| *** EinstCrazy has joined #openstack-nova | 00:01 | |
| *** pratikma_ has quit IRC | 00:01 | |
| *** yamahata has quit IRC | 00:01 | |
| *** gzhai2 has joined #openstack-nova | 00:03 | |
| *** emagana has joined #openstack-nova | 00:06 | |
| *** gzhai1 has quit IRC | 00:06 | |
| *** EinstCrazy has quit IRC | 00:09 | |
| *** VW has joined #openstack-nova | 00:09 | |
| *** Daisy_ has joined #openstack-nova | 00:09 | |
| *** VW has quit IRC | 00:10 | |
| *** NightKhaos has quit IRC | 00:10 | |
| *** VW has joined #openstack-nova | 00:10 | |
| *** gzhai3 has joined #openstack-nova | 00:13 | |
| *** Daisy_ has quit IRC | 00:14 | |
| *** gzhai2 has quit IRC | 00:15 | |
| *** NightKhaos has joined #openstack-nova | 00:16 | |
| *** zzzeek has quit IRC | 00:17 | |
| *** electroc_ has quit IRC | 00:21 | |
| *** su_zhang has quit IRC | 00:23 | |
| *** dims has joined #openstack-nova | 00:24 | |
| *** ociuhandu has joined #openstack-nova | 00:27 | |
| *** angdraug has quit IRC | 00:28 | |
| *** diegows has quit IRC | 00:30 | |
| *** pratikmallya has quit IRC | 00:30 | |
| *** VW has quit IRC | 00:32 | |
| openstackgerrit | Sam Morrison proposed openstack/nova: Catch 404s from Neutron when floating IPs not enabled https://review.openstack.org/228676 | 00:32 |
|---|---|---|
| *** VW has joined #openstack-nova | 00:32 | |
| *** tsekiyama has quit IRC | 00:33 | |
| *** VW has quit IRC | 00:36 | |
| *** oomichi has joined #openstack-nova | 00:39 | |
| *** aix has quit IRC | 00:39 | |
| *** aix has joined #openstack-nova | 00:40 | |
| *** VW has joined #openstack-nova | 00:46 | |
| *** leseb_ has quit IRC | 00:46 | |
| *** gongysh has quit IRC | 00:47 | |
| *** edmondsw has quit IRC | 00:47 | |
| *** pratikmallya has joined #openstack-nova | 00:49 | |
| *** jmccrory has joined #openstack-nova | 00:49 | |
| *** VW has quit IRC | 00:51 | |
| *** VW has joined #openstack-nova | 00:51 | |
| *** smatzek has joined #openstack-nova | 00:52 | |
| *** aix has quit IRC | 00:52 | |
| *** leseb_ has joined #openstack-nova | 00:52 | |
| *** gongysh has joined #openstack-nova | 00:54 | |
| *** shakamunyi has joined #openstack-nova | 00:55 | |
| *** aix has joined #openstack-nova | 00:55 | |
| *** VW has quit IRC | 00:55 | |
| *** andreykurilin__ has quit IRC | 00:56 | |
| *** marzif has joined #openstack-nova | 00:56 | |
| *** penick has quit IRC | 00:57 | |
| *** baoli has joined #openstack-nova | 00:58 | |
| *** aysyd has quit IRC | 00:58 | |
| *** zhangjn has joined #openstack-nova | 01:01 | |
| *** EinstCrazy has joined #openstack-nova | 01:01 | |
| *** browne has quit IRC | 01:02 | |
| *** thorst has joined #openstack-nova | 01:02 | |
| *** r-daneel has quit IRC | 01:04 | |
| *** thorst has quit IRC | 01:05 | |
| *** dave-mcc_ has joined #openstack-nova | 01:07 | |
| *** mtanino has quit IRC | 01:09 | |
| *** dave-mccowan has quit IRC | 01:10 | |
| *** Daisy_ has joined #openstack-nova | 01:16 | |
| *** zhangjn_ has joined #openstack-nova | 01:17 | |
| *** Marga_ has quit IRC | 01:18 | |
| *** Marga_ has joined #openstack-nova | 01:18 | |
| *** zhangjn has quit IRC | 01:19 | |
| alex_xu | johnthetubaguy: thanks for the update | 01:21 |
| *** apoorvad_ has joined #openstack-nova | 01:27 | |
| *** apoorvad has quit IRC | 01:31 | |
| *** apoorvad_ has quit IRC | 01:31 | |
| tonyb | Anyone in stable-maint able to look at: https://review.openstack.org/#/c/219301/ ? | 01:32 |
| * tonyb eyes up mikal | 01:32 | |
| *** unicell has quit IRC | 01:33 | |
| *** tpeoples has joined #openstack-nova | 01:39 | |
| openstackgerrit | Alex Xu proposed openstack/nova: docs: rerganize the API concept docs https://review.openstack.org/226253 | 01:42 |
| oomichi | alex_xu: morning | 01:45 |
| alex_xu | oomichi: o ha yo | 01:46 |
| oomichi | alex_xu: about https://review.openstack.org/#/c/216746 | 01:46 |
| oomichi | alex_xu: nihao :) | 01:46 |
| alex_xu | oomichi: ha, we can talk with many language | 01:46 |
| oomichi | alex_xu: there is a similar patch https://review.openstack.org/#/c/228676 | 01:46 |
| oomichi | alex_xu: yeah, including python ;) | 01:46 |
| alex_xu | oomichi: heh | 01:47 |
| *** dustins has joined #openstack-nova | 01:47 | |
| oomichi | alex_xu: I prefer the later: https://review.openstack.org/#/c/228676 by comparing 216746 | 01:47 |
| *** otter768 has joined #openstack-nova | 01:47 | |
| alex_xu | oomichi: give me a sec think about | 01:47 |
| oomichi | alex_xu: that seems simple and consistent for the other nova api | 01:48 |
| oomichi | alex_xu: thanks:) | 01:48 |
| alex_xu | oomichi: yea, that sounds good pint, consistent for other api | 01:48 |
| alex_xu | s/pint/point | 01:48 |
| oomichi | alex_xu: the commit message seems too short anyways :) | 01:48 |
| alex_xu | oomichi: sounds good point, we better to match nova-network behavior | 01:49 |
| alex_xu | oomichi: as our network api support nova-network and neutron | 01:50 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] hacking check for contextlib.nested for py34 support https://review.openstack.org/228688 | 01:50 |
| oomichi | alex_xu: good feedback, thanks for taking a look :) | 01:50 |
| alex_xu | oomichi: np, you are welcome | 01:51 |
| *** otter768 has quit IRC | 01:52 | |
| alex_xu | oomichi: wait a sec | 01:52 |
| alex_xu | oomichi: have you seen this one https://review.openstack.org/#/c/200001/9/doc/source/architecture_evolution_summary.rst | 01:52 |
| alex_xu | oomichi: about line 237 and 238 | 01:52 |
| *** pixelbeat has quit IRC | 01:53 | |
| oomichi | alex_xu: "we should change nova-networks interface to match" | 01:54 |
| oomichi | alex_xu: is that? | 01:54 |
| alex_xu | oomichi: yea | 01:54 |
| *** changbl has joined #openstack-nova | 01:55 | |
| *** baoli has quit IRC | 01:55 | |
| oomichi | alex_xu: I'm not sure that means to be honest. | 01:56 |
| alex_xu | oomichi: yea...actually I'm doubt also, not sure we can match it, I more like we can remove all of those apis directly :) | 01:56 |
| oomichi | alex_xu: I see nova-networks API seems lacks of network information by comparing neutron | 01:56 |
| oomichi | alex_xu: I guess the lines means we need to add some params to nova-networks API, I guess | 01:57 |
| *** yamahata has joined #openstack-nova | 01:57 | |
| oomichi | alex_xu: ah, yeah. that is also nice direction. | 01:57 |
| alex_xu | oomichi: anyway, just match current behavior is not bad | 01:58 |
| alex_xu | oomichi: but look at this https://review.openstack.org/#/c/228676/1/nova/network/neutronv2/api.py again | 01:59 |
| alex_xu | oomichi: all the 404 return [], sounds too generic | 01:59 |
| oomichi | alex_xu: it is nice to convert 404 to [] on REST API layer? | 02:00 |
| *** Daisy_ has quit IRC | 02:00 | |
| alex_xu | oomichi: what about create and delete action? they also may return 404 | 02:01 |
| openstackgerrit | junxu proposed openstack/nova-specs: Allow force stop server in nova rest api https://review.openstack.org/195418 | 02:01 |
| alex_xu | oomichi: only index return [], others return 404? | 02:01 |
| *** Alexus has quit IRC | 02:01 | |
| *** Daisy_ has joined #openstack-nova | 02:01 | |
| oomichi | alex_xu: only index return [], you are right. | 02:02 |
| oomichi | alex_xu: but not sure 404 is the best for create | 02:02 |
| oomichi | alex_xu: because I am proposing https://review.openstack.org/#/c/221163/ to api-wg | 02:03 |
| oomichi | alex_xu: ;) | 02:03 |
| alex_xu | oomichi: ha, cool | 02:03 |
| alex_xu | oomichi: but 404 already existed in create, see line 127 https://review.openstack.org/#/c/216746/7/nova/api/openstack/compute/floating_ips.py | 02:04 |
| oomichi | alex_xu: I feel create api should not return 404 because 404 should be used for not-found resource url. | 02:04 |
| oomichi | alex_xu: yeah, nice point | 02:04 |
| oomichi | alex_xu: nova contains this kind of inconsistency | 02:04 |
| *** changbl has quit IRC | 02:04 | |
| alex_xu | oomichi: yea, that should be 400? | 02:05 |
| oomichi | alex_xu: then I hesitated -1ing for the reason. | 02:05 |
| oomichi | alex_xu: I guess so. | 02:05 |
| *** baoli has joined #openstack-nova | 02:06 | |
| alex_xu | oomichi: so finally, we return [] for index, 400 for others? | 02:06 |
| oomichi | alex_xu: after api-wg patch is merged, I can -1 the patch because we can get a guideline. | 02:06 |
| oomichi | alex_xu: 404 is also possible options at this time, because the api-wg patch is not merged yet. | 02:07 |
| oomichi | alex_xu: maybe I just want to notice the api-wg patch for you ;-) | 02:07 |
| alex_xu | oomichi: yea | 02:07 |
| oomichi | alex_xu: anyway, [] should be for index, I beleive | 02:08 |
| openstackgerrit | Tang Chen proposed openstack/nova-specs: Host maintenance mode https://review.openstack.org/228689 | 02:08 |
| alex_xu | oomichi: yea, I'm ok with it | 02:08 |
| alex_xu | oomichi: not ok, it is +1 | 02:09 |
| *** dims has quit IRC | 02:09 | |
| *** yamahata has quit IRC | 02:15 | |
| *** stevemar has joined #openstack-nova | 02:19 | |
| *** rushil has joined #openstack-nova | 02:25 | |
| *** achanda has joined #openstack-nova | 02:25 | |
| *** salv-orlando has joined #openstack-nova | 02:28 | |
| *** baoli has quit IRC | 02:32 | |
| *** zhenguo has joined #openstack-nova | 02:34 | |
| *** browne has joined #openstack-nova | 02:35 | |
| *** Marga__ has joined #openstack-nova | 02:36 | |
| *** salv-orlando has quit IRC | 02:37 | |
| *** Marga_ has quit IRC | 02:39 | |
| *** Marga__ has quit IRC | 02:40 | |
| *** oomichi has quit IRC | 02:46 | |
| *** baoli has joined #openstack-nova | 02:49 | |
| *** ijw has quit IRC | 02:51 | |
| *** haomaiwang has joined #openstack-nova | 02:51 | |
| *** asettle has quit IRC | 02:55 | |
| *** shyu has joined #openstack-nova | 02:56 | |
| *** baoli has quit IRC | 02:56 | |
| *** baoli has joined #openstack-nova | 02:57 | |
| *** ks1 has joined #openstack-nova | 02:59 | |
| *** haomaiwang has quit IRC | 03:01 | |
| *** asettle has joined #openstack-nova | 03:01 | |
| *** haomaiwang has joined #openstack-nova | 03:01 | |
| *** baoli has quit IRC | 03:03 | |
| *** VW has joined #openstack-nova | 03:05 | |
| *** achanda has quit IRC | 03:09 | |
| *** cfriesen has quit IRC | 03:09 | |
| *** dims has joined #openstack-nova | 03:09 | |
| *** rfolco has joined #openstack-nova | 03:10 | |
| *** dims has quit IRC | 03:16 | |
| *** bnemec has quit IRC | 03:17 | |
| *** jdurgin1 has joined #openstack-nova | 03:18 | |
| *** boris-42 has quit IRC | 03:19 | |
| *** hemna has joined #openstack-nova | 03:20 | |
| *** deepthi has joined #openstack-nova | 03:23 | |
| *** dave-mcc_ has quit IRC | 03:24 | |
| *** hitalia has quit IRC | 03:27 | |
| *** Daisy_ has quit IRC | 03:27 | |
| *** Daisy has joined #openstack-nova | 03:28 | |
| *** zerda-bis has joined #openstack-nova | 03:29 | |
| *** boris-42 has joined #openstack-nova | 03:29 | |
| *** rfolco has quit IRC | 03:30 | |
| *** smatzek has quit IRC | 03:30 | |
| *** baoli has joined #openstack-nova | 03:31 | |
| *** asettle is now known as asettle-afk | 03:32 | |
| *** Daisy has quit IRC | 03:32 | |
| *** dustins has quit IRC | 03:33 | |
| *** links has joined #openstack-nova | 03:36 | |
| *** baoli has quit IRC | 03:37 | |
| *** vilobhmm111 has quit IRC | 03:40 | |
| *** ks1 has quit IRC | 03:41 | |
| *** asettle-afk is now known as asettle | 03:42 | |
| *** Kennan2 is now known as Kennnan | 03:45 | |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'dummy' status with migration_status.DUMMY. https://review.openstack.org/228177 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'failed' status with migration_status.FAILED. https://review.openstack.org/228176 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'post-migrating' status with migration_status.POST_MIGRATING. https://review.openstack.org/228166 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'finished' statuses with migration_status.FINISHED. https://review.openstack.org/228167 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'migrating' status with migration_status.MIGRATING. https://review.openstack.org/228164 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'running' status with migration_status.MIGRATING. https://review.openstack.org/228165 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'preparing' status with migration_status.PREPARING. https://review.openstack.org/228162 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'pre-migrating' status with migration_status.PRE_MIGRATING. https://review.openstack.org/228163 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Introduce constants to indicate live migration status. https://review.openstack.org/228161 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'reverted' status with migration_status.REVERTED. https://review.openstack.org/228174 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'error' status with migration_status.ERROR. https://review.openstack.org/228175 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'confirmed' status with migration_status.CONFIRMED. https://review.openstack.org/228172 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'reverting' status with migration_status.REVERTING. https://review.openstack.org/228173 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'queued' status with migration_status.QUEUED. https://review.openstack.org/228170 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'confirming' status with migration_status.CONFIRMING. https://review.openstack.org/228171 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'completed' status with migration_status.FINISHED. https://review.openstack.org/228168 | 03:50 |
| openstackgerrit | Tang Chen proposed openstack/nova: Replace 'done' status with migration_status.FINISHED. https://review.openstack.org/228169 | 03:50 |
| *** vilobhmm11 has joined #openstack-nova | 03:53 | |
| *** zhangjn_ has quit IRC | 03:54 | |
| *** vilobhmm11 has quit IRC | 03:54 | |
| *** markvoelker has quit IRC | 03:56 | |
| *** vilobhmm11 has joined #openstack-nova | 03:56 | |
| *** ociuhandu has quit IRC | 04:00 | |
| *** haomaiwang has quit IRC | 04:01 | |
| *** haomaiwang has joined #openstack-nova | 04:01 | |
| *** rushil has quit IRC | 04:02 | |
| *** hemna has quit IRC | 04:06 | |
| *** wanghua has quit IRC | 04:06 | |
| *** wanghua has joined #openstack-nova | 04:06 | |
| *** zhangjn has joined #openstack-nova | 04:10 | |
| *** VikasC has joined #openstack-nova | 04:12 | |
| *** Marga_ has joined #openstack-nova | 04:12 | |
| *** nikhil_k has joined #openstack-nova | 04:16 | |
| *** nikhil has quit IRC | 04:18 | |
| *** pratikmallya has quit IRC | 04:18 | |
| *** jdurgin1 has quit IRC | 04:21 | |
| *** pratikmallya has joined #openstack-nova | 04:24 | |
| *** pratikmallya has joined #openstack-nova | 04:27 | |
| *** salv-orlando has joined #openstack-nova | 04:29 | |
| *** achanda has joined #openstack-nova | 04:29 | |
| *** pratikmallya has quit IRC | 04:32 | |
| *** Marga_ has quit IRC | 04:34 | |
| *** Marga_ has joined #openstack-nova | 04:35 | |
| *** unicell has joined #openstack-nova | 04:35 | |
| *** Sree has joined #openstack-nova | 04:38 | |
| *** salv-orlando has quit IRC | 04:38 | |
| *** Sree has quit IRC | 04:38 | |
| *** Sree has joined #openstack-nova | 04:39 | |
| *** Marga_ has quit IRC | 04:39 | |
| openstackgerrit | venkatesh proposed openstack/nova: make the driver.Scheduler as abstract class https://review.openstack.org/212016 | 04:45 |
| *** Marga_ has joined #openstack-nova | 04:50 | |
| *** shyu has quit IRC | 04:51 | |
| *** su_zhang has joined #openstack-nova | 04:51 | |
| *** zhangjn has quit IRC | 04:52 | |
| *** zhangjn has joined #openstack-nova | 04:53 | |
| *** markvoelker has joined #openstack-nova | 04:57 | |
| *** vilobhmm11 has left #openstack-nova | 04:58 | |
| *** vilobhmm11 has joined #openstack-nova | 04:59 | |
| *** vilobhmm11 has left #openstack-nova | 04:59 | |
| *** vilobhmm11 has joined #openstack-nova | 05:00 | |
| *** haomaiwang has quit IRC | 05:01 | |
| *** haomaiwang has joined #openstack-nova | 05:01 | |
| *** markvoelker has quit IRC | 05:01 | |
| *** VW has quit IRC | 05:02 | |
| *** rotbeard has joined #openstack-nova | 05:02 | |
| *** VW has joined #openstack-nova | 05:02 | |
| *** stevemar has quit IRC | 05:03 | |
| *** nkrinner has joined #openstack-nova | 05:03 | |
| *** hemna has joined #openstack-nova | 05:03 | |
| *** hemna has quit IRC | 05:03 | |
| *** dims has joined #openstack-nova | 05:13 | |
| *** Marga_ has quit IRC | 05:15 | |
| *** Marga_ has joined #openstack-nova | 05:15 | |
| *** dims has quit IRC | 05:18 | |
| *** VW_ has joined #openstack-nova | 05:20 | |
| *** tpeoples has quit IRC | 05:20 | |
| *** VW_ has quit IRC | 05:22 | |
| *** VW_ has joined #openstack-nova | 05:23 | |
| *** krtaylor has quit IRC | 05:23 | |
| *** VW has quit IRC | 05:24 | |
| openstackgerrit | Tang Chen proposed openstack/nova-specs: Host maintenance mode https://review.openstack.org/228689 | 05:26 |
| openstackgerrit | Tang Chen proposed openstack/nova-specs: Host maintenance mode https://review.openstack.org/228689 | 05:27 |
| *** deepakcs has joined #openstack-nova | 05:32 | |
| *** garyk has quit IRC | 05:32 | |
| *** krtaylor has joined #openstack-nova | 05:32 | |
| *** sileht has quit IRC | 05:33 | |
| *** zhangjn has quit IRC | 05:36 | |
| *** achanda has quit IRC | 05:36 | |
| *** stevemar has joined #openstack-nova | 05:36 | |
| *** moshele has joined #openstack-nova | 05:37 | |
| *** _gryf has quit IRC | 05:38 | |
| gmann | alex_xu: oomichi: regarding https://review.openstack.org/#/c/228676/ | 05:41 |
| gmann | I think empty list in index gives kindda false impression that there is not FIPs instead of some plugin missing. | 05:42 |
| *** stevemar_ has joined #openstack-nova | 05:42 | |
| gmann | I feel index should raise exception in l3 plugin missing where list FIP through the same | 05:43 |
| *** stevema__ has joined #openstack-nova | 05:43 | |
| *** stevemar has quit IRC | 05:45 | |
| *** stevemar_ has quit IRC | 05:46 | |
| *** otter768 has joined #openstack-nova | 05:49 | |
| *** Marga_ has quit IRC | 05:49 | |
| *** sahid has joined #openstack-nova | 05:50 | |
| *** otter768 has quit IRC | 05:53 | |
| *** VW_ has quit IRC | 05:54 | |
| *** VW has joined #openstack-nova | 05:54 | |
| *** lsmola has joined #openstack-nova | 05:58 | |
| *** VW has quit IRC | 05:59 | |
| *** garyk has joined #openstack-nova | 05:59 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** haomaiwang has joined #openstack-nova | 06:01 | |
| *** Murali has joined #openstack-nova | 06:01 | |
| *** browne has quit IRC | 06:03 | |
| *** Daisy has joined #openstack-nova | 06:06 | |
| *** jamielennox is now known as jamielennox|away | 06:12 | |
| *** irenab has quit IRC | 06:13 | |
| *** shuquan has joined #openstack-nova | 06:14 | |
| *** su_zhang has quit IRC | 06:15 | |
| *** openstackgerrit has quit IRC | 06:16 | |
| *** openstackgerrit has joined #openstack-nova | 06:17 | |
| *** Alexus has joined #openstack-nova | 06:20 | |
| *** oomichi has joined #openstack-nova | 06:20 | |
| *** fawadkhaliq has joined #openstack-nova | 06:21 | |
| *** armax has quit IRC | 06:24 | |
| *** oomichi has quit IRC | 06:25 | |
| *** e0ne has joined #openstack-nova | 06:28 | |
| *** stevema__ has quit IRC | 06:28 | |
| *** stevemar has joined #openstack-nova | 06:28 | |
| *** lpetrut has joined #openstack-nova | 06:28 | |
| *** fawadkhaliq has quit IRC | 06:30 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/227138 | 06:32 |
| *** stevemar has quit IRC | 06:33 | |
| *** e0ne has quit IRC | 06:34 | |
| openstackgerrit | Rui Chen proposed openstack/nova-specs: Task states improvement https://review.openstack.org/216578 | 06:35 |
| *** deepakcs has quit IRC | 06:38 | |
| *** tdurakov has joined #openstack-nova | 06:38 | |
| tdurakov | mikal, hi are you around? | 06:38 |
| *** Marga_ has joined #openstack-nova | 06:39 | |
| *** Daisy has quit IRC | 06:39 | |
| tdurakov | https://review.openstack.org/#/c/183057/ - left comment about spec readiness, could you check that? | 06:39 |
| *** Daisy has joined #openstack-nova | 06:40 | |
| *** deepakcs has joined #openstack-nova | 06:42 | |
| *** Daisy has quit IRC | 06:42 | |
| *** Daisy has joined #openstack-nova | 06:42 | |
| *** salv-orlando has joined #openstack-nova | 06:44 | |
| *** markvoelker has joined #openstack-nova | 06:45 | |
| *** irenab has joined #openstack-nova | 06:46 | |
| *** salv-orlando has quit IRC | 06:47 | |
| *** markvoelker has quit IRC | 06:50 | |
| *** tdurakov has quit IRC | 06:51 | |
| *** tdurakov has joined #openstack-nova | 06:53 | |
| rajesht | any stable core around ? | 06:56 |
| *** rdopiera has joined #openstack-nova | 06:59 | |
| rajesht | garyk: could you please review stable/kilo patch https://review.openstack.org/#/c/223968/ | 07:00 |
| rajesht | garyk: thanks :) | 07:00 |
| *** haomaiwang has quit IRC | 07:01 | |
| gmann | alex_xu: even this introduced new error code (need microversion :( )- https://review.openstack.org/#/c/216746/ | 07:01 |
| *** haomaiwang has joined #openstack-nova | 07:01 | |
| alex_xu | gmann: emm....maybe oomichi suggestion is right, return emtpy list | 07:02 |
| *** Sree has quit IRC | 07:02 | |
| gmann | alex_xu: but thats does not looks good for me.. we are returning positive response when neutron says floating IPs are not supported | 07:02 |
| gmann | alex_xu: empty list will convey to user that there is no FIP and then they will try to create one and get 404 | 07:03 |
| *** jwcroppe has joined #openstack-nova | 07:04 | |
| gmann | because FIP not supported from neutron tells clearly that requested resource is not available | 07:05 |
| gmann | and same nova should tell to user | 07:05 |
| alex_xu | gmann: sorry, I'm in a meeting, will be back later | 07:05 |
| bauzas | good morning Nova | 07:05 |
| gmann | alex_xu: sure | 07:05 |
| alex_xu | gmann: http://docs.openstack.org/developer/nova/api_microversion_dev.html there is Footnotes[2], the infrastture always return 404 | 07:06 |
| *** e0ne has joined #openstack-nova | 07:07 | |
| *** FL1SK has quit IRC | 07:08 | |
| gmann | alex_xu: that for 400 and 403 only right? | 07:09 |
| *** e0ne has quit IRC | 07:09 | |
| *** Daisy has quit IRC | 07:09 | |
| *** Daisy has joined #openstack-nova | 07:10 | |
| *** matrohon has joined #openstack-nova | 07:10 | |
| *** vilobhmm11 has quit IRC | 07:12 | |
| openstackgerrit | bharaththiruveedula proposed openstack/nova-specs: Returning Compute Node IP address in hypervisor list https://review.openstack.org/215207 | 07:13 |
| alex_xu | gmann: oops, I missed that | 07:13 |
| *** Daisy has quit IRC | 07:14 | |
| *** e0ne has joined #openstack-nova | 07:21 | |
| *** salv-orlando has joined #openstack-nova | 07:28 | |
| *** alexpilotti has joined #openstack-nova | 07:31 | |
| *** Sree_ has joined #openstack-nova | 07:31 | |
| *** Sree_ is now known as Guest22185 | 07:32 | |
| *** salv-orlando has quit IRC | 07:32 | |
| *** Guest22185 has quit IRC | 07:33 | |
| *** ildikov has joined #openstack-nova | 07:34 | |
| *** Sree has joined #openstack-nova | 07:35 | |
| *** yamahata has joined #openstack-nova | 07:35 | |
| *** exploreshaifali has joined #openstack-nova | 07:40 | |
| *** jichen has joined #openstack-nova | 07:41 | |
| openstackgerrit | Sachi King proposed openstack/nova: DNM: Tox testing https://review.openstack.org/228764 | 07:42 |
| *** alex_klimov has joined #openstack-nova | 07:42 | |
| *** romainh has joined #openstack-nova | 07:43 | |
| *** acc__ has joined #openstack-nova | 07:44 | |
| *** acc_ has quit IRC | 07:44 | |
| *** jlanoux has joined #openstack-nova | 07:45 | |
| *** garyk has quit IRC | 07:47 | |
| *** garyk has joined #openstack-nova | 07:47 | |
| *** otter768 has joined #openstack-nova | 07:50 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/python-novaclient: Add --metadata as optional input when do nova image-create https://review.openstack.org/228769 | 07:51 |
| *** alexpilotti has quit IRC | 07:52 | |
| *** shuquan has quit IRC | 07:54 | |
| *** _gryf has joined #openstack-nova | 07:54 | |
| *** otter768 has quit IRC | 07:54 | |
| *** karimb has joined #openstack-nova | 07:56 | |
| *** shuquan has joined #openstack-nova | 07:56 | |
| openstackgerrit | Rui Chen proposed openstack/nova-specs: Split network plane for live migration https://review.openstack.org/194990 | 07:57 |
| *** asettle has quit IRC | 08:00 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** haomaiwang has joined #openstack-nova | 08:01 | |
| openstackgerrit | Timofey Durakov proposed openstack/nova: Wrap setup_networks_on_host with try/except https://review.openstack.org/227897 | 08:02 |
| alex_xu | johnthetubaguy: bauzas, good morning, just worked out the back port for upgrade bug https://review.openstack.org/228774 | 08:03 |
| *** xgsa has joined #openstack-nova | 08:03 | |
| bauzas | alex_xu: ack, you could perhaps get a -2 because the RC2 is not yet open, but it should be done soon as per johnthetubaguy said in his email | 08:04 |
| bauzas | RC2 *window | 08:04 |
| alex_xu | bauzas: ok | 08:05 |
| *** Sree has quit IRC | 08:05 | |
| *** Sree has joined #openstack-nova | 08:06 | |
| *** lpetrut has quit IRC | 08:06 | |
| *** yassine__ has joined #openstack-nova | 08:07 | |
| *** derekh has joined #openstack-nova | 08:07 | |
| *** markus_z has joined #openstack-nova | 08:07 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/python-novaclient: Add --metadata as optional input when do nova image-create https://review.openstack.org/228769 | 08:08 |
| *** nithyag_ has quit IRC | 08:15 | |
| *** lpetrut has joined #openstack-nova | 08:15 | |
| *** jistr has joined #openstack-nova | 08:16 | |
| *** mpaolino has joined #openstack-nova | 08:17 | |
| *** marzif has quit IRC | 08:18 | |
| openstackgerrit | Radoslav Gerganov proposed openstack/nova-specs: ConsoleAuthTokens should work for all console types https://review.openstack.org/228778 | 08:19 |
| *** danpb has joined #openstack-nova | 08:23 | |
| *** Alexus has quit IRC | 08:31 | |
| *** Daisy has joined #openstack-nova | 08:32 | |
| *** pixelbeat has joined #openstack-nova | 08:34 | |
| *** salv-orlando has joined #openstack-nova | 08:35 | |
| kashyap | On Nova live block migration: | 08:36 |
| kashyap | If you use these flags in nova.conf: | 08:36 |
| kashyap | live_migration_flag=VIR_MIGRATE_UNDEFINE_SOURCE, VIR_MIGRATE_PEER2PEER, VIR_MIGRATE_LIVE | 08:36 |
| kashyap | block_migration_flag=VIR_MIGRATE_UNDEFINE_SOURCE, VIR_MIGRATE_PEER2PEER, VIR_MIGRATE_LIVE, VIR_MIGRATE_NON_SHARED_INC | 08:36 |
| kashyap | Performing a `nova live-migration --block-migrat` *still* uses old style 'migrate inc' method (as I realize libvirt TUNNELLED migration + NBD does not play well yet) instead of using NBD-based approach. Any hints why? | 08:36 |
| kashyap | I know that the current upstream default for 'live_migration_flag' still has TUNNELLED in it. But in my test above, I explicitly removed it in my nova.conf. . . | 08:37 |
| *** julien_ZTE has joined #openstack-nova | 08:38 | |
| *** VW has joined #openstack-nova | 08:40 | |
| *** VW has quit IRC | 08:40 | |
| *** VW has joined #openstack-nova | 08:41 | |
| * kashyap patches local driver.py manually and tests again | 08:41 | |
| kashyap | Because looking at my n-cpu.log, seems like my change in nova.conf didn't take effect -- the default flags from upstream are enforced. | 08:43 |
| kashyap | sahid (or anyone): Hi, when you have a moment -- shouldn't explicit values of a config variable set in nova.conf take precedence over upstream default? | 08:45 |
| *** exploreshaifali has quit IRC | 08:45 | |
| *** VW has quit IRC | 08:45 | |
| rajesht | jaypipes: I have replied to your comments on https://review.openstack.org/170118 | 08:45 |
| rajesht | jaypipes: could you please give your suggestions ? | 08:46 |
| rajesht | jaypipes: thanks :) | 08:46 |
| *** markvoelker has joined #openstack-nova | 08:46 | |
| *** lpetrut has quit IRC | 08:48 | |
| rajesht | jaypipes: pl consider this patch set https://review.openstack.org/#/c/170118/17/nova/compute/manager.py | 08:49 |
| *** shuquan has quit IRC | 08:49 | |
| *** shuquan has joined #openstack-nova | 08:51 | |
| rajesht | mikal: needs one more +2 on stable/kilo backport, could you pls check https://review.openstack.org/223968 | 08:51 |
| *** markvoelker has quit IRC | 08:51 | |
| openstackgerrit | venkatesh proposed openstack/nova: make the driver.Scheduler as abstract class https://review.openstack.org/212016 | 08:52 |
| *** jsuchome has joined #openstack-nova | 08:53 | |
| *** Kennnan2 has joined #openstack-nova | 08:54 | |
| *** Kennnan has quit IRC | 08:54 | |
| *** MarcF has joined #openstack-nova | 08:57 | |
| *** claudiub has joined #openstack-nova | 08:58 | |
| *** haomaiwang has quit IRC | 09:01 | |
| *** haomaiwang has joined #openstack-nova | 09:01 | |
| sahid | kashyap: no you don't have, but i think that if you restart devstack it will override nova.conf | 09:02 |
| *** jichen has quit IRC | 09:02 | |
| kashyap | sahid: You mean, even if I explicitly set something that's *different* from default in config file, it'll be overridden once DevStack restarts? | 09:03 |
| kashyap | That does not sound useful | 09:03 |
| sahid | kashyap: yes i think so | 09:03 |
| johnthetubaguy | alex_xu: bauzas: this patch does stop live upgrades working, if I understand this correctly? https://review.openstack.org/#/c/228304/ | 09:04 |
| kashyap | sahid: Okay, I'll come back and test, have to step out for 20 min | 09:04 |
| bauzas | johnthetubaguy: ? | 09:05 |
| bauzas | johnthetubaguy: what do you mean ? | 09:05 |
| sahid | kashyap: i think here is a way to twek noca.conf in localerc | 09:05 |
| sahid | tweak | 09:05 |
| *** FL1SK has joined #openstack-nova | 09:05 | |
| johnthetubaguy | bauzas: that patch you were talking to alex_xu about, its basically blocking upgrade right? | 09:05 |
| bauzas | johnthetubaguy: the problem is that an instance backport was providing an InstanceV1 | 09:06 |
| johnthetubaguy | well, live upgrade | 09:06 |
| bauzas | johnthetubaguy: where our checks were wrong | 09:06 |
| johnthetubaguy | full of cold with a brain like mush, just making sure I am remembering it right | 09:06 |
| *** sambetts has quit IRC | 09:07 | |
| bauzas | johnthetubaguy: so dansmith provided a new v2 object for instance | 09:07 |
| *** haypo has joined #openstack-nova | 09:07 | |
| johnthetubaguy | bauzas: OK, I am confused now, do you have the patch for that? | 09:07 |
| *** sambetts has joined #openstack-nova | 09:07 | |
| sahid | kashyap: local.sh sorry | 09:07 |
| *** ndipanov has quit IRC | 09:07 | |
| johnthetubaguy | bauzas: simpler question, is it work having another RC release, just for this patch: https://review.openstack.org/#/c/228774/1 | 09:08 |
| johnthetubaguy | s/work/worth/ | 09:08 |
| bauzas | johnthetubaguy: https://github.com/openstack/nova/commit/713d8cb0777afb9fe4f665b9a40cac894b04aacb | 09:08 |
| johnthetubaguy | bauzas: yes, I remember that patch | 09:09 |
| johnthetubaguy | bauzas: thats why we need the above patch | 09:09 |
| bauzas | johnthetubaguy: good question, but I thought we were waiting for a RC2 for translations, nope? | 09:09 |
| bauzas | yup | 09:09 |
| johnthetubaguy | we can make that RC3 | 09:09 |
| bauzas | not sure I get that | 09:09 |
| johnthetubaguy | my question is really, did we work out why granade didn't find the bug | 09:10 |
| johnthetubaguy | just we can do RC2 for the above patch, then RC3 for translations | 09:10 |
| bauzas | johnthetubaguy: oh I see your question | 09:10 |
| bauzas | well the grenade question is actually a good point | 09:11 |
| bauzas | johnthetubaguy: I mean the partial n-cpu job should have found the problem | 09:11 |
| johnthetubaguy | right | 09:12 |
| *** davideagnello has quit IRC | 09:12 | |
| *** oomichi has joined #openstack-nova | 09:13 | |
| bauzas | johnthetubaguy: I guess it hasn't been found because it was not covered | 09:13 |
| bauzas | johnthetubaguy: by looking at the change, there are 4 impacted files | 09:13 |
| *** davideagnello has joined #openstack-nova | 09:13 | |
| bauzas | one is about numa topology, and we know we don't cover it | 09:13 |
| bauzas | one is about providing a request spec dict, but we're passing a new instance that we created on the Liberty controller, so I guess it's correct | 09:14 |
| bauzas | (speaking of https://review.openstack.org/#/c/228304/8/nova/scheduler/utils.py,cm) | 09:15 |
| bauzas | well, it would probably need to be checked for migrations tho | 09:15 |
| *** dims has joined #openstack-nova | 09:16 | |
| bauzas | one is within the ReqSpec object, but we haven't yet used this object | 09:16 |
| bauzas | and the last one is when providing some instance info from the compute to the scheduler | 09:16 |
| bauzas | johnthetubaguy: so I possibly think it was not covered | 09:18 |
| *** ndipanov has joined #openstack-nova | 09:18 | |
| bauzas | alex_xu: thoughts ? | 09:18 |
| johnthetubaguy | bauzas: yeah, thats possible, although the bug seemed to suggest he was unable to boot, maybe with NUMA configured | 09:18 |
| bauzas | johnthetubaguy: yeah I guess alex_xu tried something untested | 09:19 |
| bauzas | oh sec | 09:19 |
| *** sakthi has joined #openstack-nova | 09:19 | |
| bauzas | alex_xu: you said you were running 2015.1.0 | 09:19 |
| johnthetubaguy | that shouldn't impact that patch though | 09:19 |
| bauzas | alex_xu: could you please verify if the bug you mention is still valid with 2015.1.1 ? (doesn't require an upgrade | 09:19 |
| sakthi | Nova compute manager error spawning instances | 09:20 |
| sakthi | help me regarding this.\ | 09:20 |
| *** salv-orl_ has joined #openstack-nova | 09:21 | |
| *** dims has quit IRC | 09:22 | |
| bauzas | johnthetubaguy: sure | 09:22 |
| bauzas | johnthetubaguy: but I'm still wondering why grenade didn't catch up | 09:22 |
| johnthetubaguy | sakthi: there should be folks on #openstack to help with more usage questions, this is really just for development chat | 09:22 |
| bauzas | johnthetubaguy: https://bugs.launchpad.net/nova/+bug/1500289 is pretty explicit with the 2nd stacktrace | 09:23 |
| openstack | Launchpad bug 1500289 in OpenStack Compute (nova) "Booting instance failed with Kilo stable compute node and liberty controller" [Critical,Fix committed] - Assigned to Alex Xu (xuhj) | 09:23 |
| johnthetubaguy | sakthi: but what version are you running, and have you opened a bug for the issue at all? | 09:23 |
| sakthi | running kilo version | 09:23 |
| johnthetubaguy | bauzas: oh, I see, its the retry that actually breaks | 09:24 |
| *** salv-orlando has quit IRC | 09:25 | |
| oomichi | johnthetubaguy: hello :) | 09:25 |
| *** julien_ZTE has quit IRC | 09:26 | |
| oomichi | johnthetubaguy: can you take a look at https://review.openstack.org/#/c/227650/ again if available ? | 09:26 |
| johnthetubaguy | sakthi: OK, have you tried asking on #openstack channel, where some folks can help with usage questions? | 09:26 |
| *** julien_ZTE has joined #openstack-nova | 09:26 | |
| sakthi | ok | 09:27 |
| openstackgerrit | Vladyslav Drok proposed openstack/nova: DO NOT MERGE - Triggering experimental pipeline https://review.openstack.org/220229 | 09:28 |
| bauzas | johnthetubaguy: yup, because it's in build_instances but with a v1 instance | 09:28 |
| bauzas | johnthetubaguy: if it was a single instance creation, it should have been coming as a v2 | 09:28 |
| bauzas | that's my understanding | 09:29 |
| bauzas | johnthetubaguy: given the first stacktrace where the compute manager raised an exception | 09:29 |
| *** julien_ZTE has quit IRC | 09:30 | |
| *** VW has joined #openstack-nova | 09:32 | |
| bauzas | johnthetubaguy: and to be clear, as grenade was running the latest stable kilo incl. the bugfix for the pci stuff, it didn't raised an exception in the spawn() process, thus not requiring a reschedule | 09:33 |
| *** jlanoux has quit IRC | 09:33 | |
| *** digimonde has joined #openstack-nova | 09:33 | |
| rajesht | claudiub: needs one more +2 on stable/kilo backport, could you pls check https://review.openstack.org/223968 | 09:35 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add cpu schedular priority config https://review.openstack.org/197567 | 09:35 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: rework membacking config to support future features https://review.openstack.org/197566 | 09:35 |
| openstackgerrit | sahid proposed openstack/nova: virt: add contraint to handle realtime policy https://review.openstack.org/197568 | 09:35 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: add realtime support https://review.openstack.org/197569 | 09:35 |
| openstackgerrit | sahid proposed openstack/nova: hardware: improve parse_cpu_spec to handle exclusion range https://review.openstack.org/226783 | 09:35 |
| rajesht | claudiub: thanks :) | 09:35 |
| kashyap | sahid: Im' back. Yes, there _is_ a way to add Nova conf parameters in DevStack's local.conf, that's what I use: | 09:37 |
| kashyap | [[post-config|$NOVA_CONF]] | 09:37 |
| kashyap | [DEFAULT] | 09:37 |
| *** VW has quit IRC | 09:38 | |
| rajesht | johnthetubaguy: can we discuss about your comments on force-delete patch https://review.openstack.org/#/c/163798 if you have time | 09:38 |
| johnthetubaguy | oomichi: had some nits | 09:38 |
| johnthetubaguy | rajesht: sure thing | 09:38 |
| *** VW has joined #openstack-nova | 09:38 | |
| rajesht | johnthetubaguy: comment in https://review.openstack.org/#/c/163798/9/nova/compute/api.py line 1946 | 09:39 |
| johnthetubaguy | rajesht; just seems like we should raise raise exception.InstanceInvalidState on the duplicate force delete calls there | 09:40 |
| rajesht | johnthetubaguy: but here we can't add list of allowed task_states as it should allow force-delete on every task_state except force-deleting | 09:41 |
| *** MarcF has quit IRC | 09:41 | |
| danpb | c! | 09:42 |
| *** zhenguo has quit IRC | 09:43 | |
| *** lpetrut has joined #openstack-nova | 09:44 | |
| kashyap | cfg.StrOpt('live_migration_flag', | 09:44 |
| kashyap | default='VIR_MIGRATE_UNDEFINE_SOURCE, VIR_MIGRATE_PEER2PEER, ' | 09:44 |
| rajesht | johnthetubaguy: IMO we can't add task_state in check_instance_state which we want to reject. Is it true ? | 09:44 |
| *** jlanoux has joined #openstack-nova | 09:45 | |
| rajesht | johnthetubaguy: we can only have allowed states in check_instance_state decorator. | 09:45 |
| rajesht | johnthetubaguy: I am silently rejecting duplicate force-delete request in https://review.openstack.org/#/c/163798/9/nova/compute/api.py line #1581 | 09:46 |
| *** dims has joined #openstack-nova | 09:47 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: hacking check for contextlib.nested for py34 support https://review.openstack.org/228688 | 09:47 |
| *** lpetrut1 has joined #openstack-nova | 09:48 | |
| rajesht | johnthetubaguy: as the same was done for duplicate delete request in https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1689 | 09:48 |
| *** lpetrut has quit IRC | 09:48 | |
| *** lpetrut1 is now known as lpetrut | 09:48 | |
| *** otter768 has joined #openstack-nova | 09:51 | |
| *** Alexus has joined #openstack-nova | 09:51 | |
| *** VW has quit IRC | 09:51 | |
| *** VW has joined #openstack-nova | 09:52 | |
| *** otter768 has quit IRC | 09:55 | |
| *** Daisy has quit IRC | 09:56 | |
| *** Daisy has joined #openstack-nova | 09:56 | |
| *** VW has quit IRC | 09:56 | |
| *** kbyrne has quit IRC | 09:58 | |
| *** shuquan has quit IRC | 09:58 | |
| *** ZZelle has quit IRC | 10:01 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** Daisy has quit IRC | 10:01 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add pep8 check for assertEqual argument order https://review.openstack.org/227650 | 10:01 |
| *** haomaiwa_ has joined #openstack-nova | 10:01 | |
| *** yassine__ has quit IRC | 10:02 | |
| *** kbyrne has joined #openstack-nova | 10:03 | |
| oomichi | johnthetubaguy: thanks for your review, applied your comments to ^^^ | 10:03 |
| *** sudipto has joined #openstack-nova | 10:03 | |
| *** IanGovett has joined #openstack-nova | 10:04 | |
| johnthetubaguy | oomichi: thanks, they were a bit picky really | 10:07 |
| johnthetubaguy | rajesht: so I was waiting for my compute to make a noise, with the speaker on mute :( | 10:07 |
| *** haomaiwa_ has quit IRC | 10:07 | |
| rajesht | johnthetubaguy: np :) | 10:08 |
| *** asettle has joined #openstack-nova | 10:08 | |
| johnthetubaguy | rajesht: I am not talking about modifying the decorator, I just mean add the code in that method | 10:08 |
| *** asettle has quit IRC | 10:08 | |
| johnthetubaguy | rajesht: and make sure it raises an exception, rather than silently ignores the invalid state | 10:08 |
| rajesht | johnthetubaguy: In that case, I will raise exception in https://review.openstack.org/#/c/163798/9/nova/compute/api.py line 1585 | 10:09 |
| johnthetubaguy | rajesht: to be clear, duplicate delete requests are actually (normally) rejected by the the state check, L1689 is a bug, it should raise an exception, but thats a different patch | 10:09 |
| johnthetubaguy | rajesht: yep, that would be good I think | 10:09 |
| *** asettle has joined #openstack-nova | 10:09 | |
| rajesht | johnthetubaguy: ok , will upload a patch soon, thanks for your time :) | 10:09 |
| johnthetubaguy | rajesht: there was one more bit I think... hiding the new task state from the user, did that make sense, thats the most important comment really | 10:09 |
| rajesht | johnthetubaguy: Yeah, I checked that comment as well. Will do that as well :) | 10:10 |
| sudipto | flaper87, are you around? :) | 10:10 |
| *** ZZelle has joined #openstack-nova | 10:13 | |
| *** romainh has quit IRC | 10:15 | |
| *** paul-carlton has joined #openstack-nova | 10:22 | |
| *** tdurakov has quit IRC | 10:27 | |
| openstackgerrit | Paul Carlton proposed openstack/nova-specs: Query and cancel ongoing live migrations https://review.openstack.org/228828 | 10:29 |
| *** tdurakov has joined #openstack-nova | 10:29 | |
| *** jlanoux_ has joined #openstack-nova | 10:31 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: Switch to oslo.cache lib https://review.openstack.org/203049 | 10:33 |
| *** jlanoux has quit IRC | 10:34 | |
| *** smatzek has joined #openstack-nova | 10:35 | |
| openstackgerrit | Dmitry Guryanov proposed openstack/nova: libvirt: support snapshots with parallels virt_type https://review.openstack.org/184153 | 10:36 |
| *** EinstCrazy has quit IRC | 10:42 | |
| tojuvone | johnthetubaguy: Thanks for review, but still before making another patch, should I expose more host/service related things like enabled/disabled to this: https://review.openstack.org/192246/ | 10:45 |
| *** jlanoux has joined #openstack-nova | 10:45 | |
| *** markvoelker has joined #openstack-nova | 10:47 | |
| *** jlanoux_ has quit IRC | 10:48 | |
| *** jamielennox|away is now known as jamielennox | 10:49 | |
| *** markvoelker has quit IRC | 10:52 | |
| *** Daisy has joined #openstack-nova | 10:52 | |
| *** tojuvone is now known as tojuvone_away | 10:52 | |
| *** wanghua has quit IRC | 10:52 | |
| *** twinklechawla has joined #openstack-nova | 10:55 | |
| *** 18VAARFFB has joined #openstack-nova | 10:56 | |
| ndipanov | johnthetubaguy, hey should I be proposing this to stable/liberty https://review.openstack.org/#/c/226923/ ? | 10:57 |
| ndipanov | I can't seem to post the review | 10:57 |
| ndipanov | do you do it before the RCs or? | 10:57 |
| johnthetubaguy | its good to line them up now | 10:57 |
| ndipanov | so I tagged the bug | 10:58 |
| johnthetubaguy | currently, not seen anything bad enough to cut an RC before next tuesday | 10:58 |
| ndipanov | just wasn't clear on who does the actual "backport" | 10:58 |
| johnthetubaguy | ndipanov: generally the bug submitter | 10:58 |
| johnthetubaguy | ndipanov: if they want to make sure it happens | 10:58 |
| ndipanov | hmmm so I can't seem to propose it then | 10:58 |
| johnthetubaguy | what error do you get? | 10:58 |
| flaper87 | sudipto: I am now | 10:59 |
| ndipanov | johnthetubaguy, one sec | 10:59 |
| *** claudiub has quit IRC | 10:59 | |
| sudipto | flaper87, ok sure. Just wanted to follow up on the review comment on the glance v1 to v2 stuff. | 10:59 |
| johnthetubaguy | ndipanov: I am guessing you cherry picked the merged commit onto stable/liberty, then do git review stable/liberty or something like that? | 10:59 |
| ndipanov | johnthetubaguy, yeah and it acts like the change is closed | 10:59 |
| flaper87 | sudipto: sure | 10:59 |
| johnthetubaguy | ndipanov: oh, you will need a new change id, I believe | 10:59 |
| ndipanov | johnthetubaguy, that's not how stable usually works?? | 11:00 |
| flaper87 | sudipto: I'm working on the spec and I'll have a draft ready soon | 11:00 |
| twinklechawla | Hello everyone, I am interested in contributing in nova but I am totally new to this, so I need help! | 11:00 |
| ndipanov | it's very useful to have the same changeid | 11:00 |
| *** gszasz has joined #openstack-nova | 11:00 | |
| sudipto | flaper87, sure. | 11:00 |
| flaper87 | sudipto: That spec is the one that we should link on tha tchange you did | 11:00 |
| sudipto | flaper87, yeah that makes sense. Sure. | 11:00 |
| johnthetubaguy | ndipanov: good point, I don't remember ever having an issue with that | 11:00 |
| bauzas | twinklechawla: welcome :) | 11:00 |
| flaper87 | sudipto: btw, there's another guy from Glance that would like to work on that as well. It'd be great to have you guys talk to each other | 11:00 |
| bauzas | twinklechawla: you can find a few pointers in https://wiki.openstack.org/wiki/Nova/Mentoring | 11:00 |
| *** 18VAARFFB has quit IRC | 11:01 | |
| flaper87 | sudipto: would you mind joining #openstack-glance ? | 11:01 |
| ndipanov | ok doing git review stable/liberty seems to have done it | 11:01 |
| twinklechawla | bauzas: Thank you! | 11:01 |
| sudipto | flaper87, not at all, joining... | 11:01 |
| *** haomaiwang has joined #openstack-nova | 11:01 | |
| flaper87 | sudipto: awesome, thanks! | 11:01 |
| ndipanov | gerrit would normally figure out the branch but oh well | 11:01 |
| ndipanov | johnthetubaguy, ^ | 11:01 |
| johnthetubaguy | ndipanov: ah, cool, I see it there now: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/liberty,n,z | 11:01 |
| ndipanov | y | 11:02 |
| ndipanov | ok moving on... :) | 11:02 |
| *** exploreshaifali has joined #openstack-nova | 11:03 | |
| *** EinstCrazy has joined #openstack-nova | 11:04 | |
| *** jecarey has joined #openstack-nova | 11:05 | |
| twinklechawla | bauzas: I have a query! | 11:06 |
| *** yassine__ has joined #openstack-nova | 11:07 | |
| *** mpaolino has quit IRC | 11:09 | |
| twinklechawla | Till date I read about nova, cloned openstack-dev/devstack and openstack/nova. Further, what I have to do? | 11:09 |
| *** haomaiwang has quit IRC | 11:11 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Log warning for wrong migration flag config options https://review.openstack.org/228853 | 11:16 |
| *** twinklechawla has quit IRC | 11:16 | |
| *** Daisy has quit IRC | 11:19 | |
| *** Daisy has joined #openstack-nova | 11:20 | |
| alex_xu | johnthetubaguy: bauzas that bug only trigger by kilo node send InstanceV1 to liberty controller node, the case I produce is there is something wrong when booting instance on kilo node, then kilo node will send retry request back to liberty conductor with an instancev1 | 11:21 |
| alex_xu | I guess that is the case grenade didn't tested | 11:21 |
| *** salv-orl_ has quit IRC | 11:22 | |
| *** Daisy has quit IRC | 11:22 | |
| *** salv-orlando has joined #openstack-nova | 11:23 | |
| *** Daisy has joined #openstack-nova | 11:23 | |
| sdague | alex_xu: what kind of high level operation would trigger that? | 11:23 |
| *** Daisy has quit IRC | 11:23 | |
| *** Daisy has joined #openstack-nova | 11:24 | |
| alex_xu | sdague: what means high level operation? I tested the booting new instance | 11:24 |
| sdague | right, so we definitely boot a new instance on the partial upgrade test | 11:24 |
| sdague | I'm mostly curious how it ended up getting missed | 11:25 |
| alex_xu | it's only triggered when booting new instance failed on the kilo compute node | 11:25 |
| *** Daisy has quit IRC | 11:25 | |
| *** Daisy has joined #openstack-nova | 11:26 | |
| sdague | oh, so only if there is a fault | 11:26 |
| sdague | then a reschedule | 11:26 |
| alex_xu | sdague: yeah | 11:26 |
| sdague | yeh, we don't do fault injection | 11:26 |
| johnthetubaguy | yeah, its the rechedule | 11:26 |
| johnthetubaguy | I am thinking we make sure its in RC2 that we cut next week | 11:27 |
| johnthetubaguy | but not urgent enough to do a new RC this week | 11:27 |
| johnthetubaguy | does that make sense, do you think? | 11:27 |
| *** baoli has joined #openstack-nova | 11:27 | |
| *** thorst has joined #openstack-nova | 11:27 | |
| *** Daisy has quit IRC | 11:27 | |
| sdague | that seems fine to me | 11:27 |
| alex_xu | ok to me also | 11:27 |
| *** Daisy has joined #openstack-nova | 11:27 | |
| *** marzif has joined #openstack-nova | 11:28 | |
| johnthetubaguy | cool, thanks | 11:28 |
| *** Daisy has joined #openstack-nova | 11:28 | |
| *** baoli_ has joined #openstack-nova | 11:28 | |
| *** kiran-r has joined #openstack-nova | 11:29 | |
| *** deepthi has quit IRC | 11:29 | |
| *** jecarey has quit IRC | 11:31 | |
| *** baoli has quit IRC | 11:31 | |
| *** exploreshaifali has quit IRC | 11:32 | |
| *** Daisy has quit IRC | 11:33 | |
| *** stevemar has joined #openstack-nova | 11:40 | |
| *** tojuvone_away is now known as tojuvone | 11:41 | |
| *** dewsday has joined #openstack-nova | 11:43 | |
| *** asettle has quit IRC | 11:45 | |
| *** stevemar has quit IRC | 11:45 | |
| *** dewsday has quit IRC | 11:45 | |
| *** asettle has joined #openstack-nova | 11:46 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: hacking check for contextlib.nested for py34 support https://review.openstack.org/228688 | 11:50 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Remove novaclient.v1_1 module https://review.openstack.org/217131 | 11:50 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Remove novaclient.v1_1 module https://review.openstack.org/217131 | 11:51 |
| *** otter768 has joined #openstack-nova | 11:52 | |
| alex_xu | nova api meeting in 8 mins | 11:52 |
| *** VikasC has quit IRC | 11:56 | |
| *** otter768 has quit IRC | 11:57 | |
| dims | jaypipes: sdague: really easy one, doubles py34 coverage - https://review.openstack.org/#/c/228600/ - please take a look when you get a chance | 11:59 |
| alex_xu | nova api meeting is running~ | 12:01 |
| sdague | dims: one thing in there, I thought we had decided to just use .items | 12:01 |
| sdague | because the micro benchmarks are pointless here | 12:01 |
| *** markvoelker has joined #openstack-nova | 12:03 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Identify more py34 tests that already pass https://review.openstack.org/228600 | 12:03 |
| dims | sdague: duh! slipped my mind. sorry, fixed. | 12:03 |
| sdague | dims: +2 | 12:04 |
| dims | thanks! | 12:04 |
| openstackgerrit | garyk proposed openstack/nova: VMware: Use actual VM state instead of using the instance vm_state https://review.openstack.org/190627 | 12:05 |
| *** markus_z has quit IRC | 12:06 | |
| *** Kevin_Zheng has joined #openstack-nova | 12:06 | |
| *** markvoelker has quit IRC | 12:07 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Slightly better translation friendly formatting https://review.openstack.org/224009 | 12:07 |
| *** exploreshaifali has joined #openstack-nova | 12:07 | |
| *** aarefiev has quit IRC | 12:09 | |
| openstackgerrit | John Garbutt proposed openstack/nova: docs: re-organise the API concept docs https://review.openstack.org/226253 | 12:10 |
| *** romainh has joined #openstack-nova | 12:10 | |
| *** bfic has joined #openstack-nova | 12:10 | |
| *** SunSeeker has joined #openstack-nova | 12:11 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Enable options for oslo.reports https://review.openstack.org/216917 | 12:11 |
| *** aarefiev has joined #openstack-nova | 12:12 | |
| *** markus_z has joined #openstack-nova | 12:13 | |
| *** pm90_ has joined #openstack-nova | 12:13 | |
| *** deepakcs has quit IRC | 12:17 | |
| *** aysyd has joined #openstack-nova | 12:17 | |
| *** pm90_ has quit IRC | 12:20 | |
| *** pm90__ has joined #openstack-nova | 12:20 | |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: use helper method to remove version hardcoding https://review.openstack.org/228578 | 12:20 |
| *** pm90__ has quit IRC | 12:20 | |
| *** raildo-afk is now known as raildo | 12:21 | |
| bfic | Hey guys, I need to get local block device names from instance to pass as a 'migrate-disks' param to migrateToURI3. How can I proper get them? Other methods in libvirt driver has "block_device_info" as a parameter, but live_migration_operation no. Any idea ? | 12:22 |
| *** Daisy has joined #openstack-nova | 12:27 | |
| *** tdurakov has quit IRC | 12:30 | |
| *** rfolco has joined #openstack-nova | 12:30 | |
| *** Daisy has quit IRC | 12:30 | |
| *** tdurakov has joined #openstack-nova | 12:30 | |
| *** bswartz has quit IRC | 12:30 | |
| *** edmondsw has joined #openstack-nova | 12:31 | |
| *** Sree has quit IRC | 12:31 | |
| bauzas | sdague: yup, re: https://bugs.launchpad.net/nova/+bug/1500289 there is a reschedule coming from the compute (2nd stacktrace) because there was a driver exception on spawn() (1st trace) | 12:31 |
| openstack | Launchpad bug 1500289 in OpenStack Compute (nova) "Booting instance failed with Kilo stable compute node and liberty controller" [Critical,Fix committed] - Assigned to Alex Xu (xuhj) | 12:31 |
| *** claudiub has joined #openstack-nova | 12:31 | |
| *** Sree has joined #openstack-nova | 12:31 | |
| bauzas | sdague: so it called the conductor using the previous InstanceV1 | 12:32 |
| *** kiran-r has quit IRC | 12:32 | |
| *** markvoelker has joined #openstack-nova | 12:32 | |
| *** EinstCrazy has quit IRC | 12:34 | |
| *** Sree has quit IRC | 12:36 | |
| *** dims_ has joined #openstack-nova | 12:36 | |
| *** aix has quit IRC | 12:37 | |
| *** artom has quit IRC | 12:37 | |
| *** EinstCrazy has joined #openstack-nova | 12:40 | |
| *** dims has quit IRC | 12:40 | |
| *** exploreshaifali has quit IRC | 12:40 | |
| *** meh_di has joined #openstack-nova | 12:45 | |
| bauzas | johnthetubaguy: feature classification ? that's not the first time I heard of it, but could you please explain me what is it ? | 12:47 |
| meh_di | HELP// | 12:47 |
| bauzas | (I mean, I have some idea but I'm perhaps wrong) | 12:47 |
| *** oomichi_ has joined #openstack-nova | 12:47 | |
| meh_di | please tell me how can debug nova remotely | 12:47 |
| *** xyang1 has joined #openstack-nova | 12:48 | |
| bauzas | meh_di: http://docs.openstack.org/developer/nova/devref/development.environment.html#using-a-remote-debugger | 12:49 |
| openstackgerrit | Sudipta Biswas proposed openstack/nova: use helper method to remove version hardcoding https://review.openstack.org/228578 | 12:49 |
| *** Alexus has quit IRC | 12:51 | |
| *** alejandrito has joined #openstack-nova | 12:52 | |
| *** dims_ has quit IRC | 12:53 | |
| *** thangp has joined #openstack-nova | 12:54 | |
| meh_di | bauzas: I know this, I am using pycharm. I followed the  http://blog.jetbrains.com/pycharm/2010/12/python-remote-debug-with-pycharm but I need to know what is the remote host in this case? Is that Compute Node ? | 12:54 |
| *** breitz has quit IRC | 12:54 | |
| *** breitz has joined #openstack-nova | 12:54 | |
| bauzas | meh_di: from your opinion ? | 12:55 |
| meh_di | bauzas: I dont undrestand your question | 12:56 |
| bauzas | meh_di: I was just asking you to consider your question | 12:57 |
| bauzas | meh_di: so, you want to troubleshoot a process, right? | 12:57 |
| meh_di | yes | 12:57 |
| bauzas | meh_di: in that case, you know that you'll run the remote option for | 12:58 |
| bauzas | *that* process | 12:58 |
| *** Daisy has joined #openstack-nova | 12:58 | |
| bauzas | ie. nova-scheduler, nova-api, nova-conductor, nova-manager etc. | 12:58 |
| johnthetubaguy | bauzas: its written up here: https://review.openstack.org/#/c/215664/ | 12:59 |
| bauzas | johnthetubaguy: ack, CC'ing it | 12:59 |
| johnthetubaguy | bauzas: basically its the hypervisor support matrix, with more dimensions added | 12:59 |
| bauzas | johnthetubaguy: okay I need to read the spec before hassling you with silly questions | 13:00 |
| bauzas | :) | 13:00 |
| johnthetubaguy | bauzas: basically, link actual tempest test uuids, link to real docs, link to real CI results, in an ideal world | 13:00 |
| *** ijuwang has joined #openstack-nova | 13:00 | |
| *** salv-orlando has quit IRC | 13:01 | |
| johnthetubaguy | bauzas: so you have a feature rating, and a rating for each feature against a specific CI setup | 13:01 |
| sdague | bauzas: basically it's actually trying to describe to end users what they can get out of the cloud, because there are API features which are implementation specific | 13:01 |
| johnthetubaguy | sdague: yeah +1 | 13:01 |
| *** dims has joined #openstack-nova | 13:01 | |
| bauzas | sdague: I see, okay :) | 13:02 |
| bauzas | thanks for the pointer, will look into it | 13:02 |
| *** sudipto has quit IRC | 13:02 | |
| *** dave-mccowan has joined #openstack-nova | 13:03 | |
| openstackgerrit | Ankit Agrawal proposed openstack/nova: Test patch libvirt race condition (do not merge) https://review.openstack.org/191732 | 13:03 |
| oomichi_ | re: auto-generating api doc: We can get most part of api doc(resource url, normal status code, sample request/response) from tempest.log | 13:03 |
| *** ildikov has quit IRC | 13:03 | |
| oomichi_ | re: auto-generating api doc: This way is possible to be applied to all projects | 13:03 |
| *** mdrabe has joined #openstack-nova | 13:04 | |
| oomichi_ | re: auto-generating api doc: In nova, we can create the descriptions of api parameters of request from json-schema | 13:04 |
| *** moshele has quit IRC | 13:04 | |
| johnthetubaguy | oomichi_: yeah, json schema works | 13:05 |
| oomichi_ | re: auto-generating api doc: there is json-schema of nova *response* in Tempest now | 13:05 |
| johnthetubaguy | yeah | 13:05 |
| *** VW has joined #openstack-nova | 13:05 | |
| oomichi_ | re: auto-generating api doc: we will be able to migrate response schema also from Tempest to nova with Tempest's external plugin interface | 13:05 |
| oomichi_ | re: auto-generating api doc: After the migration, we can get the descriptions of api parameters of response also | 13:06 |
| oomichi_ | re: auto-generating api doc: then, we can get complete api doc automatically | 13:07 |
| *** mriedem_away is now known as mriedem | 13:07 | |
| meh_di | bauzas: replace the IP of Compute node insteed of the IP of remote server | 13:07 |
| bauzas | meh_di: that depends on what you want to introspect | 13:09 |
| bauzas | meh_di: that's just a remote debugger | 13:10 |
| johnthetubaguy | oomichi_: seems like a very sound approach to me | 13:10 |
| bauzas | meh_di: so you need to put a breakpoint on where you want to introspect | 13:10 |
| *** annegentle has joined #openstack-nova | 13:11 | |
| bauzas | meh_di: meaning that you should point your IDE to the nova service you want to debug (again, either the API, or the conductor, or the scheduler, or one of the compute managers) | 13:11 |
| pkholkin | Hello guys! We are working on patch 'switching to oslo_db.enginefacade' in nova. We already have init patch https://review.openstack.org/#/c/221310/12 and two other patches that use enginefacade | 13:11 |
| meh_di | i am studing nova code already | 13:12 |
| pkholkin | Could you please look, we are not sure that we configure enginefacade correctly | 13:12 |
| bauzas | meh_di: so you don't need a remote debugger, static code analysis is enough IMHO | 13:12 |
| pkholkin | we add 'configure' method in /cmd/manage.py but we don't really configure in services (in production code) | 13:12 |
| jsuchome | mriedem: hi, could I ask about https://review.openstack.org/#/c/225119/ ? Do you really think it has to wait until some discussion with cinder developers? | 13:12 |
| bauzas | meh_di: begin with a very simple scenario : "I want to boot a single VM with just a flavor and an image" | 13:13 |
| pkholkin | jaypipes: Hello Jay! Could you help me with the question above | 13:13 |
| bauzas | meh_di: so, start with the API resource corresponding to that and enter the rabbit hole | 13:13 |
| mriedem | jsuchome: this is the fix we want https://review.openstack.org/#/c/227564/ | 13:13 |
| pkholkin | and advise whom to ask too | 13:13 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add pep8 check for assertEqual argument order https://review.openstack.org/227650 | 13:13 |
| bauzas | johnthetubaguy: oh, we haven't yet merged jogo's changes for the architecture ? | 13:15 |
| bauzas | (devref, I mean) | 13:15 |
| jsuchome | mriedem: ah, cool! | 13:15 |
| bauzas | meh, pointing to the Kilo doc | 13:16 |
| mriedem | jsuchome: i also duplicated the bug on yours to something else | 13:16 |
| jsuchome | mriedem: yes, I was wondering, if your patch shouldn't also reference that bug | 13:16 |
| bauzas | johnthetubaguy: nvm, my cookie was pointing me to a stable doc branch | 13:17 |
| johnthetubaguy | bauzas: yeah, not sure we have, do you have the review link? | 13:17 |
| bauzas | johnthetubaguy: nah it's good http://docs.openstack.org/developer/nova/architecture.html | 13:17 |
| bauzas | for some reason, my doc footer was telling me I was looking at 2015.1.1 devref | 13:17 |
| bauzas | so I had to go up to the main docs site, and go back | 13:18 |
| mriedem | jsuchome: it depends on the case, in bug 1497253 you're talking about cross_az_attach being True and False | 13:18 |
| openstack | bug 1489575 in Cinder "duplicate for #1497253 Cinder always tries to create volumes in Cinder AZ with same name as Nova AZ" [Medium,Fix released] https://launchpad.net/bugs/1489575 - Assigned to Edmund Rhudy (erhudy) | 13:18 |
| bauzas | meh_di: look at this http://docs.openstack.org/developer/nova/architecture.html | 13:18 |
| mriedem | in the False case, the AZ is set and has to be on the cinder side | 13:18 |
| pkholkin | mriedem: Good day Matt! Could you please help with patch related to enginefacade? https://review.openstack.org/#/c/221310/12 | 13:18 |
| mriedem | jsuchome: if not set in cinder, then you either have to create the AZ in cinder or set allow_availability_zone_fallback=True in cinder.conf | 13:18 |
| mriedem | pkholkin: why do we want this engine facade thing? | 13:19 |
| bauzas | meh, catching up some convo about AZs | 13:19 |
| bauzas | my AZ radar just blinked on my screen | 13:19 |
| pkholkin | mriedem: it related to this spec http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/oslo_db-enginefacade.html | 13:20 |
| mriedem | pkholkin: why don't we have a specless trivial blueprint for the enginefacade stuff? | 13:20 |
| *** dustins has joined #openstack-nova | 13:21 | |
| pkholkin | mriedem: we communicate with Matthew Booth, he promised to update it for Mitaka | 13:21 |
| mriedem | pkholkin: ok, seems that should be reproposed for mitaka as a trivial re-propose | 13:21 |
| pkholkin | mriedem: yes, right | 13:21 |
| mriedem | pkholkin: and your changes aren't linked to the bp | 13:21 |
| jsuchome | mriedem: hm, my case is when cross_az_attach is False, and as I look at your last patch, I think it will behave the same way as it does now | 13:21 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Log warning for wrong migration flag config options https://review.openstack.org/228853 | 13:21 |
| *** obedmr has joined #openstack-nova | 13:21 | |
| openstackgerrit | garyk proposed openstack/nova: VMware: raise NotImplementedError for live migration methods https://review.openstack.org/228893 | 13:21 |
| mriedem | jsuchome: yes, my patch is really for the cross_az_attach case | 13:21 |
| *** links has quit IRC | 13:22 | |
| pkholkin | mriedem: yes, correct, it will be done, we already work on code | 13:22 |
| mriedem | jsuchome: like i said, if cross_az_attach is False you either have to have the same AZ in cinder as you're booting into in nova or you have to set allow_availability_zone_fallback=True in cinder.conf to avoid the error | 13:22 |
| *** jdurgin1 has joined #openstack-nova | 13:23 | |
| pkholkin | mriedem: we have a question related to the patch, could you try to help us | 13:23 |
| jsuchome | mriedem: I think that fallback does not help, it's there for cases when wanted cinder AZ would not be available, right? | 13:23 |
| mriedem | jsuchome: right | 13:23 |
| mriedem | i personally don't see a good reason for that fallback option in cinder, it's clearly a hack in cinder to workaround the bug(s) in nova related to that cross_az_attach option | 13:23 |
| mriedem | jsuchome: if you haven't noticed yet, there is a big thread in the ML about this | 13:24 |
| jsuchome | which is not my case, the problem is, volume is correcly created in existing AZ, instance would be correctly created in different nova's AZ,but user doesn't want this to happen | 13:24 |
| mriedem | then set cinder.cross_az_attach=True in nova.conf | 13:25 |
| pkholkin | mriedem: there are two facades there for main and api databases, so we need to configure them exactly once with their config options, we inserted it here https://review.openstack.org/#/c/221310/12/nova/cmd/manage.py | 13:25 |
| pkholkin | mriedem: but we are not sure in this code, because we don't call configure in services like nova-api (in production code) | 13:25 |
| pkholkin | mriedem: do you know how it should be and what services need api database? | 13:26 |
| jsuchome | mriedem: ok, customer requires cross_az_attach=False... and it works if he provides explicit AZ to nova when booting instance, but not with no AZ provided | 13:26 |
| jsuchome | (yes, I've seen the ML) | 13:26 |
| mriedem | jsuchome: i guess you should provide more details in your bug report then. is this a boot from volume case where the volume is provided (bdm.source=volume)? | 13:27 |
| mriedem | pkholkin: that's a good question for alaski but the api db would be for the nova-api service i believe | 13:27 |
| *** Daisy has quit IRC | 13:27 | |
| oomichi_ | johnthetubaguy: sdague: gmann: If we can implement complete api doc and the same single API between clouds for interoperability, | 13:27 |
| oomichi_ | johnthetubaguy: sdague: gmann: the value of json-home is not so high | 13:27 |
| jsuchome | it's boot from volume, but user doesn't want to explicitely provide AZ, just wants to nova to pick the one, and let cinder use the same name in cinder | 13:27 |
| jsuchome | mriedem: I'll update bug report, it's likely I wasn't clear enough | 13:28 |
| oomichi_ | johnthetubaguy: sdague: gmann: because api users can know how to use api from the doc completely | 13:28 |
| alaski | mriedem: pkholkin reading scrollback | 13:28 |
| *** alejandrito has quit IRC | 13:28 | |
| mriedem | jsuchome: yeah, but my point is nova doesn't know the volume AZ unless the volume is provided in the boot from volume case | 13:28 |
| *** rlrossit has joined #openstack-nova | 13:29 | |
| pkholkin | mriedem: ok, thanks for adding him, do you think we should add such configure in some other services? | 13:29 |
| mriedem | jsuchome: if the case is, cinder.cross_az_attach=False, nova boot w/o nova AZ provided, and bdm.source=volume and volume has an AZ set - and you want to boot the instance into that same AZ that the volume is in, then that's a new thing | 13:29 |
| alaski | pkholkin: as mriedem said it will need to be configured for nova-api, and also nova-conductor. For now. That may expand later | 13:30 |
| jsuchome | mriedem: I think volume hasn't AZ set, several boots can can end in different AZ's, the requirement is just that they are same for nova and cinder side | 13:30 |
| mriedem | jsuchome: and nova is creating the volume? | 13:31 |
| *** twinklechawla has joined #openstack-nova | 13:31 | |
| jsuchome | yes | 13:31 |
| *** salv-orlando has joined #openstack-nova | 13:31 | |
| mriedem | jsuchome: then if an AZ isn't specified on nova boot, then it should be None and there shouldn't be an AZ used for either the server or volume | 13:31 |
| pkholkin | alaski: ok, we will add 'configure' to /cmd for these services too | 13:31 |
| *** ctrath has joined #openstack-nova | 13:31 | |
| mriedem | jsuchome: if cinder is randomly assigning an AZ to the volume then that seems like a cinder bug | 13:31 |
| mriedem | otherwise if you want to be in a specific AZ for the server, that has to be provided at boot | 13:32 |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack/nova: libvirt: Log warning for wrong migration flag config options https://review.openstack.org/228853 | 13:32 |
| *** oomichi_ has quit IRC | 13:32 | |
| jsuchome | mriedem: I have to leave now, thanks for help, I'll look at cinder conf more closely | 13:33 |
| pkholkin | mriedem, alaski: another question, about nova-cells, we wanted to deploy devstack with nova-cells and try to test our patch but be can't | 13:33 |
| *** jsuchome has quit IRC | 13:33 | |
| pkholkin | mriedem, alaski: we found several guides and this doc http://docs.openstack.org/kilo/config-reference/content/section_compute-cells.html, is there something else? | 13:34 |
| alaski | pkholkin: why not? There are instructions in the readme for Devstack | 13:34 |
| *** stevemar has joined #openstack-nova | 13:34 | |
| mriedem | http://docs.openstack.org/developer/devstack/configuration.html?highlight=cells#cells | 13:35 |
| pkholkin | alaski: could you please some link? | 13:35 |
| pkholkin | mriedem, alaski: should it be multinode rigth? | 13:35 |
| *** stevemar_ has joined #openstack-nova | 13:36 | |
| mriedem | it's single node in the gate | 13:36 |
| alaski | pkholkin: https://github.com/openstack-dev/devstack/blob/master/doc/source/configuration.rst#cells | 13:36 |
| alaski | I guess it moved out of the Readme | 13:36 |
| markus_z | The cell feature is done via a service? I thought it was solely done via the REST API? | 13:36 |
| *** haomaiwang has joined #openstack-nova | 13:36 | |
| mriedem | let's all gather round and let old man laski explain cells v1 and cells v2 | 13:37 |
| mriedem | markus_z: http://docs.openstack.org/developer/nova/cells.html | 13:37 |
| alaski | markus_z: it's done by two main mechanisms. The compute api is subclassed and methods or RPC methods are overridden. They are overriden to send RPC messages to a cells service which proxies them to the appropriate cell | 13:37 |
| *** stevemar_ has quit IRC | 13:37 | |
| *** stevemar_ has joined #openstack-nova | 13:38 | |
| pkholkin | I also don't understand waht should be in nova_cell database and what it nova_api database, could you please explain? | 13:38 |
| markus_z | alaski: I see, thanks. I must have noted down something wrong. | 13:38 |
| alaski | pkholkin: you seem to be interested in the nova_api database which is not used in cells v1. And devstack is only setup for cellsv1 | 13:39 |
| *** stevemar has quit IRC | 13:39 | |
| alaski | pkholkin: so you may be able to ignore that | 13:39 |
| mriedem | although devstack does create the nova-api db | 13:39 |
| mriedem | to run the migrations | 13:39 |
| *** ildikov has joined #openstack-nova | 13:39 | |
| alaski | yeah | 13:39 |
| *** alejandrito has joined #openstack-nova | 13:40 | |
| *** haomaiwang has quit IRC | 13:40 | |
| lxsli | Now Mitaka is open, anyone fancy https://review.openstack.org/#/c/215568 and https://review.openstack.org/#/c/215613 re: bug 1428639 please? They've been up a while and had some +1s | 13:40 |
| openstack | bug 1428639 in openstack-ansible "Nova requires python-libguestfs in Kilo" [Medium,Fix released] https://launchpad.net/bugs/1428639 - Assigned to Kevin Carter (kevin-carter) | 13:40 |
| alaski | one key point that I should probably write down is that cellv2 will sort of just happen. it shouldn't require much in the way of a different configuration for devstack | 13:40 |
| alaski | just a few additional calls to setup the new db, and some migrations of data at some poitn | 13:40 |
| *** aix has joined #openstack-nova | 13:40 | |
| *** haomaiwa_ has joined #openstack-nova | 13:40 | |
| *** tdurakov has quit IRC | 13:40 | |
| twinklechawla | Hello everyone, Till date I read about nova, cloned openstack-dev/devstack and openstack/nova. I am new to this, so any one who can help me with it? | 13:41 |
| pkholkin | mriedem, alaski: sorry but I don't fully understand what information is stored in nova_api and what in nova_cell dbs and how are they related | 13:41 |
| *** tdurakov has joined #openstack-nova | 13:42 | |
| alaski | pkholkin: they are not is the short answer | 13:42 |
| pkholkin | alaski: ok) and could be describe also short about data in them) | 13:42 |
| pkholkin | alaski: ok) and could you describe also short about data in them) | 13:42 |
| alaski | pkholkin: the nova_api database will store information for the nova API to use to talk to the proper cell, as well as "global" data like flavors | 13:42 |
| alaski | that's for cellsv2 | 13:43 |
| alaski | the nova_cells db is for cellsv1, and has the local cell copy of the nova db. In other words the nova db is an aggregate of all nova_cell dbs in a deployment | 13:43 |
| *** burt has joined #openstack-nova | 13:44 | |
| bauzas | ... too ... much ... convos ... to ... read | 13:44 |
| *** jaypipes has quit IRC | 13:44 | |
| markus_z | twinklechawla: A good starting point is here: https://wiki.openstack.org/wiki/How_To_Contribute#If_you.27re_a_developer | 13:45 |
| *** julien_ZTE has joined #openstack-nova | 13:46 | |
| pkholkin | alaski: try to understand) hm, we will look through you given links and messages, thanks a lot!!! | 13:46 |
| *** mdbooth_ has joined #openstack-nova | 13:46 | |
| pkholkin | alaski: one last question right now, do you think if we add configuring facade in nova-api and nova-conductor services too should everything works ok? | 13:47 |
| pkholkin | I mean what should we try or test | 13:47 |
| jroll | bauzas: that's what you get for looking away from IRC for 2 seconds :D | 13:48 |
| alaski | pkholkin: the main thing right now is that the nova_api db migrations continue to run. they're being run in devstack so you should see failures there if that breaks | 13:48 |
| bauzas | jroll: well, my radar screen gave me lots of plots about AZs and cells | 13:48 |
| jroll | heh | 13:49 |
| mriedem | dims: i see a few places in nova where context is passed as a kwarg to logging methods, like the instance kwarg - is there a reason for that? | 13:49 |
| pkholkin | alaski: ok (this job gate-tempest-dsvm-cell is for cells v1?) | 13:49 |
| *** marzif has quit IRC | 13:49 | |
| *** jaypipes has joined #openstack-nova | 13:49 | |
| *** mdbooth has quit IRC | 13:49 | |
| *** mdbooth_ is now known as mdbooth | 13:49 | |
| mriedem | pkholkin: yes | 13:50 |
| *** erlon has joined #openstack-nova | 13:50 | |
| pkholkin | mriedem, alaski: Thanks you guys a lot for the answers! We will continue working in it | 13:50 |
| dims | mriedem: i think when we elevate up the privilege to admin we have to log the original context (if i remember right) | 13:51 |
| openstackgerrit | Chuck Carmack proposed openstack/nova-specs: Allow user to set and retrieve the server Description https://review.openstack.org/228908 | 13:51 |
| pkholkin | mdbooth: Hello Matthew, could tou please update your spec related to enginefacade for Mitaka please http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/oslo_db-enginefacade.html | 13:51 |
| mriedem | dims: because the request info is in the context? | 13:51 |
| dims | y | 13:52 |
| pkholkin | mdbooth: we continue working on our patches https://review.openstack.org/#/c/221310/12 | 13:52 |
| *** otter768 has joined #openstack-nova | 13:52 | |
| mriedem | dims: seems that if we're using oslo.context's RequestContext, then we don't need to pass in the context kwarg http://docs.openstack.org/developer/oslo.log/usage.html#passing-context | 13:53 |
| alaski | pkholkin: mriedem the migration gets run if nnova-api is enabled https://github.com/openstack-dev/devstack/blob/master/lib/nova#L722 | 13:53 |
| mriedem | dims: if i'm reading that correctly | 13:53 |
| dims | mriedem: a real easy one - https://review.openstack.org/#/c/228600/ - doubles the py34 test coverage | 13:53 |
| *** nkrinner has quit IRC | 13:53 | |
| dims | mriedem: let me go back and read the code | 13:53 |
| mriedem | dims: b/c otherwise i think we actually have problems in nova b/c i've seen elevated contexts passed as kwargs to logging methods | 13:53 |
| alaski | pkholkin: which means every nova tempest job will run it. There are also functional tests in tree for the nova_api db | 13:53 |
| *** Sree has joined #openstack-nova | 13:54 | |
| *** stevemar_ is now known as stevemar | 13:54 | |
| *** irenab has quit IRC | 13:54 | |
| dims | mriedem: right, if code is elevating, then the original context must be used as context= parameter | 13:55 |
| pkholkin | alaski, mriedem: the problem was that jenkins marked +1 but we were not sure that everything is configured correctly (mostly related to api db) | 13:56 |
| dims | this is clearly wrong mriedem | 13:56 |
| dims | context = context.elevated() | 13:56 |
| dims | LOG.info(_LI("Rebooting instance"), context=context, instance=instance) | 13:56 |
| mriedem | dims: yeah, i guess i'm trying to simplify it and just figure out (1) if you're using oslo.context in your app then (2) you shouldn't have to pass the context kwarg to logging methods | 13:56 |
| mriedem | because if ^ is true we could have a hacking check for it in nova | 13:57 |
| alaski | pkholkin: I'm not actually sure if a failure would cause a test failure. It depends on what the 'nova-manage db_sync' returns for a failure | 13:57 |
| *** otter768 has quit IRC | 13:57 | |
| dims | mriedem: even if you are using oslo.context when you elevate you should still be able to specify the original context to log | 13:57 |
| mriedem | dims: but do you need to? | 13:58 |
| *** pm90_ has joined #openstack-nova | 13:58 | |
| dims | mriedem: i believe so | 13:58 |
| pkholkin | alaski: yes, that's why we are asking) do you have thoughts how to test it for sure (with devstack e.g.) | 13:59 |
| dims | mriedem: hang on, reading the elevated method in nova's context.py | 13:59 |
| alaski | mriedem: oslo.logging will use the context that's set in http://git.openstack.org/cgit/openstack/oslo.context/tree/oslo_context/context.py#n71. If you want to log a different context you would pass it | 13:59 |
| alaski | get_admin_context defaults to not overwriting that | 14:00 |
| mriedem | i don't know why we would pass a different one in though | 14:00 |
| mriedem | elevated() just sets the admin bit | 14:00 |
| mriedem | request id and user/project stay the same | 14:00 |
| dims | alaski: if you are using elevated() looks like it will work fine. if you use get_admin_context() then it wont | 14:00 |
| alaski | yeah, I can't think of a reason at the moment | 14:00 |
| *** haomaiwa_ has quit IRC | 14:01 | |
| mriedem | dims: yeah using get_admin_context() instead would be bad | 14:01 |
| *** haomaiwa_ has joined #openstack-nova | 14:01 | |
| alaski | get_admin_context will create a new context, but not overwrite the one that logging would use. so I guess it depends on what you intend on logging | 14:02 |
| *** iurygregory has joined #openstack-nova | 14:02 | |
| dims | right, it's there in oslo.log if nova chooses to use it | 14:02 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Manifesto for working on Tasks https://review.openstack.org/221280 | 14:03 |
| dims | +1 to check all usages and remove context= if it's not needed | 14:03 |
| mriedem | yeah....i just did a search | 14:03 |
| mriedem | at least 36 hits on the same line | 14:03 |
| *** annegentle has quit IRC | 14:03 | |
| mriedem | i can clean those up in a bit, i'm getting sidetracked | 14:04 |
| *** tdurakov has quit IRC | 14:04 | |
| kashyap | Can I just respond to launchpad bug emails from my mail client? IIRC, LP allows that. | 14:05 |
| kashyap | Hmm, this is off-topic here. | 14:05 |
| markus_z | kashyap: Some bug comments look like that this is possible | 14:05 |
| twinklechawla | I cloned the code but I am unable to understand it --> /devstack/nova | 14:06 |
| kashyap | markus_z: Yeah, I'm trying to confirm if I can just do that, it'd be much nicer rather than loading a million bugs in my browser. | 14:06 |
| *** eharney has quit IRC | 14:06 | |
| mriedem | dims: can you confirm the description in this bug makes sense? https://bugs.launchpad.net/nova/+bug/1500896 | 14:07 |
| openstack | Launchpad bug 1500896 in OpenStack Compute (nova) "It's not necessary to pass context as kwarg to oslo.log in most cases" [Low,Triaged] | 14:07 |
| *** mc_nair has joined #openstack-nova | 14:07 | |
| dims | mriedem: ++ | 14:07 |
| mriedem | ctrath: ccarmack: rlrossit: doffm_: low hanging fruit bug https://bugs.launchpad.net/nova/+bug/1500896 | 14:08 |
| dims | mriedem: is cloudpipe stuff a thing anyone uses? | 14:10 |
| dims | and the nova-objectstore | 14:10 |
| markus_z | twinklechawla: I'm afraid you have to ask very specific questions. | 14:10 |
| *** mdrabe has quit IRC | 14:11 | |
| mriedem | dims: don't know | 14:11 |
| twinklechawla | markus_z: I am very new this, so need help! | 14:12 |
| raildo | hi guys, I'm with a doubt here, there is some tests like this: https://github.com/openstack/nova/blob/master/nova/tests/unit/test_quota.py#L1209-L1214 that expect return the remains quotas for a project but the quota calculation sounds odd to me. | 14:12 |
| markus_z | twinklechawla: Did you already have the change to read https://wiki.openstack.org/wiki/How_To_Contribute ? | 14:13 |
| raildo | like... I have a limit = 5, in_use =2, reserved = 2, how remains can be equal zero? | 14:13 |
| markus_z | *chance | 14:13 |
| twinklechawla | markus_z:I read this! | 14:13 |
| markus_z | twinklechawla: OK, cool, any questions to that? | 14:14 |
| *** achanda has joined #openstack-nova | 14:15 | |
| dims | mriedem: alaski: couple of easy ones if you get a chance https://review.openstack.org/#/c/228600/ and https://review.openstack.org/#/c/228688/ - first one doubles the py34 coverage and the 2nd one is a mechanical replacement with a hacking rule for contextlib.nested | 14:15 |
| dims | twinklechawla: opw? | 14:15 |
| twinklechawla | markus_z: I went to "Getting_the_code" and cloned 'nova' to my filesystem but there are many files to understand. | 14:15 |
| dims | twinklechawla: are you trying to pick a project to work on? | 14:15 |
| twinklechawla | dims: yes! | 14:16 |
| raildo | melwitt, ping, are you around? | 14:16 |
| dims | twinklechawla: so yes, Nova is a large codebase, i usually recommend picking up a small bug/problem and try to figure out the code just for that piece, as you go along you will learn more and more of it | 14:17 |
| *** thangp has quit IRC | 14:19 | |
| twinklechawla | dims: okay, I got it. But can you plz help me to get that 'problem' which I can solve? | 14:19 |
| *** rfolco has quit IRC | 14:19 | |
| *** rfolco has joined #openstack-nova | 14:19 | |
| markus_z | dims: twinklechawla: Maybe that one: https://bugs.launchpad.net/nova/+bug/1357751 ? | 14:20 |
| openstack | Launchpad bug 1357751 in OpenStack Compute (nova) "nova.tests.network.test_manager.AllocateTestCase should use mock" [Low,In progress] | 14:20 |
| raildo | twinklechawla, there is a low-hanging-fruit bug list here: https://bugs.launchpad.net/nova/+bugs?field.tag=low-hanging-fruit | 14:20 |
| *** thangp has joined #openstack-nova | 14:21 | |
| raildo | twinklechawla, usually this bugs are easy to solve | 14:21 |
| *** achanda has quit IRC | 14:22 | |
| *** SunSeeker has quit IRC | 14:23 | |
| twinklechawla | markus_z: raildo: dims: Please, make me understand, what kind of bugs are there in ''nova. | 14:23 |
| mriedem | twinklechawla: there are 582 bugs that aren't in progress, so you can take your pick :) | 14:24 |
| mriedem | twinklechawla: there is newcomer stuff in the nova wiki here https://wiki.openstack.org/wiki/Nova/Mentoring | 14:25 |
| *** Murali has quit IRC | 14:25 | |
| *** changbl has joined #openstack-nova | 14:25 | |
| dims | alaski: thanks! | 14:25 |
| alaski | dims: np. will find time for the second one in a bit | 14:26 |
| dims | mriedem: ah! looks like i could use a mentor :) | 14:26 |
| *** artom has joined #openstack-nova | 14:28 | |
| twinklechawla | well! I got very confused while watching these bugs :| | 14:29 |
| *** asettle has quit IRC | 14:29 | |
| *** annegentle has joined #openstack-nova | 14:29 | |
| markus_z | twinklechawla: That's normal. | 14:30 |
| twinklechawla | Really? | 14:30 |
| markus_z | twinklechawla: yes. Get's better after a few months. :) | 14:31 |
| raildo | twinklechawla, what kind of bug do you expect resolve? docs? database? some exception error? | 14:31 |
| raildo | markus_z, ++ | 14:32 |
| *** gszasz has quit IRC | 14:33 | |
| twinklechawla | markus_z: Is it necessary to submit a patch, before Application Deadline? | 14:33 |
| dims | twinklechawla: wrong channel for that question, you should ask the opw admins on their irc channel | 14:33 |
| markus_z | what's "opw"? | 14:34 |
| twinklechawla | markus_z: Outreachy Program for Women :) | 14:35 |
| *** phschwartz has quit IRC | 14:36 | |
| *** phschwartz has joined #openstack-nova | 14:36 | |
| dims | twinklechawla: this may be a easier bug https://bugs.launchpad.net/nova/+bug/1398629, there are several tests that use REQUIRES_LOCKING = True, you can pick one of those tests and fix it as mentioned in the bug report | 14:36 |
| openstack | Launchpad bug 1398629 in OpenStack Compute (nova) "Fix buggy tests that use REQUIRES_LOCKING=True" [Medium,Confirmed] - Assigned to Sean Dague (sdague) | 14:36 |
| *** MVenesio has joined #openstack-nova | 14:38 | |
| *** mdrabe has joined #openstack-nova | 14:39 | |
| dims | twinklechawla: example see nova/tests/unit/virt/vmwareapi/test_configdrive.py, flip the REQUIRES_LOCKING variable to False and you will see that when you run tests using "tox -e py27" it will fail, you need to figure out how to fix the test and submit a patch | 14:39 |
| *** salv-orlando has quit IRC | 14:39 | |
| dims | markus_z: agree? ^ | 14:40 |
| twinklechawla | dims: okey.. now I understood the thing! Thanks dim! Now I'll check it out! | 14:41 |
| markus_z | dims: yeah, sounds like a good start | 14:41 |
| *** su_zhang has joined #openstack-nova | 14:43 | |
| *** electrocucaracha has joined #openstack-nova | 14:43 | |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: WIP: Use legacy engineFacade from oslo_db.enginefacade https://review.openstack.org/221310 | 14:44 |
| openstackgerrit | Sergey Nikitin proposed openstack/nova: WIP: use oslo_db.enginefacade https://review.openstack.org/228979 | 14:44 |
| *** electrocucaracha has quit IRC | 14:44 | |
| openstackgerrit | Chuck Carmack proposed openstack/nova-specs: Allow user to set and retrieve the server Description https://review.openstack.org/228908 | 14:45 |
| *** electrocucaracha has joined #openstack-nova | 14:45 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add checkpoint logging when building an instance in compute manager https://review.openstack.org/228980 | 14:45 |
| mriedem | alaski: following up from logging talk yesterday with the large ops bug ^ | 14:45 |
| *** rfolco has quit IRC | 14:45 | |
| *** rfolco has joined #openstack-nova | 14:47 | |
| alaski | mriedem: there's a StopWatch thing you can use from oslo.utils if you wish http://git.openstack.org/cgit/openstack/oslo.utils/tree/oslo_utils/timeutils.py#n325 but I'm not -1 on that | 14:47 |
| *** mtanino has joined #openstack-nova | 14:47 | |
| *** digimonde has quit IRC | 14:48 | |
| alaski | looks reasonable otherwise | 14:48 |
| mriedem | alaski: seems like a sledehammer | 14:48 |
| alaski | yeah, just wanted to mention it | 14:48 |
| dansmith | mriedem: do you really want info on all those except the timing one? | 14:48 |
| dansmith | mriedem: for a busy cloud that's a lot of noise | 14:49 |
| mriedem | the timing one is debug | 14:49 |
| dansmith | right | 14:49 |
| mriedem | Starting instance is already info | 14:49 |
| openstackgerrit | Chuck Carmack proposed openstack/nova-specs: Allow user to set and retrieve the server Description https://review.openstack.org/228908 | 14:49 |
| dansmith | right, but you're adding several more infos for each instance | 14:49 |
| dansmith | if alaski thinks it's cool, then that's good enough for me | 14:49 |
| mriedem | let me consult the all knowing and all powerful openstack logging guidelines handbook | 14:49 |
| mriedem | https://wiki.openstack.org/wiki/LoggingStandards#Messages_at_INFO_should_be_a_unit_of_work | 14:50 |
| alaski | well I might be a bit biased since we run on debug anyways. but it's a fair point | 14:50 |
| dansmith | heh | 14:50 |
| dansmith | and also wow | 14:50 |
| dansmith | although not surprising | 14:50 |
| mriedem | i guess according to https://wiki.openstack.org/wiki/LoggingStandards#Messages_at_INFO_should_be_a_unit_of_work and what i'm adding, it'd fall into the 'bad' category | 14:51 |
| mriedem | although https://wiki.openstack.org/wiki/LoggingStandards#Messages_at_INFO_should_be_a_unit_of_work has always confused me | 14:51 |
| alaski | it's a lot of data. but I really can't imagine not having it right now | 14:51 |
| mriedem | good = thing is done | 14:51 |
| mriedem | bad = starting thing | 14:51 |
| *** achanda has joined #openstack-nova | 14:51 | |
| dansmith | right, I think that info for "building instance" or "instance is built" is a unit of work | 14:51 |
| *** armax has joined #openstack-nova | 14:51 | |
| dansmith | "doing step 12" is not | 14:51 |
| *** twinklechawla has quit IRC | 14:52 | |
| mriedem | i can change to all being debug | 14:52 |
| alaski | dansmith: you're recommending switching them all? | 14:52 |
| dansmith | alaski: well, I'm just asking, but they feel like debugs to me | 14:52 |
| *** lykinsbd_ has joined #openstack-nova | 14:52 | |
| alaski | yeah. although now you've almost convinced me that the "Took x seconds..." one should be info | 14:52 |
| mriedem | do you need a biased 3rd party opinion? | 14:52 |
| dansmith | alaski: really? | 14:53 |
| mriedem | given https://wiki.openstack.org/wiki/LoggingStandards#Messages_at_INFO_should_be_a_unit_of_work the timing one would be info | 14:53 |
| alaski | well that one is a unit of work right? | 14:53 |
| mriedem | since it's basically 'built instance' | 14:53 |
| mriedem | 'build instance in x seconds' | 14:53 |
| mriedem | *built | 14:53 |
| dansmith | there is an existing info for that right? or is that "starting build" ? | 14:53 |
| mriedem | that's 'Starting instance' | 14:54 |
| tojuvone | dansmith: I was gathering these several comments to update get-valid-server-state review.. but to proceed better couple of questions | 14:54 |
| mriedem | at the top of that method | 14:54 |
| mriedem | which is arguably not info level | 14:54 |
| *** salv-orlando has joined #openstack-nova | 14:54 | |
| dansmith | mriedem: maybe the existing should be debug and your timing one should be info? | 14:54 |
| dansmith | tojuvone: sure, shoot | 14:54 |
| mriedem | dansmith: i think that's what alaski is saying | 14:54 |
| mriedem | and i'd be ok with that | 14:54 |
| alaski | yeah, that seems the best to me | 14:54 |
| mriedem | and changing 'starting instance' to debug | 14:54 |
| dansmith | yep, okay | 14:54 |
| mriedem | hot damn! | 14:55 |
| *** lpetrut has quit IRC | 14:55 | |
| tojuvone | dansmith one would be if to expose even more from service, like enabled/disabled. Would that be good? | 14:55 |
| dansmith | tojuvone: yeah, I think I said that in one of my comments, and that seems to be the desire from the others | 14:55 |
| dansmith | tojuvone: I think it'd be good to think about what things to expose there and make them more abstract than the actual names we have for them | 14:55 |
| dansmith | tojuvone: meaning, maybe not enabled/disabled/force-down/unknown but something like 'normal/maintenance/down/unknown' | 14:56 |
| tojuvone | yes, for now I was thinking: host_state and host_status | 14:56 |
| tojuvone | enabled/disabled + up,forced_down and unknown | 14:57 |
| dansmith | tojuvone: hmm, so two keys? | 14:57 |
| *** irenab has joined #openstack-nova | 14:57 | |
| tojuvone | host_status that is straight enabled/disabled | 14:58 |
| *** tpeoples has joined #openstack-nova | 14:58 | |
| tojuvone | host_state that is up,forced_down or unknown | 14:58 |
| dansmith | tojuvone: but that becomes six possibilities instead of four | 14:58 |
| dansmith | tojuvone: I dunno, I feel like a single key is all we really need, but if you feel strongly, you can propose it and write up the reasoning and we can have a look | 14:59 |
| tojuvone | ok... so you mean disabled to be translateed as maintenance? | 14:59 |
| *** annegentle has quit IRC | 14:59 | |
| *** achanda has quit IRC | 14:59 | |
| dansmith | tojuvone: yeah, or something suitable | 14:59 |
| openstackgerrit | Ankit Agrawal proposed openstack/nova: Test patch libvirt race condition (do not merge) https://review.openstack.org/191732 | 14:59 |
| *** angdraug has joined #openstack-nova | 14:59 | |
| openstackgerrit | Marian Horban proposed openstack/nova: [WIP]Remove wsgi functionality https://review.openstack.org/219248 | 14:59 |
| dansmith | tojuvone: some non-operator looking at "disabled" may think it means more down-ness than it does in a nova context | 14:59 |
| *** annegentle has joined #openstack-nova | 15:00 | |
| tojuvone | well, I also think future if need to expose more, but still one key might do | 15:00 |
| dansmith | tojuvone: well, if you can think of additional things we might want in the future, that's good for the reasoning part :) | 15:00 |
| tojuvone | Like if there would be "going_for_maintenance_soon" later | 15:00 |
| *** haomaiwa_ has quit IRC | 15:01 | |
| tojuvone | it could be exposed to user in similar fashion | 15:01 |
| dansmith | maybe | 15:01 |
| *** haomaiwang has joined #openstack-nova | 15:01 | |
| dansmith | but write it up and we can look at it in the spec -- that's what it's for :) | 15:01 |
| alaski | +1 for details in the spec. I'd like to offer comments but am multitasking currently | 15:01 |
| tojuvone | lyes, just to write things bit better to skip week of review ;) | 15:02 |
| *** moshele has joined #openstack-nova | 15:02 | |
| dansmith | well, that's where we should be having the bulk of the discussion, because it's recorded | 15:02 |
| dansmith | well, that's where we should be having the bulk of the discussion, because it's recorded | 15:02 |
| dansmith | oops :) | 15:02 |
| *** pm90__ has joined #openstack-nova | 15:03 | |
| *** Piet_ has quit IRC | 15:03 | |
| *** pm90___ has joined #openstack-nova | 15:03 | |
| tojuvone | dansmith: Other things is if to use policy, I need to figure what to write to that as not familiar | 15:04 |
| *** dims has quit IRC | 15:04 | |
| *** pm90_ has quit IRC | 15:04 | |
| dansmith | tojuvone: okay if you look through the other api code for policy-based authorizations that should help | 15:05 |
| *** rdopiera has quit IRC | 15:05 | |
| tojuvone | dansmith: Anyhow thanks for this, was helpful, might save a round of patch set ;) | 15:06 |
| *** n0ano has joined #openstack-nova | 15:06 | |
| *** lpetrut has joined #openstack-nova | 15:06 | |
| dansmith | tojuvone: there is a call authorize() that should be easy to grep for | 15:06 |
| dansmith | tojuvone: cool, I look forward to the revision :) | 15:06 |
| dansmith | tojuvone: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/keypairs.py#L275-L276 | 15:07 |
| *** pm90__ has quit IRC | 15:07 | |
| dansmith | tojuvone: that is a conditional exposure of a thing based on policy | 15:07 |
| openstackgerrit | Matt Riedemann proposed openstack/nova: Add checkpoint logging when building an instance in compute manager https://review.openstack.org/228980 | 15:08 |
| tojuvone | dansmith: Thanks :) | 15:08 |
| *** rushil has joined #openstack-nova | 15:09 | |
| *** julien_ZTE has quit IRC | 15:09 | |
| *** amotoki has joined #openstack-nova | 15:10 | |
| *** hemnafk is now known as hemna | 15:10 | |
| ctrath | mriedem: Regarding https://review.openstack.org/228980 , is self.driver.spawn(…) synchronous? | 15:10 |
| *** zzzeek has joined #openstack-nova | 15:10 | |
| *** changbl has quit IRC | 15:11 | |
| *** julien_ZTE has joined #openstack-nova | 15:11 | |
| *** zerda-bis has quit IRC | 15:12 | |
| *** markus_z has quit IRC | 15:13 | |
| dansmith | ctrath: kinda has to be or the resource tracking and error recovery bits in that function wouldn't work, right? | 15:13 |
| *** dims__ has joined #openstack-nova | 15:16 | |
| *** haomai___ has joined #openstack-nova | 15:16 | |
| *** sudipto has joined #openstack-nova | 15:16 | |
| *** browne has joined #openstack-nova | 15:17 | |
| *** pm90___ has quit IRC | 15:17 | |
| ctrath | dansmith: derp.. yup | 15:18 |
| *** haomaiwang has quit IRC | 15:18 | |
| ccarmack | mriedem: I'm getting SSH connection timeouts in gate-dvsm-tempest-full, test_snapshot_pattern. I see a bug about that: https://bugs.launchpad.net/grenade/+bug/1355573. | 15:19 |
| openstack | Launchpad bug 1355573 in grenade "grenade volume scenarios periodically fail /w SSHTimeout" [Undecided,New] | 15:19 |
| ccarmack | mriedem: is this something I should look at? | 15:20 |
| *** EinstCrazy has quit IRC | 15:21 | |
| mriedem | ccarmack: intermittent? | 15:25 |
| *** doffm_ is now known as doffm | 15:25 | |
| *** pm90_ has joined #openstack-nova | 15:25 | |
| ccarmack | mriedem: It failed on a commit yesterday and then again on a recheck... so 2 in a row | 15:26 |
| ccarmack | FYI - https://review.openstack.org/#/c/225199/ | 15:26 |
| *** pm90__ has joined #openstack-nova | 15:27 | |
| *** stevemar has quit IRC | 15:27 | |
| jaypipes | dims__: is there an oslo-specs repository somewhere, or do oslo specs go to openstack-specs? | 15:27 |
| *** stevemar has joined #openstack-nova | 15:27 | |
| dims__ | yep, we have one openstack/oslo-specs | 15:28 |
| jaypipes | ah, k | 15:28 |
| mriedem | ccarmack: well if it's always failing, and you're not providing security groups, there is probably an issue there | 15:29 |
| mriedem | like it can't ssh into port 22 | 15:29 |
| mriedem | b/c you didn't setup security group rules | 15:29 |
| *** stevemar_ has joined #openstack-nova | 15:29 | |
| *** pm90_ has quit IRC | 15:31 | |
| *** stevemar has quit IRC | 15:32 | |
| *** VW has quit IRC | 15:33 | |
| *** VW has joined #openstack-nova | 15:33 | |
| ccarmack | mriedem: hmm, the test could ssh into the server earlier in the write_timestamp step | 15:33 |
| *** stevemar_ has quit IRC | 15:34 | |
| mriedem | ccarmack: oh, well that's weird then | 15:34 |
| ccarmack | mriedem: maybe its something I did to the test, I temporarily back out my change | 15:35 |
| ccarmack | I'll .. | 15:35 |
| *** haomai___ has quit IRC | 15:35 | |
| *** rushil has quit IRC | 15:37 | |
| *** Daisy has joined #openstack-nova | 15:39 | |
| openstackgerrit | Ryan McNair proposed openstack/nova: Add retry logic for removing device with LibVirt https://review.openstack.org/227851 | 15:42 |
| *** salv-orl_ has joined #openstack-nova | 15:43 | |
| *** READ10 has joined #openstack-nova | 15:43 | |
| *** Daisy has quit IRC | 15:43 | |
| *** salv-orlando has quit IRC | 15:46 | |
| *** mwagner_ has quit IRC | 15:48 | |
| *** julien_ZTE has quit IRC | 15:48 | |
| *** eharney has joined #openstack-nova | 15:48 | |
| *** julien_ZTE has joined #openstack-nova | 15:49 | |
| mriedem | sahid: are you still -1 on this? https://review.openstack.org/#/c/226925/ | 15:49 |
| *** jdurgin1 has quit IRC | 15:49 | |
| *** wanghua has joined #openstack-nova | 15:50 | |
| *** vilobhmm11 has joined #openstack-nova | 15:52 | |
| *** otter768 has joined #openstack-nova | 15:53 | |
| *** marzif has joined #openstack-nova | 15:54 | |
| mriedem | dansmith: johnthetubaguy: err, aren't we missing db migration placeholders in stable/liberty? https://github.com/openstack/nova/tree/stable/liberty/nova/db/sqlalchemy/migrate_repo/versions | 15:54 |
| *** vilobhmm111 has joined #openstack-nova | 15:54 | |
| dansmith | mriedem: we add them for liberty in master | 15:54 |
| dansmith | but yeah, we need to do that | 15:54 |
| mriedem | liberty in master? liberty is no longer master | 15:55 |
| dansmith | they're on the release checklist so I assume johnthetubaguy was waiting for the final RC | 15:55 |
| dansmith | mriedem: right | 15:55 |
| mriedem | ok, i thought they went into rc1 | 15:55 |
| dansmith | mriedem: if we add them to liberty itself then people on liberty will burn them immediately | 15:55 |
| dansmith | we need them to reserve spots in master in case we have to backport something to liberry | 15:55 |
| dansmith | er, liberty | 15:55 |
| mriedem | no, i mean when we opened liberty we did the placeholders in that change https://review.openstack.org/#/c/172185/ | 15:56 |
| mriedem | for mitaka we just bumped the version | 15:56 |
| mriedem | so we need a thing like https://review.openstack.org/#/c/172185/ in stable/liberty now | 15:56 |
| mriedem | that goes into rc2 | 15:56 |
| mriedem | anyway, i think we agree, i can push that change | 15:56 |
| *** vilobhmm11 has quit IRC | 15:56 | |
| johnthetubaguy | but we need them in mitaka not liberty, I thought? | 15:57 |
| dansmith | johnthetubaguy: right | 15:57 |
| dansmith | mriedem: we do not want them *in* rc2 | 15:57 |
| johnthetubaguy | I remember getting totally mixed up about this last time | 15:57 |
| dansmith | the patch above was adding them to master(liberty) right after we cut kilo | 15:57 |
| johnthetubaguy | as dansmith mentioned, its all in here, to try and reduce the chance of screwing up: https://wiki.openstack.org/wiki/Nova/ReleaseChecklist | 15:57 |
| johnthetubaguy | totally should get that into master ASAP though | 15:58 |
| sahid | mriedem: yes, i'm not with you on this, we should do not break authorship of a spec | 15:58 |
| *** otter768 has quit IRC | 15:58 | |
| mriedem | sahid: the author of the spec is still the author | 15:58 |
| mriedem | they aren't implementing the code | 15:58 |
| mriedem | sahid: like a backlog spec basically | 15:58 |
| *** fawadkhaliq has joined #openstack-nova | 15:59 | |
| *** claudiub has quit IRC | 15:59 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova-specs: spec document for sriov-physical-function-passthrough blueprint https://review.openstack.org/212472 | 16:00 |
| mriedem | sahid: the alternative is i can abandon the spec, the author won't implement it, and it will just never happen | 16:00 |
| *** su_zhang has quit IRC | 16:00 | |
| *** electrocucaracha has quit IRC | 16:00 | |
| sahid | mriedem: what is the problem to put his name on other contributors? | 16:00 |
| johnthetubaguy | mriedem: what about adding co-authored note on the spec commit? not sure we have ever done that before | 16:00 |
| mriedem | sahid: b/c it's in the implementation section | 16:00 |
| johnthetubaguy | the contributors are the folks intending to submit code, usually | 16:00 |
| mriedem | and the author isn't implementing it | 16:01 |
| mriedem | johnthetubaguy: right | 16:01 |
| *** electrocucaracha has joined #openstack-nova | 16:01 | |
| mriedem | i can add co-authored-by in the commit message if that helps move this along | 16:01 |
| openstackgerrit | Terry Wilson proposed openstack/nova: network: Add external id on OVS port/interface for bridge lookup https://review.openstack.org/203835 | 16:01 |
| sahid | i remember to have implemented some specs from danpb, i'm not sure he will be happy if i removed his name for the specs | 16:01 |
| openstackgerrit | Terry Wilson proposed openstack/nova: network: Add external id on OVS port/interface for bridge lookup https://review.openstack.org/203835 | 16:02 |
| mriedem | christ i'm not sure why we're even nit picking over this | 16:02 |
| mriedem | i'll add him then | 16:02 |
| sahid | we always have worked like that, i do not understand why you do not want to keep original author, it's a recognition and help new contributors to join community even if they are not developers | 16:03 |
| danpb | mriedem: i tend to think that the author of the spec should stay listed a a co-author in general, even if they don't write code, because its useful to know who came up with the design when reviewing it | 16:03 |
| mriedem | this is a re-proposal of an approved spec | 16:04 |
| mriedem | the original authors name is all over the blueprint in LP | 16:04 |
| *** marzif has quit IRC | 16:05 | |
| mriedem | maybe we should just drop 'other contributors' from the spec in general b/c it doesn't actually matter | 16:05 |
| mriedem | chewbacca could show up and push a change in my bp series and i wouldn't care | 16:05 |
| mriedem | -2 - you're not listed in the spec chewy, sorry | 16:05 |
| * dansmith snickers | 16:05 | |
| *** mudassirlatif has joined #openstack-nova | 16:06 | |
| danpb | mriedem: the whole primary vs other contributors split is kind of crude | 16:06 |
| meh_di | bauzas: I want to undrestand how the nova code is run on moreover, I have already studied codes but debuging the code gives you more opportunity to undrestand how methods are called. | 16:06 |
| *** vishwanathj has joined #openstack-nova | 16:06 | |
| danpb | as many of the features are the work of several people over time | 16:06 |
| *** pixelbeat has quit IRC | 16:06 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova-specs: Re-propose spec for refresh-quotas-usage https://review.openstack.org/226925 | 16:08 |
| *** raildo is now known as raildo-afk | 16:08 | |
| *** kiran-r has joined #openstack-nova | 16:08 | |
| ndipanov | oh btw if anyone's interested in an easy re-post of a spec: https://review.openstack.org/#/c/228528/ | 16:08 |
| ndipanov | danpb, ^ | 16:08 |
| ndipanov | johnthetubaguy, ^ | 16:08 |
| *** pixelbeat has joined #openstack-nova | 16:10 | |
| *** EinstCrazy has joined #openstack-nova | 16:10 | |
| *** alexpilotti has joined #openstack-nova | 16:11 | |
| *** jlanoux has quit IRC | 16:12 | |
| *** twinklechawla has joined #openstack-nova | 16:12 | |
| *** Sree has quit IRC | 16:12 | |
| *** annegentle has quit IRC | 16:12 | |
| *** pm90__ has quit IRC | 16:15 | |
| *** meh_di has left #openstack-nova | 16:15 | |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: SameHostFilter should fail if host does not have instances https://review.openstack.org/229030 | 16:18 |
| *** vilobhmm111 has left #openstack-nova | 16:18 | |
| *** matrohon has quit IRC | 16:19 | |
| *** EinstCrazy has quit IRC | 16:21 | |
| *** raildo-afk is now known as raildo | 16:22 | |
| *** Sree has joined #openstack-nova | 16:23 | |
| *** digimonde has joined #openstack-nova | 16:23 | |
| *** kairo has joined #openstack-nova | 16:23 | |
| *** alex_klimov has quit IRC | 16:24 | |
| *** unicell has quit IRC | 16:24 | |
| *** jistr has quit IRC | 16:26 | |
| *** ijw has joined #openstack-nova | 16:26 | |
| *** coreywright has quit IRC | 16:27 | |
| *** Sree has quit IRC | 16:29 | |
| dims__ | sdague: mriedem: seeing a bunch of large ops failures | 16:29 |
| *** yassine__ has quit IRC | 16:30 | |
| *** Sree has joined #openstack-nova | 16:30 | |
| openstackgerrit | Sudipta Biswas proposed openstack/nova-specs: Report host memory bandwidth as a metric in Nova https://review.openstack.org/229034 | 16:30 |
| mriedem | dims__: known issue | 16:31 |
| mriedem | dims__: https://review.openstack.org/#/c/228636/ | 16:31 |
| dims__ | mriedem: k thanks. | 16:31 |
| *** Sree_ has joined #openstack-nova | 16:32 | |
| *** Sree_ is now known as Guest78262 | 16:32 | |
| *** angdraug has quit IRC | 16:33 | |
| *** twinklechawla has quit IRC | 16:33 | |
| *** angdraug has joined #openstack-nova | 16:34 | |
| *** Guest78262 has quit IRC | 16:34 | |
| *** Sree has quit IRC | 16:34 | |
| *** matrohon has joined #openstack-nova | 16:37 | |
| *** romainh has left #openstack-nova | 16:37 | |
| *** apoorvad has joined #openstack-nova | 16:38 | |
| *** dims__ has quit IRC | 16:38 | |
| *** dims__ has joined #openstack-nova | 16:39 | |
| *** karimb has quit IRC | 16:40 | |
| openstackgerrit | Pawel Koniszewski proposed openstack/nova-specs: Force live migration https://review.openstack.org/229040 | 16:40 |
| mriedem | johnthetubaguy: there is a question in this re-proposed spec that probably didn't come up in the liberty review https://review.openstack.org/#/c/226925/1/specs/mitaka/approved/refresh-quotas-usage.rst | 16:41 |
| mriedem | johnthetubaguy: not sure what to do about that | 16:41 |
| *** stevemar has joined #openstack-nova | 16:42 | |
| johnthetubaguy | mriedem: I think its OK to ask that question now, we miss things sometimes | 16:42 |
| johnthetubaguy | mriedem: I wish I had questioned the hierarchical quota one when I could, rather than fast approving it, if that helps | 16:42 |
| *** kiranr has joined #openstack-nova | 16:43 | |
| *** READ10 has quit IRC | 16:43 | |
| *** nic has joined #openstack-nova | 16:44 | |
| *** alexpilotti has quit IRC | 16:44 | |
| mriedem | yeah, just wish i had a good answer for the question | 16:44 |
| *** pkoniszewski has joined #openstack-nova | 16:46 | |
| *** rotbeard has quit IRC | 16:47 | |
| johnthetubaguy | mriedem: ah, yeah, thats the harder bit | 16:47 |
| raildo | johnthetubaguy, mriedem I'm trying figure out how hierarchical quotas should works for users... since we didn't had quota for users on cinder, so we din't thought a lot on this case... | 16:50 |
| *** digimonde has quit IRC | 16:51 | |
| raildo | can we alllow more quotas for users that the hard limit, right? | 16:52 |
| raildo | allow* | 16:52 |
| mriedem | i don't understand the question | 16:52 |
| *** kiran-r has quit IRC | 16:52 | |
| raildo | mriedem, I mean... in quotas, we set the limit for the resource, for example the user can create 10 instances. with nested quotas, if I want allocate 5 instances for a sub-project, I'll "remove" 5 instances to the parent project, and allocated for your subproject. But for users this doesn't apply, If I have 10 instances in a project, I can apply quota for a user to 20 instances on this project, right? | 16:55 |
| *** Sree has joined #openstack-nova | 16:55 | |
| *** mwagner_ has joined #openstack-nova | 16:56 | |
| *** derekh has quit IRC | 16:56 | |
| mriedem | raildo: do you mean today w/o nested quotas? | 16:57 |
| raildo | mriedem, today, without nested quotas. | 16:57 |
| mriedem | i thought limit applied the same to users and projects | 16:57 |
| mriedem | unless you change it via the api | 16:57 |
| *** ctrath has quit IRC | 16:59 | |
| *** unicell has joined #openstack-nova | 17:00 | |
| *** READ10 has joined #openstack-nova | 17:00 | |
| *** Piet has joined #openstack-nova | 17:01 | |
| *** su_zhang has joined #openstack-nova | 17:01 | |
| *** danpb has quit IRC | 17:01 | |
| *** su_zhang has quit IRC | 17:03 | |
| *** su_zhang has joined #openstack-nova | 17:03 | |
| *** tonytan4ever has joined #openstack-nova | 17:05 | |
| raildo | mriedem, ok, so using the API I can say that a project have quota for 10 instances and a user in this project have quota for 15 instances. | 17:06 |
| *** AJaeger has joined #openstack-nova | 17:06 | |
| *** smatzek has quit IRC | 17:06 | |
| raildo | using this API call http://developer.openstack.org/api-ref-compute-v2.1.html#updatesquotauser | 17:06 |
| *** Sree has quit IRC | 17:06 | |
| AJaeger | Could I get another +2 and approval for a translation import from our bot for master, please? https://review.openstack.org/#/c/227138/ | 17:07 |
| mriedem | raildo: yeah | 17:07 |
| openstackgerrit | Dan Smith proposed openstack/nova: Add get_minimum_version() to Service object and DB API https://review.openstack.org/223320 | 17:07 |
| openstackgerrit | Dan Smith proposed openstack/nova: Allow automatic determination of compute_rpc version by service https://review.openstack.org/223321 | 17:07 |
| openstackgerrit | Dan Smith proposed openstack/nova: Make service create/update fail if version is too old https://review.openstack.org/229049 | 17:07 |
| mriedem | raildo: like this http://docs.openstack.org/cli-reference/content/novaclient_commands.html#novaclient_subcommand_quota-update | 17:07 |
| *** apoorvad_ has joined #openstack-nova | 17:08 | |
| *** ildikov has quit IRC | 17:09 | |
| *** thorst has quit IRC | 17:10 | |
| raildo | mriedem, great! so, thinking now in nested projects, we have to change something on this behavior? since a project_admin can set a quota for a subproject, maybe he can set quotas for a user in a subproject? | 17:10 |
| mriedem | seems reasonable | 17:11 |
| *** apoorvad has quit IRC | 17:12 | |
| *** pm90_ has joined #openstack-nova | 17:13 | |
| *** lsmola has quit IRC | 17:14 | |
| *** wanghua has quit IRC | 17:14 | |
| raildo | mriedem, Is there any other issue or suggestion that you see? (sorry if I'm boring you, I'll repropose the nested quota spec and I'm trying do figure out what I can improve) | 17:14 |
| *** PaulMurray has quit IRC | 17:15 | |
| mriedem | raildo: honestly i'm not really familiar with the nested quotas stuf | 17:15 |
| mriedem | *stuff | 17:15 |
| *** sudipto has quit IRC | 17:16 | |
| raildo | mriedem, ok, no problem, thanks for your time! | 17:16 |
| *** ctrath has joined #openstack-nova | 17:17 | |
| *** tjones has joined #openstack-nova | 17:17 | |
| *** tjones has left #openstack-nova | 17:17 | |
| *** ccarmack has quit IRC | 17:17 | |
| *** salv-orl_ has quit IRC | 17:18 | |
| *** ijw has quit IRC | 17:18 | |
| *** marzif has joined #openstack-nova | 17:19 | |
| *** cfriesen has joined #openstack-nova | 17:20 | |
| *** ccarmack has joined #openstack-nova | 17:21 | |
| openstackgerrit | Mark Doffman proposed openstack/nova: Point to cinder options in nova block alloc docs. https://review.openstack.org/229056 | 17:22 |
| *** Sree has joined #openstack-nova | 17:24 | |
| *** emagana has quit IRC | 17:25 | |
| *** armax has quit IRC | 17:25 | |
| mriedem | doffm: i'd like more of your commit message titles to sound like dr seuss books | 17:25 |
| *** asselin has quit IRC | 17:27 | |
| *** ijw has joined #openstack-nova | 17:27 | |
| *** sseverson has quit IRC | 17:27 | |
| *** sahid has quit IRC | 17:27 | |
| *** asselin has joined #openstack-nova | 17:28 | |
| *** sseverson has joined #openstack-nova | 17:28 | |
| *** achanda has joined #openstack-nova | 17:29 | |
| *** tdurakov has joined #openstack-nova | 17:29 | |
| *** fawadkhaliq has quit IRC | 17:30 | |
| doffm | mriedem: I love Dr seuss books! Did that come through? | 17:31 |
| *** Piet has quit IRC | 17:31 | |
| *** electrocucaracha has quit IRC | 17:32 | |
| *** sudipto has joined #openstack-nova | 17:33 | |
| *** coolsvap has quit IRC | 17:33 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: WIP: Virtual device role tagging https://review.openstack.org/226409 | 17:35 |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Make the server group invalid format message more verbose https://review.openstack.org/218468 | 17:35 |
| openstackgerrit | Dan Smith proposed openstack/nova-specs: Add service-version-behavior spec https://review.openstack.org/229062 | 17:35 |
| dansmith | alaski: ^ | 17:35 |
| dansmith | alaski: worst-named spec ever | 17:36 |
| *** lpetrut has quit IRC | 17:36 | |
| doffm | ndipanov: I've been looking at bug https://bugs.launchpad.net/nova/+bug/1497343. In looking at it i'm tempted to do some more cleanup around this, largely moving the detach functionality to block_device.py. | 17:36 |
| openstack | Launchpad bug 1497343 in OpenStack Compute (nova) "Need to consolidate duplicated volume detach code between compute manager and block_device" [Low,Triaged] - Assigned to Mark Doffman (mjdoffma) | 17:36 |
| *** thorst has joined #openstack-nova | 17:37 | |
| alaski | dansmith: what about "service version manifesto"? | 17:37 |
| dansmith | alaski: so. played. | 17:37 |
| doffm | ndipanov: Worth doing? or was there a reason that was left in compute/mananger.py | 17:37 |
| *** coolsvap has joined #openstack-nova | 17:37 | |
| raildo | mriedem, btw, do you know how the remains quota calculation works? looking on the tests, sounds odd to me: https://github.com/openstack/nova/blob/master/nova/tests/unit/test_quota.py#L1209-L1214 | 17:37 |
| *** pixelbeat has quit IRC | 17:37 | |
| *** mriedem has quit IRC | 17:38 | |
| *** salv-orlando has joined #openstack-nova | 17:38 | |
| *** exploreshaifali has joined #openstack-nova | 17:40 | |
| *** exploreshaifali has quit IRC | 17:41 | |
| *** krtaylor has quit IRC | 17:42 | |
| *** rlrossit has quit IRC | 17:42 | |
| *** mriedem has joined #openstack-nova | 17:42 | |
| *** emagana has joined #openstack-nova | 17:43 | |
| *** ijw has quit IRC | 17:44 | |
| *** sileht_ has joined #openstack-nova | 17:44 | |
| *** moshele has quit IRC | 17:44 | |
| *** sileht_ is now known as sileht | 17:45 | |
| *** sileht has quit IRC | 17:45 | |
| *** sileht has joined #openstack-nova | 17:45 | |
| *** baoli_ has quit IRC | 17:46 | |
| *** rlrossit has joined #openstack-nova | 17:47 | |
| *** jichen has joined #openstack-nova | 17:48 | |
| *** sudipto has quit IRC | 17:48 | |
| *** Sree has quit IRC | 17:49 | |
| *** penick has joined #openstack-nova | 17:51 | |
| AJaeger | thanks, melwitt for the approval of the import! | 17:54 |
| *** krtaylor has joined #openstack-nova | 17:54 | |
| *** otter768 has joined #openstack-nova | 17:55 | |
| melwitt | AJaeger: np | 17:55 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Use os-testr for py34 tox target https://review.openstack.org/229069 | 17:55 |
| *** Marga_ has quit IRC | 17:56 | |
| *** tdurakov has quit IRC | 17:57 | |
| dims__ | mtreinish: ^^ yay! | 17:57 |
| *** electrocucaracha has joined #openstack-nova | 17:58 | |
| *** annegentle has joined #openstack-nova | 17:58 | |
| *** penick has quit IRC | 17:58 | |
| *** sdague has quit IRC | 17:58 | |
| *** otter768 has quit IRC | 17:59 | |
| *** dims has joined #openstack-nova | 18:00 | |
| *** ctrath has quit IRC | 18:01 | |
| *** dims__ has quit IRC | 18:02 | |
| *** sdague has joined #openstack-nova | 18:02 | |
| *** apoorvad_ has quit IRC | 18:04 | |
| *** penick has joined #openstack-nova | 18:04 | |
| *** apoorvad has joined #openstack-nova | 18:04 | |
| *** ctrath has joined #openstack-nova | 18:04 | |
| *** ctrath has left #openstack-nova | 18:06 | |
| *** penick has quit IRC | 18:06 | |
| *** AJaeger has left #openstack-nova | 18:06 | |
| *** READ10 has quit IRC | 18:09 | |
| *** marzif has quit IRC | 18:09 | |
| *** claudiub has joined #openstack-nova | 18:13 | |
| *** penick has joined #openstack-nova | 18:13 | |
| *** ijw has joined #openstack-nova | 18:13 | |
| *** ijw has quit IRC | 18:14 | |
| *** vilobhmm11 has joined #openstack-nova | 18:16 | |
| *** emagana has quit IRC | 18:17 | |
| *** Marga_ has joined #openstack-nova | 18:19 | |
| *** ildikov has joined #openstack-nova | 18:19 | |
| *** ijw has joined #openstack-nova | 18:23 | |
| *** moshele has joined #openstack-nova | 18:24 | |
| *** tdurakov has joined #openstack-nova | 18:25 | |
| *** boris-42 has quit IRC | 18:29 | |
| *** armax has joined #openstack-nova | 18:29 | |
| mtreinish | dims__: so the reqs file is wrong on that patch, but other than that LGTM | 18:34 |
| *** moshele has quit IRC | 18:34 | |
| *** su_zhang has quit IRC | 18:36 | |
| *** Vek_ has joined #openstack-nova | 18:37 | |
| *** Vek has quit IRC | 18:37 | |
| *** setuid has quit IRC | 18:37 | |
| *** aix has quit IRC | 18:38 | |
| *** flwang has quit IRC | 18:38 | |
| *** flwang has joined #openstack-nova | 18:38 | |
| *** setuid has joined #openstack-nova | 18:42 | |
| *** nikhil has joined #openstack-nova | 18:42 | |
| *** nikhil_k has quit IRC | 18:44 | |
| *** Piet has joined #openstack-nova | 18:46 | |
| *** electrocucaracha has quit IRC | 18:46 | |
| *** jasondotstar is now known as jasondotstar_afk | 18:47 | |
| *** pixelbeat has joined #openstack-nova | 18:48 | |
| *** whenry has quit IRC | 18:50 | |
| *** electrocucaracha has joined #openstack-nova | 18:50 | |
| *** electrocucaracha has quit IRC | 18:51 | |
| *** kiranr has quit IRC | 18:52 | |
| *** sahid has joined #openstack-nova | 18:53 | |
| *** e0ne has quit IRC | 18:54 | |
| *** eharney has quit IRC | 18:55 | |
| *** ijuwang has quit IRC | 18:56 | |
| *** eharney has joined #openstack-nova | 19:01 | |
| *** stevemar has quit IRC | 19:01 | |
| *** ijuwang has joined #openstack-nova | 19:02 | |
| *** artom has quit IRC | 19:03 | |
| *** whenry has joined #openstack-nova | 19:05 | |
| *** stevemar has joined #openstack-nova | 19:08 | |
| *** davideagnello has quit IRC | 19:09 | |
| alaski | dansmith: commented on https://review.openstack.org/#/c/229062, calling it out here since I didn't vote yet | 19:10 |
| *** ctrath has joined #openstack-nova | 19:10 | |
| dansmith | cool thanks | 19:11 |
| *** angdraug has quit IRC | 19:14 | |
| *** pixelbeat has quit IRC | 19:16 | |
| *** Daisy has joined #openstack-nova | 19:18 | |
| *** davideagnello has joined #openstack-nova | 19:20 | |
| *** baoli has joined #openstack-nova | 19:21 | |
| *** tsekiyama has joined #openstack-nova | 19:22 | |
| *** Daisy has quit IRC | 19:23 | |
| mriedem | ccarmack: http://lists.openstack.org/pipermail/openstack-dev/2015-August/072797.html | 19:24 |
| mriedem | if you want to post devil's advocate questions | 19:24 |
| *** ijw has quit IRC | 19:24 | |
| *** apoorvad has quit IRC | 19:25 | |
| *** digimonde has joined #openstack-nova | 19:28 | |
| *** baoli has quit IRC | 19:31 | |
| *** diegows has joined #openstack-nova | 19:31 | |
| *** baoli has joined #openstack-nova | 19:32 | |
| *** eharney has quit IRC | 19:32 | |
| *** sileht has quit IRC | 19:39 | |
| *** angdraug has joined #openstack-nova | 19:39 | |
| *** krtaylor has quit IRC | 19:40 | |
| *** krtaylor has joined #openstack-nova | 19:43 | |
| *** eharney has joined #openstack-nova | 19:45 | |
| *** sileht has joined #openstack-nova | 19:45 | |
| *** baoli has quit IRC | 19:45 | |
| mriedem | anyone have thoughts on this? https://review.openstack.org/#/c/218975/2/nova/compute/manager.py | 19:45 |
| *** clduser_ has joined #openstack-nova | 19:46 | |
| *** baoli has joined #openstack-nova | 19:46 | |
| *** clduser has quit IRC | 19:46 | |
| *** sileht has quit IRC | 19:47 | |
| *** sileht has joined #openstack-nova | 19:48 | |
| *** tonytan4ever has quit IRC | 19:50 | |
| *** tonytan4ever has joined #openstack-nova | 19:51 | |
| *** electrocucaracha has joined #openstack-nova | 19:51 | |
| tdurakov | alaski, hi, this could be interesting to you: https://review.openstack.org/#/c/225910/ | 19:53 |
| *** otter768 has joined #openstack-nova | 19:55 | |
| tdurakov | melwitt, hi, are you around? could we discuss this: https://review.openstack.org/#/c/227897/ | 19:56 |
| *** tjones has joined #openstack-nova | 19:56 | |
| alaski | tdurakov: thanks. opened to look at later | 19:56 |
| *** baoli has quit IRC | 19:57 | |
| tdurakov | alaski, thank you | 19:57 |
| *** erlon has quit IRC | 19:58 | |
| mriedem | ctrath: apparently reset-state only allows the admin to put the instance in active or error state | 19:58 |
| mriedem | https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/schemas/reset_server_state.py#L23 | 19:58 |
| *** ZZelle_ has joined #openstack-nova | 19:59 | |
| mtreinish | dims__: yes, the requirements job would fail on that patch, but it doesn't mean the requirements you listed were correct | 19:59 |
| alaski | mriedem: I think that situation in that review is messy no matter what. But I agree with your point about bringing state from the hypervisor into Nova. I'll comment on the review | 19:59 |
| mtreinish | we need to bump g-r to reflect that nova will have a hard dep on ostestr >=0.4.0 | 19:59 |
| openstackgerrit | Dan Smith proposed openstack/nova-specs: Add service-version-behavior spec https://review.openstack.org/229062 | 19:59 |
| dims | mtreinish: :) | 19:59 |
| *** su_zhang has joined #openstack-nova | 20:00 | |
| dims | mtreinish: right | 20:00 |
| *** tdurakov has quit IRC | 20:00 | |
| *** otter768 has quit IRC | 20:00 | |
| mtreinish | dims: I just like making your life difficult :) | 20:00 |
| dims | hahaha | 20:00 |
| mtreinish | although you're g-r core now, I'll handle that patch so you can +2 it | 20:00 |
| *** penick has quit IRC | 20:01 | |
| *** earlephilhower has quit IRC | 20:01 | |
| dims | mtreinish: probably need a few days, i think we just cut a stable/liberty this morning for requirements | 20:01 |
| mriedem | alaski: so ctrath had an idea about putting the instance into error state and then the user can do something to it, | 20:01 |
| mriedem | like either reset-state or reboot it | 20:02 |
| mriedem | hard-reboot in this case https://github.com/openstack/nova/blob/master/nova/compute/vm_states.py#L51 | 20:02 |
| *** tdurakov has joined #openstack-nova | 20:02 | |
| mriedem | we could also do reset-state and put it in active state automatically if it's running on the hypervisor | 20:02 |
| mtreinish | dims: heh, that means the gates are open | 20:02 |
| mriedem | that'd basically be the same flow as the task putting it in error state, then the admin doing reset-state active on it | 20:02 |
| mtreinish | dims: well after we cut the branches for devstack and grenade tomorrow I guess | 20:02 |
| dims | right | 20:03 |
| melwitt | tdurakov: yes, I was looking over it earlier | 20:03 |
| ctrath | mriedem, alaski: the main issue that I see with putting in error state is how to tell the user the reasoning why it's in error state | 20:03 |
| mtreinish | dims: https://review.openstack.org/#/c/229129/ | 20:03 |
| tdurakov | melwitt, so, i've left comment for you, what do you think? | 20:04 |
| mriedem | ctrath: i think we'd store an instance fault with the error state if we did that | 20:04 |
| dims | mtreinish: haven't gotten my wings yet. so +1 for now :) | 20:04 |
| *** yamahata has quit IRC | 20:04 | |
| mtreinish | dims: well let's poke dhellmann to make that happen | 20:05 |
| *** penick has joined #openstack-nova | 20:05 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Use os-testr for py34 tox target https://review.openstack.org/229069 | 20:06 |
| dims | threw in a Depends-On ^^ | 20:06 |
| mtreinish | cool | 20:07 |
| mtreinish | +1 | 20:07 |
| melwitt | tdurakov: my initial thought is that we can't assume dhcp because the network might not be using it | 20:08 |
| melwitt | tdurakov: but overall it's not clear to me from the bug details what state the instance is left in when it fails network teardown during a rollback. is it in error state, etc | 20:09 |
| dims | mtreinish: love it when a plan comes together :) | 20:09 |
| *** diegows has quit IRC | 20:09 | |
| tdurakov | melwitt, nope, it could be in active state | 20:10 |
| *** twinklechawla has joined #openstack-nova | 20:10 | |
| *** Marga_ has quit IRC | 20:12 | |
| *** Marga_ has joined #openstack-nova | 20:13 | |
| *** e0ne has joined #openstack-nova | 20:14 | |
| tdurakov | melwitt, for example if nova-network accidentally stops, rollback method would fail before driver cleanup resources, while with this patch driver try to clean them up anyway | 20:15 |
| *** mrda has quit IRC | 20:18 | |
| raildo | melwitt, ping I think that you can help me with a doubt :) Do you know how the remains quotas is calculated in this tests here? https://github.com/openstack/nova/blob/master/nova/tests/unit/test_quota.py#L1209-L1214, the values sounds strange to me | 20:19 |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Add timing logging around each filter being ran https://review.openstack.org/224355 | 20:20 |
| melwitt | tdurakov: yeah. so with this patch it would leave the network allocated on the destination instead of both the network and the instance dir left allocated. I'm was thinking if we can do something better, because this would silently leave network allocated | 20:21 |
| melwitt | tdurakov: and it seems like in either case, there's no total recovery in that, there's no way to free that network allocated in the case of non-dhcp | 20:22 |
| *** diegows has joined #openstack-nova | 20:22 | |
| tdurakov | melwitt, non-dhcp for nova-network is? | 20:22 |
| melwitt | tdurakov: for example FlatManager, not using dhcp | 20:23 |
| melwitt | raildo: I don't know off the top of my head. I'd have to figure it out from what's there | 20:23 |
| raildo | melwitt, I know that this value is calculated here: https://github.com/openstack/nova/blob/master/nova/quota.py#L206 but I didn't understand this values on this tests =/ | 20:25 |
| *** stevemar has quit IRC | 20:25 | |
| *** apoorvad has joined #openstack-nova | 20:26 | |
| *** salv-orlando has quit IRC | 20:27 | |
| *** dims has quit IRC | 20:28 | |
| tdurakov | melwitt, yep, there is no way to clean up network properly, IMO we could improve this now by at least trying to clean up driver resources. there is a spec for live-migration, that will improve this situation(https://review.openstack.org/#/c/225910/) | 20:29 |
| *** stevemar has joined #openstack-nova | 20:29 | |
| *** dims has joined #openstack-nova | 20:29 | |
| *** apoorvad has quit IRC | 20:31 | |
| *** Marga_ has quit IRC | 20:32 | |
| *** Marga_ has joined #openstack-nova | 20:33 | |
| *** salv-orlando has joined #openstack-nova | 20:33 | |
| *** diegows has quit IRC | 20:34 | |
| *** sahid has quit IRC | 20:35 | |
| *** yamahata has joined #openstack-nova | 20:36 | |
| melwitt | raildo: so, it looks like "remains" starts out equal to "limit" and it's calculated each _process_quotas as "remains -= hard_limit" and in the test you can find the quota is stubbed out to set instances hard_limit as 5 and cores hard_limit as 2 | 20:36 |
| alaski | ctrath: mriedem sorry, got pulled into a call | 20:37 |
| alaski | I'm looking at the options and none of them seem good | 20:37 |
| melwitt | raildo: so for the first one instances it's 5 - 5 = 0 and the second one cores 10 - 2 = 8 and the rest didn't have hard_limit set so they are just the limit | 20:37 |
| mriedem | alaski: well there is always punt and do nothing, but that kind of sucks too | 20:38 |
| alaski | ctrath: mriedem I think the best course would be to understand how an instance gets to a state where Nova thinks it's stopped but it's not | 20:38 |
| mriedem | apparently we just had a customer have their instances stopped by that task | 20:38 |
| *** baoli has joined #openstack-nova | 20:38 | |
| mriedem | well garyk's case was HA something or other in the backend vcenter | 20:39 |
| melwitt | tdurakov: yeah, I understand. I'll think about it more and we'll get some more people to take a look | 20:39 |
| tdurakov | melwitt, thanks:) | 20:39 |
| alaski | mriedem: yeah. I wonder if we should distinguish between a user initiated stop and a sync_power_state stop | 20:39 |
| raildo | melwitt, I'm trying to find this hard_limit=2 for core.... this is hard coded in somewhere? | 20:40 |
| *** electrocucaracha has quit IRC | 20:40 | |
| alaski | mriedem: and let the former continue to use force_stop, then solve for the latter | 20:40 |
| *** electrocucaracha has joined #openstack-nova | 20:40 | |
| mriedem | i guess i'm not totally following | 20:41 |
| mriedem | like an action event or something? | 20:41 |
| *** thangp has quit IRC | 20:41 | |
| melwitt | raildo: it's here https://github.com/openstack/nova/blob/master/nova/tests/unit/test_quota.py#L1094-L1100 and you can see _stub_get_by_project is called in the beginning of the test | 20:41 |
| alaski | mriedem: like have another elif in sync_power_state for vm_state == vm_states.SYNC_STOPPED or something | 20:42 |
| *** karimb has joined #openstack-nova | 20:42 | |
| *** twinklechawla has quit IRC | 20:42 | |
| alaski | then a reboot seems more reasonable to me | 20:42 |
| *** ildikov has quit IRC | 20:42 | |
| raildo | melwitt, oh! now this makes sense! thanks melwitt, you're the best :D | 20:42 |
| mriedem | alaski: oh so you're thinking the 2nd pass | 20:43 |
| melwitt | raildo: haha, you're welcome | 20:43 |
| alaski | mriedem: yeah | 20:43 |
| *** twinklechawla has joined #openstack-nova | 20:43 | |
| mriedem | but what does that buy us? we just ignore SYNC_STOPPED? | 20:43 |
| mriedem | or reboot on SYNC_STOPPED? | 20:44 |
| alaski | then we shut down the instance if it's user initiated, and reboot on SYNC_STOPPED so we end up active again | 20:44 |
| mriedem | i'm not sure we know it's user initiated though | 20:44 |
| mriedem | if it were user initiated we could always just say don't do that and reject the bug | 20:44 |
| mriedem | unless you're grouping user initiated with 3rd party HA software initiated | 20:45 |
| mriedem | idk, i kind of like the error state idea because that puts the next action in the hands of the operator | 20:45 |
| alaski | the bug is because the vm_state ends up stopped because of the blip in vmware, right | 20:45 |
| mriedem | yeah, it's like the reboot case with the libvirt driver and the lifecycle event stuff, | 20:46 |
| mriedem | that was stopping the instance then starting it but it raced with the task and it'd be running in libvirt and the task would stop it | 20:46 |
| mriedem | the fix there was adding a 15 second sleep to the libvirt driver in the reboot case | 20:46 |
| alaski | my thinking is that we don't know the right thing to do in the periodic because we don't know if the instance should go down, or stay up | 20:47 |
| alaski | we just assume now that if the vm_state says it should be down we kill it | 20:47 |
| mriedem | alaski: yeah, and i think we can agree that's not good | 20:47 |
| *** su_zhang has quit IRC | 20:48 | |
| alaski | so I thought we could distinguish between a user calling the stop api, and a vm that's stopped because the hypervisor said it should be | 20:48 |
| *** ildikov has joined #openstack-nova | 20:48 | |
| alaski | in the first case we don't want the instance coming back up, in the second we do | 20:48 |
| alaski | (assuming now the power state says it's up) | 20:48 |
| mriedem | yeah, was going to say, in the nova-api user case the vm_state should be stopped and power state should be shutdown | 20:49 |
| mriedem | the only way power state goes to running is if the user restarted it on the backend | 20:49 |
| mriedem | which is the bug that we reject | 20:49 |
| alaski | right | 20:49 |
| mriedem | so.....i'm still not really seeing a difference with just making that assumption | 20:50 |
| *** su_zhang has joined #openstack-nova | 20:50 | |
| mriedem | i might just need the steps written down in the hypervisor blip case | 20:51 |
| mriedem | and how a new SYNC_STOPPED state makes it better | 20:51 |
| alaski | I'm just thinking of a way to make it clear what the right corrective action is | 20:52 |
| alaski | is vm_state == stopped, but hypervisor says it's running | 20:52 |
| alaski | which one is correct? | 20:52 |
| mriedem | are you thinking on pass 1 we stop it and set SYNC_STOPPED, and on pass 2 we see SYNC_STOPPED and hypervisor says it's running, so we reboot? | 20:52 |
| alaski | yes | 20:53 |
| *** pixelbeat has joined #openstack-nova | 20:53 | |
| mriedem | so at most by default your pet is only down for 10 minutes | 20:53 |
| mriedem | but it's not actually down | 20:53 |
| alaski | yeah, just the API view of it is wrong | 20:54 |
| mriedem | yeah...i still think operators are going to be pissed that nova stops it in the first place (the SYNC_STOPPED case) | 20:54 |
| alaski | well, that's the real fix here. but people seemed convinced that couldn't be fixed | 20:54 |
| mriedem | which is why i was thinking putting it in error state and letting the operator handle correcting it with reset-state is marginally better | 20:54 |
| mriedem | b/c if it's actually running and you just want it showing as ACTIVE (w/o rebooting it), you can do that with nova reset-state --active | 20:55 |
| *** VW_ has joined #openstack-nova | 20:55 | |
| mriedem | if it's 'running' but not really, you could hard reboot it | 20:55 |
| *** VW has quit IRC | 20:55 | |
| mriedem | putting it in error state from nova's end is basically saying, we don't know wtf happened, but something is wrong, you sort it out | 20:56 |
| mriedem | we could of course make that configurable (the action to take on the instance when we get into this case) | 20:56 |
| ctrath | mriedem: is there a way to gracefully sync the state? | 20:57 |
| mriedem | gracefully? | 20:57 |
| alaski | sure, that works too. I was simply considering that in one case (user called stop api) we know what happened and how to fix it | 20:57 |
| ctrath | meaning, can somebody run a command after the instance is in error state in order to update Nova with the correct state? | 20:57 |
| mriedem | ctrath: nova reset-state | 20:58 |
| ctrath | ok, cool | 20:58 |
| mriedem | but you can only put it in active or error state with that api | 20:58 |
| *** ijw has joined #openstack-nova | 20:58 | |
| *** VW_ has quit IRC | 20:58 | |
| mriedem | alaski: well if the user stops the instance and then it's running, they want the task to stop the instance again so they aren't billed right? | 20:59 |
| mriedem | or whatever | 20:59 |
| *** VW has joined #openstack-nova | 20:59 | |
| alaski | right. we don't want the resources still in use. | 20:59 |
| alaski | it's actually moot for us because we don't allow stop, but I'll gloss over that | 21:00 |
| mriedem | heh | 21:00 |
| alaski | my concern is that the new config would change that | 21:00 |
| mriedem | well, the config could default to stop as it does today for backwards compat | 21:01 |
| mriedem | i'm basically thinking actions are stop, reboot, error, noop | 21:01 |
| alaski | yeah, but changing it to fix the "we don't know why it stopped" case breaks the "user stopped it" case | 21:01 |
| *** earlephilhower has joined #openstack-nova | 21:01 | |
| mriedem | true | 21:01 |
| *** apoorvad has joined #openstack-nova | 21:02 | |
| mriedem | we know it stopped b/c larry stopped it explicitly and larry wants to keep it that way | 21:02 |
| mriedem | b/c gdi larry hates that thing running for whatever reason | 21:02 |
| alaski | a good pet listens to its owner | 21:02 |
| mriedem | shiiiiit | 21:02 |
| ctrath | gdi larry | 21:02 |
| mriedem | alaski: you've heard about mortimer right? | 21:02 |
| alaski | I've heard some of the tales | 21:03 |
| *** hemna is now known as hemnafk | 21:03 | |
| mriedem | tales of woe | 21:03 |
| *** ijw has quit IRC | 21:03 | |
| mriedem | i'm not crazy about a new vm_state, could we just set a flag on the instance? | 21:03 |
| mriedem | sysmeta? | 21:03 |
| alaski | sure. new vm_state isn't pleasant to me either | 21:03 |
| sdague | mriedem: https://review.openstack.org/#/c/227135/1 - backport of the name relaxation to kilo | 21:03 |
| mriedem | alaski: i feel like this is getting into spec territory | 21:04 |
| mriedem | else, let's just ram this through w/o consensus when everyone is in tokyo :) | 21:04 |
| *** rlrossit has left #openstack-nova | 21:04 | |
| mriedem | sdague: ok sec | 21:04 |
| alaski | heh, if we can get a third party to write it we've got two +2s | 21:05 |
| *** whenry has quit IRC | 21:05 | |
| alaski | mriedem: honestly short term a config option seems reasonable if this is causing problems. I was just thinking we could solve both cases eventually | 21:06 |
| mriedem | yeah, i think covering both is a spec | 21:06 |
| mriedem | hell, a config option might be a specless blueprint, idk | 21:06 |
| mriedem | maybe i'll update the WIP i have to see what the new config option would look with the stop/reboot/error/noop actions | 21:07 |
| *** twinklechawla has quit IRC | 21:07 | |
| *** mudassirlatif_ has joined #openstack-nova | 21:08 | |
| *** twinklechawla has joined #openstack-nova | 21:08 | |
| *** aysyd has quit IRC | 21:08 | |
| *** hemna has joined #openstack-nova | 21:09 | |
| mriedem | alaski: now can you just like, you know, summarize all of that discussion as a comment in the change so we dont' forget about it? | 21:09 |
| mriedem | :P | 21:09 |
| *** raildo is now known as raildo-afk | 21:10 | |
| *** ociuhandu has joined #openstack-nova | 21:10 | |
| *** mudassirlatif has quit IRC | 21:11 | |
| *** mudassirlatif_ is now known as mudassirlatif | 21:11 | |
| *** twinklechawla has quit IRC | 21:12 | |
| *** alexpilotti has joined #openstack-nova | 21:14 | |
| alaski | heh, I'll try | 21:16 |
| *** MVenesio has quit IRC | 21:17 | |
| mriedem | i can also write it down | 21:17 |
| mriedem | was just joking | 21:17 |
| mriedem | sort of | 21:17 |
| alaski | well now I've already started | 21:17 |
| alaski | so I'm not stopping | 21:17 |
| *** baoli has quit IRC | 21:22 | |
| *** baoli has joined #openstack-nova | 21:22 | |
| *** haypo has left #openstack-nova | 21:23 | |
| *** mudassirlatif has quit IRC | 21:27 | |
| *** thorst has quit IRC | 21:28 | |
| *** mudassirlatif has joined #openstack-nova | 21:28 | |
| *** baoli has quit IRC | 21:30 | |
| *** mudassirlatif has quit IRC | 21:30 | |
| *** baoli has joined #openstack-nova | 21:32 | |
| *** baoli has quit IRC | 21:32 | |
| *** baoli has joined #openstack-nova | 21:32 | |
| mriedem | alaski: we have a logic problem i think with "are you thinking on pass 1 we stop it and set SYNC_STOPPED, and on pass 2 we see SYNC_STOPPED and hypervisor says it's running, so we reboot?" | 21:32 |
| *** mudassirlatif has joined #openstack-nova | 21:33 | |
| mriedem | if we stop it and set the flag, in pass 2 even if the flag is set, if we stopped it in pass 1 it's probably no longer running on the hypervisor | 21:33 |
| alaski | I'm not sure on the libvirt case, but for vmware my understanding was that on one pass the power state was stopped and then on the next run it wasn't | 21:34 |
| mriedem | oh right, vm_state is active and hypervisor power state is stopped, so we stop it | 21:34 |
| mriedem | 2nd pass vm_state is stopped and hypervisor power state is running | 21:34 |
| alaski | yeah | 21:35 |
| mriedem | today we stop, but if flag is set we (reboot) | 21:35 |
| alaski | yep | 21:35 |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Avoid needless list copy in 'scheduler_host_subset_size' case https://review.openstack.org/226443 | 21:35 |
| alaski | I don't know if this will help the libvirt case though | 21:35 |
| *** mudassirlatif has quit IRC | 21:35 | |
| alaski | I'm not clear on what happens there yet | 21:36 |
| mriedem | ok, clarified in the comments | 21:37 |
| mriedem | so garyk can enjoy all of this light reading when he's up | 21:37 |
| mriedem | honestly i don't think this is really a problem in the libvirt case | 21:37 |
| mriedem | the problem in the libvirt case was the race with reboot and the lifecycle event stuff, which has been fixed | 21:38 |
| *** e0ne has quit IRC | 21:38 | |
| alaski | okay cool | 21:39 |
| openstackgerrit | Alvaro Lopez Garcia proposed openstack/nova: SameHostFilter should fail if host does not have instances https://review.openstack.org/229030 | 21:39 |
| *** ccarmack has quit IRC | 21:40 | |
| *** harlowja has quit IRC | 21:40 | |
| *** harlowja has joined #openstack-nova | 21:40 | |
| *** salv-orl_ has joined #openstack-nova | 21:43 | |
| *** mudassirlatif has joined #openstack-nova | 21:45 | |
| *** salv-orlando has quit IRC | 21:46 | |
| *** amotoki has quit IRC | 21:46 | |
| *** annegentle has quit IRC | 21:47 | |
| *** ccarmack has joined #openstack-nova | 21:47 | |
| *** annegentle has joined #openstack-nova | 21:48 | |
| *** jichen has quit IRC | 21:48 | |
| *** tdurakov has quit IRC | 21:49 | |
| *** ctrath has quit IRC | 21:49 | |
| *** mudassirlatif has quit IRC | 21:49 | |
| *** devlaps has joined #openstack-nova | 21:52 | |
| *** ijw has joined #openstack-nova | 21:52 | |
| *** zzzeek has quit IRC | 21:52 | |
| *** mudassirlatif has joined #openstack-nova | 21:53 | |
| *** digimonde has quit IRC | 21:53 | |
| *** ctrath has joined #openstack-nova | 21:56 | |
| *** hemna has quit IRC | 21:56 | |
| *** otter768 has joined #openstack-nova | 21:56 | |
| *** ijw has quit IRC | 21:57 | |
| *** ijw has joined #openstack-nova | 21:57 | |
| *** gyee has joined #openstack-nova | 21:59 | |
| *** asettle has joined #openstack-nova | 22:00 | |
| *** otter768 has quit IRC | 22:01 | |
| *** zzzeek has joined #openstack-nova | 22:01 | |
| *** hemna has joined #openstack-nova | 22:01 | |
| *** Piet has quit IRC | 22:02 | |
| *** hemna has quit IRC | 22:04 | |
| *** hemnafk is now known as hemna | 22:04 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: vmware: set default value in fake _db_content when creating objects https://review.openstack.org/229164 | 22:05 |
| mriedem | ^ the most productive thing i've done today | 22:05 |
| *** ociuhandu has quit IRC | 22:06 | |
| *** burt has quit IRC | 22:07 | |
| *** annegentle has quit IRC | 22:08 | |
| *** mdrabe has quit IRC | 22:08 | |
| *** mriedem is now known as mriedem_away | 22:09 | |
| *** mriedem_away has quit IRC | 22:09 | |
| *** tonytan_brb has joined #openstack-nova | 22:10 | |
| *** matrohon has quit IRC | 22:11 | |
| *** boris-42 has joined #openstack-nova | 22:12 | |
| *** baoli has quit IRC | 22:12 | |
| *** baoli has joined #openstack-nova | 22:13 | |
| *** tonytan4ever has quit IRC | 22:14 | |
| *** marcusvrn_ has quit IRC | 22:15 | |
| *** mwagner_ has quit IRC | 22:20 | |
| *** yamahata has quit IRC | 22:24 | |
| *** yamahata has joined #openstack-nova | 22:24 | |
| *** rfolco has quit IRC | 22:26 | |
| *** tonytan_brb has quit IRC | 22:27 | |
| *** VW has quit IRC | 22:28 | |
| *** VW has joined #openstack-nova | 22:28 | |
| *** VW has quit IRC | 22:28 | |
| *** VW has joined #openstack-nova | 22:29 | |
| *** VW has quit IRC | 22:29 | |
| *** rfolco has joined #openstack-nova | 22:29 | |
| *** VW has joined #openstack-nova | 22:29 | |
| *** tonytan4ever has joined #openstack-nova | 22:33 | |
| *** mc_nair has quit IRC | 22:34 | |
| *** stevemar has quit IRC | 22:38 | |
| *** rfolco has quit IRC | 22:42 | |
| *** baoli has quit IRC | 22:44 | |
| *** baoli has joined #openstack-nova | 22:44 | |
| *** thorst has joined #openstack-nova | 22:46 | |
| *** zzzeek has quit IRC | 22:46 | |
| *** amotoki has joined #openstack-nova | 22:46 | |
| *** mudassirlatif has quit IRC | 22:47 | |
| *** rfolco has joined #openstack-nova | 22:49 | |
| *** ijuwang has quit IRC | 22:50 | |
| *** emagana has joined #openstack-nova | 22:51 | |
| *** amotoki has quit IRC | 22:51 | |
| *** edev has joined #openstack-nova | 22:52 | |
| edev | #openstack | 22:53 |
| *** edev has left #openstack-nova | 22:53 | |
| *** alejandrito has quit IRC | 22:56 | |
| *** ctrath has quit IRC | 22:56 | |
| *** mwagner_ has joined #openstack-nova | 22:57 | |
| *** su_zhang has quit IRC | 22:58 | |
| *** mudassirlatif has joined #openstack-nova | 22:59 | |
| *** jhesketh has quit IRC | 23:01 | |
| *** jhesketh has joined #openstack-nova | 23:02 | |
| *** browne has quit IRC | 23:05 | |
| *** browne has joined #openstack-nova | 23:05 | |
| *** su_zhang has joined #openstack-nova | 23:06 | |
| *** Piet has joined #openstack-nova | 23:07 | |
| *** signed8bit has joined #openstack-nova | 23:07 | |
| *** rfolco has quit IRC | 23:12 | |
| *** julien_ZTE has quit IRC | 23:13 | |
| *** julien_ZTE has joined #openstack-nova | 23:13 | |
| *** yamahata has quit IRC | 23:15 | |
| *** tonytan4ever has quit IRC | 23:16 | |
| *** baoli has quit IRC | 23:16 | |
| *** rfolco has joined #openstack-nova | 23:17 | |
| *** ijw has quit IRC | 23:18 | |
| *** mnestratov|2 has quit IRC | 23:18 | |
| *** salv-orl_ has quit IRC | 23:19 | |
| *** mnestratov|2 has joined #openstack-nova | 23:19 | |
| *** xyang1 has quit IRC | 23:19 | |
| *** dims_ has joined #openstack-nova | 23:19 | |
| *** mudassirlatif has quit IRC | 23:21 | |
| *** dims has quit IRC | 23:22 | |
| *** thorst has quit IRC | 23:22 | |
| *** eharney has quit IRC | 23:23 | |
| *** mudassirlatif has joined #openstack-nova | 23:23 | |
| *** tpeoples has quit IRC | 23:24 | |
| *** su_zhang has quit IRC | 23:24 | |
| *** aloga has quit IRC | 23:24 | |
| *** su_zhang has joined #openstack-nova | 23:24 | |
| *** aloga has joined #openstack-nova | 23:25 | |
| *** ZZelle_ has quit IRC | 23:25 | |
| *** markmcclain has quit IRC | 23:25 | |
| *** apoorvad has quit IRC | 23:26 | |
| *** Apoorva has joined #openstack-nova | 23:30 | |
| *** irenab_ has joined #openstack-nova | 23:30 | |
| *** markmcclain has joined #openstack-nova | 23:31 | |
| *** ccarmack1 has joined #openstack-nova | 23:31 | |
| *** mdbooth_ has joined #openstack-nova | 23:31 | |
| *** dims_ has quit IRC | 23:31 | |
| *** bauzas_ has joined #openstack-nova | 23:32 | |
| *** dims_ has joined #openstack-nova | 23:32 | |
| *** andrearo1 has joined #openstack-nova | 23:32 | |
| *** sambetts_ has joined #openstack-nova | 23:32 | |
| *** mdbooth has quit IRC | 23:32 | |
| *** flaper87 has quit IRC | 23:32 | |
| *** bauzas has quit IRC | 23:32 | |
| *** rfolco has quit IRC | 23:32 | |
| *** sambetts has quit IRC | 23:32 | |
| *** htruta has quit IRC | 23:32 | |
| *** johnthetubaguy has quit IRC | 23:32 | |
| *** tristanC has quit IRC | 23:32 | |
| *** bswartz has joined #openstack-nova | 23:32 | |
| *** mdbooth_ is now known as mdbooth | 23:32 | |
| *** bauzas_ is now known as bauzas | 23:32 | |
| *** htruta` has joined #openstack-nova | 23:32 | |
| *** flaper87 has joined #openstack-nova | 23:32 | |
| *** tristanC has joined #openstack-nova | 23:33 | |
| *** ccarmack has quit IRC | 23:33 | |
| *** irenab has quit IRC | 23:33 | |
| *** albertom has quit IRC | 23:33 | |
| *** mcarden has quit IRC | 23:33 | |
| *** andrearosa has quit IRC | 23:33 | |
| *** brad[] has quit IRC | 23:33 | |
| *** rfolco has joined #openstack-nova | 23:33 | |
| *** irenab_ is now known as irenab | 23:33 | |
| *** Apoorva is now known as apoorvad | 23:33 | |
| *** aloga has quit IRC | 23:33 | |
| *** keekz has quit IRC | 23:33 | |
| *** loquacities has quit IRC | 23:33 | |
| *** aloga has joined #openstack-nova | 23:33 | |
| *** loquacities has joined #openstack-nova | 23:33 | |
| *** mcarden has joined #openstack-nova | 23:34 | |
| *** albertom has joined #openstack-nova | 23:34 | |
| *** johnthetubaguy has joined #openstack-nova | 23:34 | |
| *** keekz has joined #openstack-nova | 23:34 | |
| *** asettle has quit IRC | 23:37 | |
| *** rfolco has quit IRC | 23:40 | |
| *** edmondsw has quit IRC | 23:41 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 23:41 | |
| *** mudassirlatif has quit IRC | 23:43 | |
| *** hemna is now known as hemnafk | 23:43 | |
| *** rfolco has joined #openstack-nova | 23:44 | |
| *** achanda_ has joined #openstack-nova | 23:44 | |
| *** julien_ZTE has quit IRC | 23:44 | |
| *** coreywright_ has joined #openstack-nova | 23:44 | |
| *** achanda_ has quit IRC | 23:45 | |
| *** coreywright_ is now known as coreywright | 23:46 | |
| *** pm90_ has quit IRC | 23:46 | |
| *** achanda has quit IRC | 23:48 | |
| *** emagana has quit IRC | 23:48 | |
| *** signed8bit_ZZZzz is now known as signed8bit | 23:49 | |
| *** mudassirlatif has joined #openstack-nova | 23:51 | |
| *** otter768 has joined #openstack-nova | 23:57 | |
| *** Piet has quit IRC | 23:57 | |
| *** Piet has joined #openstack-nova | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!