10:24 |
| *** pbrooko has joined #openstack-nova | 10:24 | |
| ndipanov | alex_xu, how about removing that -1 | 10:24 |
| alex_xu | ndipanov: or, that may not happened https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L192 | 10:24 |
| ndipanov | alex_xu, ^ | 10:24 |
| ndipanov | drop the -1 ps | 10:25 |
| ndipanov | pls | 10:25 |
| *** romainh has quit IRC | 10:25 | |
| alex_xu | ndipanov: sorry, what your mean? why removing that -1 | 10:25 |
| ndipanov | alex_xu, well because obviously it's a "I don't really get the code so -1" | 10:26 |
| *** zhangjn has joined #openstack-nova | 10:26 | |
| ndipanov | aka - I am asking a question and -1ing a patch | 10:26 |
| ndipanov | I mean leave it up to you | 10:26 |
| ndipanov | but other than writing up a code walkthrough I don't know how to answer it better | 10:26 |
| *** markvoelker has joined #openstack-nova | 10:28 | |
| *** EinstCrazy has joined #openstack-nova | 10:28 | |
| alex_xu | ndipanov: sorry, I'm not asking question, I just think it is wrong, and I didn't get explain. | 10:28 |
| ndipanov | alex_xu, whatever - I'll try to explain it better | 10:28 |
| *** claudiub has joined #openstack-nova | 10:28 | |
| alex_xu | ndipanov: cool, thanks for the explain! | 10:29 |
| *** yangyp has joined #openstack-nova | 10:31 | |
| *** markvoelker has quit IRC | 10:32 | |
| ndipanov | alex_xu, np - happy to read the code for you anytime :/ | 10:33 |
| *** zhangjn has quit IRC | 10:35 | |
| *** zhangjn has joined #openstack-nova | 10:35 | |
| johnthetubaguy | ndipanov: can I check about setting the migration to failed vs error, is there a reason for that? | 10:37 |
| *** yangyp has quit IRC | 10:38 | |
| *** yangyp has joined #openstack-nova | 10:38 | |
| ndipanov | johnthetubaguy, so dansmith complained about that in a previous PS | 10:39 |
| ndipanov | and he was right | 10:39 |
| ndipanov | error means that instance files will be deleted by a periodic task | 10:39 |
| ndipanov | which is not something we want here (ever?) | 10:39 |
| ndipanov | failed in this case means basically - we could be using the scheduler to retry but in case of evacuate we don't do it as we assume an external process doing the evacuation will decide what to do in that case | 10:41 |
| ndipanov | I left a comment that we don't actually issue a retry like build_and_run does | 10:41 |
| johnthetubaguy | ndipanov: yeah, I am really thinking about this one: https://review.openstack.org/#/c/215483/2 | 10:41 |
| johnthetubaguy | superdan: I think this one might want a -2 from you: https://review.openstack.org/#/c/215483/2 | 10:42 |
| johnthetubaguy | ndipanov: I remember those loops you mean now, yeah, the clean up is assuming the revert style, something live-migrate is also missing | 10:42 |
| ndipanov | johnthetubaguy, yes I can comment on that one too | 10:43 |
| ndipanov | I was gonna do this for live migrate but I failed this cycle | 10:43 |
| *** baoli has joined #openstack-nova | 10:44 | |
| ndipanov | hopefully coming early in Mitaka | 10:44 |
| *** moshele has quit IRC | 10:44 | |
| yangyp | test | 10:44 |
| johnthetubaguy | ndipanov: yeah, that would be cool to get in | 10:44 |
| *** yangyp is now known as yangyapeng | 10:44 | |
| *** rajesht_ has joined #openstack-nova | 10:46 | |
| *** aagrawal has joined #openstack-nova | 10:46 | |
| *** baoli has quit IRC | 10:48 | |
| *** wanghua has quit IRC | 10:49 | |
| *** ankit_ag has quit IRC | 10:50 | |
| *** rajesht has quit IRC | 10:50 | |
| *** tojuvone has quit IRC | 10:51 | |
| *** smatzek has joined #openstack-nova | 10:52 | |
| ndipanov | johnthetubaguy, well that was part of the same BP/bug in theory | 10:53 |
| * johnthetubaguy nods | 10:53 | |
| *** Marga_ has joined #openstack-nova | 10:55 | |
| alex_xu | ndipanov: I think I see now, you aren't good for my first comment, I just say we shouldn't drop , but I didn't say why, right? | 10:56 |
| *** sayali has quit IRC | 10:57 | |
| *** sayali has joined #openstack-nova | 10:58 | |
| *** Marga_ has quit IRC | 10:59 | |
| *** salv-orlando has joined #openstack-nova | 11:00 | |
| ndipanov | alex_xu, no - you are wrong - we HAVE TO drop it | 11:00 |
| ndipanov | I am adding a comment now give me 5 min | 11:01 |
| *** haomaiwa_ has quit IRC | 11:01 | |
| ndipanov | it should explain it | 11:01 |
| ndipanov | I was confirming in the code | 11:01 |
| *** 6JTAB24P4 has joined #openstack-nova | 11:01 | |
| alex_xu | ndipanov: ok, just want to say sorry, I feel I should explain more clearly | 11:01 |
| alex_xu | ndipanov: yea, waiting for your replied | 11:02 |
| *** raorn_ is now known as raorn | 11:05 | |
| *** bradjones has joined #openstack-nova | 11:05 | |
| *** bradjones has quit IRC | 11:05 | |
| *** bradjones has joined #openstack-nova | 11:05 | |
| rajesht_ | johnthetubaguy, ndipanov: IMO instance files will not be deleted by periodic task if migration status is set to 'error' during live-migration failure | 11:07 |
| *** atuvenie_ has quit IRC | 11:07 | |
| *** davideagnello has joined #openstack-nova | 11:08 | |
| *** bradjones has quit IRC | 11:08 | |
| rajesht_ | johnthetubaguy, ndipanov: because periodic task will only delete instance files from compute node (source/dest) only if instance is deleted already | 11:08 |
| *** bradjones has joined #openstack-nova | 11:08 | |
| *** bradjones has quit IRC | 11:08 | |
| *** bradjones has joined #openstack-nova | 11:08 | |
| rajesht_ | johnthetubaguy, ndipanov: so there is no harm in setting migration status to 'error' during live-migration failure. | 11:09 |
| johnthetubaguy | rajesht_: I think its about this code: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6457 | 11:10 |
| *** pbrooko has quit IRC | 11:10 | |
| rajesht_ | johnthetubaguy: in above link, it will get all the migration record with migration status 'error' | 11:11 |
| rajesht_ | johnthetubaguy: but it will only list instance which are deleted | 11:11 |
| rajesht_ | johnthetubaguy: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6467 | 11:11 |
| *** asettle has joined #openstack-nova | 11:12 | |
| rajesht_ | johnthetubaguy: so only deleted instance which are having migration records with error status are filtered in https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6471 | 11:12 |
| *** davideagnello has quit IRC | 11:13 | |
| rajesht_ | johnthetubaguy, ndipanov: setting migration status to 'error' during live-migration failure will cover below case as well. | 11:13 |
| rajesht_ | johnthetubaguy, ndipanov: if instance live-migration fails after copying instance files from source to dest node and migration status is set to 'error' | 11:14 |
| johnthetubaguy | rajesht_: I think its more about that loop never needs to clean these up | 11:14 |
| *** bradjones has quit IRC | 11:14 | |
| rajesht_ | after that if instance gets deleted, the delete call will only delete instance files from instance host (not from other host which might be either source or dest node). | 11:14 |
| *** bradjones has joined #openstack-nova | 11:14 | |
| *** bradjones has quit IRC | 11:14 | |
| *** bradjones has joined #openstack-nova | 11:14 | |
| johnthetubaguy | rajesht_: so lets turn this upside down, they are different on purpose, what is the issue that them being different is causing? | 11:15 |
| *** asettle has quit IRC | 11:16 | |
| *** asettle has joined #openstack-nova | 11:16 | |
| ndipanov | alex_xu, commented | 11:16 |
| ndipanov | on ps 22 | 11:16 |
| alex_xu | ndipanov: cool, reading now | 11:16 |
| rajesht_ | johnthetubaguy: first it is not consistent with other migration api (resize, revert-resize, confirm-resize) and second it will not cover scenario which I have mentioned just now. | 11:17 |
| rajesht_ | if instance live-migration fails after copying instance files from source to dest node and migration status is set to 'failed' | 11:17 |
| alex_xu | ndipanov: thanks for such long replied frist | 11:18 |
| ndipanov | rajesht_, so | 11:18 |
| *** pixelb has joined #openstack-nova | 11:18 | |
| ndipanov | the problem with live migration is that it did not use claiming at all | 11:18 |
| rajesht_ | after that if instance gets deleted, the delete call will only delete instance files from instance host (not from other host which might be either source or dest node). and there is no other way to delete those files except that periodic task | 11:18 |
| ndipanov | and migration objects are really part of resource tracking | 11:18 |
| ndipanov | rajesht_, so any talk about consistency is pointless here - we are already pretty inconsistent | 11:19 |
| *** aagrawal has quit IRC | 11:19 | |
| johnthetubaguy | so I am seeing the error state and the failed state expressing a different intent, which while odd, seems OK to me | 11:20 |
| *** ankit_ag has joined #openstack-nova | 11:20 | |
| ndipanov | yes johnthetubaguy ^ | 11:20 |
| rajesht_ | ndipanov: ok, leaving consistency apart, but how to delete instance files from other node for above scenario ? | 11:21 |
| rajesht_ | ndipanov: other node (i.e not instance.host ) | 11:21 |
| *** salv-orlando has quit IRC | 11:23 | |
| *** otter768 has joined #openstack-nova | 11:24 | |
| *** salv-orlando has joined #openstack-nova | 11:24 | |
| *** otter768 has quit IRC | 11:28 | |
| ndipanov | rajesht_, you mean in case of a failed migration | 11:29 |
| *** lucasagomes is now known as lucas-hungry | 11:29 | |
| *** markvoelker has joined #openstack-nova | 11:29 | |
| ndipanov | rajesht_, that's actually a good point | 11:29 |
| rajesht_ | ndipanov: yes, | 11:29 |
| ndipanov | I was just thinking that | 11:29 |
| *** asettle has quit IRC | 11:30 | |
| *** pradipta has quit IRC | 11:30 | |
| ndipanov | so the migration is failed -> we restart it to a different host, -> it succeeds -> files are left | 11:30 |
| ndipanov | rajesht_, ^ right? | 11:30 |
| rajesht_ | ndipanov: I am explaining different case. | 11:31 |
| rajesht_ | ndipanov: migration is failed after copying instance files from source to dest -> we delete instance -> delete call remove instance files from instance.host -> how to delete instance files from other host | 11:32 |
| ndipanov | ok that too | 11:32 |
| *** markvoelker has quit IRC | 11:33 | |
| *** romainh has joined #openstack-nova | 11:34 | |
| *** Sree has joined #openstack-nova | 11:35 | |
| *** yassine_ has joined #openstack-nova | 11:35 | |
| ndipanov | rajesht_, so to be fair - that problem existed even before the migration record right? | 11:36 |
| alex_xu | ndipanov: just replied | 11:36 |
| *** pbrooko has joined #openstack-nova | 11:36 | |
| rajesht_ | ndipanov: yes | 11:36 |
| ndipanov | rajesht_, well that is an entirely different bug that what you reported | 11:37 |
| ndipanov | alex_xu, thanks | 11:37 |
| alex_xu | ndipanov: you are welcome | 11:37 |
| rajesht_ | ndipanov: so how can I proceed ? | 11:38 |
| *** bswartz has quit IRC | 11:39 | |
| ndipanov | rajesht_, let me comment on the bug | 11:41 |
| ndipanov | tbh I'd like to actually reproduce the issue first | 11:41 |
| rajesht_ | ndipanov: sure | 11:41 |
| *** 6JTAB24P4 has quit IRC | 11:41 | |
| *** moshele has joined #openstack-nova | 11:45 | |
| alex_xu | ndipanov: sorry, I need take a rest for a while, actually I'm ill today, get up do some review for boring...I feel unaffordable now...will be back to check after few hours | 11:47 |
| ndipanov | alex_xu, OK - checking on something | 11:48 |
| ndipanov | you may have a point actually | 11:48 |
| *** lpetrut has quit IRC | 11:51 | |
| *** abhishekk has joined #openstack-nova | 11:52 | |
| *** piyanai has joined #openstack-nova | 11:53 | |
| *** aysyd has joined #openstack-nova | 11:53 | |
| *** Marga_ has joined #openstack-nova | 11:56 | |
| alex_xu | ndipanov: :) | 11:58 |
| *** Marga_ has quit IRC | 12:00 | |
| *** maurosr is now known as murilovf | 12:02 | |
| *** murilovf is now known as maurosr | 12:02 | |
| *** tutj7 has quit IRC | 12:03 | |
| *** artom has joined #openstack-nova | 12:04 | |
| *** thorst has joined #openstack-nova | 12:04 | |
| *** moshele has quit IRC | 12:08 | |
| *** markvoelker has joined #openstack-nova | 12:11 | |
| *** ildikov has quit IRC | 12:12 | |
| *** piyanai has quit IRC | 12:15 | |
| *** piyanai has joined #openstack-nova | 12:16 | |
| *** atuvenie_ has joined #openstack-nova | 12:16 | |
| *** baoli has joined #openstack-nova | 12:17 | |
| *** mwagner_lap has joined #openstack-nova | 12:18 | |
| *** piyanai has quit IRC | 12:18 | |
| *** jecarey has quit IRC | 12:20 | |
| *** ferest has joined #openstack-nova | 12:21 | |
| *** yangyapeng has quit IRC | 12:23 | |
| *** lalit has joined #openstack-nova | 12:24 | |
| *** ildikov has joined #openstack-nova | 12:25 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: New API for get service info from id https://review.openstack.org/172412 | 12:25 |
| lalit | bauzas: are you there? | 12:26 |
| *** _gryf has quit IRC | 12:26 | |
| *** VikasC has quit IRC | 12:26 | |
| bauzas | lalit: I do | 12:26 |
| bauzas | lalit: but I'll run into a meeting in the next mins | 12:26 |
| lalit | bauzas: ohk..when I can get your time if possible? | 12:27 |
| *** ferest has quit IRC | 12:27 | |
| bauzas | lalit: what do you want to discuss ? | 12:28 |
| lalit | It is regarding patch https://review.openstack.org/#/c/213041/ | 12:28 |
| lalit | I have put global methods of flavor in Flavor/FlavorList objects but some of the test are failing..the issue might be the way of mocking which is not clear for me.. | 12:30 |
| *** lucas-hungry is now known as lucasagomes | 12:30 | |
| *** Daisy has joined #openstack-nova | 12:31 | |
| bauzas | lalit: mmm, ok | 12:31 |
| * bauzas needs to look at the chanvge | 12:31 | |
| *** edmondsw has joined #openstack-nova | 12:31 | |
| lalit | bauzas: in file https://review.openstack.org/#/c/213041/17/nova/tests/unit/api/openstack/compute/test_flavor_access.py line 144 I want to mock _flavor_get_all_db method which is @staticmethod and @require_context method , I tried like self.stubs.Set(objects.FlavorList, '_flavor_get_all_db', fake_get_all_flavors_sorted_list) but test is failing with TypeError: unbound method fake_get_all_flavors_sorted_list() must be called wi | 12:37 |
| lalit | bazuas: How I can mock some static method of class, which is called within the class methods | 12:39 |
| *** rlrossit has joined #openstack-nova | 12:39 | |
| *** rlrossit has left #openstack-nova | 12:39 | |
| *** rlrossit has joined #openstack-nova | 12:39 | |
| *** dave-mccowan has joined #openstack-nova | 12:40 | |
| *** Daisy has quit IRC | 12:41 | |
| *** sudipto has quit IRC | 12:43 | |
| *** e0ne has quit IRC | 12:44 | |
| *** dimsum__ has quit IRC | 12:47 | |
| *** dimsum__ has joined #openstack-nova | 12:47 | |
| *** xyang1 has joined #openstack-nova | 12:48 | |
| *** dimsum__ is now known as dims | 12:48 | |
| *** lalitd has joined #openstack-nova | 12:48 | |
| *** dims is now known as Guest31942 | 12:49 | |
| *** Guest31942 is now known as dims__ | 12:49 | |
| *** rfolco has joined #openstack-nova | 12:50 | |
| *** deepakcs has quit IRC | 12:53 | |
| *** deepthi has quit IRC | 12:53 | |
| *** Marga_ has joined #openstack-nova | 12:56 | |
| bauzas | lalit: back | 13:00 |
| *** Marga_ has quit IRC | 13:01 | |
| bauzas | lalit: could you please explain me again your problem ? | 13:01 |
| *** ildikov has quit IRC | 13:03 | |
| *** ildikov has joined #openstack-nova | 13:04 | |
| *** jaypipes has joined #openstack-nova | 13:05 | |
| *** dustins has joined #openstack-nova | 13:05 | |
| *** ctrath has joined #openstack-nova | 13:05 | |
| bauzas | lalit: oh I see your problem | 13:05 |
| *** mdrabe has joined #openstack-nova | 13:06 | |
| *** moshele has joined #openstack-nova | 13:06 | |
| bauzas | lalit: well, it's pretty clear, nope ? you're mocking a module method, not an object | 13:06 |
| *** exploreshaifali has joined #openstack-nova | 13:07 | |
| *** renatoarmani has joined #openstack-nova | 13:08 | |
| bauzas | lalit: here, flavor_obj is the module reference, not the generated object | 13:08 |
| *** davideagnello has joined #openstack-nova | 13:09 | |
| *** renatoarmani has quit IRC | 13:09 | |
| *** fawadkhaliq has quit IRC | 13:10 | |
| *** EinstCrazy has quit IRC | 13:11 | |
| *** davideagnello has quit IRC | 13:13 | |
| *** pbrooko has quit IRC | 13:14 | |
| *** pkoniszewski has quit IRC | 13:14 | |
| dhellmann | johnthetubaguy: does novaclient need a new release before we create the stable branch? http://paste.openstack.org/show/469795/ | 13:17 |
| johnthetubaguy | dhellmann: I thought we were good, but I should sure check | 13:18 |
| *** mfisch has quit IRC | 13:18 | |
| lalit | bauzas: I have changed in nova/objects/flavor.py in my local machine ..that file and line no I gave for the reference ... and what changes I have mentioned above I have done on my local machine which is not yet uploaded. | 13:18 |
| johnthetubaguy | dhellmann: I guess we sure need the global requirements update | 13:19 |
| *** fawadkhaliq has joined #openstack-nova | 13:19 | |
| *** mfisch has joined #openstack-nova | 13:19 | |
| *** mfisch is now known as Guest61243 | 13:19 | |
| *** raildo-afk is now known as raildo | 13:19 | |
| bauzas | lalit: well, looking at your change, I think you can't do all of that like this way | 13:19 |
| bauzas | lalit: because you're wanting to modify a remotable classmethod | 13:20 |
| *** lascii is now known as alaski | 13:20 | |
| bauzas | lalit: a remotable method (classmethod or not) means that the code is actually executed by the conductor manager | 13:21 |
| bauzas | lalit: which means that you can't change something before making sure that you don't break the compatibility with the previous versions | 13:21 |
| *** jaypipes has quit IRC | 13:22 | |
| dhellmann | johnthetubaguy: check what it is, you might be ok. We're going to try for stable branches today, so if you submit a release patch asap I can take care of it. | 13:22 |
| bauzas | lalit: https://github.com/openstack/nova/blob/master/nova/objects/base.py#L289-L362 | 13:23 |
| *** haomaiwang has joined #openstack-nova | 13:23 | |
| bauzas | lalit: https://github.com/openstack/nova/blob/master/nova/cmd/compute.py#L65-L68 | 13:23 |
| *** otter768 has joined #openstack-nova | 13:25 | |
| bauzas | and https://github.com/openstack/oslo.versionedobjects/blob/master/oslo_versionedobjects/base.py#L156-L217 | 13:25 |
| lalit | bauzas: ok, so in this case how I can put the global methods in Flavor objects ? can you suggest | 13:25 |
| *** VW has joined #openstack-nova | 13:25 | |
| bauzas | lalit: I need to understand first what you want to do | 13:26 |
| bauzas | lalit: so, given that now there are 2 DBs, you're wanting to compare both ? | 13:26 |
| lalit | bazuas: in https://review.openstack.org/#/c/213041/17/nova/objects/flavor.py line 298 . I want to put method _flavor_get_all_db in side FlavorList | 13:27 |
| lalit | as @staticmethod and @require_context | 13:28 |
| *** otter768 has quit IRC | 13:29 | |
| lalit | bazuas: I have changed it and put _flavor_get_all_db method in FlavorList object and it is working fine. but some tests like nova.tests.unit.api.openstack.compute.test_flavor_access.FlavorAccessTestV21.test_list_flavor_with_no_admin_ispublic_none is failing with "TypeError: unbound method fake_get_all_flavors_sorted_list() must be called with FlavorList instance as first argument (got FakeRequestContext instance instead)" | 13:31 |
| *** leitan has joined #openstack-nova | 13:32 | |
| PaulMurray | bauzas, was there a plan to skip this weeks scheduler meeting? | 13:32 |
| *** ankit_ag has quit IRC | 13:33 | |
| bauzas | PaulMurray: yup, we agreed on that last week | 13:33 |
| leitan | Hi everyone, has anyone ever tried to use another notification topic than "notifications" ? , cause im seeing that nova ... is just publishing the compute.instance.update events to the other topic on the queue, and the compute.instance.create.end and delete.start are just published to the topic "notifications" no matter what | 13:33 |
| bauzas | lalit: I asked you *why*, not how :) | 13:33 |
| PaulMurray | bauzas, is there a wiki page or etherpad with stuff like that on now? The eavesdrop page is pretty useless | 13:33 |
| *** smatzek has quit IRC | 13:33 | |
| bauzas | PaulMurray: you mean, knowing the cancellations ? | 13:33 |
| *** superdan is now known as dansmith | 13:34 | |
| bauzas | PaulMurray: well we could provide a wiki if needed | 13:34 |
| PaulMurray | bauzas, yes, actually I notice some of the meetings have a link to agendas or logs etc., maybe something similar can be done on the nova scheduler entry | 13:35 |
| *** eharney has joined #openstack-nova | 13:35 | |
| *** sayali has quit IRC | 13:35 | |
| *** abhishekk has quit IRC | 13:36 | |
| *** obedmr has joined #openstack-nova | 13:36 | |
| PaulMurray | see: http://eavesdrop.openstack.org/#Nova_Scheduler_(ex._Gantt)_Team_Meeting | 13:36 |
| lxsli | PaulMurray: n0ano did post to os-dev saying no scheduler meeting this week | 13:37 |
| *** zhangjn has quit IRC | 13:37 | |
| PaulMurray | lxsli - thinking of something more permanent | 13:37 |
| PaulMurray | but thanks anyway | 13:37 |
| bauzas | PaulMurray: yeah, agreed, we should have made a change for removing that specific entry | 13:38 |
| bauzas | PaulMurray: not sure tho if it's possible | 13:38 |
| bauzas | I mean, exclude some specific dates | 13:38 |
| *** renatoarmani has joined #openstack-nova | 13:38 | |
| *** mriedem has joined #openstack-nova | 13:39 | |
| *** burt has joined #openstack-nova | 13:41 | |
| johnthetubaguy | mriedem: curious what you think about releasing another python-novaclient before the stable branch is cut? | 13:41 |
| *** sayali has joined #openstack-nova | 13:41 | |
| mriedem | johnthetubaguy: is there some specific change in mind? | 13:42 |
| *** jaypipes has joined #openstack-nova | 13:42 | |
| johnthetubaguy | mriedem: I was wondering about getting this one included, I guess, otherwise, not sure we have much to add: https://github.com/openstack/python-novaclient/commit/b1204446ef17e7df244b23d7552da24a6d86b53f | 13:42 |
| *** vladikr has joined #openstack-nova | 13:43 | |
| johnthetubaguy | we could just backport, but nice to avoid some of that if we can | 13:43 |
| *** sayali has quit IRC | 13:43 | |
| mriedem | stable will be branched from the latest release tag? | 13:43 |
| mriedem | or HEAD of master? | 13:43 |
| alex_xu | johnthetubaguy: mriedem do we still have time to backport some api bug fix? | 13:44 |
| mriedem | alex_xu: to which branch? | 13:44 |
| alex_xu | mriedem: kilo | 13:45 |
| johnthetubaguy | mriedem: I think the last tag | 13:45 |
| mriedem | alex_xu: stable/kilo is still around for awhile | 13:45 |
| mriedem | alex_xu: so i don't see why not | 13:45 |
| alex_xu | mriedem: ok, got it, thanks | 13:45 |
| mriedem | johnthetubaguy: there was a feature change that went into master after 2.29 | 13:45 |
| mriedem | not sure if you want that in stable/liberty or not | 13:46 |
| *** salv-orl_ has joined #openstack-nova | 13:46 | |
| *** r-daneel has joined #openstack-nova | 13:46 | |
| *** e0ne has joined #openstack-nova | 13:47 | |
| *** erlon has joined #openstack-nova | 13:47 | |
| mriedem | johnthetubaguy: i'd be fine with a 2.30 release and then branch | 13:47 |
| *** salv-orlando has quit IRC | 13:47 | |
| *** mmedvede has joined #openstack-nova | 13:48 | |
| mriedem | these are the diffs https://github.com/openstack/python-novaclient/compare/2.29.0...master | 13:49 |
| *** dane-fichter has joined #openstack-nova | 13:50 | |
| *** haypo has left #openstack-nova | 13:50 | |
| *** fawadkhaliq has quit IRC | 13:51 | |
| *** kylek3h has joined #openstack-nova | 13:52 | |
| *** su_zhang has joined #openstack-nova | 13:53 | |
| ndipanov | alex_xu, responded - it seems that you are correct | 13:54 |
| dansmith | johnthetubaguy: so on this: https://review.openstack.org/#/c/215483/2 | 13:54 |
| *** sayali has joined #openstack-nova | 13:54 | |
| dansmith | johnthetubaguy: It looks like you were back and forth a little, but I'm not sure where you landed | 13:55 |
| dansmith | johnthetubaguy: I think that probably the right thing to do is commit this for consistency, and then in mitaka ndipanov has plans to actually add a state field that follows a state machine so that all movements are consistnt | 13:55 |
| dansmith | johnthetubaguy: given that liberty would be the first release to expose migration records for live migration, it might be good to have it be consistent | 13:56 |
| dansmith | however, I'm also fine with the "meh, leave it as is and move to the state thing later anyway" | 13:56 |
| *** Marga_ has joined #openstack-nova | 13:57 | |
| *** sseverson has quit IRC | 13:57 | |
| ndipanov | dansmith, well | 13:58 |
| ndipanov | in case of live migrate that may be OK | 13:58 |
| ndipanov | I am not sure - what do we do when the migration fails | 13:59 |
| ndipanov | as soon as we've ran migrate disk and power off | 13:59 |
| ndipanov | we can't really delete files anymore on the destination | 13:59 |
| ndipanov | in the general case | 14:00 |
| *** salv-orl_ has quit IRC | 14:00 | |
| ndipanov | idk | 14:00 |
| markus_z | @cores: Needs a 2nd +2: https://review.openstack.org/#/c/215102/ (if possible) | 14:00 |
| dansmith | ndipanov: well, I'm not sure what has to do with this, I thought this was just visibility through the api | 14:00 |
| *** haomaiwang has quit IRC | 14:01 | |
| *** david-ly_ is now known as david-lyle | 14:01 | |
| dansmith | ndipanov: but, I'm looking and realizing that my patch to get most of this exposed in the API never made it because I think we were going to delay that until after the state machine | 14:01 |
| *** haomaiwang has joined #openstack-nova | 14:01 | |
| ndipanov | dansmith, right | 14:01 |
| ndipanov | so the problem ans why this patch was posted even though the bug does not say so | 14:01 |
| *** Marga_ has quit IRC | 14:02 | |
| ndipanov | is that it leaves files on destination hosts in some cases (afaiu the guy who reported it) | 14:02 |
| johnthetubaguy | dansmith: I am kinda on the fence, I like the consistency, but I recon it ndipanov's claims patch should be consistent with the live-migrate case | 14:02 |
| dansmith | ndipanov: because we're triggering something on the status field/ | 14:02 |
| *** smatzek has joined #openstack-nova | 14:03 | |
| ndipanov | dansmith, yes that cleanup task | 14:03 |
| ndipanov | taht you pointed out on my patch | 14:03 |
| dansmith | ndipanov: but that shouldn't be doing cleanups from live migrations, should it? | 14:03 |
| ndipanov | dansmith, let me check | 14:04 |
| dansmith | looks like it would do anything I guess, but not sure it *should* | 14:04 |
| *** jlvillal has joined #openstack-nova | 14:05 | |
| ndipanov | yeah it would do anything | 14:05 |
| dansmith | is that the right thing to do though? | 14:05 |
| johnthetubaguy | interestingly, it sets the migration to failed to stop that migration getting picked up again, or something like that | 14:06 |
| dansmith | right, | 14:06 |
| dansmith | which is why a live migration going straight ->failed as it does now means this task wouldn't run on it | 14:06 |
| dansmith | but if we apply the patch, then it would | 14:06 |
| dansmith | which is probably what the poster is aiming for, but I'm not sure it's the right thing to do | 14:07 |
| *** zhangjn has joined #openstack-nova | 14:07 | |
| ndipanov | dansmith, well consider it like this | 14:07 |
| *** Daisy has joined #openstack-nova | 14:07 | |
| ndipanov | if we fail and then re-migrate elsewhere | 14:07 |
| ndipanov | we may be left with orphan files | 14:07 |
| *** dane-fichter has quit IRC | 14:07 | |
| ndipanov | so we definitely need to consider this going forward | 14:07 |
| ndipanov | right now - I am not sure what happens if live mig fails | 14:08 |
| dansmith | what I don't want to have happen, | 14:08 |
| dansmith | is that we fail in the middle of migrating, don't set host=hostB, and then have this task delete the actual instance files | 14:08 |
| *** pbrooko has joined #openstack-nova | 14:08 | |
| dansmith | I'd rather have orphans than zombies, or whatever :) | 14:08 |
| dansmith | one of those cases in the patch, | 14:09 |
| dansmith | he sets it to error if pre_live_migration fails | 14:09 |
| ndipanov | I think data loss is the word you're looking for :) | 14:09 |
| dansmith | which would be before we've moved anything.. | 14:09 |
| dansmith | heh, yeah | 14:09 |
| *** e0ne has quit IRC | 14:10 | |
| openstackgerrit | sahid proposed openstack/nova: libvirt: verify cpu bw policy capability for host https://review.openstack.org/224726 | 14:10 |
| *** thangp has joined #openstack-nova | 14:10 | |
| ndipanov | but dansmith but we set it to 'failed' when the actual live_migration fails too | 14:10 |
| dansmith | yeah, maybe it's right to do it in that case | 14:11 |
| *** jecarey has joined #openstack-nova | 14:11 | |
| *** burgerk has joined #openstack-nova | 14:11 | |
| dansmith | it really depends I think.. if live migration fails and we're doing block migration, does it really leave orphans around? | 14:11 |
| *** Daisy has quit IRC | 14:11 | |
| dansmith | also, how does this periodic task handle instances on shared storage? | 14:11 |
| dansmith | does driver.delete_instance_files() check shared storage? | 14:12 |
| *** Sree has quit IRC | 14:12 | |
| *** zhangjn has quit IRC | 14:12 | |
| *** lpetrut has joined #openstack-nova | 14:12 | |
| ndipanov | i think (hope) so | 14:12 |
| *** nelsnelson has joined #openstack-nova | 14:12 | |
| *** krtaylor has quit IRC | 14:13 | |
| dansmith | not that I see :( | 14:13 |
| ndipanov | :( | 14:13 |
| dansmith | https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L7087-7146 | 14:13 |
| *** sayali has quit IRC | 14:13 | |
| dansmith | oops | 14:13 |
| dansmith | https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L7087-L7146 | 14:13 |
| *** e0ne has joined #openstack-nova | 14:14 | |
| ndipanov | ooops | 14:14 |
| mriedem | johnthetubaguy: did you want me to move forward with the 2.30 novaclient release proposal to the release team? | 14:14 |
| dansmith | not sure that's the right place to do it though.. the periodic probably needs to do that check | 14:14 |
| danpb | dansmith: what's that you say ? a bug in nova migration code ? i don't believe it ;-P | 14:15 |
| mriedem | dansmith: https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L914 | 14:15 |
| mriedem | there is a check for shared before calling delete_instance_files | 14:15 |
| *** kiran-r has quit IRC | 14:15 | |
| dansmith | mriedem: that's not what we're talking about | 14:16 |
| mriedem | i think the destroy_disks kwarg is checked in the compute manager based on shared storage | 14:16 |
| ndipanov | mriedem, not in the periodic task | 14:16 |
| mriedem | oh | 14:16 |
| dansmith | mriedem: https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6447-L6490 | 14:16 |
| mriedem | then that seems like a problem... | 14:16 |
| dansmith | yeah | 14:16 |
| dansmith | danpb: this is not a bug in the migration code | 14:16 |
| dansmith | ndipanov: so I'll put up a patch for this, but the question is still "if this wasn't broken, under what conditions would we want to do this for live migration" | 14:17 |
| *** sayali has joined #openstack-nova | 14:18 | |
| *** ijuwang has joined #openstack-nova | 14:18 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Correct memory validation for live migration https://review.openstack.org/180151 | 14:18 |
| ndipanov | dansmith, well - live migration should take care of this on it's own | 14:19 |
| ndipanov | but that may not always happen | 14:19 |
| ndipanov | similar to resize I think | 14:19 |
| ndipanov | so the answer is probably that we want to do this for live mig if we think it's a good idea | 14:19 |
| dansmith | johnthetubaguy: I'm going to target this data loss bug straight for rc1 if that's okay | 14:19 |
| *** zhangjn has joined #openstack-nova | 14:19 | |
| dansmith | ndipanov: but not if pre-live-migration fails right? | 14:19 |
| *** jhesketh has quit IRC | 14:20 | |
| ndipanov | afaict this guards against, during the whole dance of migration one node went awol | 14:20 |
| ndipanov | well even then the message could get lost right? | 14:20 |
| dansmith | yeah, but do we have anything to clean up at that point? | 14:20 |
| dansmith | either way, if you think it's safe, then that's fine | 14:20 |
| ndipanov | that task I think :) | 14:20 |
| *** jhesketh has joined #openstack-nova | 14:20 | |
| dansmith | just makes me nervous | 14:20 |
| ndipanov | I don't - I'm jus thinking what it was supposed to do | 14:20 |
| bauzas | jaypipes: around ? | 14:21 |
| bauzas | jaypipes: currently reviewing https://review.openstack.org/#/c/213591/ | 14:21 |
| ndipanov | I think we definitely need a clear way to decide if something failed | 14:21 |
| ndipanov | before we can decide to clean it up | 14:21 |
| ndipanov | there's a lot of guessing | 14:21 |
| *** pbrooko_ has joined #openstack-nova | 14:22 | |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: WIP: Use legacy engineFacade from oslo_db.enginefacade https://review.openstack.org/221310 | 14:22 |
| *** krtaylor has joined #openstack-nova | 14:22 | |
| *** david-lyle has quit IRC | 14:22 | |
| bauzas | jaypipes: so, if I read correctly, you changed the test to be a negative one ? coolness | 14:22 |
| openstackgerrit | Pavel Kholkin proposed openstack/nova: WIP: use oslo_db.enginefacade https://review.openstack.org/225065 | 14:23 |
| *** pbrooko has quit IRC | 14:23 | |
| dansmith | ndipanov: I'm actually worried this is not an easy fix.. I'm not sure we have interfaces out of the driver to make the call at this point...still looking | 14:23 |
| dansmith | ndipanov: to fix the task for shared storage I mean | 14:23 |
| ndipanov | the periodic task you mean? | 14:24 |
| *** browne has joined #openstack-nova | 14:24 | |
| jaypipes | bauzas: ya. | 14:24 |
| bauzas | jaypipes: ok, I just left a comment about except: but that's just a comment, +1'd | 14:25 |
| johnthetubaguy | dansmith: +1 for targeting a data loss bug | 14:25 |
| *** Daisy has joined #openstack-nova | 14:25 | |
| dansmith | ndipanov: yeah, but found some things that might give me what I want | 14:25 |
| dansmith | johnthetubaguy: k | 14:25 |
| *** vlaza has quit IRC | 14:26 | |
| dansmith | ndipanov: ah, yeah, no I don't think I can make this determination from the periodic | 14:26 |
| openstackgerrit | Monty Taylor proposed openstack/gantt: Change ignore-errors to ignore_errors https://review.openstack.org/225734 | 14:26 |
| *** david-lyle has joined #openstack-nova | 14:26 | |
| bauzas | jaypipes: last point, is this intentional ? https://review.openstack.org/#/c/213592/11 | 14:26 |
| mriedem | claudiub: want to set a priority on this bug? https://bugs.launchpad.net/nova/+bug/1466056 | 14:27 |
| openstack | Launchpad bug 1466056 in OpenStack Compute (nova) kilo "Hyper-V: serial ports issue on Windows Threshold" [Undecided,In progress] - Assigned to Claudiu Belu (cbelu) | 14:27 |
| dansmith | ndipanov: because live migration does that during the pre phase, which we'd have to re-execute in order to get the answer | 14:27 |
| bauzas | jaypipes: quite funny, do you want an ATC badge maybe ? :D | 14:27 |
| *** Marga_ has joined #openstack-nova | 14:27 | |
| ndipanov | dansmith, sounds like a job for migration context? | 14:27 |
| *** Marga_ has quit IRC | 14:28 | |
| *** armax has joined #openstack-nova | 14:28 | |
| dansmith | ndipanov: yeah, but not sure that's a change I want to make at this point in the release ... | 14:28 |
| *** Marga_ has joined #openstack-nova | 14:28 | |
| dansmith | ndipanov: and actually, it can't be in the migration context, because the instance could have been moved again since this migration that we're cleaning up was run, so we need it on the migration record itself probably, which is even more of a change | 14:29 |
| ndipanov | dansmith, ugh | 14:30 |
| ndipanov | well at that point we need full blown history of everything really | 14:30 |
| ndipanov | how about if we made the PT a bit more defensive | 14:30 |
| *** tsekiyama has joined #openstack-nova | 14:31 | |
| ndipanov | as in - the instance is clearly gone but has artifacts on our host | 14:31 |
| ndipanov | that may not be so easy either... | 14:32 |
| jaypipes | bauzas: abandoned. no idea how I f'd that up. :) | 14:32 |
| dansmith | ndipanov: yeah, not sure how we can do that in a reasonable way | 14:32 |
| bauzas | jaypipes: there will be an Upstream Training in Tokyo, if you want to attend :p | 14:32 |
| dansmith | hah | 14:32 |
| ndipanov | well it would either have to be very defensive (i.e. do it only for deleted instances) | 14:32 |
| ndipanov | LOL | 14:32 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/nova: Ironic: Add support for microversion https://review.openstack.org/221848 | 14:33 |
| dansmith | ndipanov: yeah, we could do that I guess | 14:33 |
| ndipanov | not sure how useful that is | 14:33 |
| dansmith | ndipanov: but that means the residue stays until the instance is deleted | 14:33 |
| dansmith | right | 14:33 |
| ndipanov | but it's better than nothing | 14:33 |
| * bauzas jk, heh | 14:33 | |
| dansmith | ndipanov: well, the other option is to disable this by default, document it in the release notes heavily if people want to turn it on for non-shared cases | 14:33 |
| dansmith | ndipanov: because if you're on shared storage, this isn't a problem for you anyway (right?) | 14:33 |
| ndipanov | yeah I think so, but shared in this case means ceph too | 14:34 |
| openstackgerrit | Monty Taylor proposed openstack/nova: Change ignore-errors to ignore_errors https://review.openstack.org/225773 | 14:34 |
| dansmith | ndipanov: right, but ceph should be unaffected by this just like shared-file-on-nfs is right? | 14:34 |
| mriedem | rajesht_: this isn't appropriate for stable as a backport https://review.openstack.org/#/c/223968/ | 14:35 |
| mriedem | rajesht_: it's just code cleanup | 14:35 |
| ndipanov | yeah I guess so | 14:35 |
| dansmith | ndipanov: problem is, we have to add a new config variable for this, since the original patch shared one with instance_delete_interval for backportability | 14:37 |
| dansmith | johnthetubaguy: ^^ | 14:37 |
| ndipanov | sigh | 14:37 |
| dansmith | if this weren't a CVE I would say we just revert this code, honestly | 14:37 |
| ndipanov | how is it a CVE? | 14:37 |
| dansmith | ndipanov: leaving instance data laying around on unrelated nodes | 14:38 |
| dansmith | it was weak at best, but it was handled as an embargoed patch, so... :( | 14:38 |
| ndipanov | yeah :( | 14:39 |
| dansmith | full disclosure, I reviewed it and didn't catch this along with the few others that reviewed it | 14:40 |
| ndipanov | our CVE process blows I am not at all surprised | 14:41 |
| dansmith | yeah, well, regardless | 14:41 |
| dansmith | adding another conf for this seems like the easiest thing to do at this point | 14:42 |
| ndipanov | yeah sadly | 14:42 |
| ndipanov | which is bound to raise questions - are you opening people up to a security issue | 14:42 |
| ndipanov | which I don't think this is unless I am missing something | 14:42 |
| claudiub | mriedem: sorry, I was afk. Thanks for setting the priority and reviewing it. :) | 14:43 |
| dansmith | ndipanov: well, I think the security exposure was weak to begin with, but as I understand it, the most important thing is that there is a way to not be exposed | 14:44 |
| *** eharney has quit IRC | 14:44 | |
| dansmith | ndipanov: if we can't reasonably have a safe default (deleting data != safe) then I think this is the next best thing | 14:44 |
| *** zzzeek has joined #openstack-nova | 14:45 | |
| *** zhangjn has quit IRC | 14:45 | |
| johnthetubaguy | dansmith: reading through the context here | 14:46 |
| johnthetubaguy | ndipanov: would in interested in ideas to improve the embargo review, I know some folks are working on a security patch only gerrit system | 14:48 |
| *** jlanoux has quit IRC | 14:49 | |
| ndipanov | johnthetubaguy, well I don't have any better than that :) | 14:50 |
| mriedem | claudiub: can you reply to the comment in https://review.openstack.org/#/c/223290/ ? | 14:50 |
| *** zhangjn has joined #openstack-nova | 14:51 | |
| johnthetubaguy | ndipanov: I think its blocked on us upgrading gerrit, which is taking time after it broke the world a few months back, dunno any more than that in terms of status I am afraid | 14:51 |
| *** achanda has joined #openstack-nova | 14:51 | |
| johnthetubaguy | dansmith: are you thinking turning the periodic task off by default, or something like that? https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L6448 | 14:52 |
| *** cfriesen has joined #openstack-nova | 14:52 | |
| dansmith | johnthetubaguy: yes, just a sec | 14:52 |
| openstackgerrit | Dan Smith proposed openstack/nova: Disable the _cleanup_incomplete_migrations() periodic task by default https://review.openstack.org/225811 | 14:52 |
| dansmith | johnthetubaguy: ndipanov ^ | 14:52 |
| *** xek has quit IRC | 14:53 | |
| johnthetubaguy | dansmith: I am missing a bit, its only working on deleted instances, is the problem that with shared storage that disk could be used elsewhere somehow? | 14:54 |
| openstackgerrit | Monty Taylor proposed openstack/python-novaclient: Change ignore-errors to ignore_errors https://review.openstack.org/225821 | 14:54 |
| dansmith | lol | 14:54 |
| andreykurilin | johnthetubaguy: hi! | 14:55 |
| dansmith | ndipanov: so, this doesn't do general cleanup at all :D | 14:55 |
| johnthetubaguy | andreykurilin: hi | 14:55 |
| andreykurilin | johnthetubaguy: I need your help. When novaclient 2.29.0 was released, global-requirements was not updated and all projects use previous release 2.28.1 | 14:56 |
| claudiub | mriedem: done. | 14:57 |
| mriedem | andreykurilin: https://github.com/openstack/requirements/blob/master/global-requirements.txt#L163 | 14:57 |
| xyang1 | mriedem: hi, can you take a look of this patch? https://review.openstack.org/#/c/225278/ | 14:57 |
| mriedem | novaclient 2.29 can be used | 14:57 |
| mriedem | andreykurilin: https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L260 | 14:57 |
| andreykurilin | mriedem: can be, but 2.28.1 is installed by default | 14:57 |
| *** EinstCrazy has joined #openstack-nova | 14:57 | |
| mriedem | andreykurilin: huh? | 14:57 |
| mriedem | that would only happen if someone capped at 2.28.1 | 14:57 |
| dansmith | johnthetubaguy: the problem is I'm an idiot apparently | 14:57 |
| mriedem | oh you mean 2.28.1 was already installed, it's not going to force the update to 2.29 | 14:57 |
| mriedem | andreykurilin: but why would we need to raise the minimum required novaclient in g-r to 2.29? | 14:58 |
| mriedem | people can upgrade if they need to | 14:58 |
| johnthetubaguy | dansmith: did you change coffee supplier? | 14:59 |
| *** mdrabe has quit IRC | 14:59 | |
| *** EinstCrazy has quit IRC | 14:59 | |
| *** electrocucaracha has joined #openstack-nova | 14:59 | |
| dansmith | johnthetubaguy: no, BUT, I haven't had much yet this morning, so I guess I get a pass, right? | 14:59 |
| johnthetubaguy | dansmith: sure :) | 14:59 |
| *** albertom has quit IRC | 14:59 | |
| *** emagana has joined #openstack-nova | 14:59 | |
| johnthetubaguy | dansmith: I subscribe to the "data loss is bad" point of view, so thats cool | 15:00 |
| dansmith | heh | 15:00 |
| andreykurilin | mriedem: Please look at latest results of gate-rally-dsvm-rally-nova job | 15:00 |
| andreykurilin | mriedem: http://logs.openstack.org/63/225663/2/check/gate-rally-dsvm-rally-nova/610db4d/logs/devstacklog.txt.gz#_2015-09-21_14_12_32_448 | 15:00 |
| dansmith | johnthetubaguy: lemme wait for ndipanov to agree and then I'll drop the bug and the patch and work on that next cup of coffee | 15:00 |
| *** haomaiwang has quit IRC | 15:01 | |
| johnthetubaguy | andreykurilin: that seems out of date, somehow | 15:01 |
| *** achanda has quit IRC | 15:01 | |
| johnthetubaguy | upper-constraints.txt | 15:01 |
| ndipanov | agree with what dansmith ? | 15:01 |
| *** haomaiwang has joined #openstack-nova | 15:01 | |
| dansmith | ndipanov: there is an only-for-deleted-instances filter in that periodic :D | 15:01 |
| *** devlaps has joined #openstack-nova | 15:02 | |
| ndipanov | there is | 15:02 |
| andreykurilin | johnthetubaguy: I post a change today https://review.openstack.org/#/c/225593/ . But dhellmann said "The constraints file was updated in another patch syncing the constraints for all of our releases, so that part of this change isn't needed." | 15:02 |
| ndipanov | so then we should be good | 15:02 |
| ndipanov | although that makes the fix even more bizarre | 15:02 |
| andreykurilin | so we have 2.28.1 used in the gates now. at least in Rally gates | 15:02 |
| johnthetubaguy | andreykurilin: but mriedem pointed out this: https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L260 | 15:03 |
| johnthetubaguy | andreykurilin: maybe rally needs to sync the latest requirements? | 15:03 |
| *** pbrooko has joined #openstack-nova | 15:03 | |
| andreykurilin | johnthetubaguy: https://github.com/openstack/rally/blob/master/requirements.txt#L26 | 15:04 |
| andreykurilin | it is equal to global-requirements | 15:04 |
| *** pbrooko_ has quit IRC | 15:04 | |
| johnthetubaguy | andreykurilin: its not that one, its this one: https://github.com/openstack/requirements/blob/master/upper-constraints.txt#L260 | 15:04 |
| andreykurilin | mriedem: "but why would we need to raise the minimum required novaclient in g-r to 2.29?" because 2.29.0 release contains several critical bug-fixes | 15:04 |
| johnthetubaguy | andreykurilin: the log you posted seemed to disagree with that | 15:04 |
| johnthetubaguy | ndipanov: +1 on that fix being even more bizarre | 15:05 |
| *** zhangjn has quit IRC | 15:05 | |
| *** albertom has joined #openstack-nova | 15:05 | |
| johnthetubaguy | andreykurilin: the lots said this, which seems incorrect given the current state of the global requirements repo: Collecting python-novaclient===2.28.1 (from -c /opt/stack/new/requirements/upper-constraints.txt (line 263)) | 15:05 |
| *** zhangjn has joined #openstack-nova | 15:06 | |
| johnthetubaguy | andreykurilin: but I am far from expert with this stuff, but seems odd to me | 15:06 |
| *** zhangjn has quit IRC | 15:06 | |
| dhellmann | andreykurilin: we don't raise minimum requirements just for bug fixes | 15:07 |
| dhellmann | johnthetubaguy, andreykurilin : we *just* merged the constraints change, so it may not be showing up in test jobs yet | 15:07 |
| johnthetubaguy | dhellmann: ah, that would do it | 15:07 |
| *** zhangjn has joined #openstack-nova | 15:07 | |
| dhellmann | if you refresh your requirements sandbox you'll see 2.29.0 there now | 15:08 |
| *** zhangjn has quit IRC | 15:08 | |
| andreykurilin | johnthetubaguy: http://logs.openstack.org/63/225263/2/check/gate-tempest-dsvm-heat/a3ed66d/logs/devstacklog.txt.gz#_2015-09-21_13_40_04_796 log from one of the latest patch in heat | 15:08 |
| *** annegentle has joined #openstack-nova | 15:08 | |
| johnthetubaguy | andreykurilin: yeah, the required updated only just merged ^ | 15:08 |
| *** zhangjn has joined #openstack-nova | 15:09 | |
| *** zhangjn has quit IRC | 15:09 | |
| *** ZZelle has joined #openstack-nova | 15:09 | |
| *** zhangjn has joined #openstack-nova | 15:10 | |
| openstackgerrit | Mike Dorman proposed openstack/nova: Cells v1: Support attach/detach interface events https://review.openstack.org/215459 | 15:11 |
| *** lpetrut has quit IRC | 15:11 | |
| andreykurilin | dhellmann: what is requirement to change version? what about broken one of the main functional(boot a server with bdmV1)? | 15:11 |
| *** lpetrut has joined #openstack-nova | 15:11 | |
| andreykurilin | dhellmann: 2.29. | 15:11 |
| *** zhangjn has quit IRC | 15:11 | |
| *** eharney has joined #openstack-nova | 15:12 | |
| *** Piet has joined #openstack-nova | 15:12 | |
| *** zhangjn has joined #openstack-nova | 15:12 | |
| dhellmann | andreykurilin: the latest version will be installed by default in unit tests and other jobs that do not yet honor the constraints file. jobs that do use the constraints file will also get 2.29.0 the next time they are run, now that the file is updated. | 15:13 |
| andreykurilin | johnthetubaguy: ok, great. I expect that some jobs will be green after recheck | 15:13 |
| dhellmann | andreykurilin: we raise the minimum version when a project depends on having a feature that is not implemented in an earlier version | 15:13 |
| johnthetubaguy | mriedem: sorry, I forgot to reply, if you could do the 2.30 that would be great | 15:14 |
| andreykurilin | dhellmann: I suppose that we can raise the minimum version when we don't want to recommend to use previous minimum version. | 15:15 |
| *** kmARC_ has quit IRC | 15:16 | |
| *** mdrabe has joined #openstack-nova | 15:16 | |
| *** hemnafk is now known as hemna | 15:16 | |
| *** devlaps has quit IRC | 15:19 | |
| mriedem | johnthetubaguy: ok i'll do that | 15:20 |
| mriedem | xyang1: will take a look sometime today | 15:20 |
| xyang1 | mriedem: great, thanks! | 15:20 |
| lalit | bauzas: yaa, that was commented by alaski and melwitt for to put global methods as the static methods of Flavor objects | 15:20 |
| lalit | bauzas: sorry for the late reply, was in meeting | 15:21 |
| openstackgerrit | Daniel Berrange proposed openstack/nova-specs: Libvirt driver emulator threads placement policy https://review.openstack.org/225893 | 15:22 |
| *** rushil has joined #openstack-nova | 15:22 | |
| *** stevemar has joined #openstack-nova | 15:22 | |
| *** dustins is now known as dustins|away | 15:23 | |
| *** jlanoux has joined #openstack-nova | 15:24 | |
| *** otter768 has joined #openstack-nova | 15:26 | |
| *** devlaps has joined #openstack-nova | 15:28 | |
| *** jdurgin1 has joined #openstack-nova | 15:29 | |
| johnthetubaguy | alex_xu: hey, I just spotted the -2 here: https://review.openstack.org/#/c/220791/15 do we have patches up for that already? | 15:29 |
| *** otter768 has quit IRC | 15:30 | |
| *** ildikov has quit IRC | 15:32 | |
| *** pixelb has quit IRC | 15:33 | |
| bauzas | lalit: otp | 15:33 |
| *** pixelb has joined #openstack-nova | 15:34 | |
| *** pbrooko has quit IRC | 15:34 | |
| *** jdurgin1 has quit IRC | 15:35 | |
| *** mtanino has joined #openstack-nova | 15:35 | |
| *** pbrooko has joined #openstack-nova | 15:35 | |
| *** tjones has joined #openstack-nova | 15:39 | |
| *** tjones has left #openstack-nova | 15:39 | |
| *** EinstCrazy has joined #openstack-nova | 15:40 | |
| alaski | lalit: I only caught part of the conversation so maybe my suggestion won't help but you could look at https://review.openstack.org/#/c/161904/9/nova/tests/unit/objects/test_instance_mapping.py,cm for inspiration on how testing was done in a similar situation | 15:40 |
| ctrath | alaski: is cells v2 useable yet in master? | 15:41 |
| *** exploreshaifali has quit IRC | 15:41 | |
| alaski | ctrath: it's not | 15:41 |
| *** nkrinner has quit IRC | 15:42 | |
| ctrath | alaski: thanks.. is the plan to make is useable in the mitaka release? | 15:42 |
| mriedem | the plan was to make it usable in liberty | 15:43 |
| alaski | ctrath: yep, that's definitely the plan. we spent more time in some areas than I was expecting this cycle, but I think we can do it this time | 15:43 |
| mriedem | patches welcome | 15:43 |
| *** ssurana has joined #openstack-nova | 15:43 | |
| mriedem | a better question is, what needs to be done before it's usable? | 15:43 |
| ctrath | mriedem: +1 | 15:43 |
| mriedem | there are outstanding patches for cells v2 in https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 15:44 |
| mriedem | which are going to be deferred to mitaka | 15:44 |
| *** atuvenie_ has quit IRC | 15:45 | |
| mriedem | ctrath: there is also a weekly meeting https://wiki.openstack.org/wiki/Meetings/NovaCellsv2 | 15:45 |
| bauzas | mriedem: no longer weekly | 15:45 |
| alaski | persisting the request spec is a big piece. Then there's adding a buildrequest object/table, though I'm rethinking this and will have an alternate proposal soon. and then there's some db/RPC switching work | 15:45 |
| ctrath | mriedem: thanks.. I'll start attending that. | 15:45 |
| alaski | but I'm going to write all of this up before summit so people can read up if they wish | 15:46 |
| alaski | yeah, the meeting is temporarily biweekly during freeze. we'll go back to weekly after the summit | 15:46 |
| bauzas | alaski: mriedem: btw. I saw you made very valuable feedback on my changes but without leaving a note, so consequently missed it :) | 15:48 |
| * bauzas should lobby for a +0 CR | 15:48 | |
| mriedem | bauzas: the az devref change? | 15:48 |
| bauzas | mriedem: yup | 15:48 |
| *** matrohon has quit IRC | 15:49 | |
| bauzas | mriedem: since my dash is pretty messy, I missed that | 15:49 |
| mriedem | i didn't want to -1 and make you ragequit :) | 15:49 |
| bauzas | mriedem: I'm resilient | 15:49 |
| bauzas | mriedem: I'm rather A/P | 15:49 |
| bauzas | you kill me, I respawn | 15:49 |
| *** zhangjn has quit IRC | 15:50 | |
| mriedem | will silver kill you? | 15:50 |
| mriedem | maybe don't tell me | 15:50 |
| *** yassine_ has quit IRC | 15:50 | |
| openstackgerrit | Daniel Berrange proposed openstack/nova-specs: Libvirt real time instances https://review.openstack.org/139688 | 15:50 |
| *** stevemar has quit IRC | 15:51 | |
| *** sayali has quit IRC | 15:52 | |
| bauzas | mriedem: I haven't yet found my kryptonite | 15:52 |
| *** zhangjn has joined #openstack-nova | 15:53 | |
| *** haomaiwang has quit IRC | 15:54 | |
| *** haomaiwang has joined #openstack-nova | 15:55 | |
| *** stevemar has joined #openstack-nova | 15:57 | |
| *** sseverson has joined #openstack-nova | 15:58 | |
| *** emagana has quit IRC | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-nova | 16:01 | |
| *** stevemar has quit IRC | 16:02 | |
| *** emagana has joined #openstack-nova | 16:02 | |
| *** diogogmt has joined #openstack-nova | 16:02 | |
| *** jlanoux has quit IRC | 16:02 | |
| *** salv-orlando has joined #openstack-nova | 16:03 | |
| *** alejandrito has joined #openstack-nova | 16:03 | |
| *** apuimedo has quit IRC | 16:06 | |
| *** ildikov has joined #openstack-nova | 16:06 | |
| *** atuvenie_ has joined #openstack-nova | 16:07 | |
| *** alex_klimov has quit IRC | 16:10 | |
| *** su_zhang has quit IRC | 16:11 | |
| *** Daisy has quit IRC | 16:11 | |
| *** scheuran has quit IRC | 16:12 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Add version parameter for servers schema extension points https://review.openstack.org/222032 | 16:12 |
| openstackgerrit | John Garbutt proposed openstack/nova: Filter leading/trailing spaces for name field in v2.1 compat mode https://review.openstack.org/220791 | 16:12 |
| *** apuimedo has joined #openstack-nova | 16:13 | |
| *** zhangjn has quit IRC | 16:14 | |
| openstackgerrit | Andreas Jaeger proposed openstack/nova: Fix capitalization of IP https://review.openstack.org/225923 | 16:15 |
| *** EinstCrazy has quit IRC | 16:15 | |
| *** atuvenie__ has joined #openstack-nova | 16:17 | |
| *** atuvenie_ has quit IRC | 16:17 | |
| *** moshele has quit IRC | 16:18 | |
| mriedem | doffm: there are still two with hits in here https://review.openstack.org/#/c/223751/ | 16:19 |
| mriedem | so let's not remove those, otherwise it looks good | 16:19 |
| openstackgerrit | Mark Doffman proposed openstack/nova: libvirt: Don't attempt to convert initrd images. https://review.openstack.org/225289 | 16:20 |
| johnthetubaguy | ndipanov: are you still happy with this one? I think I get why it should work now. https://review.openstack.org/#/c/200485/23 | 16:21 |
| johnthetubaguy | dansmith: I am wondering if you wanted to re-apply your +2 to this one: https://review.openstack.org/#/c/200485/23 | 16:21 |
| markus_z | @cores: A 2nd +2 would be great: https://review.openstack.org/#/c/215102/ (libvirt, live-migration, console) | 16:23 |
| *** nic has joined #openstack-nova | 16:25 | |
| *** rushil has quit IRC | 16:26 | |
| mriedem | doffm: looks like you need to add the -V option to lsof here https://review.openstack.org/#/c/223125/ | 16:26 |
| mriedem | to see why it's failing | 16:26 |
| *** whenry_ has joined #openstack-nova | 16:27 | |
| *** Apoorva has joined #openstack-nova | 16:27 | |
| mriedem | johnthetubaguy: this seems to be cutting it very close to rc1 for string freeze https://review.openstack.org/#/c/225923/ | 16:28 |
| johnthetubaguy | mriedem: yeah, seems too late | 16:29 |
| johnthetubaguy | looks like they are reaching out to them | 16:29 |
| *** mdrabe_ has joined #openstack-nova | 16:30 | |
| mriedem | johnthetubaguy: there are 2 approved novaclient changes going through zuul now, i'll push the 2.30 release request once those are merged | 16:32 |
| ndipanov | johnthetubaguy, so finishing up a comment | 16:32 |
| ndipanov | alex_xu, is right but we have the same race in resize too | 16:32 |
| ndipanov | johnthetubaguy, give me 5 minutes | 16:33 |
| johnthetubaguy | ndipanov: ah, I missed a comment somewhere | 16:33 |
| *** mdrabe has quit IRC | 16:33 | |
| johnthetubaguy | mriedem: sounds OK to me. If we miss the window, we can always backport the fixes folks might need | 16:34 |
| *** dfisher has joined #openstack-nova | 16:34 | |
| *** sahid has quit IRC | 16:34 | |
| openstackgerrit | Mark Doffman proposed openstack/nova: Move logging outside of LibvirtConfigObject.to_xml https://review.openstack.org/224216 | 16:35 |
| johnthetubaguy | ndipanov: the way I read it, you don't decrement the resource usage as you don't pass the flavor in, but I have honestly got quite confused around all those details | 16:36 |
| *** stevemar has joined #openstack-nova | 16:37 | |
| johnthetubaguy | ndipanov: I am wondering if we should let this slip into mitaka? i.e. should we delay RC1 for this? | 16:37 |
| *** romainh has left #openstack-nova | 16:38 | |
| *** annegentle has quit IRC | 16:38 | |
| *** annegentle has joined #openstack-nova | 16:39 | |
| *** e0ne has quit IRC | 16:41 | |
| *** VW has quit IRC | 16:41 | |
| *** rushil has joined #openstack-nova | 16:42 | |
| *** VW has joined #openstack-nova | 16:42 | |
| *** garyk has quit IRC | 16:42 | |
| *** coreywright has quit IRC | 16:44 | |
| *** dustins|away is now known as dustins | 16:45 | |
| *** VW has quit IRC | 16:47 | |
| *** browne has quit IRC | 16:47 | |
| *** VW has joined #openstack-nova | 16:48 | |
| dfisher | as a first-time contributor to nova, I'm struggling to decipher a test failure. For https://review.openstack.org/#/c/225267/ and the gate-nova-python27 run, I'm getting http://paste.openstack.org/show/472969/ Can somebody help me out with this? | 16:52 |
| *** coreywright has joined #openstack-nova | 16:52 | |
| nic | First things first: welcome, and thanks for the help | 16:52 |
| dfisher | :) | 16:52 |
| openstackgerrit | Mark Doffman proposed openstack/nova: DO NOT MERGE: Extra logging for lxc teardown fail. https://review.openstack.org/223125 | 16:54 |
| ndipanov | johnthetubaguy, I commented on https://review.openstack.org/#/c/200485/ | 16:55 |
| *** tongli has joined #openstack-nova | 16:55 | |
| ndipanov | the issue is extremely subtle | 16:55 |
| ndipanov | I think we should merge it ASAP really as this can't be backported and makes evacuate better | 16:55 |
| ndipanov | but up to you | 16:55 |
| *** derekh has quit IRC | 16:55 | |
| *** ijw has joined #openstack-nova | 16:56 | |
| *** markus_z has quit IRC | 16:57 | |
| *** haomaiwang has quit IRC | 16:58 | |
| *** haomaiwang has joined #openstack-nova | 16:59 | |
| *** haomaiwang has quit IRC | 17:01 | |
| *** haomaiwang has joined #openstack-nova | 17:01 | |
| *** haomaiwang has quit IRC | 17:01 | |
| mriedem | dfisher: the test is checking explicitly if yo'uve changed a value in an enum field | 17:01 |
| mriedem | which HVType is | 17:01 |
| mriedem | b/c it changes the contract for the objects that use that enum | 17:02 |
| dfisher | ah. | 17:02 |
| mriedem | "Some objects have changed; please make sure the versions have been bumped, and then update their hashes here." | 17:02 |
| * dfisher nods | 17:02 | |
| mriedem | it's talking about the objects under nova.objects that use that field | 17:02 |
| *** haomaiwang has joined #openstack-nova | 17:02 | |
| *** haomaiwang has quit IRC | 17:03 | |
| mriedem | dfisher: are there going to be more changes upstream for solaris support in nova? or is this just enabling things to work out of tree at oracle? | 17:03 |
| *** jamespd has quit IRC | 17:03 | |
| dansmith | ndipanov: so after your long comment on your patch, I'm not sure where you stand.. you want to land that as-is with bugs reported for future tweaks, yes? | 17:03 |
| *** haomaiwang has joined #openstack-nova | 17:03 | |
| dfisher | mriedem: we have a driver that we're *trying* to get in but getting CI set up on our network is an IT nightmare. | 17:04 |
| *** rdopiera has quit IRC | 17:04 | |
| ndipanov | dansmith, actually I am not sure myself tbh :( | 17:04 |
| mriedem | dfisher: is there a spec for that? | 17:04 |
| dfisher | no, not yet. | 17:04 |
| *** jamespd has joined #openstack-nova | 17:04 | |
| dansmith | ndipanov: okay | 17:04 |
| dfisher | this is the only prereq for us, however. this little patch. | 17:04 |
| ndipanov | let me think about it some more | 17:04 |
| mriedem | dfisher: then i'm not sure that you're trying to get a driver in :) | 17:04 |
| dansmith | ndipanov: in case it's not obvious, I'm appreciative of you working on making it better :) | 17:05 |
| ndipanov | it's surprisingly tricky | 17:05 |
| dfisher | we're trying to get CI in place *before* coming in. | 17:05 |
| dfisher | there's no sense in doing specs/blueprints if we can't participate long-term | 17:05 |
| mriedem | dfisher: that's kind of getting the cart before the horse | 17:05 |
| mriedem | it's probably in your best interest to get the spec and discussion going on a new virt driver before spending a lot of time on a 3rd party CI system | 17:06 |
| *** unicell has joined #openstack-nova | 17:06 | |
| dfisher | ok. I'll work with my team on that piece. | 17:06 |
| ndipanov | dansmith, thanks :) | 17:06 |
| *** sudipto has joined #openstack-nova | 17:07 | |
| *** su_zhang has joined #openstack-nova | 17:09 | |
| *** danpb has quit IRC | 17:10 | |
| dfisher | mriedem: ok, so it looks like I need to bump nova/objects/hv_spec.py to v1.2 .. right? | 17:11 |
| dansmith | dfisher: that patch doesn't pass unit tests | 17:11 |
| dansmith | dfisher: it's not cool to recheck it for that reason, as you just spent a bunch of resources re-running a thing that won't pass | 17:11 |
| *** haomaiwang has quit IRC | 17:11 | |
| dfisher | dansmith: understood. As I said, I'm new to contributing to Nova. Apologies for the wasted resources. | 17:12 |
| dansmith | dfisher: yep, understand | 17:12 |
| *** asselin has quit IRC | 17:12 | |
| *** devlaps has quit IRC | 17:12 | |
| ndipanov | dansmith, OK I' | 17:12 |
| ndipanov | ve made up my mind | 17:13 |
| mriedem | dfisher: well anything that uses https://github.com/openstack/nova/blob/master/nova/objects/fields.py#L634 in nova.objects. | 17:13 |
| ndipanov | I'll update the patch | 17:13 |
| mriedem | dfisher: and this https://github.com/openstack/nova/blob/master/nova/objects/fields.py#L658 | 17:13 |
| dansmith | ndipanov: heh, okay :) | 17:14 |
| dfisher | mriedem: ok. | 17:14 |
| *** pbrooko has quit IRC | 17:14 | |
| *** asselin has joined #openstack-nova | 17:17 | |
| *** moshele has joined #openstack-nova | 17:19 | |
| dfisher | mriedem: would you recommend that we do a spec for solaris zones before merging this patch? | 17:20 |
| *** achanda has joined #openstack-nova | 17:20 | |
| *** electrocucaracha has quit IRC | 17:21 | |
| mriedem | dfisher: well, if your goal is to add a solaris virt driver to nova, you're going to need a spec for that, | 17:21 |
| mriedem | and that would make this change one of the work items (of many) for that spec | 17:21 |
| mriedem | dfisher: there are other out of tree things in that hvtype though, like zvm, so there is precedent for it not needing to be in tree | 17:22 |
| mriedem | for better or worse | 17:22 |
| mriedem | having said all that, this is going to need to wait for mitaka anyway because of the rpc version bump and the fact we're tagging liberty rc1 tomorrow | 17:22 |
| dfisher | oh absolutely. | 17:22 |
| mriedem | dfisher: btw, is your virt driver code already out on github somewhere? | 17:22 |
| dfisher | it's not on github, but it's open on java.net. want the link? | 17:23 |
| mriedem | sure | 17:23 |
| *** davideagnello has joined #openstack-nova | 17:23 | |
| dfisher | digging … please hold :) | 17:23 |
| *** e0ne has joined #openstack-nova | 17:24 | |
| dfisher | https://java.net/projects/solaris-userland/sources/gate/show/components/openstack/nova/files/solariszones | 17:27 |
| dansmith | dfisher: just a preview of the future, I'm going to expect a very detailed explanation of why you can't support all of solaris' boutique virtualization things in the libvirt library before I'm on board with you adding another driver to nova | 17:27 |
| *** otter768 has joined #openstack-nova | 17:27 | |
| dfisher | it's incredibly slow for some reason. | 17:27 |
| * dansmith thinks: maybe it's because: java | 17:27 | |
| dfisher | :) | 17:27 |
| *** thedodd has joined #openstack-nova | 17:28 | |
| dfisher | dansmith: moving all of our functionality to libvirt would (likely) be impossible. We'd end up turning libvirt into a simple proxy (take some libvirt call and transform it into the proper zones call). | 17:29 |
| dfisher | not that modifying libvirt is impossible. getting that level of buyoff from the internal teams would be impossible. | 17:29 |
| dansmith | dfisher: that's pretty much exactly what libvirt is | 17:29 |
| *** aix has quit IRC | 17:29 | |
| mriedem | dfisher: how accurate is this anymore? https://www.redhat.com/archives/libvir-list/2008-November/msg00007.html | 17:30 |
| dansmith | dfisher: I get that the shortest path to glory is dumping a bunch of code on us, but that's not really a thing that matters to me :) | 17:30 |
| dfisher | xVM died when Sun got bought by Oracle. | 17:30 |
| dansmith | mriedem: xvm is xen.. a different thing | 17:30 |
| dfisher | dansmith: please don't think we're dumping and running here. | 17:30 |
| dfisher | the LDOMs teams have largely moved away from libvirt AFAIK | 17:31 |
| dansmith | dfisher: the reality is even if you're going to stick around, you're very likely to duplicate a lot of code that already exists in the other virt drivers, which wouldn't be duplicated if we accessed you via libvirt | 17:31 |
| *** otter768 has quit IRC | 17:31 | |
| *** hitalia has joined #openstack-nova | 17:32 | |
| mriedem | there is much beautiful xml to be had in this repo | 17:32 |
| dfisher | honestly, what's the difference? It's a ton of bolt-on code to libvirt or a driver which uses the proper Nova pieces in a self-contained format | 17:32 |
| dfisher | there's already all of the existing virt/driver.py code to support hyper-v, libvirt, vmware, etc. This is another driver in that fashion | 17:33 |
| dansmith | dfisher: it's a pretty big difference, especially since you said it would just be libvirt proxying through to your stuff | 17:33 |
| *** browne has joined #openstack-nova | 17:33 | |
| dansmith | dfisher: right, and many of those duplicate a lot of each other | 17:34 |
| dansmith | dfisher: lots of image management stuff, some networking bits, etc | 17:34 |
| dfisher | i know we looked at the libvirt angle when we started this back during Grizzly and it was deemed to be unfeasible for us. | 17:34 |
| dansmith | dfisher: fwiw, everyone says that :) | 17:34 |
| dfisher | what's the point then in having a modular framework then? Your framework works *great* and we're able to simply drop in and go. | 17:35 |
| ndipanov | dansmith, super easy | 17:35 |
| ndipanov | https://review.openstack.org/#/c/224013/4 | 17:35 |
| mriedem | i for one love the plethora of virt type and version conditionals all over the libvirt driver in nova | 17:35 |
| ijw | mriedem, you have no sense of adventure | 17:35 |
| dansmith | dfisher: our driver model is not very concrete | 17:35 |
| dansmith | dfisher: I do get that plugging in at our level is less work for you :) | 17:36 |
| ijw | dansmith: you say that but it's not terribly variable at this point. It's just not committed as a stable, backward compatible interdface | 17:36 |
| dansmith | ijw: our virt interface has very very little abstraction between the compute manager and the virt driver, IMHO | 17:37 |
| *** subscope has joined #openstack-nova | 17:37 | |
| *** breitz has joined #openstack-nova | 17:37 | |
| ijw | true nuff | 17:38 |
| dhellmann | johnthetubaguy, mriedem: what did you decide about a novaclient release today? should I wait to create your stable/liberty branch? | 17:39 |
| dfisher | so, what about the other non-libvirt drivers? Are you pushing those owners/vendors to adopt libvirt? | 17:39 |
| mriedem | dhellmann: i was waiting for 225821 to merge | 17:39 |
| mriedem | it's in the gate now | 17:39 |
| dansmith | dfisher: no, but other people that have shown up since have, yes, which is why we consume lxc from libvirt for example | 17:39 |
| dhellmann | mriedem: ok, thanks | 17:39 |
| dfisher | the major difference here is those other drivers are typically non-linux based. | 17:40 |
| dfisher | like solaris zones | 17:40 |
| dfisher | whereas lxc is not. | 17:40 |
| dansmith | I think you lost me, but when I was talking to the oracle guy in HK about all your virtualization things, they all sounded nearly equivalent in model to linux containers or qemu at one level or another | 17:41 |
| doffm | mriedem: In https://review.openstack.org/#/c/223751/ I can only find one that is still failing. 1249065. That started failing again today. I have a feeling that last week it had been 10 days since that last failure. | 17:42 |
| doffm | mriedem: Is there anything we do to indicate that this test is intermittent / shouldn't be removed for now? | 17:42 |
| dfisher | dansmith: I'll work with my team on a Mitaka spec for this. We can hopefully address the lack of libvirt usage and move this forward. | 17:43 |
| *** garyk has joined #openstack-nova | 17:43 | |
| mriedem | doffm: sometimes we put a comment in the query yaml file, but it's probably not necessary right now, it could be failing on stable/kilo for example and we just haven't backported a fix there | 17:44 |
| mriedem | doffm: the other was http://status.openstack.org/elastic-recheck/index.html#1461172 | 17:44 |
| *** lpetrut has quit IRC | 17:44 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: rebuild: RPC sends additional args and claims are done https://review.openstack.org/200485 | 17:45 |
| dansmith | ndipanov: yep, got it | 17:45 |
| ndipanov | johnthetubaguy, dansmith ^ also I updated the patch and basically followed alex_xu 's advice | 17:45 |
| dansmith | ndipanov: okay I have a call starting soon, but I'll put it in my tab queue | 17:45 |
| doffm | mriedem: I'm going blind. Ok, will remove both from patch. | 17:46 |
| *** harlowja has joined #openstack-nova | 17:46 | |
| mriedem | doffm: on bug 1249065 - it was check queue and experimental, so might just be someone's bad change | 17:48 |
| openstack | bug 1249065 in OpenStack Compute (nova) "Nova throws 400 when attempting to add floating ip (instance.info_cache.network_info is empty)" [Medium,Fix released] https://launchpad.net/bugs/1249065 - Assigned to Matt Riedemann (mriedem) | 17:48 |
| mriedem | doffm: yeah, there were 2 changes that hit that bug, one was just a busted change that failed everything. the other was an ironic patch and the bug showed up on the kilo side of the upgrade job, so that makes sense. there was a change for that bug which we didn't backport to kilo b/c of rpc api version changes | 17:50 |
| mriedem | so leaving that query around for now is ok | 17:50 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: rebuild: RPC sends additional args and claims are done https://review.openstack.org/200485 | 17:53 |
| *** Apoorva_ has joined #openstack-nova | 17:57 | |
| *** baoli has quit IRC | 17:59 | |
| *** Apoorva has quit IRC | 18:00 | |
| *** baoli has joined #openstack-nova | 18:01 | |
| *** moshele has quit IRC | 18:04 | |
| *** tonytan4ever has joined #openstack-nova | 18:05 | |
| *** fawadkhaliq has joined #openstack-nova | 18:11 | |
| openstackgerrit | Andreas Jaeger proposed openstack/nova: Fix capitalization of IP https://review.openstack.org/225923 | 18:14 |
| *** tsekiyam_ has joined #openstack-nova | 18:15 | |
| *** fawadkhaliq has quit IRC | 18:15 | |
| *** tsekiyama has quit IRC | 18:18 | |
| *** wverdugo500 has joined #openstack-nova | 18:19 | |
| *** tsekiyam_ has quit IRC | 18:19 | |
| *** electrocucaracha has joined #openstack-nova | 18:21 | |
| *** angdraug has joined #openstack-nova | 18:23 | |
| *** fawadkhaliq has joined #openstack-nova | 18:25 | |
| openstackgerrit | garyk proposed openstack/nova: libvirt: add unit tests for the designer utility methods https://review.openstack.org/225975 | 18:27 |
| *** Apoorva_ has quit IRC | 18:30 | |
| *** tonytan4ever has quit IRC | 18:30 | |
| *** Apoorva has joined #openstack-nova | 18:31 | |
| *** pixelb has quit IRC | 18:31 | |
| *** VW has quit IRC | 18:35 | |
| *** electrocucaracha has quit IRC | 18:36 | |
| *** vilobhmm_11 has joined #openstack-nova | 18:36 | |
| *** VW has joined #openstack-nova | 18:36 | |
| *** vladikr has quit IRC | 18:40 | |
| *** VW has quit IRC | 18:41 | |
| *** fawadkhaliq has quit IRC | 18:41 | |
| *** su_zhang has quit IRC | 18:45 | |
| *** dfisher has left #openstack-nova | 18:48 | |
| *** su_zhang has joined #openstack-nova | 18:49 | |
| *** tonytan4ever has joined #openstack-nova | 18:49 | |
| *** renatoarmani has quit IRC | 18:49 | |
| *** rushil has quit IRC | 18:55 | |
| *** sudipto has quit IRC | 18:57 | |
| *** rushil has joined #openstack-nova | 18:57 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: live-migration: Logs exception if operation failed https://review.openstack.org/217822 | 18:58 |
| *** vladikr has joined #openstack-nova | 18:58 | |
| *** su_zhang has quit IRC | 19:00 | |
| *** vilobhmm_11 has quit IRC | 19:00 | |
| *** vilobhmm_11 has joined #openstack-nova | 19:00 | |
| *** Marga_ has quit IRC | 19:00 | |
| *** baoli has quit IRC | 19:02 | |
| *** gabriel-bezerra has quit IRC | 19:02 | |
| *** nllrte has joined #openstack-nova | 19:02 | |
| *** nllrte has left #openstack-nova | 19:03 | |
| *** matrohon has joined #openstack-nova | 19:03 | |
| *** annegentle has quit IRC | 19:03 | |
| *** bradjones has quit IRC | 19:04 | |
| mriedem | apmelton: i was going to work on the functional test changes for https://review.openstack.org/#/c/216895/ unless you have that worked out locally | 19:09 |
| *** gabriel-bezerra has joined #openstack-nova | 19:10 | |
| apmelton | mriedem: sounds good, I really appreciate it | 19:10 |
| *** ociuhandu has quit IRC | 19:10 | |
| apmelton | and no, I don't have anything worked out locally | 19:10 |
| *** lucasagomes is now known as lucas-dinner | 19:12 | |
| claudiub | mriedem: hi. you've amended one of my patches and you removed the exception's message from the log: https://review.openstack.org/#/c/217822/ | 19:13 |
| mriedem | claudiub: it'll be in the LOG.exception trace right? | 19:13 |
| *** Piet has quit IRC | 19:14 | |
| mriedem | claudiub: see everywhere else that LOG.exception is used in the compute manager | 19:14 |
| claudiub | so LOG.exception logs the original exception's message? | 19:15 |
| *** ijw has quit IRC | 19:16 | |
| mriedem | "Exception info is always added to the logging message." | 19:16 |
| mriedem | yes | 19:16 |
| erlon | ndipanov: ping | 19:17 |
| dansmith | ndipanov: still around? | 19:18 |
| mriedem | claudiub: http://paste.openstack.org/show/473153/ | 19:19 |
| claudiub | mriedem: yeah, just tested myself. sorry, my bad. | 19:20 |
| mriedem | np | 19:20 |
| *** electrocucaracha has joined #openstack-nova | 19:21 | |
| *** ZZelle_ has joined #openstack-nova | 19:24 | |
| *** obedmr has quit IRC | 19:25 | |
| *** VW has joined #openstack-nova | 19:27 | |
| *** otter768 has joined #openstack-nova | 19:28 | |
| *** obedmr has joined #openstack-nova | 19:28 | |
| *** ociuhandu has joined #openstack-nova | 19:29 | |
| *** otter768 has quit IRC | 19:32 | |
| *** breitz has quit IRC | 19:35 | |
| *** Marga_ has joined #openstack-nova | 19:35 | |
| *** hitalia has quit IRC | 19:36 | |
| *** breitz has joined #openstack-nova | 19:37 | |
| *** ctrath has quit IRC | 19:44 | |
| erlon | ndipanov: all: I'm trying to add a fix on horizon, to allow volume type selection in the 'Boot from image(Creates a new volume)', this is handled in os-volumes_boot in nova, but, along the code path there are several mentions to removing this in version 2 of RPC API. Do you think this is an acceptable fix? | 19:44 |
| *** Marga_ has quit IRC | 19:46 | |
| *** Marga_ has joined #openstack-nova | 19:46 | |
| *** ctrath has joined #openstack-nova | 19:46 | |
| mriedem | apmelton: so i think the problem in the functional tests is it's using the vlanmanager via a fixture | 19:48 |
| mriedem | apmelton: so i tries to ensure a fake vlan100 interface | 19:48 |
| mriedem | *it tries | 19:48 |
| mriedem | so i guess we just have to stub out the call to netifaces.ifaddresses.... | 19:50 |
| *** ijw has joined #openstack-nova | 19:51 | |
| *** baoli has joined #openstack-nova | 19:52 | |
| *** baoli has quit IRC | 19:53 | |
| *** baoli has joined #openstack-nova | 19:53 | |
| apmelton | mriedem: ah, interesting | 19:53 |
| *** VW_ has joined #openstack-nova | 19:53 | |
| apmelton | mriedem: where's that fixture get defined? | 19:53 |
| mriedem | nova.test | 19:53 |
| mriedem | https://github.com/openstack/nova/blob/master/nova/test.py#L74 | 19:54 |
| *** ijw has quit IRC | 19:56 | |
| *** VW has quit IRC | 19:57 | |
| *** VW has joined #openstack-nova | 19:57 | |
| apmelton | mriedem: if it's using the actual VlanManager does that mean it's actually creating those interfaces? | 19:57 |
| openstackgerrit | Andreas Jaeger proposed openstack/nova: Fix capitalization of IP https://review.openstack.org/225923 | 19:57 |
| *** VW has quit IRC | 19:57 | |
| *** Piet has joined #openstack-nova | 19:57 | |
| *** VW_ has quit IRC | 19:57 | |
| *** VW has joined #openstack-nova | 19:58 | |
| mriedem | apmelton: something must fake that stuff out | 19:58 |
| *** hitalia has joined #openstack-nova | 19:59 | |
| *** e0ne has quit IRC | 19:59 | |
| *** e0ne has joined #openstack-nova | 20:00 | |
| apmelton | mriedem: https://github.com/openstack/nova/blob/74f0b29233c26bb747aded13d069247a50df8419/nova/network/linux_net.py#L1262-L1268 | 20:00 |
| mriedem | apmelton: yeah, https://github.com/openstack/nova/blob/74f0b29233c26bb747aded13d069247a50df8419/nova/tests/unit/conf_fixture.py#L48 | 20:05 |
| mriedem | set in the ConfFixture | 20:05 |
| mriedem | which is set for all nova tests here https://github.com/openstack/nova/blob/74f0b29233c26bb747aded13d069247a50df8419/nova/test.py#L225: | 20:06 |
| mriedem | oops https://github.com/openstack/nova/blob/74f0b29233c26bb747aded13d069247a50df8419/nova/test.py#L225 | 20:06 |
| apmelton | interesting | 20:06 |
| mriedem | so if i just patch netifaces.ifaddresses in the SampleNetworks fixture, things are happy | 20:07 |
| apmelton | should be | 20:07 |
| mriedem | so maybe rather than copy/paste this pattern everywhere https://review.openstack.org/#/c/216895/1/nova/tests/unit/network/test_manager.py | 20:07 |
| mriedem | we can just add that into linux_net like _execute is done | 20:08 |
| mriedem | i'll take a look at doing that globally if CONF.fake_network | 20:08 |
| *** emagana_ has joined #openstack-nova | 20:08 | |
| *** jichen has quit IRC | 20:08 | |
| *** emagana has quit IRC | 20:08 | |
| *** su_zhang has joined #openstack-nova | 20:09 | |
| *** tongli has quit IRC | 20:09 | |
| *** rushil has quit IRC | 20:09 | |
| apmelton | sounds good mriedem, thanks for walking me through the test fixtures! | 20:09 |
| *** annegentle has joined #openstack-nova | 20:10 | |
| *** lpetrut has joined #openstack-nova | 20:13 | |
| *** su_zhang has quit IRC | 20:13 | |
| *** smatzek has quit IRC | 20:14 | |
| *** baoli has quit IRC | 20:20 | |
| *** thangp has quit IRC | 20:21 | |
| *** e0ne has quit IRC | 20:29 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova: WIP: Disable security group tests in tempest-dsvm-cells-rc using a config option https://review.openstack.org/226043 | 20:30 |
| *** Apoorva_ has joined #openstack-nova | 20:30 | |
| mriedem | ccarmack: that series can all be closes-bug ^ | 20:31 |
| mriedem | since each change is in a repo that is associated with that bug | 20:31 |
| *** tonytan4ever has quit IRC | 20:32 | |
| ccarmack | mriedem: so if a fix crosses multiple repos, its close-bug on each repo commit? | 20:32 |
| mriedem | ccarmack: if it closes the work for the bug on that project, then yes | 20:33 |
| mriedem | if you have mulitiple changes for the same bug in the same project, then partial-bug or related-bug is OK until the final change in that project which actually closes the bug for that project, | 20:33 |
| mriedem | but i don't think that's the case here | 20:33 |
| *** Apoorva has quit IRC | 20:34 | |
| *** su_zhang has joined #openstack-nova | 20:37 | |
| *** ijw has joined #openstack-nova | 20:38 | |
| ccarmack | mriedem: ok, btw some of the scenario tests that create security groups are blacklisted for cells because neutron is disabled. Ie. if I un-blacklist them, I get "No floating IP pool found" error. So even if I add a skip check for security groups in tempest, the test is still blacklisted for the floating ip pool problem. | 20:39 |
| *** markvoelker has quit IRC | 20:41 | |
| *** rlrossit has left #openstack-nova | 20:43 | |
| mriedem | ccarmack: then devstack should set service_available.neutron=False for cells | 20:43 |
| mriedem | you could make that a dependent change for your devstack change | 20:44 |
| *** tonytan4ever has joined #openstack-nova | 20:44 | |
| mriedem | ccarmack: you shouldn't be trying to test cells with neutron anyway | 20:44 |
| mriedem | the gate-tempest-dsvm-cells job runs with nova-network | 20:44 |
| ccarmack | mriedem: yes, nova-network is used for the cells gate -- http://logs.openstack.org/04/225204/1/experimental/gate-tempest-dsvm-cells/34812c8/logs/localrc.txt.gz | 20:45 |
| bauzas | ccarmack: I'm about to look at reducing the tempest rc file by writing an etherpad, could you just lemme know which changes you're writing now for that? | 20:45 |
| ccarmack | mriedem: I think that's why I get "no floating IP pool found" when I unblacklist some of the tests for cells | 20:46 |
| mriedem | bauzas: just follow the dependency chain here https://review.openstack.org/#/c/226043/ | 20:47 |
| mriedem | goes nova -> devstack -> tempest | 20:47 |
| mriedem | ccarmack: yeah, so if you're doing local testing, don't use neutron | 20:47 |
| bauzas | mriedem: okay, will look at that tomorrow | 20:48 |
| ccarmack | mriedem: its failing with the experimental cells CI | 20:48 |
| ccarmack | I stopped trying to run this locally | 20:48 |
| bauzas | mriedem: my goal is to look at Tempest and provide some devref for helping some new contributors to have low-hanging-bugs | 20:48 |
| bauzas | by creating a bug for each of those | 20:48 |
| bauzas | and using an etherpad for tracking | 20:48 |
| mriedem | bauzas: there is the cells testing etherpad that alaski had | 20:49 |
| mriedem | i'd just track there probably | 20:49 |
| bauzas | mriedem: yeah, was thinking about that one | 20:49 |
| mriedem | https://etherpad.openstack.org/p/nova-cells-testing | 20:49 |
| bauzas | yup, know it :) | 20:49 |
| mriedem | could probably delete most of that and just point at the rc file in nova's tree now | 20:49 |
| bauzas | yup... | 20:50 |
| mriedem | ccarmack: https://review.openstack.org/#/c/225199/3/tempest/scenario/test_stamp_pattern.py | 20:50 |
| mriedem | you can't test with test_stamp_pattern | 20:50 |
| mriedem | ccarmack: # @decorators.skip_because(bug="1205344") | 20:50 |
| ccarmack | yes, I saw that, ... mriedem | 20:50 |
| mriedem | there are reasons why test_stamp_pattern is skipped | 20:50 |
| mriedem | not cells reasons | 20:50 |
| mriedem | big fat ugly reasons https://review.openstack.org/#/c/218355/ | 20:51 |
| mriedem | that kashyap hasn't fixed yet | 20:51 |
| mriedem | :P | 20:51 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Add some devref for AZs https://review.openstack.org/223802 | 20:51 |
| *** pumaranikar has joined #openstack-nova | 20:52 | |
| ccarmack | mriedem: but if you look at test_minimum_basic_scenario, thats in the blacklist. I tried to unblacklist, thinking it doesn't even create a security group, but then I go "no floating IP pool found" | 20:53 |
| ccarmack | mriedem: So adding the skip test is kind of pointless if the test won't run due to neutron is disabled (ie for tests that do create a security group) | 20:54 |
| mriedem | ccarmack: in the nova change with the rc file, you enabled test_volume_boot_pattern | 20:54 |
| mriedem | not test_minimum_basic_scenario | 20:54 |
| ccarmack | mriedem: yes I'm seeing if that runs and I can remove the blacklist | 20:54 |
| *** matrohon has quit IRC | 20:55 | |
| mriedem | note that in tempest, things like this: | 20:55 |
| mriedem | @test.services('compute', 'volume', 'image', 'network') | 20:55 |
| mriedem | that means it's going to check that service_available config option and skip those tests if one of those services aren't available | 20:55 |
| mriedem | ^ is in test_minimum_basic_scenario, so that won't run | 20:55 |
| mriedem | in gate-tempest-dsvm-cells anyway | 20:56 |
| mriedem | network is confusing b/c it really means neutron | 20:56 |
| ccarmack | mriedem: oh, thanks for the heads up. so many ways to block a test | 20:56 |
| *** pumaranikar has quit IRC | 20:57 | |
| *** aix has joined #openstack-nova | 20:57 | |
| *** pumaranikar has joined #openstack-nova | 20:57 | |
| ccarmack | mriedem: my question is, should I add the security group skip checks even if the test is not running for other reasons? | 20:57 |
| mriedem | ccarmack: yes, because neutron has the port_security_enabled config option | 20:58 |
| mriedem | which can be FALSE | 20:58 |
| mriedem | sorry, False - didn't mean to shout | 20:58 |
| *** angdraug has quit IRC | 20:59 | |
| mriedem | or it's a flag on the network provider, i'm not sure exactly, but it's a thing that basically says no security groups | 21:00 |
| mriedem | sc68cal: could explain it i'm sure | 21:00 |
| * sc68cal reads scrollback | 21:00 | |
| mriedem | "create_port:port_security_enabled": "rule:admin_or_network_owner or rule:context_is_advsvc", | 21:00 |
| mriedem | looks like it's a policy thing | 21:00 |
| *** dustins has quit IRC | 21:01 | |
| *** andrewbogott has joined #openstack-nova | 21:01 | |
| mriedem | once rax upstreams their neutron driver then people can use cells with neutron :) | 21:01 |
| mriedem | alaski: ^ something like that, right? | 21:01 |
| mriedem | btw, what does cern/godaddy use for network? nova-net? | 21:02 |
| alaski | mriedem: heh, something like that | 21:02 |
| sc68cal | mriedem: nova-net | 21:02 |
| alaski | yeah, I think everyone else is nova-net with cells | 21:02 |
| *** raildo is now known as raildo-afk | 21:02 | |
| sc68cal | mriedem: RAX upstreaming their neutron driver won't add neutron-cells support by itself | 21:02 |
| sc68cal | mriedem: mlavelle is supposed to be working on neutron cells support | 21:02 |
| *** vilobhmm_11 has quit IRC | 21:03 | |
| *** pumaranikar has quit IRC | 21:03 | |
| *** lpetrut has quit IRC | 21:03 | |
| sc68cal | mriedem: he works at RAX, but it's a separate effort | 21:03 |
| *** ijw has quit IRC | 21:03 | |
| mriedem | i thought there would be magical leprechauns or something once the driver was there? easy button? | 21:03 |
| mriedem | c'mon | 21:03 |
| sc68cal | mriedem: :) | 21:03 |
| ccarmack | mriedem: I'm not clear if adding the security group skip checks is worth it since some test are not running due to neutron is disabled, or some bug is blocking the test | 21:03 |
| *** rfolco has quit IRC | 21:04 | |
| alaski | the reason cells works for us is that we don't expose the networking api on its own. that's the challenging part we haven't solved yet, getting the two to interact properly | 21:04 |
| mriedem | ccarmack: ignore cells when doing the tempest change | 21:04 |
| *** angdraug has joined #openstack-nova | 21:04 | |
| mriedem | ccarmack: you can also run nova-network w/o cells and no security groups if you want | 21:04 |
| mriedem | mordred: jeblair: is community infra running neutron or nova-net? | 21:04 |
| andrewbogott | Before I log a bug… am I totally misunderstanding how "openstack security group rule delete” should work? It seems like it must take two positionals (group, rule) but it only takes one, and then fails... | 21:04 |
| *** vilobhmm_11 has joined #openstack-nova | 21:05 | |
| mriedem | mordred: jeblair: or depends on what the backing hp/rax clouds are using? | 21:05 |
| mordred | aroo? | 21:05 |
| mordred | mriedem: we do not run either | 21:05 |
| mordred | mriedem: yet | 21:05 |
| mordred | mriedem: but infra-cloud will run neutron | 21:05 |
| mriedem | yeah, i walked it back | 21:05 |
| mordred | :) | 21:05 |
| mriedem | neutron with security groups? | 21:05 |
| mordred | mriedem: fwiw, all of our cloud providers wuse neutron | 21:05 |
| mordred | we do not have any use for securityu-groups no | 21:05 |
| mriedem | ccarmack: ^ that's your answer then | 21:06 |
| mordred | but at least one of our public clouds does implement them | 21:06 |
| *** eharney has quit IRC | 21:06 | |
| mriedem | andrewbogott: is that a python-openstackclient question? | 21:06 |
| mordred | and wejust have to remeber to go set them to accept al as afirst step in using that cloud :) | 21:06 |
| *** burgerk has quit IRC | 21:06 | |
| andrewbogott | mriedem: yes… is that its own #room these days? | 21:06 |
| mriedem | andrewbogott: i just find dtroyer or stevemar on -dev | 21:07 |
| stevemar | andrewbogott: yes, #openstack-sdks | 21:07 |
| mriedem | or there | 21:07 |
| stevemar | ;) | 21:07 |
| andrewbogott | ok, thanks | 21:07 |
| andrewbogott | There is /also/ a nova bug underneath that command, but I have to get a proper demonstration of the bug first :) | 21:08 |
| mriedem | nova does'nt have bugs | 21:10 |
| mriedem | sorry | 21:10 |
| stevemar | mriedem: truth | 21:10 |
| mriedem | except what this fixes https://review.openstack.org/#/c/223203/ | 21:11 |
| * mriedem wishes some brave cores would approve that | 21:11 | |
| *** r-daneel has quit IRC | 21:14 | |
| andrewbogott | mriedem: if you feel like indulging me, a simple question: Can you delete security group rules, at all? I can’t because "Out of range value for column 'deleted' at row 1" | 21:15 |
| andrewbogott | Which I fear is going to turn out to be a schema-migration issue and hence impossible to reproduce :( | 21:15 |
| mriedem | umm | 21:15 |
| mriedem | i assume tempest has a test to add/remove secgroup rules | 21:15 |
| andrewbogott | I’d think | 21:15 |
| claudiub | neutron security groups? | 21:15 |
| andrewbogott | nova | 21:16 |
| mriedem | https://github.com/openstack/tempest/blob/master/tempest/api/compute/security_groups/test_security_group_rules.py | 21:16 |
| mordred | andrewbogott: yes | 21:17 |
| mordred | https://github.com/openstack-infra/shade/blob/master/shade/__init__.py#L3044-L3056 too | 21:17 |
| andrewbogott | So most likely my db is in an unexpected state due to a broken migration script :( | 21:17 |
| mordred | andrewbogott: yes | 21:17 |
| andrewbogott | ok :( Time to read a couple year’s worth of commit logs I guess | 21:18 |
| *** ijw has joined #openstack-nova | 21:19 | |
| *** stevemar has quit IRC | 21:19 | |
| *** stevemar has joined #openstack-nova | 21:20 | |
| *** markvoelker has joined #openstack-nova | 21:20 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: net: explicitly set mac on linux bridge https://review.openstack.org/216895 | 21:22 |
| mriedem | apmelton: ^ that should do it | 21:22 |
| *** stevemar has quit IRC | 21:24 | |
| *** achanda has quit IRC | 21:24 | |
| *** kylek3h has quit IRC | 21:26 | |
| *** bnemec has quit IRC | 21:27 | |
| mriedem | alaski: bauzas: in case you haven't seen this https://etherpad.openstack.org/p/PAO-LDT-cells-patches | 21:27 |
| mriedem | shows up on the operators ML | 21:27 |
| * bauzas looking | 21:28 | |
| *** promulo has joined #openstack-nova | 21:28 | |
| bauzas | mriedem: I know that CERN runs a few hacks | 21:28 |
| bauzas | mriedem: mostly for sync'ing flavors and other stuff | 21:28 |
| mriedem | looks like this is mostly NeCTAR | 21:28 |
| alaski | hadn't seen that, thanks | 21:28 |
| *** otter768 has joined #openstack-nova | 21:28 | |
| mriedem | http://lists.openstack.org/pipermail/openstack-operators/2015-September/008171.html | 21:29 |
| bauzas | mriedem: alaski: belmoreira provided some github links for their sutff | 21:29 |
| alaski | godaddy seems to be pushing up a lot of their stuff | 21:29 |
| *** thorst has quit IRC | 21:29 | |
| alaski | stuff from nectar I mean | 21:29 |
| bauzas | and I guess they probably unattended the PAO midcycle, saw they're running for an EU Ops meetup | 21:29 |
| mriedem | i was wondering if those were the same | 21:29 |
| bauzas | lemme find the CERN githubs | 21:30 |
| mriedem | https://github.com/cernops/nova ? | 21:30 |
| bauzas | https://github.com/cernops/nova-flavor-create that kind of stuff | 21:31 |
| bauzas | but they have other stuff | 21:31 |
| mriedem | yeah | 21:32 |
| mriedem | db purge | 21:32 |
| mriedem | quota sync | 21:32 |
| *** alexpilotti has quit IRC | 21:32 | |
| bauzas | https://github.com/cernops?utf8=%E2%9C%93&query=nova | 21:32 |
| bauzas | yeah that | 21:32 |
| mriedem | some of that i think we're trying to work into nova-manage | 21:32 |
| mriedem | like db purge | 21:32 |
| mriedem | and i thought there was some quota sync spec for liberty | 21:32 |
| mriedem | http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/refresh-quotas-usage.html | 21:33 |
| *** Piet has quit IRC | 21:33 | |
| mriedem | heh, code never landed https://blueprints.launchpad.net/nova/+spec/refresh-quotas-usage | 21:33 |
| bauzas | mmm | 21:33 |
| *** promulo has left #openstack-nova | 21:33 | |
| bauzas | romain hardouin | 21:33 |
| bauzas | I know that guy | 21:33 |
| *** otter768 has quit IRC | 21:33 | |
| mriedem | they landed a spec but no code | 21:33 |
| mriedem | :( | 21:33 |
| mriedem | for once it's not our fault! | 21:33 |
| bauzas | IIRC, he works for cloudwatt | 21:33 |
| bauzas | you know, our French Amazon :p | 21:34 |
| bauzas | that has been deadshot | 21:34 |
| bauzas | headshot | 21:34 |
| *** promulo has joined #openstack-nova | 21:35 | |
| bauzas | our "sovereign cloud" ha ha | 21:36 |
| *** promulo has left #openstack-nova | 21:36 | |
| bauzas | mriedem: any idea if the Nectar patches are Kilo or later ? | 21:40 |
| bauzas | mriedem: nvm https://github.com/NeCTAR-RC/nova/blob/nectar/juno/setup.cfg#L3 | 21:40 |
| mriedem | at least they keep their patches in public github | 21:41 |
| bauzas | yeah | 21:41 |
| mriedem | https://github.com/NeCTAR-RC/nova/blob/nectar/kilo/setup.cfg | 21:41 |
| bauzas | I'm just worried by the rebase | 21:41 |
| mriedem | that's their problem | 21:41 |
| *** bnemec has joined #openstack-nova | 21:41 | |
| mriedem | heh, forked from comstud-nova https://github.com/comstud/nova | 21:42 |
| bauzas | yay | 21:42 |
| bauzas | but it seems they rebased on top of it | 21:42 |
| mriedem | yes | 21:42 |
| mriedem | version = 2013.2 | 21:42 |
| bauzas | because comstud's repo is older | 21:42 |
| mriedem | i would hope so | 21:42 |
| mriedem | all of this comstud talk is bringing up painful memories and now i have to leavce | 21:43 |
| bauzas | so I guess they forked from it and then rebased from somewhere else | 21:43 |
| mriedem | *leave | 21:43 |
| bauzas | hah | 21:43 |
| *** mriedem is now known as mriedem_away | 21:43 | |
| *** achanda has joined #openstack-nova | 21:44 | |
| *** achanda has quit IRC | 21:44 | |
| *** achanda has joined #openstack-nova | 21:44 | |
| *** jecarey has quit IRC | 21:46 | |
| *** ctrath has quit IRC | 21:49 | |
| *** eglynn_ has quit IRC | 21:50 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: Make instance snapshots to respect the force_raw_images setting https://review.openstack.org/226069 | 21:51 |
| *** alex_klimov has joined #openstack-nova | 21:56 | |
| *** shihanzhang has quit IRC | 21:58 | |
| *** lalitd has quit IRC | 22:03 | |
| *** nic has quit IRC | 22:03 | |
| *** nic has joined #openstack-nova | 22:04 | |
| *** jaypipes has quit IRC | 22:05 | |
| *** alejandrito has quit IRC | 22:05 | |
| *** breitz has quit IRC | 22:08 | |
| *** breitz has joined #openstack-nova | 22:08 | |
| *** mdrabe_ has quit IRC | 22:09 | |
| *** asettle has joined #openstack-nova | 22:12 | |
| *** burt has quit IRC | 22:13 | |
| *** leitan has quit IRC | 22:13 | |
| *** yamahata has joined #openstack-nova | 22:15 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: WIP: Hyper-V: adds os-win library https://review.openstack.org/212102 | 22:15 |
| openstackgerrit | Claudiu Belu proposed openstack/nova: WIP: Hyper-V: removes *Utils modules and unit tests https://review.openstack.org/215917 | 22:15 |
| *** stevemar has joined #openstack-nova | 22:16 | |
| *** yuner027 has quit IRC | 22:19 | |
| *** yuner027 has joined #openstack-nova | 22:19 | |
| *** electrocucaracha has quit IRC | 22:20 | |
| *** stevemar has quit IRC | 22:20 | |
| *** tonytan4ever has quit IRC | 22:21 | |
| *** bswartz has joined #openstack-nova | 22:22 | |
| *** Piet has joined #openstack-nova | 22:23 | |
| *** xyang1 has quit IRC | 22:24 | |
| *** alex_klimov has quit IRC | 22:26 | |
| *** su_zhang has quit IRC | 22:27 | |
| *** su_zhang has joined #openstack-nova | 22:27 | |
| *** alexpilotti has joined #openstack-nova | 22:31 | |
| *** VW has quit IRC | 22:33 | |
| *** VW has joined #openstack-nova | 22:33 | |
| *** diogogmt has quit IRC | 22:35 | |
| *** su_zhang_ has joined #openstack-nova | 22:36 | |
| *** dims__ has quit IRC | 22:37 | |
| *** ccarmack has left #openstack-nova | 22:37 | |
| *** thedodd has quit IRC | 22:37 | |
| *** vladikr has quit IRC | 22:37 | |
| *** su_zhang has quit IRC | 22:38 | |
| *** VW has quit IRC | 22:38 | |
| *** ijuwang has quit IRC | 22:39 | |
| *** annegentle has quit IRC | 22:39 | |
| *** pixelb has joined #openstack-nova | 22:44 | |
| *** obedmr has quit IRC | 22:46 | |
| *** markvoelker has quit IRC | 22:48 | |
| *** ctrath has joined #openstack-nova | 22:56 | |
| *** electrocucaracha has joined #openstack-nova | 22:58 | |
| *** cfriesen has quit IRC | 22:58 | |
| *** electrocucaracha has quit IRC | 22:59 | |
| *** cfriesen has joined #openstack-nova | 23:00 | |
| *** derekh has joined #openstack-nova | 23:00 | |
| *** electrocucaracha has joined #openstack-nova | 23:00 | |
| *** aix has quit IRC | 23:02 | |
| *** su_zhang has joined #openstack-nova | 23:03 | |
| *** electrocucaracha has quit IRC | 23:04 | |
| *** mtanino has quit IRC | 23:05 | |
| *** hemna is now known as hemnafk | 23:06 | |
| *** su_zhang_ has quit IRC | 23:06 | |
| *** electrocucaracha has joined #openstack-nova | 23:06 | |
| *** vilobhmm_11 has quit IRC | 23:07 | |
| *** ZZelle_ has quit IRC | 23:08 | |
| *** su_zhang has quit IRC | 23:09 | |
| *** su_zhang has joined #openstack-nova | 23:10 | |
| *** emagana_ has quit IRC | 23:11 | |
| *** obedmr has joined #openstack-nova | 23:13 | |
| *** edmondsw has quit IRC | 23:13 | |
| *** alexpilotti has quit IRC | 23:16 | |
| *** ijw has quit IRC | 23:16 | |
| *** dims has joined #openstack-nova | 23:17 | |
| *** vilobhmm_11 has joined #openstack-nova | 23:21 | |
| *** dims has quit IRC | 23:23 | |
| *** jamielennox|away is now known as jamielennox | 23:23 | |
| *** dims has joined #openstack-nova | 23:23 | |
| *** cfriesen_ has joined #openstack-nova | 23:23 | |
| *** ijw has joined #openstack-nova | 23:23 | |
| *** cfriesen has quit IRC | 23:24 | |
| *** cfriesen__ has joined #openstack-nova | 23:25 | |
| *** hitalia has quit IRC | 23:26 | |
| *** obedmr has quit IRC | 23:27 | |
| *** IanGovett has quit IRC | 23:28 | |
| *** cfriesen_ has quit IRC | 23:28 | |
| *** obedmr has joined #openstack-nova | 23:28 | |
| *** asettle is now known as asettle-afk | 23:29 | |
| *** otter768 has joined #openstack-nova | 23:29 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: Fix model query for non-admin flavor lookups https://review.openstack.org/211809 | 23:30 |
| *** derekh has quit IRC | 23:32 | |
| *** obedmr has quit IRC | 23:33 | |
| *** otter768 has quit IRC | 23:34 | |
| *** dims has quit IRC | 23:39 | |
| *** dims has joined #openstack-nova | 23:40 | |
| *** Apoorva_ has quit IRC | 23:42 | |
| *** dims_ has joined #openstack-nova | 23:43 | |
| *** Apoorva has joined #openstack-nova | 23:43 | |
| *** dims has quit IRC | 23:45 | |
| *** lucas-dinner has quit IRC | 23:47 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova: WIP: Hyper-V: adds os-win library https://review.openstack.org/212102 | 23:48 |
| *** zhangjn has joined #openstack-nova | 23:54 | |
| *** zhangjn has quit IRC | 23:54 | |
| *** zhangjn has joined #openstack-nova | 23:54 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova: Fix model query for non-admin flavor lookups https://review.openstack.org/211809 | 23:55 |
| *** zhangjn has joined #openstack-nova | 23:55 | |
| *** zhangjn has quit IRC | 23:55 | |
| *** zhangjn has joined #openstack-nova | 23:56 | |
| *** markvoelker has joined #openstack-nova | 23:56 | |
| *** thorst has joined #openstack-nova | 23:57 | |
| *** electrocucaracha has quit IRC | 23:58 | |
| *** asettle-afk has quit IRC | 23:59 | |
| *** achanda_ has joined #openstack-nova | 23:59 | |
| *** aysyd has quit IRC | 23:59 | |
| *** EinstCrazy has joined #openstack-nova | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!