| *** achanda has joined #openstack-nova | 00:00 | |
| *** y_sawai has joined #openstack-nova | 00:00 | |
| *** y_sawai has quit IRC | 00:00 | |
| *** y_sawai has joined #openstack-nova | 00:01 | |
| *** y_sawai has quit IRC | 00:01 | |
| *** achanda has quit IRC | 00:04 | |
| *** SourabhP has quit IRC | 00:07 | |
| *** y_sawai has joined #openstack-nova | 00:07 | |
| *** vilobhmm1 has quit IRC | 00:09 | |
| *** yuntongjin has joined #openstack-nova | 00:12 | |
| *** achanda has joined #openstack-nova | 00:12 | |
| *** lucas-dinner has quit IRC | 00:12 | |
| *** pixelb has quit IRC | 00:15 | |
| *** salv-orlando has quit IRC | 00:15 | |
| *** dims__ has quit IRC | 00:16 | |
| *** promulo__ has joined #openstack-nova | 00:18 | |
| *** shoutm has quit IRC | 00:19 | |
| *** electrocucaracha has quit IRC | 00:21 | |
| *** promulo has quit IRC | 00:22 | |
| *** shoutm has joined #openstack-nova | 00:23 | |
| *** penick has quit IRC | 00:25 | |
| *** ijw_ has joined #openstack-nova | 00:26 | |
| *** tpeoples has joined #openstack-nova | 00:26 | |
| *** tjones has joined #openstack-nova | 00:27 | |
| *** baoli has quit IRC | 00:27 | |
| *** apuimedo has quit IRC | 00:27 | |
| *** Apoorva has quit IRC | 00:28 | |
| *** y_sawai has quit IRC | 00:28 | |
| *** zhenguo has joined #openstack-nova | 00:30 | |
| *** vilobhmm_11 has joined #openstack-nova | 00:36 | |
| *** sayali_ has joined #openstack-nova | 00:37 | |
| *** sayali_ has quit IRC | 00:37 | |
| *** sayali_ has joined #openstack-nova | 00:37 | |
| *** VW_ has joined #openstack-nova | 00:37 | |
| *** sayali has quit IRC | 00:37 | |
| *** dguitarbite has quit IRC | 00:37 | |
| *** dbite has joined #openstack-nova | 00:37 | |
| *** diogogmt has joined #openstack-nova | 00:39 | |
| *** claudiub has quit IRC | 00:39 | |
| *** VW_ has quit IRC | 00:39 | |
| *** VW_ has joined #openstack-nova | 00:39 | |
| *** vivekd has quit IRC | 00:40 | |
| *** dannywil_ has quit IRC | 00:43 | |
| *** dannywilson has joined #openstack-nova | 00:43 | |
| *** VW_ has quit IRC | 00:44 | |
| *** zhangjn has joined #openstack-nova | 00:44 | |
| *** dannywilson has quit IRC | 00:45 | |
| *** morgabra has quit IRC | 00:50 | |
| *** morgabra has joined #openstack-nova | 00:50 | |
| *** dims__ has joined #openstack-nova | 00:51 | |
| *** ijw_ has quit IRC | 00:51 | |
| *** tjones has quit IRC | 00:52 | |
| *** ijw_ has joined #openstack-nova | 00:53 | |
| *** Sukhdev has quit IRC | 00:55 | |
| *** dims__ has quit IRC | 00:56 | |
| *** xyang1 has joined #openstack-nova | 00:57 | |
| *** ijw_ has quit IRC | 00:57 | |
| *** kaisers has quit IRC | 01:01 | |
| *** kaisers has joined #openstack-nova | 01:03 | |
| *** tpeoples has quit IRC | 01:03 | |
| *** asettle is now known as asettle-afk | 01:03 | |
| *** r-daneel has quit IRC | 01:04 | |
| *** browne has quit IRC | 01:04 | |
| *** kjelly has quit IRC | 01:07 | |
| *** achanda has quit IRC | 01:09 | |
| *** hemna has joined #openstack-nova | 01:10 | |
| *** nic has quit IRC | 01:13 | |
| *** mtanino has quit IRC | 01:13 | |
| *** vivekd has joined #openstack-nova | 01:16 | |
| *** shuquan has joined #openstack-nova | 01:17 | |
| *** thorst has joined #openstack-nova | 01:17 | |
| *** wanghua has joined #openstack-nova | 01:21 | |
| *** armax has quit IRC | 01:22 | |
| *** shuquan_ has joined #openstack-nova | 01:23 | |
| *** shuquan has quit IRC | 01:23 | |
| *** shuquan_ is now known as shuquan | 01:23 | |
| *** baoli has joined #openstack-nova | 01:27 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add secgroup param checks for Neutron https://review.openstack.org/192043 | 01:28 |
|---|---|---|
| *** baoli has quit IRC | 01:28 | |
| *** baoli has joined #openstack-nova | 01:28 | |
| *** haomaiwa_ has joined #openstack-nova | 01:30 | |
| *** tri2sing has quit IRC | 01:30 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add secgroup param checks for Neutron https://review.openstack.org/192043 | 01:30 |
| *** ijw_ has joined #openstack-nova | 01:35 | |
| *** armax has joined #openstack-nova | 01:35 | |
| *** asettle-afk has quit IRC | 01:36 | |
| *** baoli has quit IRC | 01:37 | |
| *** baoli has joined #openstack-nova | 01:38 | |
| *** vivekd has quit IRC | 01:39 | |
| *** shuquan has quit IRC | 01:40 | |
| *** rlrossit has joined #openstack-nova | 01:41 | |
| *** rlrossit has left #openstack-nova | 01:41 | |
| *** vilobhmm_11 has quit IRC | 01:43 | |
| *** armax has quit IRC | 01:46 | |
| *** diogogmt has quit IRC | 01:46 | |
| *** Kevin_Zheng has joined #openstack-nova | 01:47 | |
| *** diogogmt has joined #openstack-nova | 01:48 | |
| gmann | oomichi: alex_xu : - https://etherpad.openstack.org/p/liberty-nova-api-testing | 01:49 |
| *** vivekd has joined #openstack-nova | 01:49 | |
| gmann | oomichi: alex_xu : i will push cleanup patches for old jobs removal | 01:50 |
| *** lyanchih has joined #openstack-nova | 01:50 | |
| oomichi | gmann: thanks for pointing it out | 01:51 |
| oomichi | gmann: did you write it? | 01:51 |
| oomichi | gmann: or sdague? | 01:51 |
| gmann | oomichi: sdague did and posted patches also | 01:52 |
| gmann | oomichi: just cleanup of old jobs left | 01:52 |
| alex_xu | gmann: cool | 01:52 |
| gmann | oomichi: sdague : alex_xu : all looks good just for project_config patch m concern about whether we run v2 compatible job for tempest too | 01:53 |
| oomichi | gmann: that seems lack of legacy v2.0 code without v2.1 code | 01:53 |
| gmann | oomichi: it has that also - gate-tempest-dsvm-nova-v20-api-legacy in https://review.openstack.org/#/c/219370/1/zuul/layout.yaml | 01:54 |
| gmann | oomichi: is that you mean? | 01:54 |
| oomichi | gmann: that is, thanks | 01:55 |
| *** meghal1 has quit IRC | 01:55 | |
| *** zhangjn_ has joined #openstack-nova | 01:57 | |
| *** zhangjn has quit IRC | 01:58 | |
| *** liusheng has joined #openstack-nova | 02:02 | |
| *** achanda has joined #openstack-nova | 02:04 | |
| *** shyu has joined #openstack-nova | 02:04 | |
| *** humble_ has joined #openstack-nova | 02:05 | |
| *** wanghua has quit IRC | 02:08 | |
| *** tpeoples has joined #openstack-nova | 02:08 | |
| *** achanda has quit IRC | 02:14 | |
| *** yuntongjin has quit IRC | 02:16 | |
| *** thorst has quit IRC | 02:17 | |
| *** achanda has joined #openstack-nova | 02:23 | |
| *** salv-orlando has joined #openstack-nova | 02:27 | |
| *** baoli has quit IRC | 02:29 | |
| *** baoli_ has joined #openstack-nova | 02:30 | |
| *** unicell has quit IRC | 02:31 | |
| *** salv-orlando has quit IRC | 02:32 | |
| *** yuntongjin has joined #openstack-nova | 02:34 | |
| *** leizhang has joined #openstack-nova | 02:35 | |
| *** ijw_ has quit IRC | 02:38 | |
| eliqiao | hello, will the items in nova tables being deleted automaticly? for EX: in instances table, when we deleting a instance, nova just mark 'delete'=True instead of delete it from table , so I wonder when will this record being delete from table? | 02:38 |
| eliqiao | and who will trigger that? | 02:38 |
| *** ijw_ has joined #openstack-nova | 02:38 | |
| *** ijw_ has quit IRC | 02:39 | |
| *** Sree has quit IRC | 02:41 | |
| *** annegentle has quit IRC | 02:41 | |
| *** yuntongjin1 has joined #openstack-nova | 02:42 | |
| *** annegentle has joined #openstack-nova | 02:43 | |
| *** yuntongjin has quit IRC | 02:43 | |
| *** Sukhdev_ has joined #openstack-nova | 02:45 | |
| *** hparekh has joined #openstack-nova | 02:46 | |
| *** yuntongjin has joined #openstack-nova | 02:47 | |
| *** gyee has quit IRC | 02:47 | |
| *** hhuang has joined #openstack-nova | 02:48 | |
| *** yuntongjin1 has quit IRC | 02:48 | |
| *** achanda has quit IRC | 02:49 | |
| *** hakimo_ has joined #openstack-nova | 02:52 | |
| *** hemna has quit IRC | 02:54 | |
| *** hakimo has quit IRC | 02:54 | |
| *** leizhang has quit IRC | 02:55 | |
| *** annegentle has quit IRC | 02:57 | |
| *** meghal has joined #openstack-nova | 03:02 | |
| *** achanda has joined #openstack-nova | 03:02 | |
| *** yuntongjin1 has joined #openstack-nova | 03:02 | |
| *** changbl has joined #openstack-nova | 03:02 | |
| *** hemna has joined #openstack-nova | 03:03 | |
| *** yuntongjin has quit IRC | 03:03 | |
| *** meghal1 has joined #openstack-nova | 03:04 | |
| *** humble__ has joined #openstack-nova | 03:05 | |
| *** meghal has quit IRC | 03:06 | |
| *** humble_ has quit IRC | 03:06 | |
| *** humble_ has joined #openstack-nova | 03:07 | |
| *** achanda has quit IRC | 03:08 | |
| *** dave-mccowan has quit IRC | 03:08 | |
| *** asettle has joined #openstack-nova | 03:11 | |
| *** humble__ has quit IRC | 03:11 | |
| *** mmedvede has quit IRC | 03:11 | |
| *** yuntongjin has joined #openstack-nova | 03:12 | |
| *** yuntongjin1 has quit IRC | 03:14 | |
| *** rajesht has joined #openstack-nova | 03:14 | |
| rajesht | tristanC: hi | 03:16 |
| openstackgerrit | apporc proposed openstack/nova: change qemu guest agent socket path https://review.openstack.org/214935 | 03:17 |
| openstackgerrit | Eli Qiao proposed openstack/nova: API: Handle InstanceUnknownCell exceptions https://review.openstack.org/212965 | 03:18 |
| *** shuquan has joined #openstack-nova | 03:19 | |
| *** vladikr has quit IRC | 03:20 | |
| *** yuntongjin1 has joined #openstack-nova | 03:22 | |
| *** yuntongjin has quit IRC | 03:25 | |
| *** haomaiwa_ has quit IRC | 03:28 | |
| *** haomaiwang has joined #openstack-nova | 03:29 | |
| alex_xu | gmann: is there any way we run a test for specific job and skip for another job? | 03:31 |
| *** rajesht has quit IRC | 03:31 | |
| *** Sree has joined #openstack-nova | 03:35 | |
| *** baoli_ has quit IRC | 03:36 | |
| gmann | alex_xu: on gerrit, i do not think we have that option | 03:41 |
| *** salv-orlando has joined #openstack-nova | 03:41 | |
| *** salv-orlando has quit IRC | 03:46 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Fix incomplete error message of quota exceeded https://review.openstack.org/197460 | 03:46 |
| *** links has joined #openstack-nova | 03:48 | |
| *** armax has joined #openstack-nova | 03:52 | |
| *** humble__ has joined #openstack-nova | 03:55 | |
| *** humble_ has quit IRC | 03:59 | |
| *** zhangjn has joined #openstack-nova | 04:00 | |
| *** zhangjn_ has quit IRC | 04:02 | |
| *** dims__ has joined #openstack-nova | 04:03 | |
| *** deepthi has joined #openstack-nova | 04:04 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add UniqueConstraint for quota usages table https://review.openstack.org/212908 | 04:05 |
| *** SourabhP has joined #openstack-nova | 04:06 | |
| *** ishant has joined #openstack-nova | 04:08 | |
| *** unicell has joined #openstack-nova | 04:10 | |
| *** Nikolay_St has joined #openstack-nova | 04:12 | |
| *** yuntongjin1 has quit IRC | 04:12 | |
| *** shuquan has quit IRC | 04:13 | |
| *** unicell has quit IRC | 04:16 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: add support for muti_statuses_query from migration https://review.openstack.org/169631 | 04:16 |
| *** xyang1 has quit IRC | 04:17 | |
| *** dims__ is now known as dims | 04:17 | |
| *** dims has quit IRC | 04:24 | |
| *** unicell has joined #openstack-nova | 04:24 | |
| *** hparekh has quit IRC | 04:26 | |
| *** garyk has quit IRC | 04:27 | |
| *** hparekh has joined #openstack-nova | 04:30 | |
| *** hemna has quit IRC | 04:31 | |
| *** lyanchih has quit IRC | 04:31 | |
| *** irenab has quit IRC | 04:33 | |
| *** hemna has joined #openstack-nova | 04:36 | |
| *** ijw_ has joined #openstack-nova | 04:37 | |
| *** achanda has joined #openstack-nova | 04:42 | |
| *** meghal1 has quit IRC | 04:43 | |
| *** Sukhdev_ has quit IRC | 04:44 | |
| *** jcoufal has joined #openstack-nova | 04:52 | |
| alex_xu | gmann: yea | 04:53 |
| *** salv-orlando has joined #openstack-nova | 04:56 | |
| *** mtreinish has quit IRC | 04:57 | |
| *** lyanchih has joined #openstack-nova | 04:58 | |
| *** VikasC has joined #openstack-nova | 04:59 | |
| *** salv-orlando has quit IRC | 05:01 | |
| gmann | oomichi: sdague : alex_xu : cleanup patches for v2.1 jobs - https://review.openstack.org/#/c/219555/ https://review.openstack.org/#/c/219553/ | 05:03 |
| alex_xu | gmann: thanks! | 05:03 |
| gmann | alex_xu: np | 05:03 |
| *** mtreinish has joined #openstack-nova | 05:04 | |
| *** meghal has joined #openstack-nova | 05:05 | |
| *** markvoelker has joined #openstack-nova | 05:06 | |
| *** unicell has quit IRC | 05:07 | |
| *** jichen has joined #openstack-nova | 05:09 | |
| *** unicell has joined #openstack-nova | 05:09 | |
| *** markvoelker has quit IRC | 05:11 | |
| *** markvoelker has joined #openstack-nova | 05:12 | |
| *** lyanchih has quit IRC | 05:14 | |
| *** yuntongjin has joined #openstack-nova | 05:15 | |
| *** Sukhdev_ has joined #openstack-nova | 05:17 | |
| *** IanGovett has quit IRC | 05:25 | |
| *** Sree has quit IRC | 05:25 | |
| *** IanGovett has joined #openstack-nova | 05:25 | |
| *** vilobhmm_11 has joined #openstack-nova | 05:27 | |
| *** irenab has joined #openstack-nova | 05:27 | |
| *** rotbeard has joined #openstack-nova | 05:28 | |
| *** Sree has joined #openstack-nova | 05:28 | |
| *** Murali has joined #openstack-nova | 05:30 | |
| *** IanGovett has quit IRC | 05:30 | |
| *** vilobhmm_11 has quit IRC | 05:32 | |
| *** nkrinner has joined #openstack-nova | 05:35 | |
| *** shuquan has joined #openstack-nova | 05:36 | |
| *** shuquan has quit IRC | 05:41 | |
| *** shuquan has joined #openstack-nova | 05:41 | |
| *** rushiagr_away is now known as rushiagr | 05:44 | |
| openstackgerrit | IWAMOTO Toshihiro proposed openstack/nova: Omit RescheduledException in instance_fault.message https://review.openstack.org/165751 | 05:49 |
| *** jcoufal has quit IRC | 05:50 | |
| *** salv-orlando has joined #openstack-nova | 05:52 | |
| *** amotoki has joined #openstack-nova | 05:54 | |
| *** amotoki has quit IRC | 05:56 | |
| *** lyanchih has joined #openstack-nova | 05:59 | |
| *** amotoki has joined #openstack-nova | 06:00 | |
| *** amotoki_ has joined #openstack-nova | 06:02 | |
| *** hhuang has quit IRC | 06:02 | |
| *** Sukhdev_ has quit IRC | 06:04 | |
| *** amotoki has quit IRC | 06:06 | |
| *** garyk has joined #openstack-nova | 06:07 | |
| *** diogogmt has quit IRC | 06:11 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Fix API directories on the doc https://review.openstack.org/219570 | 06:15 |
| *** sahid has joined #openstack-nova | 06:18 | |
| *** lpetrut has joined #openstack-nova | 06:20 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Fix API directories on the doc https://review.openstack.org/219570 | 06:20 |
| *** jichen has quit IRC | 06:21 | |
| *** browne has joined #openstack-nova | 06:23 | |
| alex_xu | oomichi: ping | 06:23 |
| *** asettle has quit IRC | 06:23 | |
| *** markvoelker has quit IRC | 06:23 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Transifex https://review.openstack.org/212367 | 06:24 |
| oomichi | alex_xu: pong | 06:24 |
| alex_xu | oomichi: hey, just want to talk about https://review.openstack.org/#/c/217727 | 06:24 |
| oomichi | alex_xu: yeah, ok | 06:25 |
| *** ildikov has quit IRC | 06:25 | |
| alex_xu | oomichi: extend schema for scheduler-hints by extension, looks like it isn't a option for the future, as we will return extension | 06:25 |
| oomichi | alex_xu: s/return/remove/ ? | 06:26 |
| alex_xu | oomichi: yea, sorry for that | 06:26 |
| oomichi | alex_xu: yeah, that is right way. but before saying that, do we need to consider out-of-tree features? | 06:26 |
| *** lpetrut has quit IRC | 06:27 | |
| alex_xu | oomichi: the original concern is the v2.1 api compatible mode should allow extened hints, otherwise the v2.1 comaptible mode isn't compatible with v2 | 06:27 |
| oomichi | alex_xu: the commit message said "allow for out of tree extensions" | 06:28 |
| alex_xu | oomichi: initially I don't think we should consider out of tree features | 06:28 |
| openstackgerrit | lyanchih proposed openstack/nova: libvirt - Flavor's quota instance resource not observed for non-ephemeral storage https://review.openstack.org/201019 | 06:29 |
| *** vilobhmm_11 has joined #openstack-nova | 06:29 | |
| alex_xu | oomichi: but there is good point from alaski , you can find the original comment at https://review.openstack.org/#/c/215436/ | 06:29 |
| oomichi | alex_xu: that was different from the same validation of legacy v2 api | 06:29 |
| alex_xu | oomichi: it's more about v2 api allow extend scheduler-hints, and in the real world a lot of use like that | 06:30 |
| *** mpaolino has joined #openstack-nova | 06:30 | |
| oomichi | alex_xu: before that, they need to update their own out-of-tree features for applying them to v2.1 code | 06:31 |
| *** Nikolay_St has quit IRC | 06:31 | |
| oomichi | alex_xu: I think they can extend the corresponding schema also during that | 06:31 |
| *** tojuvone has joined #openstack-nova | 06:32 | |
| *** venkat_p has joined #openstack-nova | 06:32 | |
| alex_xu | oomichi: emm...yea, they need update their own out of tree features even with v2.1 compat mode | 06:32 |
| alex_xu | oomichi: but change the schema, really like change the API contract | 06:32 |
| * alex_xu actually on the side not merge 217727 | 06:33 | |
| oomichi | alex_xu: out-of-tree features always change the API contract | 06:33 |
| alex_xu | oomichi: emm....we should say out of tree features should extend the API contract? | 06:34 |
| *** browne has quit IRC | 06:34 | |
| alex_xu | oomichi: emm...I probably get yoru point | 06:35 |
| alex_xu | s/yoru/your | 06:35 |
| oomichi | alex_xu: ideally, it is great that all features are implemented at upstream without out-of-tree because of the interoperability | 06:35 |
| oomichi | alex_xu: that will be the API contract | 06:35 |
| alex_xu | oomichi: yea | 06:36 |
| alex_xu | oomichi: actually it's not | 06:36 |
| oomichi | alex_xu: but I can see out-of-tree features also in the real world | 06:36 |
| oomichi | alex_xu: so I'd like to avoid add-hoc way for out-of-tree feature for clean code at this time | 06:37 |
| *** hhuang has joined #openstack-nova | 06:37 | |
| oomichi | alex_xu: we already provide clean way(but that is a little difficult) | 06:37 |
| oomichi | alex_xu: how about writting the way to extend schemas as the doc | 06:39 |
| oomichi | alex_xu: now we don't have enough doc about json-schema in tree | 06:39 |
| oomichi | alex_xu: I will post a patch for that | 06:40 |
| *** ndipanov has quit IRC | 06:40 | |
| oomichi | alex_xu: https://review.openstack.org/219570 is the first step | 06:40 |
| alex_xu | oomichi: sounds make sense to me. Whatever the user need update their out-of-tree extension. | 06:40 |
| alex_xu | oomichi: and we support extend the server create's json-schema | 06:41 |
| alex_xu | oomichi: cool, thanks | 06:41 |
| oomichi | alex_xu: I will imply out-of-tree extension but they will be able to know the way from the doc | 06:41 |
| oomichi | alex_xu: as upstream doc | 06:42 |
| alex_xu | oomichi: ok | 06:42 |
| alex_xu | oomichi: thanks for the discussion | 06:43 |
| oomichi | alex_xu: welcome always :) | 06:44 |
| alex_xu | oomichi: :) | 06:44 |
| *** rushil has joined #openstack-nova | 06:46 | |
| *** achanda has quit IRC | 06:47 | |
| *** rushil_ has joined #openstack-nova | 06:48 | |
| *** exploreshaifali has joined #openstack-nova | 06:50 | |
| *** vilobhmm_11 has quit IRC | 06:50 | |
| *** dims has joined #openstack-nova | 06:50 | |
| *** stefan_amann has joined #openstack-nova | 06:50 | |
| alex_xu | sdague: johnthetubaguy ^ hope you notice the discussion as above, it's about https://review.openstack.org/#/c/217727/ | 06:50 |
| *** lyanchih has quit IRC | 06:51 | |
| *** rushil has quit IRC | 06:51 | |
| *** afazekas__ has joined #openstack-nova | 06:52 | |
| *** dims has quit IRC | 06:54 | |
| *** venkat_p has quit IRC | 06:55 | |
| openstackgerrit | Zhenyu Zheng proposed openstack/nova: Add UniqueConstraint for quota usages table https://review.openstack.org/212908 | 06:56 |
| *** ildikov has joined #openstack-nova | 06:58 | |
| *** venkat_p has joined #openstack-nova | 07:08 | |
| *** tpeoples has quit IRC | 07:08 | |
| *** rdopiera has joined #openstack-nova | 07:11 | |
| *** markus_z has joined #openstack-nova | 07:13 | |
| openstackgerrit | garyk proposed openstack/nova: Revert "Remove references to suds" https://review.openstack.org/219225 | 07:15 |
| garyk | mriedem_away: ^ - i am even struggling to open the log file from the nova log to past additional information to the bug | 07:16 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add JSON-Schema note to api_plugins.rst https://review.openstack.org/219580 | 07:16 |
| openstackgerrit | garyk proposed openstack/nova: Revert "Remove references to suds" https://review.openstack.org/219225 | 07:17 |
| garyk | oomichi: please see a comment on https://review.openstack.org/219580 | 07:19 |
| *** meghal has quit IRC | 07:19 | |
| oomichi | garyk: thanks, so quick :) | 07:19 |
| *** ndipanov has joined #openstack-nova | 07:19 | |
| garyk | oomichi: early bird catches the worm :) | 07:20 |
| ndipanov | johnthetubaguy, once you're on - https://review.openstack.org/#/c/216733/ - dansmith -1ed it out of the gate for some changes, missing a second +2 now | 07:20 |
| *** alexschm has joined #openstack-nova | 07:21 | |
| *** lyanchih has joined #openstack-nova | 07:21 | |
| *** coolsvap_ has joined #openstack-nova | 07:21 | |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add JSON-Schema note to api_plugins.rst https://review.openstack.org/219580 | 07:21 |
| *** Nikolay_St has joined #openstack-nova | 07:22 | |
| oomichi | alex_xu: I posted a patch https://review.openstack.org/219580 for the doc, I'm happy if you see it | 07:23 |
| *** achanda has joined #openstack-nova | 07:24 | |
| alex_xu | oomichi: cool, checking now | 07:24 |
| *** cfriesen has quit IRC | 07:25 | |
| *** coolsvap_ has quit IRC | 07:26 | |
| *** rushil_ has quit IRC | 07:27 | |
| *** pbrooko has joined #openstack-nova | 07:28 | |
| *** vivekd has quit IRC | 07:28 | |
| *** rushil has joined #openstack-nova | 07:28 | |
| *** kylek3h has quit IRC | 07:28 | |
| *** kylek3h has joined #openstack-nova | 07:29 | |
| *** karimb has joined #openstack-nova | 07:32 | |
| *** kylek3h has quit IRC | 07:33 | |
| *** armax has quit IRC | 07:34 | |
| *** alex_klimov has joined #openstack-nova | 07:36 | |
| *** eglynn has quit IRC | 07:37 | |
| *** yuntongjin has quit IRC | 07:39 | |
| *** jlanoux has joined #openstack-nova | 07:41 | |
| *** dims has joined #openstack-nova | 07:44 | |
| *** jlanoux has quit IRC | 07:45 | |
| *** Vek has quit IRC | 07:45 | |
| *** SourabhP has quit IRC | 07:45 | |
| *** dims has quit IRC | 07:49 | |
| *** shuquan has quit IRC | 07:51 | |
| *** achanda has quit IRC | 07:51 | |
| *** Nikolay_St has quit IRC | 07:52 | |
| *** shuquan has joined #openstack-nova | 07:52 | |
| *** BobBall-AWOL is now known as BobBall | 07:54 | |
| *** jlanoux has joined #openstack-nova | 07:56 | |
| *** lpetrut has joined #openstack-nova | 08:03 | |
| openstackgerrit | garyk proposed openstack/nova: Compute: update finish_revert_resize log to have some context https://review.openstack.org/218656 | 08:03 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack/nova: Add JSON-Schema note to api_plugins.rst https://review.openstack.org/219580 | 08:05 |
| *** rubasov has quit IRC | 08:06 | |
| *** ildikov has quit IRC | 08:08 | |
| *** jistr has joined #openstack-nova | 08:11 | |
| *** yassine_ has joined #openstack-nova | 08:11 | |
| *** eglynn has joined #openstack-nova | 08:12 | |
| *** lucas-dinner has joined #openstack-nova | 08:12 | |
| *** lucas-dinner has quit IRC | 08:12 | |
| *** lucasagomes has joined #openstack-nova | 08:12 | |
| *** romainh has joined #openstack-nova | 08:16 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/216968 | 08:18 |
| bauzas | alex_xu: morning | 08:19 |
| alex_xu | bauzas: morning | 08:20 |
| bauzas | alex_xu: I have a question for you re: exception messages vs. v2 API | 08:20 |
| bauzas | alex_xu: https://review.openstack.org/#/c/218468/2/nova/api/openstack/compute/legacy_v2/contrib/server_groups.py,cm | 08:20 |
| bauzas | alex_xu: since ^changes the output message, is it acceptable? | 08:20 |
| alex_xu | bauzas: I think it is ok | 08:21 |
| alex_xu | bauzas: the message field just something human readable string | 08:21 |
| *** ildikov has joined #openstack-nova | 08:21 | |
| bauzas | alex_xu: yeah, it should be not consumable | 08:21 |
| alex_xu | bauzas: yeah | 08:21 |
| *** vivekd has joined #openstack-nova | 08:23 | |
| *** rubasov has joined #openstack-nova | 08:24 | |
| *** shuquan has quit IRC | 08:24 | |
| *** EinstCrazy has joined #openstack-nova | 08:25 | |
| bauzas | alex_xu: okay, thanks for the heads-up | 08:27 |
| alex_xu | bauzas: np, you are welcome | 08:27 |
| *** shuquan has joined #openstack-nova | 08:27 | |
| *** earlephilhower has quit IRC | 08:28 | |
| openstackgerrit | Rui Chen proposed openstack/nova: Fix booting fail when unlimited project quota https://review.openstack.org/219596 | 08:30 |
| *** venkat_p has quit IRC | 08:31 | |
| *** matrohon has quit IRC | 08:35 | |
| *** shoutm has quit IRC | 08:36 | |
| *** dims has joined #openstack-nova | 08:38 | |
| markus_z | bauzas: I'll be in vacation, starting tomorrow. Would you give the status in the next two nova meetings? | 08:41 |
| *** marzif has joined #openstack-nova | 08:43 | |
| *** venkat_p has joined #openstack-nova | 08:43 | |
| *** dims has quit IRC | 08:43 | |
| markus_z | bauzas: I'm talking about the bugs. | 08:44 |
| *** Marga_ has joined #openstack-nova | 08:44 | |
| bauzas | markus_z: on a call, sorry | 08:44 |
| *** humble__ has quit IRC | 08:45 | |
| *** lyanchih has quit IRC | 08:46 | |
| *** exploreshaifali has quit IRC | 08:48 | |
| *** Marga_ has quit IRC | 08:48 | |
| *** shuquan has quit IRC | 08:49 | |
| *** Marga_ has joined #openstack-nova | 08:49 | |
| *** derekh has joined #openstack-nova | 08:50 | |
| *** lyanchih has joined #openstack-nova | 08:50 | |
| *** danpb has joined #openstack-nova | 08:51 | |
| *** markvoelker has joined #openstack-nova | 08:57 | |
| *** amotoki_ has quit IRC | 08:58 | |
| *** rotbeard has quit IRC | 08:58 | |
| *** rotbeard has joined #openstack-nova | 08:58 | |
| *** rotbeard has quit IRC | 09:00 | |
| *** lalitd has joined #openstack-nova | 09:01 | |
| *** matrohon has joined #openstack-nova | 09:01 | |
| *** IanGovett has joined #openstack-nova | 09:01 | |
| *** markvoelker has quit IRC | 09:02 | |
| *** rushil has quit IRC | 09:03 | |
| bauzas | markus_z: I'm back, you want me to handover the bug reporting ? | 09:04 |
| bauzas | markus_z: sure, can do | 09:04 |
| markus_z | bauzas: Yes, that would be great. | 09:04 |
| markus_z | bauzas: Awesome, thanks! | 09:05 |
| bauzas | markus_z: 2 weeks ? | 09:05 |
| markus_z | Just tomorrow and next week. I'm back at 15th of September. | 09:05 |
| *** Marga_ has quit IRC | 09:05 | |
| bauzas | markus_z: okay, np | 09:07 |
| bauzas | markus_z: enjoy your holidays :) | 09:07 |
| markus_z | bauzas: Thanks! I'll do :) | 09:07 |
| *** salv-orlando has quit IRC | 09:07 | |
| bauzas | andreykurilin: morning :) | 09:10 |
| bauzas | andreykurilin: I have a question for you re: https://github.com/openstack/python-novaclient/blob/master/novaclient/v2/contrib/host_evacuate_live.py | 09:10 |
| *** takedakn has joined #openstack-nova | 09:10 | |
| bauzas | andreykurilin: I just wonder when it was added in the tree | 09:11 |
| *** shoutm has joined #openstack-nova | 09:11 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: virtuozzo instance rescue mode support https://review.openstack.org/217679 | 09:12 |
| bauzas | andreykurilin: nvm, found it | 09:15 |
| bauzas | andreykurilin: https://github.com/openstack/python-novaclient/commit/a400b58d0344b50462cb25b84044f110a4a36d9a | 09:15 |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: virtuozzo instance rescue mode support https://review.openstack.org/217679 | 09:15 |
| *** kaisers1 has joined #openstack-nova | 09:16 | |
| *** zhangjn has quit IRC | 09:17 | |
| *** amotoki has joined #openstack-nova | 09:19 | |
| *** mnestratov has joined #openstack-nova | 09:19 | |
| *** kiran-r has joined #openstack-nova | 09:20 | |
| *** marzif has quit IRC | 09:22 | |
| *** takedakn has quit IRC | 09:24 | |
| *** VikasC has quit IRC | 09:28 | |
| *** amotoki has quit IRC | 09:30 | |
| *** venkatesh_ has joined #openstack-nova | 09:32 | |
| *** venkat_p has quit IRC | 09:32 | |
| *** dims has joined #openstack-nova | 09:32 | |
| *** shuquan has joined #openstack-nova | 09:35 | |
| *** dims has quit IRC | 09:37 | |
| *** jhesketh has quit IRC | 09:39 | |
| *** ociuhandu has quit IRC | 09:41 | |
| *** jhesketh has joined #openstack-nova | 09:41 | |
| *** karmatronic has joined #openstack-nova | 09:44 | |
| *** rotbeard has joined #openstack-nova | 09:44 | |
| *** karimb has quit IRC | 09:45 | |
| *** panbalag has joined #openstack-nova | 09:46 | |
| *** EinstCrazy has quit IRC | 09:46 | |
| *** salv-orlando has joined #openstack-nova | 09:50 | |
| *** alex_klimov has quit IRC | 09:54 | |
| *** alex_klimov has joined #openstack-nova | 09:54 | |
| *** gszasz has joined #openstack-nova | 09:54 | |
| *** alex_klimov has quit IRC | 09:54 | |
| *** alex_klimov has joined #openstack-nova | 09:55 | |
| *** shuquan has quit IRC | 09:55 | |
| *** sudipto has joined #openstack-nova | 09:56 | |
| *** RuiChen has quit IRC | 09:58 | |
| *** Vek has joined #openstack-nova | 09:58 | |
| *** kylek3h has joined #openstack-nova | 09:59 | |
| *** karimb has joined #openstack-nova | 10:00 | |
| *** karmatronic has quit IRC | 10:01 | |
| *** karimb has quit IRC | 10:01 | |
| *** paboldin has quit IRC | 10:01 | |
| *** karimb has joined #openstack-nova | 10:01 | |
| *** panbalag has left #openstack-nova | 10:03 | |
| *** kylek3h has quit IRC | 10:04 | |
| *** ociuhandu has joined #openstack-nova | 10:05 | |
| *** marzif has joined #openstack-nova | 10:06 | |
| *** amotoki has joined #openstack-nova | 10:16 | |
| *** karimb has quit IRC | 10:16 | |
| *** kiran-r has quit IRC | 10:19 | |
| *** shoutm has quit IRC | 10:19 | |
| *** zhangjn has joined #openstack-nova | 10:23 | |
| *** zhangjn has quit IRC | 10:26 | |
| *** dims has joined #openstack-nova | 10:26 | |
| *** vponomaryov has joined #openstack-nova | 10:28 | |
| *** alex_klimov has quit IRC | 10:31 | |
| *** dims has quit IRC | 10:31 | |
| *** sayali_ has quit IRC | 10:32 | |
| *** VikasC has joined #openstack-nova | 10:33 | |
| *** sayali has joined #openstack-nova | 10:34 | |
| *** qianli has quit IRC | 10:34 | |
| *** jichen has joined #openstack-nova | 10:36 | |
| *** shyu has quit IRC | 10:36 | |
| *** hparekh has quit IRC | 10:43 | |
| *** hparekh has joined #openstack-nova | 10:45 | |
| *** ZZelle has quit IRC | 10:48 | |
| *** ZZelle has joined #openstack-nova | 10:48 | |
| alexschm | sahid: so I think a solution to avoid the changes in serial.py would be to acquire serial ports in spawn() and store them in instance.system_metadata | 10:51 |
| *** venkatesh_ has quit IRC | 10:51 | |
| *** lyanchih has quit IRC | 10:51 | |
| *** venkatesh_ has joined #openstack-nova | 10:51 | |
| alexschm | sahid: is this what you have in mind? | 10:51 |
| *** dave-mccowan has joined #openstack-nova | 10:51 | |
| sahid | alexschm: not really - from what i have understand you have modified serial to make able to return same ports for an instance if we call "acquire" more than once, right ? | 10:53 |
| alexschm | sahid yes that's right | 10:53 |
| sahid | alexschm: ok and case of live migration we call acquire for a given instance on target host in pre_live_migration and when we spawn the instance on target, right ? | 10:55 |
| sahid | (just to be sure we are on a same way) | 10:56 |
| *** pixelb has joined #openstack-nova | 10:56 | |
| alexschm | sahid: yeah that's also right, and right now we also call acquire() in hard_reboot(), and multiple times during live migration | 10:56 |
| *** alex_klimov has joined #openstack-nova | 10:56 | |
| *** claudiub has joined #openstack-nova | 10:57 | |
| *** amotoki has quit IRC | 10:57 | |
| sahid | alexschm: but all of these acquires come fom the same caller | 10:58 |
| sahid | which is... | 10:58 |
| sahid | _get_guest_config | 10:58 |
| alexschm | right | 10:59 |
| *** takedakn has joined #openstack-nova | 11:01 | |
| sahid | alexschm: my thinking is to check whether domainxml is already configured with serial interfaces | 11:02 |
| sahid | that should take care of all other methods like hard_reboot... | 11:02 |
| sahid | i do not know/understand why we pass a more tha once in _get_guest_config but until we fix that i guess it is a reasonable way to fix problem | 11:02 |
| *** jgriffith has quit IRC | 11:02 | |
| *** eharney has quit IRC | 11:02 | |
| alexschm | sahid: no I think that's not possible, because _get_guest_config() creates the guest in the first place | 11:03 |
| alexschm | it's called during spawn() and the XML is generated from the result of _get_guest_config() | 11:03 |
| *** jgriffith has joined #openstack-nova | 11:03 | |
| *** eharney has joined #openstack-nova | 11:04 | |
| *** jgriffith is now known as Guest91834 | 11:04 | |
| *** amotoki has joined #openstack-nova | 11:04 | |
| sahid | alexschm: i do not say we should pass on this method. i just say we should add a condition in _create_serial_console_devices | 11:05 |
| *** promulo__ has quit IRC | 11:05 | |
| alexschm | sahid: but I don't see how _create_serial_console_devices() should find out if ports were acquired for this instance already or not... | 11:07 |
| sahid | alexschm: can _get_serial_ports_from_instance helps you? | 11:08 |
| *** amotoki has quit IRC | 11:10 | |
| alexschm | sahid _get_serial_ports_from_instance tries to use the XML to determine the serial ports... | 11:10 |
| alexschm | sahid: so maybe we could catch an exception to determine that the instance is not there yet and new ports need to be acquired | 11:10 |
| alexschm | sahid: similar to what _get_existing_domain_xml() does | 11:11 |
| sahid | alexschm: what is problem with using XML? | 11:12 |
| alexschm | sahid: no problem! | 11:13 |
| *** Anticimex has quit IRC | 11:13 | |
| *** baoli_ has joined #openstack-nova | 11:13 | |
| *** mwagner_lap has joined #openstack-nova | 11:15 | |
| *** mmedvede has joined #openstack-nova | 11:17 | |
| alexschm | sahid: how about this? http://paste.openstack.org/show/440878/ | 11:17 |
| sahid | alexschm: I prefer you change method _get_serial_ports_from_instance to pass guest object instead of instance | 11:19 |
| sahid | so that method will return an "empty" iterator | 11:20 |
| sahid | and so you can do you check without to wait for an exception | 11:20 |
| sahid | alexschm: sound good for you ? | 11:20 |
| *** dims has joined #openstack-nova | 11:21 | |
| *** kiran-r has joined #openstack-nova | 11:21 | |
| sahid | (that should be done in a separated commit) | 11:21 |
| *** chhavi has joined #openstack-nova | 11:23 | |
| *** Anticimex has joined #openstack-nova | 11:23 | |
| alexschm | sahid: so I would first change _get_serial_ports_from_instance() to get a guest object and then put the live migration fix ontop? | 11:23 |
| sahid | alexschm: yes that should be great | 11:24 |
| alexschm | sahid: ok thanks for your input! | 11:25 |
| *** dims has quit IRC | 11:26 | |
| *** lucasagomes is now known as lucas-hungry | 11:28 | |
| *** amotoki has joined #openstack-nova | 11:28 | |
| *** matrohon has quit IRC | 11:29 | |
| *** baoli has joined #openstack-nova | 11:31 | |
| *** pbrooko has quit IRC | 11:33 | |
| *** baoli_ has quit IRC | 11:33 | |
| *** VikasC has quit IRC | 11:33 | |
| andreykurilin | bauzas: hi! | 11:34 |
| andreykurilin | bauzas: sorry for late reply. did you found the answer for your question? | 11:36 |
| *** aix has quit IRC | 11:36 | |
| *** haomaiwang has quit IRC | 11:37 | |
| *** ctrath has joined #openstack-nova | 11:42 | |
| *** bapalm has quit IRC | 11:43 | |
| *** takedakn has quit IRC | 11:43 | |
| *** karimb has joined #openstack-nova | 11:44 | |
| *** rushil has joined #openstack-nova | 11:46 | |
| *** rushil_ has joined #openstack-nova | 11:47 | |
| *** kylek3h has joined #openstack-nova | 11:47 | |
| *** bapalm has joined #openstack-nova | 11:49 | |
| *** rushil has quit IRC | 11:50 | |
| *** thorst has joined #openstack-nova | 11:54 | |
| *** rfolco has joined #openstack-nova | 11:54 | |
| mkoderer | oomichi: ping | 11:59 |
| bauzas | andreykurilin: yup | 12:01 |
| bauzas | andreykurilin: I spotted the initial commit | 12:01 |
| bauzas | andreykurilin: it's just horribly named | 12:01 |
| *** mmedvede has quit IRC | 12:01 | |
| andreykurilin | :) | 12:01 |
| *** matrohon has joined #openstack-nova | 12:02 | |
| bauzas | andreykurilin: btw. do you know how the contrib CLI directory is stable ? | 12:02 |
| bauzas | I mean, could someone just modify the name of it? | 12:02 |
| bauzas | and then break the backwards compat ? | 12:03 |
| alex_xu | sdague: morning, we just left one API patch at here https://review.openstack.org/#/c/214601/ hope you can take a look at when you have time, thanks | 12:04 |
| andreykurilin | bauzas: unfortunately, changing the name will break some projects, because of direct imports:( but we can do the same as for novaclient.v1_1 module | 12:05 |
| *** mmedvede has joined #openstack-nova | 12:05 | |
| andreykurilin | bauzas: https://github.com/openstack/python-novaclient/blob/master/novaclient/v1_1/__init__.py | 12:05 |
| *** MarcF has quit IRC | 12:07 | |
| *** jaypipes has joined #openstack-nova | 12:08 | |
| andreykurilin | bauzas: also, I prefer to wait new release of novaclient and remove all direct imports and rewrite to use disciver_extension https://github.com/openstack/python-novaclient/blob/master/novaclient/client.py#L717 instead, where we can implement deprecation and renaming mechanism(maybe we can do implement it before new release ):) | 12:08 |
| *** deepthi has quit IRC | 12:08 | |
| alex_xu | sdague: thanks :) | 12:08 |
| bauzas | andreykurilin: on a call | 12:09 |
| *** aix has joined #openstack-nova | 12:09 | |
| jaypipes | morning all | 12:09 |
| openstackgerrit | Noel Nelson Dsouza proposed openstack/nova: Delete flavor should raise error, if it is inuse https://review.openstack.org/217674 | 12:09 |
| *** mwagner_lap has quit IRC | 12:10 | |
| andreykurilin | jaypipes: good morning:) | 12:10 |
| *** mwagner_lap has joined #openstack-nova | 12:10 | |
| *** irenab has quit IRC | 12:11 | |
| alex_xu | jaypipes: good morning | 12:11 |
| jaypipes | morning guys :) | 12:11 |
| *** jcoufal has joined #openstack-nova | 12:11 | |
| *** dprince has joined #openstack-nova | 12:13 | |
| *** dprince has left #openstack-nova | 12:16 | |
| mkoderer | jaypipes: hi ;) we have an issue with v2.1 .. https://bugs.launchpad.net/nova/+bug/1491325 | 12:17 |
| openstack | Launchpad bug 1491325 in OpenStack Compute (nova) "nova api v2.1 does not allow to use autodetection of volume device path" [Undecided,New] | 12:17 |
| * jaypipes looks | 12:17 | |
| mkoderer | bug in Manila: https://bugs.launchpad.net/manila/+bug/1491262 | 12:17 |
| openstack | Launchpad bug 1491262 in Manila "Gate jobs failing with InvalidInput exception" [Critical,New] | 12:17 |
| mkoderer | can we extend the regex and allow "auto"? | 12:18 |
| *** aysyd has joined #openstack-nova | 12:18 | |
| *** vladikr has joined #openstack-nova | 12:19 | |
| *** shaohe_feng has joined #openstack-nova | 12:22 | |
| *** MarcF has joined #openstack-nova | 12:22 | |
| *** jcoufal has quit IRC | 12:22 | |
| garyk | mriedem_away: please see https://review.openstack.org/219225 | 12:22 |
| vponomaryov | mkoderer: we also can omit key "device" in novaclient | 12:25 |
| *** bkopilov has quit IRC | 12:25 | |
| vponomaryov | mkoderer: but in this case we will be forced to use only latest release of novaclient | 12:25 |
| vponomaryov | mkoderer: with this fix | 12:26 |
| zigo | Could someone help me backporting the patches for CVE-2015-3241 to Icehouse? | 12:26 |
| openstackgerrit | Feodor Tersin proposed openstack/nova: Remove unused root_metadata method of BlockDeviceMappingList https://review.openstack.org/212324 | 12:26 |
| *** bkopilov has joined #openstack-nova | 12:26 | |
| zigo | My backport broke all unit tests... :( | 12:26 |
| vponomaryov | mkoderer: so, I agree, allowing 'auto' for compatibility with v2.0 - best solution | 12:26 |
| *** karimb has quit IRC | 12:26 | |
| *** rushiagr is now known as rushiagr_away | 12:27 | |
| *** david-lyle has quit IRC | 12:27 | |
| *** mordred has quit IRC | 12:27 | |
| *** dansmith has quit IRC | 12:28 | |
| *** david-lyle has joined #openstack-nova | 12:28 | |
| *** raildo-afk is now known as raildo | 12:29 | |
| *** ctrath has quit IRC | 12:29 | |
| openstackgerrit | Mikhail Feoktistov proposed openstack/nova: libvirt: virtuozzo instance rescue mode support https://review.openstack.org/217679 | 12:29 |
| *** ftersin has joined #openstack-nova | 12:29 | |
| *** rushil_ has quit IRC | 12:31 | |
| *** venkatesh_ has quit IRC | 12:31 | |
| *** venkatesh_ has joined #openstack-nova | 12:32 | |
| *** rushil has joined #openstack-nova | 12:32 | |
| *** moshele has joined #openstack-nova | 12:32 | |
| *** lucas-hungry is now known as lucasagomes | 12:32 | |
| *** dansmith has joined #openstack-nova | 12:33 | |
| *** dansmith is now known as Guest91339 | 12:33 | |
| *** chhavi has quit IRC | 12:36 | |
| *** rushil has quit IRC | 12:37 | |
| *** claudiub has quit IRC | 12:38 | |
| *** irenab has joined #openstack-nova | 12:40 | |
| *** _kiran_ has joined #openstack-nova | 12:41 | |
| *** liusheng has quit IRC | 12:42 | |
| *** kiran-r has quit IRC | 12:42 | |
| *** _kiran_ is now known as kiran-r | 12:43 | |
| *** venkatesh_ has quit IRC | 12:44 | |
| *** mpaolino_ has joined #openstack-nova | 12:44 | |
| *** mpaolino has quit IRC | 12:45 | |
| *** shaohe_feng has quit IRC | 12:45 | |
| *** rlrossit has joined #openstack-nova | 12:45 | |
| *** takedakn has joined #openstack-nova | 12:45 | |
| *** shaohe_feng has joined #openstack-nova | 12:45 | |
| *** takedakn has quit IRC | 12:47 | |
| *** zhangjn has joined #openstack-nova | 12:47 | |
| *** dims has joined #openstack-nova | 12:48 | |
| bauzas | jaypipes: morning, on a call but if you have time to review https://review.openstack.org/#/c/219182/1 it would be greatly appreciated ;) | 12:50 |
| *** annegentle has joined #openstack-nova | 12:51 | |
| jaypipes | bauzas: ah, this is the thing that alex_xu caught.. yeah. gimme a few. | 12:52 |
| bauzas | jaypipes: yep | 12:52 |
| *** shaohe_feng has quit IRC | 12:52 | |
| *** zhangjn has quit IRC | 12:54 | |
| *** mdrabe has joined #openstack-nova | 12:55 | |
| *** diogogmt has joined #openstack-nova | 12:56 | |
| *** mpaolino_ has quit IRC | 12:56 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add display-flavor-id-in-migration.rst https://review.openstack.org/219431 | 12:58 |
| *** ccarmack has joined #openstack-nova | 12:58 | |
| *** diogogmt has quit IRC | 12:59 | |
| *** mpaolino has joined #openstack-nova | 13:00 | |
| *** ildikov has quit IRC | 13:00 | |
| *** chhavi has joined #openstack-nova | 13:01 | |
| *** mwagner_lap has quit IRC | 13:01 | |
| *** mpaolino has quit IRC | 13:01 | |
| *** jcoufal has joined #openstack-nova | 13:02 | |
| *** kiran-r has quit IRC | 13:02 | |
| *** mpaolino has joined #openstack-nova | 13:02 | |
| *** diogogmt has joined #openstack-nova | 13:04 | |
| *** zhangjn has joined #openstack-nova | 13:04 | |
| *** zhangjn has quit IRC | 13:04 | |
| *** vivekd has quit IRC | 13:04 | |
| *** zhangjn has joined #openstack-nova | 13:05 | |
| *** pixelb has quit IRC | 13:06 | |
| *** vivekd has joined #openstack-nova | 13:06 | |
| *** pixelb has joined #openstack-nova | 13:08 | |
| *** ijuwang has joined #openstack-nova | 13:08 | |
| *** diogogmt has quit IRC | 13:08 | |
| *** lyanchih has joined #openstack-nova | 13:11 | |
| openstackgerrit | sahid proposed openstack/nova-specs: virt: dedicated CPUs placement policy https://review.openstack.org/219692 | 13:11 |
| *** ildikov has joined #openstack-nova | 13:12 | |
| zhenguo | jaypipes: can you please have a look at this https://review.openstack.org/#/c/204088/ when you have time? | 13:13 |
| jaypipes | bauzas: reviewed. | 13:13 |
| bauzas | jaypipes: ack, will rebase shortly then | 13:15 |
| *** mpaolino has quit IRC | 13:16 | |
| ftersin | bauzas, alaski, ndipanov: hi. a question about disk filter for volume backed instances | 13:16 |
| *** rushiagr_away is now known as rushiagr | 13:17 | |
| ndipanov | ftersin, probably broken? | 13:17 |
| *** ishant has quit IRC | 13:17 | |
| ftersin | known issue https://review.openstack.org/#/c/194059/2 https://review.openstack.org/#/c/184982/ | 13:17 |
| *** links has quit IRC | 13:17 | |
| *** dims_ has joined #openstack-nova | 13:17 | |
| ftersin | as i've understood you prefere to null out flavor.root_gb instead of analyze of instance type from inside of disk filter | 13:18 |
| ftersin | even if is_volume_backed_instance is available from inside disk filter | 13:18 |
| ftersin | is it correct? | 13:18 |
| bauzas | ftersin: I'm violently opposed to add a key in filt_props that would increase tech debt and would make difficult to be removed | 13:19 |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: [WIP] Allow to use autodetection of volume device path https://review.openstack.org/219696 | 13:19 |
| bauzas | jaypipes: yeah, your commit is good :) | 13:19 |
| ndipanov | ftersin, well it wasn't available easily until recently | 13:19 |
| ndipanov | because you need objects for that | 13:19 |
| ndipanov | ftersin, and we want to avoid db hits in filters | 13:19 |
| ftersin | we can move is_volume_backed_instance to Instance object | 13:19 |
| ndipanov | which I think unless we are passing block device mapping to filters | 13:19 |
| ndipanov | which we are not but we could be | 13:20 |
| ndipanov | is not a thing we can do easily | 13:20 |
| ndipanov | so there is some work to be done there | 13:20 |
| ndipanov | (I think) | 13:20 |
| ndipanov | jaypipes, dan had one issue with this so I had to rework it - mind +wing? https://review.openstack.org/#/c/216733/ | 13:21 |
| *** dims has quit IRC | 13:21 | |
| *** rushiagr is now known as rushiagr_away | 13:21 | |
| ftersin | i agree that db hits are not desired in filters, so it should be done from outside there | 13:21 |
| jaypipes | ndipanov: sure, as soon as I finish zhenguo's review. | 13:21 |
| ndipanov | jaypipes, awesome! | 13:22 |
| ndipanov | ftersin we should be sending block device mapping along with the instance to the scheduler | 13:22 |
| ndipanov | there is no such thing now I think | 13:23 |
| alaski | ftersin: my view is that I don't think the scheduler/resource tracker should make decisions about what resources are consumed differently for a boot from volume instance. | 13:23 |
| *** pixelb has quit IRC | 13:23 | |
| ndipanov | alaski, not sure what you're saying... | 13:23 |
| alaski | I think that info should be correctly passed in. root_gb should equal 0 if no local disk is used | 13:23 |
| *** pixelb has joined #openstack-nova | 13:23 | |
| ndipanov | alaski, I guess | 13:23 |
| alex_xu | bauzas: hi, I added comment at https://review.openstack.org/#/c/197912/26/nova/scheduler/filters/__init__.py | 13:24 |
| ftersin | what is an appropriate place to set root_gb=0? i found is build_request_spec. is it good one? | 13:24 |
| *** vivekd_ has joined #openstack-nova | 13:25 | |
| alex_xu | bauzas: it's about help people migrate their out-of-tree filters | 13:25 |
| *** mriedem_away is now known as mriedem | 13:25 | |
| bauzas | alex_xu: yeah, but we discussed that and we can't leave something only for out-of-tree usage | 13:25 |
| *** mwagner_lap has joined #openstack-nova | 13:25 | |
| *** karimb has joined #openstack-nova | 13:25 | |
| *** vivekd has quit IRC | 13:25 | |
| alaski | ftersin: it should be set when the instance is created. but if it's necessary to have root_gb be non zero for some reason then perhaps we need another resource field to pass in | 13:25 |
| bauzas | alex_xu: so I can provide an UpgradeImpact tag but that's it | 13:25 |
| *** vivekd_ is now known as vivekd | 13:26 | |
| alex_xu | bauzas: ok, cool, if we already have agreement here, so I'm ok with it :) | 13:26 |
| bauzas | ftersin: +1 to alaski, like I explained in the second change | 13:26 |
| *** shuquan has joined #openstack-nova | 13:26 | |
| bauzas | ftersin: if root_gb is wrong, then we have to fix it | 13:26 |
| alaski | I'm taking the simple view of passing in everything that the scheduler/resource tracker should consume, not passing in some stuff and saying "figure out what you should consume based on outside factors" | 13:28 |
| ftersin | alaski, bauzas: we talk about flavor.root_gb. flavor can be loaded from db in build_request_spec. | 13:28 |
| ftersin | alaski: if you mean instance.root_gb, we can set it to 0, but this is another deal | 13:29 |
| mriedem | garyk: so what is still using suds? oslo.vmware? | 13:29 |
| *** zul has quit IRC | 13:29 | |
| garyk | mriedem: oslo.vmware | 13:30 |
| *** VW_ has joined #openstack-nova | 13:30 | |
| garyk | it was never dropped even though there was/is talk about changing to pyvmomi (but that would require a rewrite of the nova driver - and you know what that entails...) | 13:31 |
| alaski | ftersin: I just mean generally. I don't know what the right solution is here, just that we should pass in correct data and not care about bfv in the scheduler | 13:31 |
| *** MarcF has quit IRC | 13:31 | |
| *** zul has joined #openstack-nova | 13:32 | |
| mriedem | danpb: ndipanov: ftersin: i could use some help in figuring out why test_stamp_pattern is failing here - it almost seems like a libvirt thing but i'm not sure https://review.openstack.org/#/c/218355/ | 13:33 |
| mriedem | it attaches a volume created from a volume snapshot, attaches at /dev/vdb but the vdb partition doesn't show up on the instance | 13:34 |
| mriedem | http://logs.openstack.org/55/218355/5/check/gate-tempest-dsvm-neutron-full/4dce003/console.html#_2015-09-01_19_37_24_955 | 13:34 |
| mriedem | it also appears to be racy, sometimes works and sometimes doesn't | 13:34 |
| *** chhavi has quit IRC | 13:35 | |
| *** rdopiera has quit IRC | 13:35 | |
| *** dustins has joined #openstack-nova | 13:36 | |
| *** amotoki has quit IRC | 13:36 | |
| *** shuquan has quit IRC | 13:37 | |
| danpb | mriedem: hmm, the assertion error there only shows a vda and sr0 | 13:38 |
| mriedem | yeah | 13:38 |
| danpb | mriedem: the vda will be the original root disk and sr0 will be the cdrom | 13:38 |
| mriedem | it's checking to see that vdb is in the partition list | 13:38 |
| openstackgerrit | Ludovic Beliveau proposed openstack/nova: Allow saving empty pci_device_pools in ComputeNode object https://review.openstack.org/216049 | 13:38 |
| danpb | so seems like perhaps the guest hasn't finished hotplug yet | 13:38 |
| mriedem | danpb: this is the instance domain xml on spawn | 13:39 |
| mriedem | http://logs.openstack.org/55/218355/5/check/gate-tempest-dsvm-neutron-full/4dce003/logs/screen-n-cpu.txt.gz#_2015-09-01_19_36_10_538 | 13:39 |
| danpb | is there anyway to capture dmesg from the guest in question ? | 13:39 |
| mriedem | this is where we start attaching the volume http://logs.openstack.org/55/218355/5/check/gate-tempest-dsvm-neutron-full/4dce003/logs/screen-n-cpu.txt.gz#_2015-09-01_19_36_16_616 | 13:39 |
| mriedem | danpb: like ssh into the guest and get that? | 13:39 |
| *** smatzek has joined #openstack-nova | 13:40 | |
| danpb | we see the XML for the attachment just a little further http://logs.openstack.org/55/218355/5/check/gate-tempest-dsvm-neutron-full/4dce003/logs/screen-n-cpu.txt.gz#_2015-09-01_19_36_19_972 | 13:40 |
| *** edmondsw has joined #openstack-nova | 13:41 | |
| danpb | so we can see we've asked libvirt to attach the volume as vdb | 13:41 |
| danpb | i wonder how long tempest is waiting for the device to appear ? | 13:41 |
| danpb | maybe it simply isn't long enough | 13:41 |
| mriedem | i actually removed the timeout | 13:41 |
| mriedem | because it was timing out at like 5 minutes | 13:41 |
| jaypipes | zhenguo: review done. | 13:42 |
| danpb | mriedem:5 minutes for the whole test, or 5 minutes waiting for it to cmplete attachment | 13:42 |
| jaypipes | ndipanov: +w'd | 13:42 |
| mriedem | for the partition to show up on the guest | 13:42 |
| danpb | oh wow, that' sway too long | 13:42 |
| zhenguo | jaypipes: thanks | 13:42 |
| mriedem | danpb: well, 196 seocnds | 13:43 |
| mriedem | *seconds | 13:43 |
| jaypipes | zhenguo: np | 13:43 |
| mriedem | still over 3 minutes | 13:43 |
| danpb | yeah, sssh into the guest and get the dmesg output | 13:43 |
| mriedem | ok | 13:43 |
| danpb | and also 'lspci' output woul dbe nice | 13:43 |
| danpb | so we see if the virtio pci device has appeared | 13:43 |
| *** annegentle has quit IRC | 13:43 | |
| mriedem | danpb: do you think we still need the loop waiting for the partition to show up? i figured it'd be there once the attach was done | 13:43 |
| ndipanov | jaypipes, awesome | 13:44 |
| mriedem | i.e. once the cinder volume went to in-use | 13:44 |
| *** MarcF has joined #openstack-nova | 13:44 | |
| danpb | mriedem: no, it is asynchronous | 13:45 |
| mriedem | ok, i'll add the loop/wait back into the test | 13:45 |
| danpb | when libvirt returns control it merely means the guest device model sees the new device | 13:45 |
| *** dbite has quit IRC | 13:45 | |
| danpb | the guest operating system will process the hotplug event asynchronously | 13:45 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests() https://review.openstack.org/199205 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix RequestSpec.instance_group hydratation https://review.openstack.org/219182 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler https://review.openstack.org/188487 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Prepare filters for using RequestSpec object https://review.openstack.org/197912 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify left filters for RequestSpec https://review.openstack.org/202678 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object https://review.openstack.org/203078 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify metric-related filters for RequestSpec https://review.openstack.org/202676 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify NUMA, PCI and num_instances filters for RequestSpec https://review.openstack.org/202677 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler https://review.openstack.org/199051 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Aggregate filters for RequestSpec https://review.openstack.org/202674 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify default filters for RequestSpec https://review.openstack.org/202675 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Affinity filter for RequestSpec https://review.openstack.org/202673 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in HostManager https://review.openstack.org/191251 | 13:47 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Remove reservation_id from the logs when a schedule fails https://review.openstack.org/217798 | 13:47 |
| Guest91339 | jaypipes: passing jenkins: https://review.openstack.org/#/c/217819/ :) | 13:48 |
| bauzas | Guest91339: heh, dansmith hacked | 13:48 |
| *** dguitarbite has joined #openstack-nova | 13:49 | |
| jaypipes | Guest91339: nice :) | 13:49 |
| mriedem | danpb: it'd be nice if we passed the instance or instance_uuid to the libvirt volume driver's connect_volume method so we could actually log the instance in context for http://logs.openstack.org/55/218355/5/check/gate-tempest-dsvm-neutron-full/4dce003/logs/screen-n-cpu.txt.gz#_2015-09-01_19_36_19_972 | 13:49 |
| *** chhavi has joined #openstack-nova | 13:49 | |
| bauzas | jaypipes: your wishes are now reality https://review.openstack.org/#/c/219182/ | 13:49 |
| mriedem | when i'm searching the n-cpu logs by instance uuid i missed that | 13:49 |
| mriedem | same with LibvirtConfigObject.to_xml | 13:50 |
| mriedem | or just move that | 13:50 |
| Guest91339 | doh | 13:51 |
| *** Guest91339 is now known as dansmith | 13:51 | |
| *** dansmith is now known as Guest36431 | 13:51 | |
| bauzas | dansmith/Guest91339: self.get_dansmith_thoughts(https://review.openstack.org/#/c/219182/2) | 13:51 |
| bauzas | hah | 13:52 |
| bauzas | Guest36431: self.get_dansmith_thoughts(https://review.openstack.org/#/c/219182/2) could be appreciated because of something we discussed yesterday (or the day before) | 13:52 |
| danpb | mriedem: its a shame we can't just setup a thread-local in the manager.py whenever we have an instance | 13:52 |
| danpb | mriedem: and then have *every* log message from that thread include the instance | 13:53 |
| danpb | having to pass around instance just for log messages sucks | 13:53 |
| danpb | because we'll forever be finding places we forgot to log it | 13:53 |
| mriedem | yeah, i can always get around this by logging after the call to connect_volume with the results | 13:53 |
| danpb | in fact this kind of thing must be possible with a thread local | 13:53 |
| danpb | as we already do that to get the request logged | 13:54 |
| mriedem | i kind of want config.to_xml() to not log the xml str on debug since that appears to be duplicated logging in a lot of places | 13:54 |
| *** karimb has quit IRC | 13:54 | |
| zhenguo | jaypipes: thanks for the review, but I don't quite understand why we don't need to get infomation in the properties collections ever? | 13:54 |
| jaypipes | zhenguo: exactly... I don't see why we ever need the properties collection for resource amounts. | 13:55 |
| ftersin | alaski: do you think that saving instance.flavor.root_gb=0 in DB is more right and straightforward than setting instance_type.root_gb=0 for request_spec in build_request_spec? | 13:55 |
| zhenguo | jaypipes: properties notion of resources will be used if the node is not being used and is not unavailable. | 13:55 |
| Guest36431 | bauzas: I will look while I wait for nickserv to catch up | 13:55 |
| bauzas | Guest36431: this time, it's not me | 13:56 |
| Guest36431 | heh | 13:56 |
| jaypipes | zhenguo: if the node is not being used and is not available, what use does Nova have for it? | 13:56 |
| zhenguo | jaypipes: the node is not used, means is not deployed by nova | 13:56 |
| *** Guest36431 is now known as dansmith | 13:56 | |
| zhenguo | jaypiples: we need ther properties resources for scheduling | 13:56 |
| jaypipes | zhenguo: right, so why should the node even be included in resource calculations by nova? | 13:56 |
| *** haomaiwang has joined #openstack-nova | 13:57 | |
| zhenguo | jaypipes: we need to collect the unused node resources, then shceduler will know which one can be deployed. | 13:58 |
| jaypipes | zhenguo: oh, you are saying that a node might have its resource amounts changed (e.g. RAM added) after an instanec is deployed on it, and if that node is deleted and reused for different instance, that is when the properties collection will come into play? | 13:58 |
| *** lbeliveau has joined #openstack-nova | 13:58 | |
| zhenguo | jaypipes: yeah. | 13:59 |
| *** thangp has joined #openstack-nova | 13:59 | |
| *** mriedem is now known as mriedem_meeting | 13:59 | |
| jaypipes | zhenguo: got it. understand now... ok, lemme approve this. thx for discussing. | 13:59 |
| zhenguo | jaypipes: thanks :) | 13:59 |
| openstackgerrit | Zhongcheng Lao proposed openstack/nova: VMware: Omit exceptions during snapshot cleanup https://review.openstack.org/216086 | 13:59 |
| bauzas | jaypipes: yeah, just reviewed zhenguo's change and I think it's okay | 14:00 |
| bauzas | jaypipes: if an instance is selected, we need to consume all the resources | 14:00 |
| zhenguo | bauzas: thanks for the review. | 14:00 |
| bauzas | zhenguo: yet some nits to look at | 14:01 |
| *** markvoelker has joined #openstack-nova | 14:01 | |
| alaski | ftersin: I would think they should match, but I think there are reasons we can't set root_gb = 0 in the db | 14:02 |
| alaski | ftersin: it needs some thought. but I think something like that is a better route than adding bdms to the request spec | 14:03 |
| *** burgerk has joined #openstack-nova | 14:04 | |
| ftersin | alaski: agree with that, and i also think that having changed flavor in db can lead to some undesirable effects, like customer questions :) | 14:05 |
| *** mordred has joined #openstack-nova | 14:05 | |
| *** marzif has quit IRC | 14:05 | |
| *** markvoelker has quit IRC | 14:06 | |
| mordred | hey all - I've got some functional tests failing consistently in shade now against nova-net devstack ... looks like maybe a behavior change, so I wanted to check and see if it's intended or accidental | 14:06 |
| *** marzif has joined #openstack-nova | 14:06 | |
| mordred | http://logs.openstack.org/55/219655/1/check/gate-shade-dsvm-functional-nova/1154770/console.html#_2015-09-02_12_10_56_071 | 14:06 |
| mordred | is this a known issue? or a new behavior for nova? (like, do I need to be passing in an ID and not he actual IP address?) | 14:07 |
| *** burt has joined #openstack-nova | 14:07 | |
| *** geekinutah has joined #openstack-nova | 14:07 | |
| *** kiran-r has joined #openstack-nova | 14:09 | |
| *** vivekd_ has joined #openstack-nova | 14:09 | |
| *** _kiran_ has joined #openstack-nova | 14:09 | |
| *** _kiran_ has quit IRC | 14:10 | |
| *** _kiran_ has joined #openstack-nova | 14:11 | |
| *** vivekd has quit IRC | 14:12 | |
| *** vivekd_ is now known as vivekd | 14:12 | |
| bauzas | dansmith: call me a fool, but do you feel that it would be difficult to remove the dict compat for InstanceV2 ? | 14:13 |
| *** Piet has quit IRC | 14:13 | |
| dansmith | bauzas: yes, and it's not part of the api, so it doesn't have anything to do with the versioning | 14:13 |
| *** kiran-r has quit IRC | 14:13 | |
| bauzas | dansmith: I haven't yet checked the callers, hence the question | 14:13 |
| *** karimb has joined #openstack-nova | 14:13 | |
| dansmith | bauzas: sudipto is working through those conversions | 14:14 |
| bauzas | dansmith: ack, good to konw | 14:14 |
| bauzas | dansmith: okay, fair enough | 14:14 |
| ftersin | alaski: i know a case when having instance.flavor.root_gb=0 doesn't help - it's resize operation | 14:14 |
| *** annegentle has joined #openstack-nova | 14:14 | |
| sudipto | bauzas, I have had a hard time with a few objects. | 14:16 |
| *** raildo is now known as raildo-afk | 14:16 | |
| ftersin | alaski: in that case nova passes in a custom flavor | 14:16 |
| markus_z | mordred: IIUC it should be the ID of the floating ip: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/floating_ips.py#L153 | 14:16 |
| *** raildo-afk is now known as raildo | 14:17 | |
| mordred | markus_z: ok. so the fact that ip has worked up until now is just an accident | 14:17 |
| mordred | I can work with that | 14:17 |
| mordred | markus_z: thanks! | 14:17 |
| markus_z | mordred: How did you call the API? Via the client? | 14:17 |
| *** _kiran_ has quit IRC | 14:18 | |
| mordred | markus_z: yah. novaclient | 14:18 |
| *** kiran-r has joined #openstack-nova | 14:18 | |
| markus_z | mordred: The novaclient does/did probably a translation of address to id | 14:18 |
| markus_z | mordred: http://docs.openstack.org/cli-reference/content/novaclient_commands.html#novaclient_subcommand_floating-ip-delete | 14:18 |
| mordred | nova_client.floating_ips.delete('x.x.x.x') | 14:18 |
| mordred | markus_z: sory - I was unclear - python client, not command line client | 14:19 |
| markus_z | mordred: Hmm... could be an issue | 14:20 |
| *** tojuvone has quit IRC | 14:20 | |
| mordred | markus_z: floating_ips.get('x.x.x.x') should get me the floating ip object, right? | 14:21 |
| alaski | ftersin: part of this issue is that there's a lot of handwaving with flavors and booting from volume. we need a clear model for what it means when a flavor specifies a root_gb but we've booted from volume. | 14:21 |
| markus_z | mordred: At least I would expect it. What does happen? | 14:22 |
| bauzas | dansmith: do you have 20 mins for clarifying https://review.openstack.org/#/c/219182/2/nova/objects/instance_group.py,cm L138 ? | 14:22 |
| bauzas | oops | 14:22 |
| bauzas | 2 mins | 14:22 |
| bauzas | lol | 14:22 |
| *** haomaiwang has quit IRC | 14:23 | |
| mordred | markus_z: haven't tried get yet - all o the code we currently have just passed the ip address to delete - so I'm thinking now I want to get first and then pass the id to delete | 14:23 |
| sudipto | dansmith, when you have some time: https://review.openstack.org/#/c/201282/ | 14:23 |
| *** bswartz has joined #openstack-nova | 14:23 | |
| dansmith | sudipto: yeah, I know, but those can continue to go in after tomorrow, but other stuff can't | 14:23 |
| bswartz | dims_: ping | 14:23 |
| dansmith | sudipto: so, have to allocate time, you know | 14:23 |
| dansmith | bauzas: sure, what about it? | 14:23 |
| bauzas | dansmith: I tried to understand but I'm still unclear | 14:24 |
| markus_z | mordred: http://docs.openstack.org/developer/python-novaclient/ref/v2/floating_ips.html | 14:24 |
| sudipto | dansmith, sure, np - sorry for asking, just don't remember the dates well. | 14:24 |
| *** zhenguo has quit IRC | 14:24 | |
| markus_z | mordred: That's all I could find. | 14:24 |
| mordred | markus_z: yah. me too | 14:24 |
| bauzas | dansmith: so, you mean it's good to raise something if passed by ? | 14:24 |
| mordred | markus_z: oh well - I'll just poke and prod for a while - thanks! | 14:25 |
| *** haomaiwa_ has joined #openstack-nova | 14:25 | |
| markus_z | mordred: sure, you're welcome | 14:25 |
| dansmith | bauzas: yeah, I think you should fail if someone tries to set .hosts and then .save() them | 14:25 |
| *** ducnc has joined #openstack-nova | 14:25 | |
| bauzas | dansmith: so he would have to unset the field first ? | 14:25 |
| bauzas | dansmith: like making the contract explicit ? | 14:26 |
| dansmith | bauzas: no, you reset the field when you load it, so it would only be in updates if someone changes that field | 14:26 |
| *** Guest91834 is now known as jgriffith | 14:26 | |
| *** dustins_ has joined #openstack-nova | 14:26 | |
| dansmith | bauzas: which is what you want -- signaling to the dev that the save isn't going to save it | 14:26 |
| dansmith | just ignoring it is more obscure | 14:26 |
| *** mdrabe has quit IRC | 14:27 | |
| *** lyanchih has quit IRC | 14:27 | |
| dansmith | if someone comes along and thinks they can change that field and .save() it, then you want to make it clear that it's not going to happen, not just silently ignore so that they think it worked | 14:27 |
| bauzas | dansmith: ah I see | 14:27 |
| bauzas | dansmith: I'm saying that because I actually set the field by hand when hydrating the object | 14:27 |
| dansmith | bauzas: yep, I know, which is why I said you should reset there :) | 14:27 |
| *** Murali has quit IRC | 14:27 | |
| bauzas | dansmith: but that makes sense to do that way yeah | 14:28 |
| bauzas | dansmith: because it would force to get the proper values if got | 14:28 |
| *** dustins has quit IRC | 14:28 | |
| bauzas | dansmith: I thnik I get it, lemme do some stuff | 14:28 |
| *** tpeoples has joined #openstack-nova | 14:28 | |
| dansmith | okay, but no need to alter that patch, just slap something else on the end or something | 14:28 |
| *** mtanino has joined #openstack-nova | 14:30 | |
| *** kiran-r has quit IRC | 14:30 | |
| bauzas | dansmith: well, I'm fine, most of the series is still queued in the pipeline | 14:30 |
| bauzas | dansmith: and I prefer nice code for the beginning | 14:31 |
| bauzas | :) | 14:31 |
| dansmith | well, your call I guess | 14:31 |
| *** afazekas__ has quit IRC | 14:31 | |
| ftersin | alaski: i think that a clear model is... a usage of bdm (but obviously not to check a booting type) | 14:34 |
| ftersin | alaski: i mean that bdm list contains all info about instance usage of host local disk | 14:35 |
| *** eharney has quit IRC | 14:36 | |
| *** masayukig has quit IRC | 14:36 | |
| ftersin | alaski: it contain root disk, ephemerals, swap... we can just calculate the usage and compare it with host capability | 14:37 |
| *** electrocucaracha has joined #openstack-nova | 14:37 | |
| *** ducnc has quit IRC | 14:38 | |
| bswartz | Does anyone know what's going on with this bug? https://bugs.launchpad.net/nova/+bug/1491325 | 14:39 |
| openstack | Launchpad bug 1491325 in OpenStack Compute (nova) "nova api v2.1 does not allow to use autodetection of volume device path" [Undecided,In progress] - Assigned to Davanum Srinivas (DIMS) (dims-v) | 14:39 |
| ftersin | alaski: an additional benefit we can have with that approach is that a user can request an ephemeral disk smaller that a selected flavor allows | 14:39 |
| bswartz | This bug is breaking the manila gate tests | 14:39 |
| *** ducnc has joined #openstack-nova | 14:39 | |
| bswartz | There is a fix for it by dims_ but it's marked [WIP] | 14:39 |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: Hyper-V: Adds Hyper-V UEFI Secure Boot https://review.openstack.org/209581 | 14:40 |
| openstackgerrit | Simona Iuliana Toader proposed openstack/nova: Hyper-V vTPM Devices https://review.openstack.org/218642 | 14:40 |
| jroll | mtreinish: I got the job changed up for full tempest and checked experimental again on https://review.openstack.org/219225 | 14:40 |
| jroll | mtreinish: I'm in meetings much of today but will try to look this afternoon sometime | 14:41 |
| *** moshele has quit IRC | 14:41 | |
| ftersin | alaski: now in that case resource consumption is estimated by flavor.ephemeral_gb, but in fact it differs from that | 14:41 |
| dhellmann | johnthetubaguy: if you ack this client release request with a +1 I can do it today: https://review.openstack.org/#/c/219283/ | 14:41 |
| mtreinish | jroll: ok cool, I'll keep an eye out for the results | 14:41 |
| jroll | thanks sir | 14:41 |
| *** ducnc has quit IRC | 14:42 | |
| *** alexpilotti has joined #openstack-nova | 14:42 | |
| johnthetubaguy | dhellmann: acked that from me, sorry for the delay | 14:42 |
| *** markvoelker has joined #openstack-nova | 14:42 | |
| dhellmann | johnthetubaguy: not a problem on my end, I just wanted to make sure you were ok with the release so I didn't surprise you :-) | 14:42 |
| openstackgerrit | Balazs Gibizer proposed openstack/nova: Fix the help text of monkey_patch config param https://review.openstack.org/219722 | 14:42 |
| *** markvoelker has quit IRC | 14:42 | |
| *** diogogmt has joined #openstack-nova | 14:42 | |
| johnthetubaguy | dhellmann: all good, yeah, I should probably find a way to make mriedem_meeting able to do that without my ack | 14:43 |
| *** markvoelker has joined #openstack-nova | 14:43 | |
| *** masayukig has joined #openstack-nova | 14:43 | |
| *** browne has joined #openstack-nova | 14:43 | |
| alaski | ftersin: I agree with that. I'm not sure bdm is the right thing to pass in, but flavor isn't really either. we want something that indicates actual usage, however that is determined | 14:43 |
| *** mdrabe has joined #openstack-nova | 14:43 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add display-flavor-id-in-migration.rst https://review.openstack.org/219431 | 14:44 |
| dhellmann | johnthetubaguy: list him as the release liaison on https://wiki.openstack.org/wiki/CrossProjectLiaisons | 14:44 |
| *** zhangjn has quit IRC | 14:45 | |
| *** artom has joined #openstack-nova | 14:45 | |
| *** zhangjn has joined #openstack-nova | 14:45 | |
| *** zhangjn has quit IRC | 14:45 | |
| *** ildikov has quit IRC | 14:46 | |
| johnthetubaguy | dhellmann: right, that would make sense, I should put the old nova-release team there I guess | 14:46 |
| *** zhangjn has joined #openstack-nova | 14:46 | |
| mtreinish | jroll: someone else triggered experimental on a patch, the job is running: https://jenkins04.openstack.org/job/gate-tempest-dsvm-ironic-pxe_ipa-full-nv/1/console | 14:46 |
| dansmith | bauzas: found another spelling bug in that bottom patch | 14:46 |
| jroll | mtreinish: awesome | 14:46 |
| *** cfriesen has joined #openstack-nova | 14:46 | |
| *** zhangjn has quit IRC | 14:46 | |
| dhellmann | johnthetubaguy: sure, if you have more than one person you can list a few, though it'd be good to make it clear who we should come to with issues as well | 14:47 |
| bauzas | dansmith: cool, thanks | 14:47 |
| * bauzas thinks that ayone should speak French | 14:47 | |
| *** ducnc has joined #openstack-nova | 14:47 | |
| *** zhangjn has joined #openstack-nova | 14:47 | |
| dansmith | voulez vous coucher avec ... uh, le nevermind | 14:48 |
| johnthetubaguy | dhellmann: +1 | 14:48 |
| *** rushil has joined #openstack-nova | 14:48 | |
| johnthetubaguy | dansmith: popular culture references, whatever next | 14:49 |
| *** zhangjn has quit IRC | 14:49 | |
| johnthetubaguy | alaski: I am guessing we should defer this one now :( https://blueprints.launchpad.net/nova/+spec/add-buildrequest-obj | 14:49 |
| alaski | johnthetubaguy: yeah, doesn't look like it's going to make it | 14:49 |
| *** zhangjn has joined #openstack-nova | 14:50 | |
| *** zhangjn has quit IRC | 14:50 | |
| *** zzzeek has joined #openstack-nova | 14:50 | |
| johnthetubaguy | alaski: yeah, the persist request spec seems maybe possible, or I am just being over optimistic? | 14:51 |
| dansmith | so is the gate jammed up because things are failing or because we're on the doorstep of a deadline? | 14:51 |
| alaski | johnthetubaguy: that one's ready IMO, unless there's feedback I haven't seen this morning | 14:51 |
| johnthetubaguy | alaski: it looks like it has a +1 I just opened it up to take a peak | 14:51 |
| bauzas | alaski: I need to look at your change | 14:51 |
| bauzas | johnthetubaguy: I can review it by today | 14:52 |
| johnthetubaguy | bauzas: +1 https://review.openstack.org/#/c/211753/ is well worth a look | 14:52 |
| bauzas | johnthetubaguy: yeah, that and jaypipes's stuff :) | 14:52 |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add display-flavor-id-in-migration.rst https://review.openstack.org/219431 | 14:52 |
| johnthetubaguy | just wondering what we can get into the gate before tomorrow morning | 14:52 |
| *** yamahata has joined #openstack-nova | 14:52 | |
| dansmith | johnthetubaguy: I assume the object and rpc bump bits will be accepted after tomorrow, right? | 14:52 |
| dansmith | I think we talked about that already, | 14:53 |
| johnthetubaguy | dansmith: yeah, I pushed the RPC bump to RC1, I can argue for that one | 14:53 |
| dansmith | johnthetubaguy: okay, I think we'll only get instance bumped for the objects, but that's the biggest one | 14:53 |
| johnthetubaguy | dansmith: I was hoping to get the objects stuff in, but I guess there are no string changes in there, so lets move that too | 14:53 |
| dansmith | johnthetubaguy: and the conductor rpc bump will be nearly a no-op I htink | 14:53 |
| johnthetubaguy | dansmith: yeah, that seems well worth it | 14:53 |
| johnthetubaguy | oh, true | 14:53 |
| dansmith | johnthetubaguy: that's all one blueprint, right? objects and rpc? | 14:53 |
| *** diogogmt has quit IRC | 14:54 | |
| dansmith | liberty-bump-object-and-rpcapi-versions | 14:54 |
| bauzas | johnthetubaguy: FWIW, the ReqSpec series is having no modified strings neither new deps, so we're prettly clear off the StringFreeze and DependencyFreeze | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests() https://review.openstack.org/199205 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Fix RequestSpec.instance_group hydration https://review.openstack.org/219182 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in the FilterScheduler https://review.openstack.org/188487 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Prepare filters for using RequestSpec object https://review.openstack.org/197912 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify left filters for RequestSpec https://review.openstack.org/202678 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object https://review.openstack.org/203078 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify metric-related filters for RequestSpec https://review.openstack.org/202676 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify NUMA, PCI and num_instances filters for RequestSpec https://review.openstack.org/202677 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler https://review.openstack.org/199051 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Aggregate filters for RequestSpec https://review.openstack.org/202674 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify default filters for RequestSpec https://review.openstack.org/202675 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Affinity filter for RequestSpec https://review.openstack.org/202673 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec object in HostManager https://review.openstack.org/191251 | 14:54 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Remove reservation_id from the logs when a schedule fails https://review.openstack.org/217798 | 14:54 |
| dansmith | does the dep freeze count against g-r or nova itself? | 14:54 |
| *** mriedem_meeting is now known as mriedem | 14:54 | |
| dansmith | because we reaaaaly need to get this merged: https://review.openstack.org/#/c/216968/13 | 14:55 |
| bauzas | (well, tbh, only the first change https://review.openstack.org/219182 is having a string changed) | 14:55 |
| bauzas | dansmith: https://review.openstack.org/219182 updated | 14:55 |
| bauzas | jaypipes: ^ as well | 14:55 |
| *** diogogmt has joined #openstack-nova | 14:55 | |
| mriedem | dansmith: dep freeze is just no new minimum required versions of things i think | 14:56 |
| johnthetubaguy | dansmith: ah, good point, I think we might get away with that, but not going to tag until the morning | 14:56 |
| mriedem | https://wiki.openstack.org/wiki/DepFreeze | 14:56 |
| dansmith | mriedem: well, that change is a new minimum | 14:56 |
| mriedem | dansmith: i meant in g-r | 14:56 |
| dansmith | ah, okay | 14:56 |
| dansmith | so we're good then | 14:56 |
| mriedem | yeah | 14:56 |
| mriedem | if the damn change ever merges | 14:56 |
| dansmith | yeah | 14:56 |
| dims_ | bswartz: i want to add a test case | 14:57 |
| dims_ | bswartz: hence the WIP | 14:57 |
| johnthetubaguy | dansmith: mriedem: I added that requirements sync at the top of here, so I try not to forget it when tagging in the morning: https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 14:57 |
| bswartz | dims_: is that going to happen soon? we can't merge anything in manila while that bug exists | 14:57 |
| *** ctrath has joined #openstack-nova | 14:58 | |
| *** ctrath has quit IRC | 14:58 | |
| dims_ | bswartz: in about an hour. need to get to work | 14:58 |
| johnthetubaguy | dims_: I keep meaning to catch you about this one: https://blueprints.launchpad.net/nova/+spec/liberty-olso-sync | 14:58 |
| lalitd | alaski: hi, today meeting will be there? | 14:58 |
| *** hemnafk is now known as hemna | 14:58 | |
| bswartz | dims_: okay thanks -- please ping me when you have another patchset up | 14:58 |
| dims_ | ack | 14:58 |
| johnthetubaguy | dims_: I marked it as partially implemented, given the timing | 14:59 |
| dims_ | johnthetubaguy: sounds good. | 14:59 |
| *** krotscheck is now known as kro_afk | 14:59 | |
| dims_ | johnthetubaguy: early in M, we need to do oslo.policy and oslo.cache | 14:59 |
| johnthetubaguy | dims_: I think we got all the stuff thats blocking you folks merged? | 14:59 |
| dims_ | johnthetubaguy: yep | 14:59 |
| *** vishy has quit IRC | 14:59 | |
| dims_ | johnthetubaguy: everything needed for L. yes. | 14:59 |
| dims_ | thanks!! | 14:59 |
| ftersin | alaski: build_request_spec seems a good place for that as well, but... | 14:59 |
| johnthetubaguy | dims_: early M sounds good, I know there were questions around policy, but yeah | 14:59 |
| *** vivekd has quit IRC | 15:00 | |
| ftersin | alaski: currently size of root bdm is always 0 for image backed instances | 15:00 |
| johnthetubaguy | dims_: thanks for keeping pushing, its good work :) | 15:00 |
| dims_ | yep. we can talk about it at summit | 15:00 |
| johnthetubaguy | +1 | 15:00 |
| alaski | lalitd: the next meeting is next week | 15:00 |
| dims_ | very welcome | 15:00 |
| *** matrohon has quit IRC | 15:00 | |
| * dims_ back in a bit | 15:00 | |
| *** mwagner_lap has quit IRC | 15:00 | |
| ftersin | alaski: and it's not clear for me how to consider a target flavor in resize operation | 15:01 |
| johnthetubaguy | dansmith: sorry, I failed to send my reply to you, yes I think both bumps are in here now: https://launchpad.net/nova/+milestone/liberty-rc1 and https://blueprints.launchpad.net/nova/+spec/liberty-bump-object-and-rpcapi-versions | 15:01 |
| ftersin | alaski: so i propose to null out root_gb in build_request_spec now based on is_volume_backed_instance | 15:02 |
| *** yamahata has quit IRC | 15:02 | |
| lalitd | alaski: it is regarding patch https://review.openstack.org/#/c/213041/ . can you have a look? it is passing unit/functional tests but having some fails in granade and tempest. can you suggest on them | 15:03 |
| *** yamahata has joined #openstack-nova | 15:03 | |
| dansmith | johnthetubaguy: okay | 15:03 |
| ftersin | alaski: probably this solution can even be backported | 15:03 |
| johnthetubaguy | dansmith: there are a few objects patches left that might help us drop more conductor things: https://review.openstack.org/#/q/topic:bp/liberty-objects,n,z | 15:03 |
| *** dave-mccowan has quit IRC | 15:03 | |
| alaski | ftersin: I'm more in favor of that than what was in the patch. But I think it needs to happen after that is all objectified, and should probably have a bp/spec | 15:04 |
| alaski | lalitd: sure, I will take a look later | 15:04 |
| lalitd | alaski: thankyou | 15:04 |
| *** kro_afk is now known as krotscheck | 15:04 | |
| *** unicell1 has joined #openstack-nova | 15:04 | |
| *** nkrinner has quit IRC | 15:04 | |
| dansmith | johnthetubaguy: ah, yheah, I had a patch to do that a while ago, but it was always blocked on something small | 15:05 |
| dansmith | cool if we can do it now | 15:05 |
| *** unicell has quit IRC | 15:06 | |
| *** ducnc has quit IRC | 15:06 | |
| jaypipes | bauzas: +Wd | 15:06 |
| johnthetubaguy | dansmith: seems easier just to get some of those in now | 15:06 |
| *** ildikov has joined #openstack-nova | 15:07 | |
| johnthetubaguy | alaski: do you know if we are likely to get these ones in now: https://blueprints.launchpad.net/nova/+spec/flavor-cell-api seems like soft_delete is back again? | 15:08 |
| dansmith | johnthetubaguy: totally.. I meant before I was held up on RT issues still needing dict-based updates or something | 15:08 |
| *** ducnc has joined #openstack-nova | 15:08 | |
| johnthetubaguy | dansmith: ah, I missed that dependency there, makes sense | 15:09 |
| alaski | johnthetubaguy: I think there are a few kinks to work out of two of the patches on that, so I think they'll miss | 15:09 |
| alaski | johnthetubaguy: soft_delete we're back and forth on, I'm pushing to remove it but obstacles have come up | 15:10 |
| alaski | so we're working through those | 15:10 |
| bauzas | jaypipes: coolness, your karma will be good | 15:10 |
| johnthetubaguy | alaski: yeah, I think I am going to apply the -2 hammer at this point, I think we should focus on the ready-ier stuff, if thats OK? | 15:10 |
| bauzas | alaski: johnthetubaguy: reviewing https://review.openstack.org/#/c/211753/ now | 15:11 |
| bauzas | FWIW | 15:11 |
| alaski | johnthetubaguy: yeah | 15:11 |
| *** dims_ has quit IRC | 15:12 | |
| *** jistr is now known as jistr|call | 15:12 | |
| *** garyk has quit IRC | 15:14 | |
| *** dannywilson has joined #openstack-nova | 15:14 | |
| bauzas | alaski: so I guess you preferred to persist the nested objects for the ReqSpec within the same table, and not leave any FK ? | 15:14 |
| bauzas | alaski: it looks good to me, just wondering if we should explain why | 15:15 |
| *** dave-mccowan has joined #openstack-nova | 15:16 | |
| bauzas | alaski: one could object that for example InstanceGroup does have a .save() method which persists in its separate table | 15:16 |
| *** dannywil_ has joined #openstack-nova | 15:16 | |
| *** dannywilson has quit IRC | 15:16 | |
| alaski | bauzas: yeah. it's basically for the same reason we store flavor with instance. that way deletes/modifications happening on the included resources don't matter | 15:17 |
| alaski | we capture the request spec as it was when it was used | 15:17 |
| bauzas | alaski: yeah I totally got that | 15:17 |
| bauzas | alaski: I'm just wondering if we should just put some comment | 15:17 |
| alaski | I can if that will help | 15:18 |
| bauzas | alaski: explaining that we snapshot those objects | 15:18 |
| bauzas | alaski: so we're not exactly duplicating, but rather making them asynchronous from their actual origin | 15:18 |
| bauzas | alaski: hence my comment here, not in the change | 15:19 |
| bauzas | alaski: but I don't want to delay, we can add those comments later on | 15:20 |
| johnthetubaguy | mriedem: thanks for sorting out the python-novaclient release, looks like that has been merged now | 15:20 |
| mriedem | \o/ | 15:20 |
| *** tpeoples has quit IRC | 15:20 | |
| bauzas | mriedem: johnthetubaguy: question about client stability, what if I want to change a very badly named method in the contrib repo ? | 15:21 |
| *** Sukhdev_ has joined #openstack-nova | 15:21 | |
| bauzas | mriedem: johnthetubaguy: ie. https://github.com/openstack/python-novaclient/blob/master/novaclient/v2/contrib/host_evacuate_live.py | 15:22 |
| *** eharney has joined #openstack-nova | 15:22 | |
| bauzas | mriedem: johnthetubaguy: I guess I would break the compat' right? | 15:22 |
| bauzas | mriedem: johnthetubaguy: so I would need a forward path ? | 15:22 |
| alaski | bauzas: okay, I can add a followup patch with a comment | 15:23 |
| *** emagana has joined #openstack-nova | 15:23 | |
| bauzas | alaski: hold on | 15:23 |
| *** ctrath has joined #openstack-nova | 15:23 | |
| bauzas | alaski: https://review.openstack.org/#/c/211753/22/nova/objects/request_spec.py,cm will conflict with https://review.openstack.org/#/c/219182/2 which was recently approved | 15:23 |
| bauzas | alaski: you should preemptively provide a 1.4 version | 15:24 |
| *** karimb has quit IRC | 15:24 | |
| johnthetubaguy | bauzas: seems like you need to add a new method, and "deprecate" the old one, or something like that | 15:24 |
| bauzas | alaski: sorry about that | 15:24 |
| bauzas | johnthetubaguy: yeah, my thoughts | 15:24 |
| johnthetubaguy | bauzas: I can't remember how we do that off hand | 15:24 |
| bauzas | johnthetubaguy: so, the subsequent question is, what is the deprecation period for that old method name ? | 15:25 |
| alaski | bauzas: okay, I'll respin. Don't worry, I'm used to these breaks by now :) | 15:25 |
| *** dustins_ has quit IRC | 15:25 | |
| *** smatzek has quit IRC | 15:25 | |
| bauzas | johnthetubaguy: since those releases are async, we can't count per cycle | 15:25 |
| *** ducnc has quit IRC | 15:26 | |
| bauzas | johnthetubaguy: so I'm just wondering how much we have to wait until we drop the legacy method | 15:27 |
| johnthetubaguy | bauzas: probably wait a few decades, and we should be good | 15:28 |
| bauzas | cool \o/ | 15:28 |
| *** mwagner_lap has joined #openstack-nova | 15:28 | |
| *** tpeoples has joined #openstack-nova | 15:28 | |
| johnthetubaguy | bauzas: I think technically, we have to bump the major version to drop anything, but would want to give folks some warning of the drop I guess (shrugs) | 15:28 |
| bauzas | johnthetubaguy: well, one could argue the CLI is not aimed to be used programatically, so no strong contract | 15:29 |
| johnthetubaguy | bauzas: we do have stable branches now though, which might change things a little, might be able to wait one release, or something, but would need to talk that though with some client folks | 15:29 |
| *** tri2sing has joined #openstack-nova | 15:30 | |
| bauzas | johnthetubaguy: the stable branch is consuming the novaclient API, right? only devstack is using the CLI (and operators) | 15:30 |
| ftersin | alaski: well, lets look on the whole problem. It refers not only to disk filters (including ExactDiskFilter), but to some other parts as well. | 15:30 |
| ftersin | alaski: e.g. MoveClaim which uses flavor.root_gb | 15:31 |
| *** yassine_ has quit IRC | 15:31 | |
| ftersin | alaski: other ones are fixed by https://review.openstack.org/#/c/200870 and mentioned in the commit message there | 15:32 |
| ftersin | alaski: i think we should have a plan to fix all of this together and completely, shouldn't we? | 15:32 |
| johnthetubaguy | bauzas: jaypipes: so are these lats patches likely to make it for liberty at this point? https://review.openstack.org/#/c/128992/27 | 15:33 |
| ftersin | alaski: what is your suggestion how to process this? | 15:34 |
| bauzas | johnthetubaguy: I leave jaypipes answer, but those are needing https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:pci-cleanup,n,z | 15:34 |
| bauzas | johnthetubaguy: wait for fencing tho | 15:34 |
| bauzas | johnthetubaguy: until jaypipes confirms | 15:34 |
| johnthetubaguy | bauzas: hmm, they don't seem to be rebased on those | 15:35 |
| jaypipes | johnthetubaguy: I have been trying to get the pci cleanup patch series done before that... :( | 15:35 |
| johnthetubaguy | OK, gotcha | 15:35 |
| bauzas | those changes are easily reviewable, could we maybe try to defer the decision until we know the possibilities for FFE ? | 15:35 |
| *** markus_z has quit IRC | 15:35 | |
| alaski | ftersin: yes, we should. | 15:35 |
| ftersin | alaski: a patch series to discuss it on review stage; a spec; or smth else? | 15:35 |
| alaski | ftersin: I would suggest a spec, or a PoC patch series leading into a spec | 15:36 |
| bauzas | johnthetubaguy: jaypipes was actually planning to deliver a new change for https://review.openstack.org/#/c/213591/9 so I could +1 it | 15:36 |
| *** tonytan4ever has joined #openstack-nova | 15:37 | |
| ftersin | alaski: ok, thanks. i'll think about it | 15:37 |
| *** karimb has joined #openstack-nova | 15:37 | |
| jaypipes | bauzas: yes, just been in meetings all morning. | 15:38 |
| bauzas | jaypipes: you should have asked your dog to code | 15:38 |
| jaypipes | bauzas: lol | 15:38 |
| *** jistr|call is now known as jistr | 15:38 | |
| * johnthetubaguy leaves bauzas to review the pug code | 15:39 | |
| alaski | ftersin: I should write some thoughts out as well. I don't have a full plan around this, just didn't want bdms in filter_properties :) | 15:39 |
| *** armax has joined #openstack-nova | 15:40 | |
| *** alexschm has quit IRC | 15:40 | |
| johnthetubaguy | jaypipes: cool, so I will leave those open, in the hope we get a +2 +W on some of those before the morning, stranger things have happened I guess | 15:40 |
| jaypipes | johnthetubaguy: trying my best... :( | 15:40 |
| johnthetubaguy | jaypipes: totally understood | 15:41 |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: RT: Migration resource tracking uses migration context https://review.openstack.org/218938 | 15:42 |
| ndipanov | dansmith, ^ | 15:43 |
| mtreinish | jroll: how many servers can we create at once in that fake ironic test env? most of the failures look to be caused by exhausting the number of available: http://logs.openstack.org/57/209957/3/experimental/gate-tempest-dsvm-ironic-pxe_ipa-full-nv/c77a341//logs/screen-n-cond.txt.gz?level=TRACE#_2015-09-02_13_31_02_334 | 15:43 |
| bauzas | alaski: reviewed https://review.openstack.org/#/c/211753/22 | 15:44 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add persistence to the RequestSpec object https://review.openstack.org/211753 | 15:44 |
| mtreinish | jroll: well aside from the run taking too long and jenkins killing it, which also happened :) | 15:44 |
| bauzas | alaski: I just wonder how the magic is made for ignore_hosts and the other ListOfStrings fields | 15:44 |
| bauzas | jaypipes: you need me on that ? | 15:45 |
| jroll | mtreinish: so, we need about 1g per "baremetal node", I think we've kept it to 2 or 3 so far due to that | 15:45 |
| bauzas | jaypipes: I mean, I can spin-off a new change for https://review.openstack.org/#/c/213591/9 | 15:45 |
| jroll | mtreinish: "baremetal node" being a VM | 15:45 |
| bauzas | jaypipes: so you could even +2 it ;) | 15:45 |
| * bauzas is bad | 15:45 | |
| mtreinish | jroll: ok, yeah we're constrained to 8gb on the test slave, and we typically use between 6-7gb just to run things, so I expect the limit is 2 | 15:46 |
| jroll | mtreinish: yeah :( | 15:46 |
| *** xiaohui has quit IRC | 15:47 | |
| *** xiaohui has joined #openstack-nova | 15:47 | |
| *** electrocucaracha has quit IRC | 15:48 | |
| jaypipes | bauzas: nah, gimme a few | 15:49 |
| lxsli | I'm messing with conductor tests and hitting a race condition after using CastAsCall | 15:49 |
| lxsli | anyone know the best way to work around this please? | 15:49 |
| *** ijw_ has quit IRC | 15:49 | |
| bauzas | lxsli: race cond ? | 15:49 |
| *** dustins has joined #openstack-nova | 15:49 | |
| mriedem | lxsli: like a thread spawn issue? | 15:50 |
| lxsli | btw does CastAsCall work with mock (as opposed to mox) tests? | 15:50 |
| bauzas | lxsli: you mean the same test that is not providing the same output ? | 15:50 |
| mriedem | lxsli: there should be stubs to disable that stuff | 15:50 |
| lxsli | The code after the cast is not supposed to execute until the cast returns; but it happens before, causing my mock assert_called_once_with to fail | 15:50 |
| mtreinish | jroll: we might be able to bump it by 1: http://paste.openstack.org/show/441603/ (from when that no free hosts failure happened) | 15:50 |
| mriedem | lxsli: you have to stub that stuff out | 15:50 |
| mtreinish | jroll: we had > 1GB free, but that doesn't leave us a lot of headroom | 15:50 |
| mriedem | trying to find an example | 15:50 |
| lxsli | mriedem: so I have to use mox? | 15:51 |
| mriedem | lxsli: i think there are actual fixtures in tree somewhere | 15:51 |
| mtreinish | jroll: another thing we could try is disabling other services, like swift to free up some more ram | 15:51 |
| *** dims has joined #openstack-nova | 15:51 | |
| *** doude has quit IRC | 15:52 | |
| lxsli | mriedem: I have self.useFixture(cast_as_call.CastAsCall(self.stubs)) | 15:53 |
| *** kiran-r has joined #openstack-nova | 15:53 | |
| lxsli | mriedem: does this rely on self.mox.StubOutWithMock to setup self.stubs? | 15:54 |
| *** nic has joined #openstack-nova | 15:55 | |
| *** browne has quit IRC | 15:55 | |
| *** dustins has quit IRC | 15:55 | |
| lxsli | mriedem: I'm trying to clean up nova/tests/unit/conductor/test_conductor.py | 15:55 |
| *** garyk has joined #openstack-nova | 15:55 | |
| *** electrocucaracha has joined #openstack-nova | 15:55 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add list-server-invalid-option-as-error.rst https://review.openstack.org/219755 | 15:56 |
| mriedem | lxsli: self.useFixture(cast_as_call.CastAsCall(self.stubs)) is what i was looking for | 15:56 |
| lxsli | ;-; | 15:57 |
| *** mtanino has quit IRC | 15:57 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add list-server-invalid-option-as-error.rst https://review.openstack.org/219755 | 15:57 |
| *** dustins has joined #openstack-nova | 15:57 | |
| *** ndipanov has quit IRC | 15:57 | |
| *** ndipanov has joined #openstack-nova | 15:57 | |
| mriedem | lxsli: i want to say there is a fixture for mocking out nova.utils.spawn_n also | 15:58 |
| lxsli | In _BaseTestCase (same file) I can see: fake_utils.stub_out_utils_spawn_n(self.stubs) | 15:58 |
| mriedem | fake_utils.stub_out_utils_spawn_n(self.stubs) | 15:58 |
| mriedem | yeah | 15:58 |
| mriedem | well if those are being used i'm all out of ideas | 15:58 |
| * mriedem goes back to being unhelpful | 15:59 | |
| lxsli | thanks for trying! | 15:59 |
| *** ijw_ has joined #openstack-nova | 16:01 | |
| openstackgerrit | Davanum Srinivas (dims) proposed openstack/nova: Allow to use autodetection of volume device path https://review.openstack.org/219696 | 16:02 |
| *** unicell1 has quit IRC | 16:02 | |
| *** dims has quit IRC | 16:02 | |
| *** kiran-r has quit IRC | 16:02 | |
| *** markvoelker has quit IRC | 16:03 | |
| *** jlanoux has quit IRC | 16:04 | |
| *** dims_ has joined #openstack-nova | 16:04 | |
| dims_ | bswartz: updated 219696 | 16:05 |
| *** browne has joined #openstack-nova | 16:05 | |
| openstackgerrit | Alexis Lee proposed openstack/nova: WIP Unify on _schedule_instances https://review.openstack.org/219761 | 16:05 |
| bswartz | dims_: thank you reviewing now | 16:06 |
| lxsli | jaypipes: if you like test cleanup you might like https://review.openstack.org/219761 ; and if you know how to solve the CastAsCall issue that'd be great | 16:08 |
| lxsli | but no rush! | 16:08 |
| jaypipes | jayqueue.push(lxsli) | 16:08 |
| lxsli | thanks :) | 16:08 |
| jaypipes | :) | 16:08 |
| *** mtanino has joined #openstack-nova | 16:08 | |
| *** sudipto has quit IRC | 16:09 | |
| *** alex_klimov has quit IRC | 16:09 | |
| openstackgerrit | Bartosz Górski proposed openstack/python-novaclient: Adds missing internationalization for help message https://review.openstack.org/219768 | 16:11 |
| mriedem | ccarmack: you can post the nova change for this now too https://review.openstack.org/#/c/219718/ | 16:12 |
| mriedem | make the nova change depend on the devstack change | 16:12 |
| mriedem | using Depends-On: Ibf1f9aaa63e5f17b7d8774b511940ba8421e0887 | 16:12 |
| mriedem | in the commit message of the nova change | 16:12 |
| *** shoutm has joined #openstack-nova | 16:12 | |
| *** stefan_amann has quit IRC | 16:13 | |
| ccarmack | mriedem: yup, will do | 16:13 |
| *** afazekas__ has joined #openstack-nova | 16:16 | |
| *** devlaps has joined #openstack-nova | 16:16 | |
| *** alexpilotti has quit IRC | 16:17 | |
| *** romainh has left #openstack-nova | 16:17 | |
| *** sahid has quit IRC | 16:17 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add display-flavor-id-in-migration.rst https://review.openstack.org/219431 | 16:18 |
| *** sachin has joined #openstack-nova | 16:18 | |
| mriedem | apmelton: have you ever seen this ? http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz?level=TRACE#_2015-09-01_22_10_50_142 | 16:19 |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add list-server-invalid-option-as-error.rst https://review.openstack.org/219755 | 16:19 |
| apmelton | mriedem: yea, something is leaking nbd devices | 16:20 |
| mriedem | ok, failed deletes | 16:20 |
| mriedem | http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz?level=TRACE#_2015-09-01_22_13_11_763 | 16:20 |
| mriedem | and those pile up so eventually n-sch pukes on overquota | 16:20 |
| *** vladikr has quit IRC | 16:21 | |
| dansmith | mriedem: jaypipes just +Wd my change that depends on the requirements change | 16:22 |
| dansmith | mriedem: if I kick it out, I have to re-run check on the full stack above it | 16:22 |
| mriedem | let's just keep watching the reqs sync change i guess | 16:22 |
| jaypipes | dansmith: sorry man, did I f*** something up? | 16:22 |
| dansmith | mriedem: okay | 16:22 |
| dansmith | jaypipes: yes, you were too helpful | 16:22 |
| dansmith | :) | 16:22 |
| jaypipes | crap | 16:23 |
| dansmith | jaypipes: no dude, it's all good | 16:23 |
| dansmith | things are flying around, hard to keep track of everything | 16:23 |
| *** jichen has quit IRC | 16:23 | |
| *** sudipto has joined #openstack-nova | 16:24 | |
| *** burgerk_ has joined #openstack-nova | 16:24 | |
| *** vilobhmm_11 has joined #openstack-nova | 16:25 | |
| mriedem | apmelton: any ideas on what's not doing the cleanup? | 16:25 |
| mriedem | or things to check? | 16:25 |
| *** tjones has joined #openstack-nova | 16:25 | |
| apmelton | mriedem: sorry, I don't | 16:26 |
| mriedem | i see something | 16:27 |
| mriedem | i spy | 16:27 |
| *** meghal has joined #openstack-nova | 16:27 | |
| mriedem | i don't see any of the error/warning logs in here http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/mount/nbd.py#n51 | 16:27 |
| openstackgerrit | Chuck Carmack proposed openstack/nova: Remove the blacklisted nova-cells shelve tests https://review.openstack.org/219778 | 16:27 |
| *** Piet has joined #openstack-nova | 16:28 | |
| *** doude_ has joined #openstack-nova | 16:28 | |
| *** meghal1 has joined #openstack-nova | 16:29 | |
| *** vladikr has joined #openstack-nova | 16:29 | |
| apmelton | mriedem: interesting, so nbd devices are in use, but they odn't have locks? | 16:30 |
| mriedem | i see the unget happening at times http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_10_46_490 | 16:30 |
| *** meghal has quit IRC | 16:31 | |
| *** gszasz has quit IRC | 16:34 | |
| *** tjones has quit IRC | 16:34 | |
| *** afazekas__ has quit IRC | 16:35 | |
| apmelton | mriedem: so I see this http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_10_46_490 | 16:37 |
| apmelton | mriedem: and then a bit further down I see http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_10_48_928 | 16:37 |
| apmelton | which means the rootfs was never unlinked from the nbd device | 16:37 |
| *** earlephilhower has joined #openstack-nova | 16:38 | |
| apmelton | mriedem: do you know if instances are soft deleted in devstack, and then cleaned up at a later date? | 16:38 |
| *** tjones has joined #openstack-nova | 16:38 | |
| mriedem | should be a straight server delete | 16:39 |
| mriedem | mtreinish: ^ | 16:39 |
| mriedem | apmelton: i think there are parts of tempest that wait for the server to actually be gone and some that move on, not sure though | 16:39 |
| apmelton | mriedem: I believe this would be something nova would do behind the scenes | 16:39 |
| apmelton | let me try to find the config opt | 16:40 |
| mtreinish | mriedem: for the most part tempest does a wait/poll loop on delete calls waiting for a 404 | 16:40 |
| mriedem | mtreinish: ok that's what i thought | 16:40 |
| mtreinish | mriedem: sometimes it's done right after the delete, and others it's later | 16:40 |
| mtreinish | but we should always be verifying that | 16:40 |
| mriedem | mtreinish: right, otherwise you can go over quota | 16:40 |
| mtreinish | well, its a bigger deal for the tenant reuse case when you use an accounts.yaml | 16:41 |
| mtreinish | isolated tenants normally protects us from quota issues | 16:41 |
| mriedem | allow_tenant_isolation=False | 16:41 |
| mriedem | is that even still an option in tempest? | 16:41 |
| mtreinish | mriedem: yes it is, but there is accounts.yaml as an alternative: http://docs.openstack.org/developer/tempest/configuration.html#credential-provider-mechanisms | 16:42 |
| mtreinish | mriedem: I've been trying to start the deprecation timer on the old 3 sets of creds in config though | 16:43 |
| *** marzif has quit IRC | 16:43 | |
| mtreinish | but I need to get some extra jobs up first | 16:43 |
| *** marzif has joined #openstack-nova | 16:43 | |
| *** jistr has quit IRC | 16:44 | |
| *** ptm_away is now known as PaulMurray | 16:44 | |
| *** angdraug has joined #openstack-nova | 16:44 | |
| mriedem | apmelton: i see a lot of these too http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_13_12_182 | 16:46 |
| *** unicell has joined #openstack-nova | 16:47 | |
| *** stefan_amann has joined #openstack-nova | 16:47 | |
| *** afazekas__ has joined #openstack-nova | 16:48 | |
| apmelton | mriedem: is there an absolute minimum set of tests we can run? I feel like starting with this huge set and trying to narrow down any and all bugs at once isn't going to be sustainable | 16:50 |
| mriedem | yeah, probably | 16:50 |
| *** meghal1 has quit IRC | 16:50 | |
| mriedem | i was going to completely disable the cinder service in the tempest run so it doesn't try any volume stuff | 16:50 |
| mriedem | mtreinish: are smoke tests still a thing ? | 16:51 |
| mtreinish | mriedem: yes, that's what we run on grenade | 16:51 |
| mtreinish | it's a small set to sanity check | 16:51 |
| *** lucasagomes is now known as lucas-dinner | 16:51 | |
| mriedem | ok, i'll try using that | 16:51 |
| mtreinish | mriedem: oh, it's also what the puppet integration tempest jobs have started running too | 16:51 |
| *** tjones has quit IRC | 16:51 | |
| *** rotbeard has quit IRC | 16:51 | |
| *** tjones has joined #openstack-nova | 16:52 | |
| mriedem | echo_summary "Running tempest scenario and smoke tests" | 16:52 |
| mriedem | grenade doesn't run scenario tests | 16:52 |
| mriedem | who are you trying to fool | 16:52 |
| *** sachin has quit IRC | 16:53 | |
| mtreinish | mriedem: it runs some scenarios | 16:54 |
| mtreinish | but not all anymore | 16:54 |
| mtreinish | patches welcom | 16:54 |
| mriedem | everytime i push a change you guys say i'm not welcome here | 16:54 |
| mtreinish | who says that? | 16:55 |
| * mtreinish glares at dansmith | 16:55 | |
| mriedem | i'll say, dtroyer | 16:55 |
| mriedem | since he's not here | 16:55 |
| *** smatzek has joined #openstack-nova | 16:55 | |
| mriedem | test_volume_boot_pattern is a smoke test? | 16:55 |
| mtreinish | I think so | 16:56 |
| mriedem | that's rhetorical, | 16:56 |
| mriedem | but i'm surprised | 16:56 |
| *** earlephilhower has quit IRC | 16:56 | |
| *** tonytan4ever has quit IRC | 16:57 | |
| *** salv-orlando has quit IRC | 16:57 | |
| *** derekh has quit IRC | 16:57 | |
| *** earlephilhower has joined #openstack-nova | 16:58 | |
| sdague | well, it is kind of a basic thing people want to do | 16:59 |
| mriedem | mtreinish: what is this for? https://github.com/openstack-dev/devstack/blob/master/lib/tempest#L609 | 17:00 |
| sdague | so, the relaxed validation of 2.0 on 2.1 isn't apparently perfect - https://bugs.launchpad.net/nova/+bug/1491512 | 17:00 |
| openstack | Launchpad bug 1491512 in OpenStack Compute (nova) "Behavior change with latest nova paste config" [Undecided,New] | 17:00 |
| mriedem | mtreinish: nvm, i guess just to ge the venv | 17:00 |
| mtreinish | mriedem: that's to create a venv for running the verify-tempest-config tool | 17:00 |
| mtreinish | it's kinda hacky | 17:00 |
| sdague | infra's shade library was incorrectly building floating ip resource urls which we now are rejecting as not a valid id (which they are not) | 17:01 |
| *** annegentle has quit IRC | 17:01 | |
| *** penick has joined #openstack-nova | 17:01 | |
| *** electrocucaracha has quit IRC | 17:01 | |
| *** nic has quit IRC | 17:02 | |
| *** sudipto has quit IRC | 17:03 | |
| openstackgerrit | Nicolas Simonds proposed openstack/nova-specs: RBD instance snapshots https://review.openstack.org/188244 | 17:04 |
| *** meghal has joined #openstack-nova | 17:04 | |
| *** Sukhdev_ has quit IRC | 17:06 | |
| mriedem | mtreinish: ah damn, it's d-g that runs tempest | 17:06 |
| mriedem | i figured that'd be the case | 17:06 |
| *** sachin_ has joined #openstack-nova | 17:07 | |
| *** electrocucaracha has joined #openstack-nova | 17:07 | |
| mriedem | i wonder if i can overwrite DEVSTACK_GATE_TEMPEST_FULL=0 from devstack so that d-g will ignore it | 17:07 |
| *** rushil has quit IRC | 17:08 | |
| *** ssurana has joined #openstack-nova | 17:10 | |
| mriedem | apmelton: well let's see what this version gives us https://review.openstack.org/#/c/219448/ | 17:10 |
| *** lpetrut has quit IRC | 17:10 | |
| *** nic has joined #openstack-nova | 17:10 | |
| *** lpetrut has joined #openstack-nova | 17:10 | |
| sdague | oh, it's not as bad as I thought. We used to return a 404 on the url they were building, now we return a 400 because the id wasn't possible to be valid | 17:12 |
| openstackgerrit | Artom Lifshitz proposed openstack/python-novaclient: Option to specify max servers for live evacuate https://review.openstack.org/219792 | 17:13 |
| *** afazekas__ has quit IRC | 17:14 | |
| *** tri2sing has quit IRC | 17:16 | |
| mriedem | apmelton: since my devstack change depends on https://review.openstack.org/#/c/216895/ and i need to add some debug logging to nova, i'm going to rebase https://review.openstack.org/#/c/216895/ on top of a logging change | 17:18 |
| *** danpb has quit IRC | 17:19 | |
| apmelton | mriedem: go for it | 17:19 |
| apmelton | mriedem: I haven't had the time to fix up the functional tests on that patch though | 17:19 |
| *** marzif has quit IRC | 17:20 | |
| mriedem | that's fine | 17:20 |
| mriedem | i want to add logging to see if disk.teardown_container is doing anything | 17:20 |
| *** tri2sing has joined #openstack-nova | 17:21 | |
| *** VW_ has quit IRC | 17:22 | |
| *** sachin_ has quit IRC | 17:22 | |
| *** VW_ has joined #openstack-nova | 17:22 | |
| openstackgerrit | Artom Lifshitz proposed openstack/nova: Make Glance proxy respect osapi_max_limit https://review.openstack.org/179947 | 17:23 |
| jroll | mtreinish: yeah, I think we could get rid of swift | 17:23 |
| *** sachin_ has joined #openstack-nova | 17:23 | |
| *** achanda has joined #openstack-nova | 17:24 | |
| mriedem | jroll: take care of? https://lh3.googleusercontent.com/-4z4fS_h7b7U/VOt-DyDiCzI/AAAAAAAAFVU/U9WvSpylXjw/w640-h400-p-k/Screenshot%2B2015-02-23%2B20.20.43.png | 17:24 |
| *** sachin_ has quit IRC | 17:25 | |
| mriedem | ooo even better http://33.media.tumblr.com/d02b989a308703c4fdb031ed18d74791/tumblr_mnlb12Foz51qifm00o2_500.gif | 17:25 |
| jroll | mriedem: https://uproxx.files.wordpress.com/2013/10/screen-shot-2013-10-31-at-9-39-23-am.png?w=650&h=351 | 17:25 |
| mriedem | i like that guy, don't particularly care for that show | 17:26 |
| *** sachin_ has joined #openstack-nova | 17:26 | |
| jroll | heh | 17:26 |
| *** VW_ has quit IRC | 17:27 | |
| *** sachin_ has quit IRC | 17:27 | |
| *** VW_ has joined #openstack-nova | 17:28 | |
| *** sachin_ has joined #openstack-nova | 17:29 | |
| *** sachin_ has quit IRC | 17:30 | |
| *** sachin_ has joined #openstack-nova | 17:31 | |
| *** sachin_ has quit IRC | 17:32 | |
| *** dims_ has quit IRC | 17:33 | |
| *** stefan_amann has quit IRC | 17:33 | |
| *** markvoelker has joined #openstack-nova | 17:35 | |
| *** markvoelker has quit IRC | 17:35 | |
| *** dims has joined #openstack-nova | 17:35 | |
| *** markvoelker has joined #openstack-nova | 17:36 | |
| openstackgerrit | Nikola Dipanov proposed openstack/nova: RT: Migration resource tracking uses migration context https://review.openstack.org/218938 | 17:39 |
| ndipanov | dansmith, ^ super embarrassed - there was a pep8 error :( | 17:39 |
| tdurakov | hi | 17:39 |
| dansmith | heh | 17:39 |
| tdurakov | is there any cons of making https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L151 | 17:39 |
| tdurakov | not a singleton | 17:40 |
| *** aix has quit IRC | 17:40 | |
| *** angdraug has quit IRC | 17:40 | |
| dansmith | ndipanov: at least it was an important pep8 error | 17:40 |
| ndipanov | it really was not - newline in a docstring, that's about as random as you get | 17:41 |
| tdurakov | jaypipes, hi, could you help me with^ ? | 17:41 |
| dansmith | ndipanov: I know, I was being sarcastic :) | 17:41 |
| dims | tdurakov: problems with auth failures? | 17:41 |
| tdurakov | dims, yep | 17:41 |
| dims | tdurakov: how would making it a singleton help? (trying to understand) | 17:41 |
| tdurakov | dims, not making it singleton allow to use it in neutron-client | 17:42 |
| dims | tdurakov: mriedem has worked a bunch on this portion of the code i believe. | 17:42 |
| tdurakov | dims, as i understand, https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L180 - it knows how to deal with 401 from neutron | 17:43 |
| mriedem | tdurakov: i'd ask jamielennox | 17:43 |
| *** emagana has quit IRC | 17:43 | |
| *** karimb has quit IRC | 17:43 | |
| dims | tdurakov: right | 17:44 |
| tdurakov | mriedem, ok already pinged him on openstack-keystone | 17:44 |
| *** yamahata has quit IRC | 17:44 | |
| tdurakov | dims, so why we are not using it in keystone client as auth plugin? | 17:45 |
| tdurakov | the only thing it's singleton, but why it is singleton? | 17:45 |
| dims | tdurakov: one sec | 17:46 |
| tdurakov | we are making https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L217 instead | 17:46 |
| *** stefan_amann has joined #openstack-nova | 17:46 | |
| dims | tdurakov: so i was doing this research for keystone v3 support. i figured out that devstack can use v3 using this new config options - https://review.openstack.org/#/c/214644/2/lib/neutron-legacy,cm | 17:48 |
| *** thangp_ has joined #openstack-nova | 17:49 | |
| *** thangp has quit IRC | 17:49 | |
| dims | tdurakov: so if you do auth_plugin "v2password" and use correct auth_url you can use the auth plugin instead of the deprecated and hard coded v2_auth.Password in https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L180 | 17:49 |
| dims | tdurakov: do you want to try that? | 17:49 |
| dims | tdurakov: i had a companion review to that to ensure that nova.conf.sample has the config options for auth plugins - https://review.openstack.org/#/c/214423/ | 17:51 |
| mriedem | apmelton: so when i'm seeing http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_14_10_063 - it's during a multiple server create test, | 17:54 |
| *** ijw_ has quit IRC | 17:54 | |
| mriedem | apmelton: the first instance created is ok, it's the 2nd that fails | 17:54 |
| mriedem | apmelton: the weird tihng is they are using different nbd devices | 17:54 |
| mriedem | the first uses /dev/nbd9 and is ok, the one that fails is using /dev/nbd12 | 17:54 |
| *** dustins has quit IRC | 17:56 | |
| *** ijw_ has joined #openstack-nova | 17:56 | |
| apmelton | mriedem: you know what..... | 17:56 |
| apmelton | mriedem: I think this may be a bigger issue than thought at first | 17:56 |
| *** armax has quit IRC | 17:57 | |
| apmelton | mriedem: I wonder if the code that's calling the link/unlink expects that's only supposed to be for initial set up and the instance isn't going to actually be running off the nbd device | 17:58 |
| apmelton | mriedem: I'm not too sure about that though | 17:59 |
| apmelton | I've got to run to a meeting, but I'll keep digging through those logs in a bit | 17:59 |
| mriedem | if you're not, i'm not | 17:59 |
| mriedem | ok | 17:59 |
| *** dustins has joined #openstack-nova | 17:59 | |
| *** stefan_amann has quit IRC | 17:59 | |
| *** geekinutah has quit IRC | 18:00 | |
| *** eglynn has quit IRC | 18:01 | |
| *** kmartin_ has joined #openstack-nova | 18:02 | |
| bswartz | nova cores, this bugfix is ready for merge: https://review.openstack.org/#/c/219696/ | 18:03 |
| bswartz | please merge it ASAP because Manila cannot merge anything until this fix lands | 18:03 |
| *** kmartin has quit IRC | 18:03 | |
| *** sachin_ has joined #openstack-nova | 18:04 | |
| mriedem | apmelton: i think i found something, ping me when you are free | 18:05 |
| *** markvoelker has quit IRC | 18:05 | |
| *** marzif has joined #openstack-nova | 18:09 | |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/api.py#n464 doesn't actually do anything | 18:11 |
| mriedem | because img._mounter is None | 18:11 |
| *** Sukhdev_ has joined #openstack-nova | 18:11 | |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/api.py#n371 | 18:12 |
| *** sachin_ has quit IRC | 18:14 | |
| *** sachin_ has joined #openstack-nova | 18:15 | |
| *** sachin_ has quit IRC | 18:19 | |
| *** annegentle has joined #openstack-nova | 18:20 | |
| *** electrocucaracha has quit IRC | 18:22 | |
| *** marzif has quit IRC | 18:22 | |
| *** tonytan4ever has joined #openstack-nova | 18:22 | |
| *** marzif has joined #openstack-nova | 18:22 | |
| mriedem | sdague: https://review.openstack.org/#/c/219696/ | 18:23 |
| *** sachin_ has joined #openstack-nova | 18:24 | |
| sdague | mriedem: yeh, seems fine to me | 18:24 |
| *** Chaitu has joined #openstack-nova | 18:24 | |
| mriedem | also a bug in kilo | 18:24 |
| Chaitu | Hay can someone one help me out with tempest code to check the status of the created instance???? | 18:24 |
| sdague | on the v2.1 stack that's not enabled by default, right? | 18:25 |
| *** electrocucaracha has joined #openstack-nova | 18:25 | |
| *** electrocucaracha has quit IRC | 18:25 | |
| mriedem | sdague: volume attachment? | 18:26 |
| *** rushil has joined #openstack-nova | 18:27 | |
| *** meghal has quit IRC | 18:28 | |
| *** artom has quit IRC | 18:28 | |
| *** marzif has quit IRC | 18:28 | |
| *** Chaitu has quit IRC | 18:29 | |
| *** sachin_ has quit IRC | 18:29 | |
| *** sachin_ has joined #openstack-nova | 18:30 | |
| *** armax has joined #openstack-nova | 18:31 | |
| *** meghal has joined #openstack-nova | 18:32 | |
| bswartz | sdague, mriedem: thank you! | 18:33 |
| sdague | mriedem: I thought this was only getting rejected on the json schema | 18:33 |
| *** aix has joined #openstack-nova | 18:34 | |
| mriedem | sdague: what i'm really confused about is why my change in tempest to enable test_stamp_pattern that attaches a volume but doesn't specify a device name wasn't failing | 18:34 |
| mriedem | and i see in the tempest logs that there is actually a device specified http://logs.openstack.org/55/218355/6/check/gate-tempest-dsvm-nova-v21-full/03cdd7a/logs/tempest.txt.gz#_2015-09-02_17_32_14_204 | 18:34 |
| mriedem | but i don't see what in tempest is specifying that | 18:35 |
| mriedem | oh nvm that's the response | 18:35 |
| mriedem | still, i'm not sure how that didn't fail the v2.1 json schema on the nova side for this bug | 18:35 |
| sdague | actually, wait a second | 18:36 |
| *** tri2sing has quit IRC | 18:37 | |
| sdague | I'm actually -2 on https://review.openstack.org/#/c/219696 | 18:37 |
| *** tri2sing has joined #openstack-nova | 18:37 | |
| *** sachin_ has quit IRC | 18:37 | |
| sdague | that's not needed at all, just don't specify the device | 18:37 |
| sdague | optional != nullable | 18:37 |
| mriedem | sdague: yeah, that's what i'm doing in tempest | 18:37 |
| mriedem | not specifying device | 18:37 |
| mriedem | so agree on -2 | 18:37 |
| sdague | I'm not sure why manila thinks they need this patch | 18:38 |
| mriedem | it's a bug in novaclient | 18:38 |
| mriedem | novaclient passes device=None | 18:38 |
| mriedem | and manila uses novaclient | 18:38 |
| mriedem | bswartz: dims: ^ | 18:38 |
| bauzas | sooooo, xen CI is wedged (c) mriedem | 18:38 |
| *** sachin_ has joined #openstack-nova | 18:38 | |
| bauzas | http://zuul.openstack.xenproject.org/scoreboard/?project=openstack/nova&user=jenkins,citrix_xenserver_ci,XenProject-CI&timeframe=24 | 18:38 |
| mriedem | BobBall: ^ | 18:38 |
| bauzas | BobBall: ^ | 18:38 |
| bswartz | mriedem: so what should we be passing if not None? | 18:38 |
| openstackgerrit | Sean Dague proposed openstack/nova: Allow to use autodetection of volume device path https://review.openstack.org/219696 | 18:39 |
| mriedem | bswartz: don't set the device key in the request | 18:39 |
| sdague | right | 18:39 |
| sdague | that's what optional means | 18:39 |
| sdague | don't give it the key | 18:39 |
| vponomaryov | sdague: mriedem is right, novaclient sends it always | 18:39 |
| bswartz | mriedem: in that case we need a bugfix to land in novaclient to unwedge us | 18:40 |
| sdague | vponomaryov: then fix that | 18:40 |
| *** sachin_ has quit IRC | 18:40 | |
| sdague | this is not a bug in the server | 18:40 |
| mriedem | we just released novaclient 2.27 so it's probably a change since then i'd guess | 18:40 |
| vponomaryov | sdague: but it is not compatible with v2.0 | 18:40 |
| *** burgerk has quit IRC | 18:40 | |
| dims | sdague: did you see the test case? when there is device=None in incoming json we fail, if there is no device, we work ok. | 18:41 |
| sdague | dims, yes, which is intentional | 18:41 |
| sdague | that's actually the behavior | 18:41 |
| dims | sdague: so with this fix, both missing and None are treated as the same | 18:41 |
| *** sachin_ has joined #openstack-nova | 18:41 | |
| sdague | dims: right, which I'm saying we *should not* do | 18:41 |
| dansmith | optional and nullable are not the same | 18:41 |
| sdague | because that's confusing | 18:41 |
| dansmith | at all. | 18:41 |
| sdague | right, exactly | 18:42 |
| bswartz | sdague: it may be intentional but if it breaks backwards compatibility it's a regression | 18:42 |
| *** sachin_ has quit IRC | 18:42 | |
| sdague | bswartz: just because you were using a library that does the wrong thing. So fix the library | 18:42 |
| dims | sdague: earlier before the schema check, both worked fine | 18:42 |
| dims | sdague: +1 to fix the library too | 18:42 |
| sdague | dims: all manner of gorp used to "work" in Nova when you sent it random strings | 18:42 |
| sdague | and it made for possible security issues and completely unpredicatable code | 18:43 |
| dansmith | strings were coerced to integers! | 18:43 |
| dansmith | by mysql! | 18:43 |
| sdague | yeh | 18:43 |
| dims | seemed like a clean fix to me :) but it's your call | 18:44 |
| *** sachin_ has joined #openstack-nova | 18:44 | |
| sdague | so, this started failing 14 hours ago? | 18:45 |
| melwitt | the interesting thing would be to check if nova api itself allows it to be left out. because there's a thing with I think sec groups description during create is supposed to be optional but nova api incorrectly requires it | 18:45 |
| sdague | melwitt: it does, tempest runs with it not there | 18:45 |
| sdague | which mriedem dug up | 18:45 |
| *** sachin_ has quit IRC | 18:46 | |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Don't pass null device when attaching a volume https://review.openstack.org/219835 | 18:46 |
| mriedem | ^ needs tests | 18:46 |
| sdague | and, I agree, the sec group description thing pisses me off :) I'll +2 anyone that actually makes that optional again like it's supposed to be | 18:46 |
| melwitt | sdague: ah, okay. late to the party :) glad this one is just a client thing | 18:46 |
| vponomaryov | sdague: merge of this commit - https://github.com/openstack/nova/commit/a9090bb5f50310702408aee340af9d4dad9474cf was the bell of a bug | 18:46 |
| *** jecarey has joined #openstack-nova | 18:46 | |
| sdague | vponomaryov: right, that's the v2.0 on v2.1 stack | 18:47 |
| bswartz | sdague: are you okay with old versions of novaclient not working with new versions of nova? when stuff breaks you'll just tell people to upgrade? | 18:47 |
| sdague | so we need to figure out if the schema relax that we believed we were doing actually is done right, this tripped up shade as well, mordred filed a bug | 18:48 |
| sdague | that was 404 -> 400 on a resource that they were building incorrectly | 18:48 |
| mordred | bswartz: I'm personally ok with neither direction breaking- but if one of them wants to break, I feel MORE strongly that new versions of novaclient be able to talk to old versions of nova | 18:49 |
| mordred | of course, that's just my loud-mouthed opinion | 18:49 |
| mriedem | because updating a library is a million times easier than updating your server | 18:49 |
| mordred | yup | 18:49 |
| mordred | and "just use the latest client library" is a thing that should pretty much always work | 18:50 |
| mordred | but if your provider is running diablo and now you have to figure out how to install a 3 year old python library ... I mean, good luck with that | 18:50 |
| mordred | HOWEVER - I've also given up on python-*client being for end users on most days - so it's possible I have 4 year old opinions that I should revisit | 18:50 |
| bswartz | yes but applications that currently work on Kilo will suddenly fail when pointed at a Liberty server, until they update their libs | 18:51 |
| *** penick has quit IRC | 18:51 | |
| bswartz | I agree updating libs is way easier than updating the server, but there are also relatively few servers and many many clients | 18:52 |
| mriedem | well we could raise the minimum required novaclient for liberty to be >=2.28 | 18:52 |
| mriedem | which would have to happen, like, today | 18:52 |
| mriedem | >=2.27.1 i should say | 18:53 |
| openstackgerrit | Matt Riedemann proposed openstack/python-novaclient: Don't pass null device when attaching a volume https://review.openstack.org/219835 | 18:54 |
| mriedem | sdague: melwitt: dansmith: ^ | 18:54 |
| mriedem | should be ready | 18:54 |
| *** harlowja has quit IRC | 18:54 | |
| sdague | mriedem: +2 | 18:54 |
| mtreinish | mriedem: you should add a functional test too :) | 18:54 |
| mriedem | hey | 18:54 |
| mriedem | no | 18:54 |
| mriedem | you should | 18:54 |
| mriedem | so once we land that we could release it as 2.27.1 today | 18:55 |
| mtreinish | I don't use novaclient, so i don't really care | 18:55 |
| mriedem | i wonder if osc has this same bug | 18:55 |
| * mriedem looks | 18:55 | |
| mtreinish | mriedem: doesn't osc just call to novaclient? | 18:56 |
| melwitt | mriedem: I checked and it appears osc doesn't do volume attach | 18:56 |
| dansmith | heh | 18:56 |
| mriedem | ah ok | 18:56 |
| mriedem | well that's, useful | 18:56 |
| dansmith | safe by lack of functionality | 18:56 |
| *** meghal1 has joined #openstack-nova | 18:56 | |
| *** meghal has quit IRC | 18:56 | |
| melwitt | mtreinish: it doesn't. it does its own things and it's better maintained than the misc clients I think. it's used by devstack to do the whole setup, so I was thinking to recommend it if it has volume attach but I don't see it in the command list in the docs | 18:57 |
| mriedem | mtreinish: test_stamp_pattern only fails b/c i'm not logging enough stuff from the guest :) | 18:57 |
| *** harlowja has joined #openstack-nova | 18:58 | |
| *** meghal1 has quit IRC | 18:58 | |
| *** meghal has joined #openstack-nova | 18:58 | |
| sdague | melwitt: openstack server add volume | 18:58 |
| sdague | that would be where we should figure it out | 18:59 |
| mtreinish | melwitt: what is http://git.openstack.org/cgit/openstack/python-openstackclient/tree/openstackclient/compute/client.py#n30 for? | 18:59 |
| mriedem | mtreinish: i think that's just their own client thing | 18:59 |
| mriedem | oh novaclient | 18:59 |
| mriedem | i see that now | 18:59 |
| *** markvoelker has joined #openstack-nova | 19:00 | |
| mtreinish | melwitt, mriedem: it's also in requirements for osc | 19:01 |
| melwitt | mtreinish, mriedem: well, I was way wrong then. sorry :( | 19:01 |
| mriedem | mtreinish: mtreinish: sdague: just confirmed osc | 19:01 |
| mriedem | compute_client.volumes.create_server_volume( | 19:01 |
| mriedem | server.id, | 19:01 |
| mriedem | volume.id, | 19:01 |
| mriedem | parsed_args.device, | 19:01 |
| mriedem | ) | 19:01 |
| mriedem | so that just calls into novaclient for the thing i just fixed | 19:01 |
| mriedem | uses novaclient python api | 19:01 |
| melwitt | I really thought it was free from the underlying *clients and that's why it was better | 19:02 |
| *** sachin_ has joined #openstack-nova | 19:02 | |
| sdague | melwitt: no, they didn't go that direction | 19:02 |
| *** pixelb has quit IRC | 19:03 | |
| sdague | so... the reason this exposed at all is because we changed the default paste.ini in nova to be the v2.0 on v2.1 compat layer, which was supposed to relax some of the validation again. And I think the question here needs to be if this should be relaxed further. | 19:03 |
| sdague | definitely not on the v2.1 side, but if v2.0 on v2.1 should allow it | 19:03 |
| mriedem | seems like it should | 19:04 |
| mriedem | otherwise it's not really v2.0 | 19:04 |
| sdague | ok, so that's the bug and possible fix that should be sorted | 19:04 |
| * dims nods | 19:04 | |
| edleafe | there is talk of moving osc to use the openstack-python SDK, but that's about as far as it's gone - talk | 19:04 |
| mriedem | not it | 19:04 |
| sdague | so the relevant information should be piled into this bug - https://bugs.launchpad.net/nova/+bug/1491512 | 19:05 |
| openstack | Launchpad bug 1491511 in OpenStack Compute (nova) "duplicate for #1491512 Behavior change with latest nova paste config" [Undecided,New] | 19:05 |
| sdague | and out of those we can decide which things to address | 19:05 |
| sdague | can someone write up this situation in a comment on that bug? | 19:06 |
| *** coolsvap_ has joined #openstack-nova | 19:09 | |
| *** matrohon has joined #openstack-nova | 19:10 | |
| mriedem | i can | 19:11 |
| mriedem | if someone can tell me if we're in string freeze today or not | 19:11 |
| mriedem | or is that thursday? | 19:12 |
| sdague | I have no idea, I'm going to assume that someone will send an email when we are. I feel like guessing on that isn't a great way to go about it. | 19:13 |
| bauzas | mriedem: https://wiki.openstack.org/wiki/Liberty_Release_Schedule I guess we can allow that | 19:13 |
| mriedem | sdague: i thought you had a psychic link with ttx? | 19:14 |
| *** annegentle has quit IRC | 19:17 | |
| sdague | is nova meeting early or late tomorrow? | 19:17 |
| *** chhavi has quit IRC | 19:18 | |
| dansmith | early | 19:18 |
| sdague | ok, I'll stick this on the agenda | 19:19 |
| *** electrocucaracha has joined #openstack-nova | 19:20 | |
| *** rushil has quit IRC | 19:23 | |
| sdague | mriedem: infra found another issue with novaclient, looking at it over there | 19:24 |
| *** ijw__ has joined #openstack-nova | 19:27 | |
| *** ijw_ has quit IRC | 19:28 | |
| *** promulo has joined #openstack-nova | 19:29 | |
| *** meghal has quit IRC | 19:31 | |
| *** jichen has joined #openstack-nova | 19:34 | |
| *** annegentle has joined #openstack-nova | 19:35 | |
| *** ssurana has quit IRC | 19:35 | |
| *** geekinutah has joined #openstack-nova | 19:36 | |
| dims | mriedem: dhellmann says thu - john said 1-3 in his email (http://lists.openstack.org/pipermail/openstack-dev/2015-June/065819.html) | 19:36 |
| *** lpetrut has quit IRC | 19:44 | |
| *** coolsvap_ has quit IRC | 19:45 | |
| bauzas | alaski: AFAICS, cells meeting in 1.15h? | 19:48 |
| alaski | bauzas: not this week, next week | 19:48 |
| bauzas | alaski: so you ran it at 1700UTC ? | 19:49 |
| bauzas | damn it | 19:49 |
| alaski | no meeting today | 19:49 |
| alaski | every two weeks for now | 19:49 |
| bauzas | ooooh right | 19:49 |
| alaski | so next week at 1700 | 19:50 |
| bauzas | we actually agreed on that | 19:50 |
| * bauzas sometimes forgets his socks... | 19:50 | |
| bauzas | ack | 19:50 |
| *** ildikov has quit IRC | 19:51 | |
| *** alex_klimov has joined #openstack-nova | 19:53 | |
| *** rods has left #openstack-nova | 19:53 | |
| openstackgerrit | Matt Riedemann proposed openstack/nova: libvirt: add debug logging for lxc teardown paths https://review.openstack.org/219859 | 19:54 |
| mriedem | apmelton: ^ turns out what i thought was probably busted is not maybe, but it sure is confusing | 19:54 |
| mriedem | in here http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/api.py#n457 | 19:54 |
| mriedem | the whole "if container_root_device:" block is odd because img.teardown() does all of that stuff | 19:55 |
| sdague | so this is neat, novaclient doesn't work with RAX because they don't respond the version url | 19:55 |
| mriedem | but only iff _DiskImage is constructed with a mounter | 19:55 |
| *** ctrath has quit IRC | 19:55 | |
| apmelton | mriedem: that is really odd.... | 19:56 |
| alaski | sdague: is that a v2.1 thing? | 19:56 |
| apmelton | mriedem: that doc string is also really odd | 19:56 |
| sdague | alaski: not really | 19:57 |
| *** ZZelle_ has joined #openstack-nova | 19:57 | |
| apmelton | it sounds like it's trying to tear down the container rootfs while the instance is still active | 19:57 |
| sdague | it's the fact that you policy disallow access to the version info on /v2 | 19:57 |
| sdague | it returns a 401 | 19:57 |
| mriedem | apmelton: yeah, that is probably wrong, that actually happens in clean_lxc_namespace | 19:57 |
| apmelton | ah gotcha | 19:57 |
| sdague | the client is attempting to learn what's available, and you block that | 19:57 |
| mriedem | which is also very bolt on looking | 19:57 |
| bauzas | mmm, 111 changes in the gate \o/ | 19:57 |
| mriedem | apmelton: in here http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/driver.py#n4395 | 19:58 |
| alaski | sdague: hmm, nova policy fail is a 403 so that must be the rate limiting proxy in front | 19:58 |
| *** ctrath has joined #openstack-nova | 19:58 | |
| sdague | alaski: it is? 401 is unauthorized | 19:58 |
| apmelton | mriedem: I think that's correct, we wan to unmount the rootfs from the host's mount namespace, but leave the nbd device attached | 19:59 |
| sdague | maybe it's set admin only or something | 19:59 |
| alaski | sdague: yeah, we return 403 Forbidden for policy fails | 19:59 |
| mriedem | apmelton: right, and that's what you were wondering about earlier right>/ | 19:59 |
| sdague | alaski: ok, so then something else | 19:59 |
| apmelton | yea, I'm kinda wondering if something is attempting to disconnect the nbd device while the instance is still running, and that crashes qemu-nbd | 20:00 |
| *** erhudy has joined #openstack-nova | 20:00 | |
| sdague | any chance you could figure out why you are blocking that request with an unauthorized | 20:00 |
| alaski | sdague: yeah. there's also no policy check there | 20:00 |
| alaski | sdague: I'll look into it | 20:00 |
| alaski | it's likely repose, which we run in front of Nova | 20:00 |
| sdague | especially given that your service catalog actually says it's where your version info is | 20:00 |
| mriedem | apmelton: note that i never see this in the logs: | 20:01 |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/disk/api.py#n478 | 20:01 |
| sdague | anyway, we should also probably talk about this at the nova meeting tomorrow | 20:02 |
| melwitt | sdague alaski: looking at the paste again, it looks like it's showing a 401 from repose with keystone uri, doesn't that imply it's getting it from keystone? | 20:03 |
| alaski | melwitt: there's a paste? | 20:04 |
| melwitt | alaski: http://paste.openstack.org/show/442204/ | 20:04 |
| *** ildikov has joined #openstack-nova | 20:04 | |
| sdague | melwitt: no, on line 17 it got the new token | 20:05 |
| *** ssurana has joined #openstack-nova | 20:05 | |
| sdague | on 18 it's using that to make the request to nova | 20:05 |
| mriedem | apmelton: yeah so we're totally trying to mount the nbd device twice | 20:05 |
| mriedem | 1. http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_10_49_893 | 20:05 |
| *** Sree has quit IRC | 20:06 | |
| *** baoli has quit IRC | 20:06 | |
| mriedem | 2. http://logs.openstack.org/48/219448/1/check/gate-tempest-dsvm-full/bc0fc2a/logs/screen-n-cpu.txt.gz#_2015-09-01_22_10_50_024 | 20:06 |
| mriedem | well, i guess that's not two mounts | 20:07 |
| mriedem | qemu-nbd -c | 20:07 |
| mriedem | and then mount | 20:07 |
| alaski | melwitt: thanks. forwarded the paste to the repose expert here. since I dislike dealing with that project | 20:07 |
| melwitt | it's almost as if the token it got back wasn't valid somehow. invalidated. I don't see in the versions extension how it can return 401, I don't see any policy enforcement in it | 20:09 |
| alaski | it's possible it wasn't valid, but repose does it's own policy enforcement and it's possible it's (mis)configured to deny that url for some reason | 20:10 |
| melwitt | oh, I see | 20:11 |
| alaski | but it's configured via 4 or 5 configs and outputs three different logs, so I will ask for help before I go digging in | 20:11 |
| *** Sukhdev_ has quit IRC | 20:11 | |
| melwitt | understandable :) | 20:12 |
| *** ssurana has quit IRC | 20:13 | |
| sdague | melwitt: no, its policy enforcement | 20:14 |
| *** vilobhmm_11 has quit IRC | 20:14 | |
| sdague | apparently because we didn't have tooling that actually touched that url before, a lot of clouds have policied it off | 20:15 |
| *** vilobhmm_11 has joined #openstack-nova | 20:15 | |
| *** lbeliveau_ has joined #openstack-nova | 20:15 | |
| *** lbeliveau_ has quit IRC | 20:15 | |
| *** lbeliveau has left #openstack-nova | 20:15 | |
| melwitt | mriedem: earlier today I closed a bug that got fixed by the move to os-brick, I think there are probably a lot of those -- I'm not sure how to know which ones we can close since that merged | 20:16 |
| mriedem | melwitt: as in the bug was fixed in os-brick and we're using os-brick now? | 20:17 |
| mriedem | i guess it would depend on the bug, b/c we're not using os-brick for all libvirt volume drivers | 20:17 |
| *** lbeliveau has joined #openstack-nova | 20:17 | |
| melwitt | sdague: at an extension level (policy off the extension)? I didn't see an entry for the versions extension in our default policy.json in tree so I wasn't sure | 20:17 |
| sdague | melwitt: sorry, not nova policy | 20:18 |
| sdague | some site level thing | 20:18 |
| sdague | mordred is dumping data from all the clouds | 20:18 |
| melwitt | sdague: oh, right, okay. I misunderstood | 20:18 |
| sdague | apparently nova version-list doesn't work on rax either | 20:19 |
| sdague | even in the old client | 20:19 |
| sdague | probably for the same reasons | 20:19 |
| melwitt | ah, interesting | 20:19 |
| *** lbeliveau has quit IRC | 20:19 | |
| *** Marga_ has joined #openstack-nova | 20:20 | |
| *** markvoelker has quit IRC | 20:21 | |
| mordred | http://paste.openstack.org/show/442335/ | 20:21 |
| mordred | there's the dump of what I've tested/learned so far across the clouds I have accounts on | 20:22 |
| *** mwagner_lap has quit IRC | 20:23 | |
| *** ildikov has quit IRC | 20:23 | |
| *** rushil has joined #openstack-nova | 20:25 | |
| *** lpetrut has joined #openstack-nova | 20:25 | |
| mordred | alaski: and yes - I am getting the errors from Repose on rax | 20:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Scheduler RPC API select_dests() https://review.openstack.org/199205 | 20:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify left filters for RequestSpec https://review.openstack.org/202678 | 20:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Prepare filters for using RequestSpec object https://review.openstack.org/197912 | 20:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify conductor to use RequestSpec object https://review.openstack.org/203078 | 20:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify metric-related filters for RequestSpec https://review.openstack.org/202676 | 20:25 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Use RequestSpec in the ChanceScheduler https://review.openstack.org/199051 | 20:26 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify NUMA, PCI and num_instances filters for RequestSpec https://review.openstack.org/202677 | 20:26 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Aggregate filters for RequestSpec https://review.openstack.org/202674 | 20:26 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify default filters for RequestSpec https://review.openstack.org/202675 | 20:26 |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Modify Affinity filter for RequestSpec https://review.openstack.org/202673 | 20:26 |
| *** smatzek has quit IRC | 20:28 | |
| mordred | http://paste.openstack.org/show/442384/ <-- more complex - added runabove, was more verbose about rackspace | 20:29 |
| *** rushil has quit IRC | 20:30 | |
| openstackgerrit | jichenjc proposed openstack/nova-specs: Add display-flavor-id-in-migration.rst https://review.openstack.org/219431 | 20:30 |
| *** bswartz has quit IRC | 20:31 | |
| mriedem | apmelton: i traced through the logs in one of these failed instance file deletes | 20:32 |
| mriedem | https://bugs.launchpad.net/nova/+bug/1491563/comments/2 | 20:32 |
| openstack | Launchpad bug 1491563 in OpenStack Compute (nova) "Failed to spawn lxc instance due to nbd mount in use or rootfs busy" [Undecided,New] | 20:32 |
| sdague | mordred: is there a bug yet? | 20:32 |
| mriedem | apmelton: i see the mount/unmount happen to the nbd device, and the qemu-nbd -d before trying to delete the rootfs | 20:32 |
| mriedem | apmelton: so i'm not sure what's causing the EBUSY unless qemu-nbd -d is async | 20:32 |
| melwitt | mriedem: yeah, in this case it was the find logic for removing a device and in os-brick the logic is different and doesn't have the bug, from what I understand https://bugs.launchpad.net/nova/+bug/1454512 | 20:32 |
| openstack | Launchpad bug 1454512 in OpenStack Compute (nova) "Device for other volume is deleted unexpected during volume detach when iscsi multipath is used " [Medium,Fix committed] | 20:32 |
| mordred | sdague: want I should file one? that'll be 2 nova bugs in one day for me | 20:33 |
| sdague | if you could, that would be great | 20:33 |
| mordred | on it | 20:34 |
| mordred | sdague: I'm not going to be polite in my words | 20:34 |
| sdague | mordred: I will appreciate the non politeness | 20:34 |
| *** vilobhmm_11 has quit IRC | 20:34 | |
| mriedem | apmelton: https://lists.nongnu.org/archive/html/qemu-devel/2011-12/msg00734.html seems to suggest qemu-nbd is asynchronous | 20:35 |
| *** markvoelker has joined #openstack-nova | 20:37 | |
| mordred | sdague: https://bugs.launchpad.net/nova/+bug/1491579 | 20:38 |
| openstack | Launchpad bug 1491579 in OpenStack Compute (nova) "against all sanity, nova needs to work around broken public clouds" [Undecided,New] | 20:38 |
| *** rfolco has quit IRC | 20:38 | |
| sdague | mordred: love it | 20:38 |
| mriedem | nice | 20:39 |
| sdague | although, it's really a novaclient bug, but that's fine | 20:39 |
| mordred | sdague: good point | 20:39 |
| mriedem | updated | 20:39 |
| sdague | I'm fixing | 20:39 |
| sdague | or mriedem is, because he's fast like that | 20:39 |
| sdague | I'm marking critical on the novaclient side | 20:40 |
| edleafe | mordred: I like | 20:40 |
| mriedem | yeah | 20:40 |
| mriedem | sdague: do you have a fix in mind? | 20:40 |
| mriedem | or waiting for the meeting in the morning? | 20:40 |
| dansmith | mordred: to be honest, I'm a bit disappointed | 20:40 |
| mordred | dansmith: was I not caustic enough? | 20:40 |
| dansmith | yeah.. let me go get a link to a good example of what I was expecting | 20:41 |
| mordred | dansmith: I think I've just become too resigned to everything in the world being broken all the time. it's hard to get properly righteously indignant anymore | 20:41 |
| mordred | dansmith: you're going to get the heat patch aren't you? | 20:41 |
| dansmith | mordred: no, but while I find this link, please dig up the heat patch for me to grade | 20:41 |
| sdague | mriedem: yeh, I have half a plan | 20:42 |
| sdague | so, a good chunk of the clouds actually respond to version list | 20:42 |
| sdague | which means we can fail up to that url which has the same info | 20:42 |
| sdague | and then the ones that don't, just assume default 2.0 | 20:42 |
| *** raildo is now known as raildo-afk | 20:43 | |
| *** raildo-afk is now known as raildo | 20:43 | |
| mordred | dansmith: https://review.openstack.org/#/c/122962/1//COMMIT_MSG <-- you have to look at PS#1 because I got like 100 -2's on it | 20:43 |
| sdague | mordred: did any cloud "hang" on version-list? or just failed | 20:43 |
| mordred | sdague: many hung | 20:43 |
| sdague | on 2.26? | 20:43 |
| mordred | sdague: runabove hangs on version-list on 2.26 | 20:44 |
| mordred | everyone else either works or fails | 20:44 |
| mordred | on 2.27, hp, dreamhost and runabove all hang on /v2 | 20:44 |
| *** matrohon has quit IRC | 20:44 | |
| sdague | ok, so, we'll have to do this with short timeouts and cache the results to be less annoying | 20:44 |
| *** Piet has quit IRC | 20:45 | |
| sdague | mriedem: anyway, seem reasonable? | 20:45 |
| *** electrocucaracha has quit IRC | 20:45 | |
| sdague | I'm about to end the day, because i've got a LUG meeting, but I'll start hacking on this in the morning | 20:45 |
| *** harlowja has quit IRC | 20:45 | |
| mriedem | over my head but sure | 20:46 |
| mriedem | i was reading mordred's heat patch | 20:46 |
| sdague | mriedem: I fail to believe that :) | 20:46 |
| mriedem | i was just going to add it to the nova meeting agenda for the morning | 20:46 |
| *** electrocucaracha has joined #openstack-nova | 20:46 | |
| sdague | mriedem: yeh, it should be on the agenda for sure | 20:46 |
| *** electrocucaracha has quit IRC | 20:47 | |
| mriedem | i'm spending most of my time trying to figure out this lxc/nbd mess :) | 20:47 |
| *** signed8bit has joined #openstack-nova | 20:47 | |
| mriedem | i will say, this has been valuable for me to read through the virt disk api and mount stuff | 20:47 |
| *** markvoelker has quit IRC | 20:48 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add persistence to the RequestSpec object https://review.openstack.org/211753 | 20:49 |
| *** tjones has left #openstack-nova | 20:52 | |
| mriedem | apmelton: on the plus side, i think snapshot works | 20:53 |
| dansmith | well, I can't find the epic monty email | 20:53 |
| mriedem | even though the hypervisor support matrix ssays it isn't http://docs.openstack.org/developer/nova/support-matrix.html#operation_snapshot_libvirt_vz_ct | 20:53 |
| dansmith | mordred: but yeah, the heat patch is pretty good | 20:53 |
| *** ociuhandu has quit IRC | 20:54 | |
| mriedem | sdague: do you think we should do a 2.27.1 novaclient release for https://review.openstack.org/#/c/219835/ ? seems that's needed to unblock manila | 20:55 |
| mriedem | i.e. do that before waiting for a solution to this version list tihng | 20:55 |
| apmelton | mriedem: sorry, just got out of another meeting | 20:55 |
| mriedem | apmelton: are you buy chance a manager? | 20:55 |
| mriedem | *by | 20:55 |
| apmelton | ha | 20:56 |
| apmelton | no, but I feel like I should be one with the amount of meetings and sync I've been in lately | 20:56 |
| *** ccarmack has left #openstack-nova | 20:59 | |
| *** ctrath has quit IRC | 20:59 | |
| *** armax has quit IRC | 21:00 | |
| *** armax has joined #openstack-nova | 21:00 | |
| *** ctrath has joined #openstack-nova | 21:02 | |
| *** lpetrut has quit IRC | 21:02 | |
| anish | mriedem: this is caught in the cracks https://blueprints.launchpad.net/nova/+spec/add-transport-support-to-iscsi | 21:02 |
| anish | can you close out that BP ? | 21:02 |
| anish | it was implemented before kilo | 21:02 |
| mriedem | anish: i superseded the blueprint | 21:04 |
| mriedem | in lp | 21:04 |
| mriedem | which is apparently a thing you can do | 21:04 |
| anish | mriedem: which also works ! thanks ! | 21:04 |
| apmelton | mriedem: finally catching up | 21:04 |
| *** pixelb has joined #openstack-nova | 21:05 | |
| apmelton | mriedem: so qemu-nbd -d appears to just exit 0, whether it works or not? | 21:06 |
| mriedem | apmelton: well it's exiting 0 for sure | 21:06 |
| bauzas | alaski: meh, it seems that you also missed the sched_hints serialization in https://review.openstack.org/#/c/211753/24/nova/objects/request_spec.py,cm | 21:06 |
| mriedem | i'm just not sure if it's async | 21:06 |
| mriedem | apmelton: i probably need to ask danpb about that | 21:06 |
| *** thangp_ has quit IRC | 21:07 | |
| *** raildo is now known as raildo-afk | 21:07 | |
| *** baoli has joined #openstack-nova | 21:07 | |
| alaski | bauzas: now you point that out | 21:07 |
| bauzas | alaski: yeah sorry I missed it when I reviewed the last patch | 21:08 |
| *** baoli has quit IRC | 21:08 | |
| *** ctrath has quit IRC | 21:08 | |
| *** baoli has joined #openstack-nova | 21:08 | |
| bauzas | alaski: gimme a few more mins, just verifying all the fields | 21:08 |
| *** achanda has quit IRC | 21:08 | |
| *** ctrath has joined #openstack-nova | 21:09 | |
| sdague | mriedem: I'm fine with that | 21:09 |
| *** thorst has quit IRC | 21:09 | |
| sdague | this other thing is going to take a couple of days to get right I think | 21:10 |
| openstackgerrit | Ryan Rossiter proposed openstack/nova: Use o.vo fingerprint generation https://review.openstack.org/219880 | 21:10 |
| sdague | and we should unblock them asap | 21:10 |
| mriedem | yeah so i'll monitor that merge tonight and do the release | 21:11 |
| mriedem | at least the release request | 21:11 |
| *** ctrath has quit IRC | 21:11 | |
| sdague | mriedem: cool | 21:12 |
| *** Sukhdev_ has joined #openstack-nova | 21:13 | |
| *** dave-mccowan has quit IRC | 21:14 | |
| *** unicell has quit IRC | 21:14 | |
| *** unicell has joined #openstack-nova | 21:14 | |
| dansmith | sdague: why do you feel that novaclient bug needs to be critical? | 21:14 |
| *** Marga_ has quit IRC | 21:15 | |
| mriedem | apmelton: interesting https://bugs.launchpad.net/ubuntu/+source/qemu/+bug/1435428 | 21:16 |
| openstack | Launchpad bug 1435428 in qemu (Ubuntu) "disconnecting qemu-nbd leaves device node behind" [Medium,Fix released] | 21:16 |
| *** Marga_ has joined #openstack-nova | 21:16 | |
| mriedem | although that's marked as a regression in vivid and we're testing on trusty | 21:17 |
| *** jwcroppe_ has joined #openstack-nova | 21:17 | |
| *** chutwig has joined #openstack-nova | 21:18 | |
| *** tonytan4ever has quit IRC | 21:18 | |
| *** jwcroppe has quit IRC | 21:19 | |
| *** dims has quit IRC | 21:19 | |
| *** baoli has quit IRC | 21:20 | |
| openstackgerrit | Andrew Laski proposed openstack/nova: Add persistence to the RequestSpec object https://review.openstack.org/211753 | 21:21 |
| *** jichen has quit IRC | 21:21 | |
| * alaski waits for bauzas to find another thing | 21:21 | |
| *** harlowja has joined #openstack-nova | 21:21 | |
| *** edmondsw has quit IRC | 21:21 | |
| bauzas | alaski: :( | 21:22 |
| *** chutwig is now known as erhudy1 | 21:22 | |
| alaski | bauzas: it's great that you're finding it in review | 21:22 |
| alaski | not good that I'm leaving them in there | 21:22 |
| bauzas | alaski: I just -1'd this morning when seeing the miss of force_hosts without having taken time to correctly review the other fields, my bad for that :/ | 21:23 |
| alaski | bauzas: the negligence was mine | 21:24 |
| *** aysyd has quit IRC | 21:25 | |
| *** ctrath has joined #openstack-nova | 21:25 | |
| *** tonytan4ever has joined #openstack-nova | 21:26 | |
| *** vilobhmm_11 has joined #openstack-nova | 21:29 | |
| *** ociuhandu has joined #openstack-nova | 21:29 | |
| *** ociuhandu has quit IRC | 21:30 | |
| *** ociuhandu has joined #openstack-nova | 21:31 | |
| *** ociuhandu has quit IRC | 21:32 | |
| *** derekh has joined #openstack-nova | 21:33 | |
| *** ociuhandu has joined #openstack-nova | 21:33 | |
| mriedem | vponomaryov: i take it that screen-m-shr.txt is not indexed in logstash.o.o? | 21:35 |
| *** lucas-dinner has quit IRC | 21:36 | |
| bauzas | alaski: FWIW, +1'd | 21:37 |
| *** achanda has joined #openstack-nova | 21:37 | |
| alaski | bauzas: thanks | 21:37 |
| bauzas | g'night, folks | 21:39 |
| *** dannywil_ has quit IRC | 21:40 | |
| *** karimb has joined #openstack-nova | 21:40 | |
| *** marzif has joined #openstack-nova | 21:44 | |
| *** dave-mccowan has joined #openstack-nova | 21:45 | |
| *** baoli has joined #openstack-nova | 21:46 | |
| *** dannywilson has joined #openstack-nova | 21:47 | |
| *** achanda has quit IRC | 21:47 | |
| *** electrocucaracha has joined #openstack-nova | 21:47 | |
| *** ftersin has quit IRC | 21:51 | |
| *** achanda has joined #openstack-nova | 21:54 | |
| *** rlrossit has left #openstack-nova | 21:55 | |
| *** electrocucaracha has quit IRC | 21:55 | |
| *** mriedem has quit IRC | 21:56 | |
| *** mtanino has quit IRC | 21:57 | |
| *** zzzeek has quit IRC | 21:58 | |
| *** burgerk_ has quit IRC | 21:59 | |
| *** bkopilov has quit IRC | 22:01 | |
| *** sachin_ has quit IRC | 22:01 | |
| *** hemna is now known as hemnafk | 22:02 | |
| *** dustins has quit IRC | 22:03 | |
| *** mtanino has joined #openstack-nova | 22:05 | |
| *** bkopilov has joined #openstack-nova | 22:06 | |
| *** Sree has joined #openstack-nova | 22:07 | |
| *** IanGovett has quit IRC | 22:07 | |
| *** signed8bit has quit IRC | 22:08 | |
| *** IanGovett has joined #openstack-nova | 22:08 | |
| *** annegentle has quit IRC | 22:08 | |
| mikal | Morning | 22:08 |
| *** jaypipes has quit IRC | 22:10 | |
| *** whenry_ has quit IRC | 22:11 | |
| *** Sree has quit IRC | 22:12 | |
| *** IanGovett has quit IRC | 22:12 | |
| *** rushil has joined #openstack-nova | 22:13 | |
| *** jcoufal has quit IRC | 22:14 | |
| *** shoutm has quit IRC | 22:14 | |
| *** electrocucaracha has joined #openstack-nova | 22:14 | |
| *** rushil_ has joined #openstack-nova | 22:14 | |
| *** rushil has quit IRC | 22:18 | |
| *** marzif has quit IRC | 22:18 | |
| *** diogogmt has quit IRC | 22:19 | |
| *** bkopilov has quit IRC | 22:22 | |
| *** achanda has quit IRC | 22:24 | |
| *** achanda_ has joined #openstack-nova | 22:24 | |
| *** sachin_ has joined #openstack-nova | 22:24 | |
| *** tonytan4ever has quit IRC | 22:24 | |
| *** jecarey has quit IRC | 22:25 | |
| *** alex_klimov has quit IRC | 22:25 | |
| *** jecarey has joined #openstack-nova | 22:25 | |
| *** emagana has joined #openstack-nova | 22:26 | |
| *** eharney has quit IRC | 22:26 | |
| *** thorst has joined #openstack-nova | 22:27 | |
| *** Marga_ has quit IRC | 22:27 | |
| *** vladikr has quit IRC | 22:28 | |
| *** bkopilov has joined #openstack-nova | 22:28 | |
| *** hemnafk is now known as hemna | 22:28 | |
| *** tpeoples has quit IRC | 22:28 | |
| *** electrocucaracha has quit IRC | 22:29 | |
| *** zhangjn has joined #openstack-nova | 22:32 | |
| *** VW__ has joined #openstack-nova | 22:35 | |
| *** dannywilson has quit IRC | 22:35 | |
| *** zhangjn has quit IRC | 22:36 | |
| *** mdrabe has quit IRC | 22:36 | |
| *** dannywilson has joined #openstack-nova | 22:37 | |
| *** mwagner_lap has joined #openstack-nova | 22:37 | |
| *** VW_ has quit IRC | 22:39 | |
| *** VW__ has quit IRC | 22:40 | |
| *** diogogmt has joined #openstack-nova | 22:40 | |
| *** unicell has quit IRC | 22:43 | |
| *** unicell has joined #openstack-nova | 22:43 | |
| *** haomaiwang has joined #openstack-nova | 22:43 | |
| *** unicell has quit IRC | 22:44 | |
| *** unicell has joined #openstack-nova | 22:44 | |
| *** thorst has quit IRC | 22:45 | |
| *** haomaiwa_ has quit IRC | 22:46 | |
| *** sachin_ has quit IRC | 22:47 | |
| *** electrocucaracha has joined #openstack-nova | 22:47 | |
| *** sachin_ has joined #openstack-nova | 22:48 | |
| *** hemna is now known as hemnafk | 22:49 | |
| *** ijuwang has quit IRC | 22:50 | |
| *** derekh has quit IRC | 22:50 | |
| *** mriedem has joined #openstack-nova | 22:55 | |
| *** diogogmt has quit IRC | 22:55 | |
| *** asettle has joined #openstack-nova | 22:57 | |
| *** mtanino_ has joined #openstack-nova | 22:59 | |
| *** ctrath has quit IRC | 23:01 | |
| *** mtanino has quit IRC | 23:02 | |
| mtreinish | oomichi, mriedem: qq is there a way we can have a GET on v2/${tenant_id} act the same as a GET on v2? | 23:02 |
| mtreinish | I'm just trying to figure out the best fix for: http://logs.openstack.org/73/219873/1/check/gate-tempest-dsvm-full/61dffed/console.html#_2015-09-02_22_15_32_439 | 23:02 |
| *** mriedem has quit IRC | 23:03 | |
| *** mriedem has joined #openstack-nova | 23:04 | |
| *** sachin_ has quit IRC | 23:07 | |
| *** sachin_ has joined #openstack-nova | 23:07 | |
| *** achanda_ has quit IRC | 23:12 | |
| *** Sukhdev_ has quit IRC | 23:13 | |
| *** baoli has quit IRC | 23:14 | |
| *** Piet has joined #openstack-nova | 23:17 | |
| *** erhudy has quit IRC | 23:19 | |
| *** baoli has joined #openstack-nova | 23:20 | |
| *** achanda has joined #openstack-nova | 23:21 | |
| *** baoli has quit IRC | 23:25 | |
| melwitt | do we consider something like this a bug, where we trace on excessive values that libvirt for example doesn't support? https://bugs.launchpad.net/nova/+bug/1407643 that is, it seems like the fix would be to do api level validation on the inputs but how would we choose what min/max should be? | 23:26 |
| openstack | Launchpad bug 1407643 in OpenStack Compute (nova) "Setting network bandwidth quota in extra_specs causes a VM creation to fail in devstack" [Low,In progress] - Assigned to Radoslaw Smigielski (radoslaw-smigielski) | 23:26 |
| *** nic has quit IRC | 23:28 | |
| *** moshele has joined #openstack-nova | 23:31 | |
| *** dannywilson has quit IRC | 23:31 | |
| *** krotscheck is now known as kro_paternity | 23:34 | |
| *** ndipanov has quit IRC | 23:35 | |
| *** sachin_ has quit IRC | 23:37 | |
| *** sachin_ has joined #openstack-nova | 23:38 | |
| *** sachin_ has quit IRC | 23:38 | |
| *** sachin_ has joined #openstack-nova | 23:39 | |
| *** baoli has joined #openstack-nova | 23:40 | |
| *** harlowja has quit IRC | 23:42 | |
| *** harlowja has joined #openstack-nova | 23:43 | |
| *** morgabra has quit IRC | 23:45 | |
| *** ZZelle_ has quit IRC | 23:45 | |
| *** morgabra has joined #openstack-nova | 23:45 | |
| mriedem | melwitt: you can't validate that in the api probably | 23:46 |
| *** markvoelker has joined #openstack-nova | 23:46 | |
| mriedem | because it doesn't fail until it hinds a compute host to run on and then libvirt pukes on it during spawn | 23:46 |
| *** geekinutah has quit IRC | 23:46 | |
| melwitt | mriedem: that's what I was thinking | 23:46 |
| mriedem | the best you could probably hope for is a reschedule | 23:47 |
| melwitt | I was thinking to mark it Invalid but started thinking if there was any way to "fix" that | 23:47 |
| mriedem | and that the libvirt error gets into the instance fault so the user sees why the boot failed | 23:47 |
| melwitt | yeah, I was thinking that too. it's not that they wouldn't see | 23:47 |
| mriedem | the fix would probably just be making sure they know why their request failed | 23:48 |
| mriedem | is possible | 23:48 |
| melwitt | mriedem: your novaclient patch hit that same failure again, it's been rechecked again | 23:48 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/216968 | 23:48 |
| mriedem | which is probably only through instance faults once it's on the compute host | 23:48 |
| mriedem | melwitt: awesome | 23:48 |
| mriedem | maybe it'll all be blocked | 23:48 |
| melwitt | mriedem: right (instance faults) | 23:49 |
| *** shoutm has joined #openstack-nova | 23:49 | |
| dansmith | mriedem: the proposal bot is pissing me off | 23:50 |
| dansmith | mriedem: so, we can kick my change out after 7.5h of grinding, | 23:50 |
| dansmith | or we can just let it go | 23:51 |
| oomichi | dansmith, alaski: hi, can you take a look at https://review.openstack.org/#/c/207029/ ? | 23:51 |
| melwitt | dansmith: is it still re-proposing before merge is possible? | 23:51 |
| dansmith | melwitt: yes | 23:51 |
| mriedem | yeah it seems to auto-rebase | 23:51 |
| dansmith | mriedem: for what it's worth, we already merged the first patch that requires 0.9.0 anyway, so holding this one hostage is only of medium value | 23:51 |
| mriedem | dansmith: which change? | 23:51 |
| mriedem | dansmith: btw, did you guys actually talk about what to do about this busted novaclient thing or just punt and make fart jokes for an hour> | 23:52 |
| mriedem | ? | 23:52 |
| dansmith | gdi you're going to make me look it up aren't you... | 23:52 |
| *** achanda has quit IRC | 23:52 | |
| mriedem | yes, i'm busy eating chips with the child | 23:52 |
| dansmith | mriedem: we did, but I mean we can't really make a call I don't think | 23:52 |
| mriedem | so no 2.28 tag from 2.26? | 23:52 |
| mriedem | that seems messy to me | 23:53 |
| mriedem | we can certainly go ahead with blacklisting 2.27 in g-r | 23:53 |
| dansmith | mriedem: https://review.openstack.org/#/c/217382/ | 23:53 |
| *** takedakn has joined #openstack-nova | 23:53 | |
| dansmith | mriedem: if you do that in g-r you'll reset the proposal bot nova change again :) | 23:53 |
| mriedem | nova doesn't require novaclient | 23:53 |
| dansmith | mriedem: clarkb would argue that doing that fixes it for us, but not for the rest of the world | 23:53 |
| dansmith | oh, right :) | 23:54 |
| mriedem | yeah i know, it's not a fix, but it's a thing to do anyway | 23:54 |
| mriedem | since manila doesn't work with 2.27 | 23:54 |
| mriedem | actually, <2.27 | 23:54 |
| dansmith | yeah, I guess that works to unblock manila | 23:54 |
| mriedem | <= | 23:54 |
| dansmith | if you think that's worth it | 23:54 |
| mriedem | i still plan on doing 2.27.1 with that earlier fix for manila | 23:55 |
| dansmith | oh, manila is just the block device thing, right? | 23:55 |
| mriedem | yes | 23:55 |
| dansmith | yeah, so do that to fix them | 23:55 |
| dansmith | and we can talk about the bigger thing in the meeting tomorrow | 23:55 |
| *** IanGovett has joined #openstack-nova | 23:55 | |
| mriedem | yeah | 23:55 |
| mriedem | actually it's not really novaclient that was the issue, it was that v2.0 on v2.1 change | 23:56 |
| mriedem | but novaclient had a bug for sure - just wasn't exposed until that server side change | 23:56 |
| *** karimb has quit IRC | 23:57 | |
| dansmith | oomichi: got it | 23:58 |
| oomichi | dansmith: thanks! | 23:58 |
| dansmith | np | 23:58 |
| *** markvoelker has quit IRC | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
