| *** dave-mccowan has joined #openstack-nova | 00:00 | |
| *** diogogmt has quit IRC | 00:02 | |
| *** promulo__ has joined #openstack-nova | 00:02 | |
| *** shoutm has joined #openstack-nova | 00:02 | |
| *** sneti has quit IRC | 00:03 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Add timing logging around each filter being ran https://review.openstack.org/224355 | 00:04 |
|---|---|---|
| *** promulo has quit IRC | 00:06 | |
| *** markvoelker has joined #openstack-nova | 00:06 | |
| *** aysyd has quit IRC | 00:07 | |
| *** zhangjn has quit IRC | 00:09 | |
| *** mdbooth has quit IRC | 00:10 | |
| *** stevemar has joined #openstack-nova | 00:11 | |
| *** markvoelker has quit IRC | 00:11 | |
| *** stevemar has quit IRC | 00:15 | |
| *** devlaps has quit IRC | 00:15 | |
| *** apuimedo has quit IRC | 00:16 | |
| *** mdbooth has joined #openstack-nova | 00:16 | |
| *** tsekiyama has quit IRC | 00:20 | |
| *** mtanino has quit IRC | 00:20 | |
| openstackgerrit | Atsushi SAKAI proposed openstack/python-novaclient: Fix nova bash-completion needs authentication https://review.openstack.org/223012 | 00:21 |
| *** achanda_ has quit IRC | 00:31 | |
| *** diogogmt has joined #openstack-nova | 00:32 | |
| *** achanda has joined #openstack-nova | 00:33 | |
| *** tpeoples has quit IRC | 00:34 | |
| *** ssurana has quit IRC | 00:35 | |
| *** achanda has quit IRC | 00:38 | |
| *** diogogmt has quit IRC | 00:41 | |
| *** diogogmt has joined #openstack-nova | 00:42 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Iteratively generate host states from 'get_all_host_states' https://review.openstack.org/224373 | 00:44 |
| *** otter768 has joined #openstack-nova | 00:46 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Iteratively generate host states from 'get_all_host_states' https://review.openstack.org/224373 | 00:46 |
| *** otter768 has quit IRC | 00:51 | |
| *** ijw_ has quit IRC | 00:54 | |
| *** EinstCrazy has joined #openstack-nova | 00:55 | |
| *** Apoorva has quit IRC | 00:56 | |
| *** pixelb has quit IRC | 00:57 | |
| *** piyanai has joined #openstack-nova | 00:58 | |
| *** zhangjn has joined #openstack-nova | 00:58 | |
| *** dims_ has quit IRC | 01:01 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Iteratively generate host states from 'get_all_host_states' https://review.openstack.org/224373 | 01:01 |
| *** browne has quit IRC | 01:03 | |
| *** veena has quit IRC | 01:04 | |
| *** piyanai has quit IRC | 01:04 | |
| *** Jeffrey4l has joined #openstack-nova | 01:06 | |
| *** markvoelker has joined #openstack-nova | 01:06 | |
| *** vilobhmm_11 has quit IRC | 01:06 | |
| *** zhangjn has quit IRC | 01:07 | |
| *** Daisy has joined #openstack-nova | 01:07 | |
| *** zzzeek has quit IRC | 01:09 | |
| *** piyanai has joined #openstack-nova | 01:09 | |
| openstackgerrit | Tony Breeds proposed openstack/nova-specs: Revises Documentation impact section in mitaka template https://review.openstack.org/198337 | 01:10 |
| *** Daisy has quit IRC | 01:11 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Add timing logging around each filter being ran https://review.openstack.org/224355 | 01:17 |
| *** achanda has joined #openstack-nova | 01:22 | |
| *** kjelly has joined #openstack-nova | 01:25 | |
| *** dims_ has joined #openstack-nova | 01:26 | |
| *** yamahata has quit IRC | 01:27 | |
| *** zhenguo has joined #openstack-nova | 01:31 | |
| *** diogogmt has quit IRC | 01:32 | |
| *** thorst has quit IRC | 01:33 | |
| *** diogogmt has joined #openstack-nova | 01:33 | |
| *** sneti has joined #openstack-nova | 01:33 | |
| *** zhangjn has joined #openstack-nova | 01:36 | |
| *** xyang1 has quit IRC | 01:39 | |
| *** achanda has quit IRC | 01:41 | |
| *** 7F1AALPHR has joined #openstack-nova | 01:43 | |
| *** edmondsw has quit IRC | 01:52 | |
| *** EinstCrazy has quit IRC | 01:53 | |
| *** EinstCrazy has joined #openstack-nova | 01:53 | |
| *** shoutm has quit IRC | 01:55 | |
| *** VW_ has joined #openstack-nova | 01:59 | |
| *** 7F1AALPHR has quit IRC | 02:01 | |
| *** haomaiwang has joined #openstack-nova | 02:01 | |
| *** r-daneel has quit IRC | 02:02 | |
| *** Jeffrey4l has quit IRC | 02:06 | |
| openstackgerrit | ZhuChunzhan proposed openstack/nova: Improve the performance of get instances https://review.openstack.org/214555 | 02:06 |
| *** VW_ has quit IRC | 02:11 | |
| *** unicell has quit IRC | 02:11 | |
| *** VW_ has joined #openstack-nova | 02:11 | |
| *** stevemar has joined #openstack-nova | 02:12 | |
| *** chenzeng has joined #openstack-nova | 02:13 | |
| *** shihanzhang has quit IRC | 02:13 | |
| *** browne has joined #openstack-nova | 02:13 | |
| *** shihanzhang has joined #openstack-nova | 02:13 | |
| *** VW_ has quit IRC | 02:16 | |
| *** stevemar has quit IRC | 02:16 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: WIP: Fix to run v2.1 comp mode tests with v2 sample files https://review.openstack.org/224386 | 02:17 |
| *** salv-orlando has joined #openstack-nova | 02:18 | |
| *** yamahata has joined #openstack-nova | 02:18 | |
| *** shoutm has joined #openstack-nova | 02:19 | |
| *** Jeffrey4l has joined #openstack-nova | 02:19 | |
| *** salv-orlando has quit IRC | 02:23 | |
| *** tpeoples has joined #openstack-nova | 02:26 | |
| *** humble__ has joined #openstack-nova | 02:30 | |
| *** Shangxdy has joined #openstack-nova | 02:31 | |
| *** humble_ has quit IRC | 02:33 | |
| *** david-lyle has joined #openstack-nova | 02:34 | |
| *** humble__ has quit IRC | 02:37 | |
| *** humble__ has joined #openstack-nova | 02:38 | |
| gmann | alex_xu_: johnthetubaguy : we have some more to fix for v2.1 comp mode - https://bugs.launchpad.net/nova/+bug/1496664 | 02:38 |
| openstack | Launchpad bug 1496664 in OpenStack Compute (nova) "V2.1 comp mode behavior should be fixed for diff of v2 and v2.1 APIs" [Undecided,New] | 02:38 |
| alex_xu_ | gmann: thanks! | 02:40 |
| alex_xu_ | let me check the detail | 02:40 |
| gmann | alex_xu_: ok | 02:40 |
| alex_xu_ | gmann: what is "extension info - namespace diff"? | 02:41 |
| *** achanda has joined #openstack-nova | 02:41 | |
| gmann | alex_xu_: https://github.com/openstack/nova/blob/master/nova/tests/functional/api_sample_tests/api_samples/all_extensions/extensions-list-resp-v2.json.tpl#L24 | 02:42 |
| openstackgerrit | felix23ma proposed openstack/nova: Fix bug: volume booted instance rebuild doesn't change operating system in root device https://review.openstack.org/223887 | 02:44 |
| gmann | alex_xu_: there are separate sample files for those cases in our sample tests - if you see https://review.openstack.org/#/c/224386/ | 02:44 |
| alex_xu_ | gmann: ah, I see now, thanks | 02:44 |
| *** Shangxdy has quit IRC | 02:44 | |
| alex_xu_ | gmann: the namespaces look like we need fix | 02:44 |
| gmann | alex_xu_: hummm | 02:45 |
| alex_xu_ | gmann: rate limits should be fine. if user add limit middleware to the wsgi stack, then it will work | 02:45 |
| gmann | alex_xu_: actually i m sure for v-f thing but rate limit and extension info am doubtful | 02:45 |
| *** achanda has quit IRC | 02:45 | |
| alex_xu_ | gmann: emm...yea, namespace just for xml, right? | 02:46 |
| gmann | alex_xu_: limit middleware to the wsgi stack ? you mean in paste.ini ? | 02:46 |
| alex_xu_ | gmann: yea | 02:46 |
| gmann | alex_xu_: but that would work as our v2.1 limit.py would not take care of that | 02:46 |
| alex_xu_ | gmann: I remember the limit controller take care of that | 02:47 |
| gmann | https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/limits.py | 02:47 |
| *** otter768 has joined #openstack-nova | 02:47 | |
| *** earlephilhower has quit IRC | 02:47 | |
| alex_xu_ | gmann: https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/limits.py#L45 | 02:48 |
| *** EinstCrazy has quit IRC | 02:48 | |
| gmann | alex_xu_: but env 'nova.limits' will not be set as done in v2 case - https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/legacy_v2/limits.py#L260 | 02:49 |
| alex_xu_ | gmann: if user add the middleware in the wsgi stack, the middleware will add that https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/legacy_v2/limits.py#L260 | 02:50 |
| alex_xu_ | gmann: what you mean 'set ad done'? | 02:51 |
| gmann | alex_xu_: let me chekc more may be m missing something | 02:52 |
| *** otter768 has quit IRC | 02:52 | |
| alex_xu_ | gmann: the vif is hard one :( | 02:52 |
| *** haomaiw__ has joined #openstack-nova | 02:52 | |
| gmann | alex_xu_: fix is simple just about to release | 02:53 |
| gmann | alex_xu_: but m not much happy to having those fix in our v2.1 code :( | 02:53 |
| *** haomaiwang has quit IRC | 02:53 | |
| alex_xu_ | gmann: yea :( | 02:53 |
| alex_xu_ | gmann: at least, that won't break the user, because we removed the entry from the extensions info | 02:55 |
| *** VW_ has joined #openstack-nova | 02:56 | |
| gmann | alex_xu_: yea, but with other bug fixes for v2.1, i honestly feeling VIF bug could have gone directly on v2.1 without microversion | 02:56 |
| gmann | alex_xu_: what you think | 02:56 |
| alex_xu_ | gmann: emm...yea, that is similar logic | 02:58 |
| gmann | alex_xu_: can we consider that now? | 02:58 |
| gmann | not sure johnthetubaguy opinion on this | 02:58 |
| alex_xu_ | gmann: at least the v2.1 compat can't equal to v2 | 02:59 |
| *** VikasC has joined #openstack-nova | 02:59 | |
| alex_xu_ | gmann: as we merged and split the some extensions, if user want to disable the vif-n extend attribut, they can't do that with v2.1 compat mode, as we will merge the extensions | 03:00 |
| *** mdrabe has joined #openstack-nova | 03:00 | |
| alex_xu_ | gmann: and we already bump version for add new attribute | 03:00 |
| *** haomaiw__ has quit IRC | 03:01 | |
| alex_xu_ | gmann: if add that back, what we can do for this https://review.openstack.org/#/c/217572/? | 03:01 |
| gmann | alex_xu_: hummm, but for this we need to fix the extension thing also for v2.1 comp mode | 03:01 |
| *** haomaiwa_ has joined #openstack-nova | 03:02 | |
| gmann | alex_xu_: that is latest microversion so can be removed but not sure lib-3 is released and it was there | 03:02 |
| alex_xu_ | gmann: for CD user, that is already released I think | 03:03 |
| *** achanda has joined #openstack-nova | 03:03 | |
| alex_xu_ | gmann: fix for 2.1, sounds not option for now | 03:05 |
| gmann | alex_xu_: humm, yea. m also not in much favor of reverting that microversion but we could have decide that time | 03:05 |
| gmann | alex_xu_: anyways so fixing for v2.1 comp mode we need to fix extension info also | 03:05 |
| gmann | alex_xu_: which should return the VIF extension? | 03:06 |
| *** garyk has quit IRC | 03:08 | |
| gmann | alex_xu_: catch you later. need to go for lunch | 03:08 |
| gmann | alex_xu_: I will put up that patches up and we can see how we ccan fix those for v2.1 comp mode | 03:08 |
| alex_xu_ | gmann: yea, thanks a lot | 03:09 |
| alex_xu_ | gmann: I will think about it more also | 03:09 |
| *** jecarey has joined #openstack-nova | 03:09 | |
| gmann | alex_xu_: thanks :) | 03:09 |
| alex_xu_ | gmann: np | 03:09 |
| *** diogogmt has quit IRC | 03:10 | |
| *** coolsvap is now known as coolsvap|holiday | 03:17 | |
| openstackgerrit | ZhuChunzhan proposed openstack/nova: Improve the performance of get instances https://review.openstack.org/214555 | 03:27 |
| *** EinstCrazy has joined #openstack-nova | 03:29 | |
| *** dave-mccowan has quit IRC | 03:33 | |
| *** claudiub has quit IRC | 03:33 | |
| *** lyanchih has joined #openstack-nova | 03:33 | |
| *** sneti has quit IRC | 03:35 | |
| *** achanda has quit IRC | 03:35 | |
| *** achanda has joined #openstack-nova | 03:35 | |
| *** VW_ has quit IRC | 03:36 | |
| *** VW_ has joined #openstack-nova | 03:36 | |
| *** penick has joined #openstack-nova | 03:40 | |
| *** VW_ has quit IRC | 03:40 | |
| *** garyk has joined #openstack-nova | 03:41 | |
| *** boris-42 has quit IRC | 03:50 | |
| *** shihanzhang has quit IRC | 03:51 | |
| *** boris-42 has joined #openstack-nova | 03:51 | |
| *** mdrabe has quit IRC | 03:53 | |
| *** vilobhmm_11 has joined #openstack-nova | 03:56 | |
| *** kumar has joined #openstack-nova | 03:57 | |
| *** yuner027 has quit IRC | 03:58 | |
| *** yuner027 has joined #openstack-nova | 03:58 | |
| *** haomaiwa_ has quit IRC | 04:01 | |
| *** piyanai has quit IRC | 04:01 | |
| *** haomaiwa_ has joined #openstack-nova | 04:02 | |
| *** piyanai has joined #openstack-nova | 04:02 | |
| *** jecarey has quit IRC | 04:03 | |
| *** unicell has joined #openstack-nova | 04:03 | |
| *** yamahata has quit IRC | 04:03 | |
| *** unicell has quit IRC | 04:03 | |
| *** unicell has joined #openstack-nova | 04:03 | |
| openstackgerrit | Rui Chen proposed openstack/nova-specs: Task states improvement https://review.openstack.org/216578 | 04:06 |
| *** shihanzhang has joined #openstack-nova | 04:09 | |
| *** ijw_ has joined #openstack-nova | 04:11 | |
| *** stevemar has joined #openstack-nova | 04:12 | |
| *** fawadkhaliq has joined #openstack-nova | 04:15 | |
| *** piyanai has quit IRC | 04:15 | |
| *** ijw_ has quit IRC | 04:15 | |
| *** stevemar has quit IRC | 04:16 | |
| *** reed_ has joined #openstack-nova | 04:16 | |
| *** shoutm_ has joined #openstack-nova | 04:17 | |
| *** reed_ has quit IRC | 04:18 | |
| *** shoutm has quit IRC | 04:19 | |
| *** stevemar has joined #openstack-nova | 04:23 | |
| *** unicell1 has joined #openstack-nova | 04:32 | |
| *** unicell has quit IRC | 04:33 | |
| *** lyanchih has quit IRC | 04:36 | |
| *** dims_ has quit IRC | 04:38 | |
| *** achanda has quit IRC | 04:39 | |
| *** dims_ has joined #openstack-nova | 04:43 | |
| *** dims_ has quit IRC | 04:44 | |
| *** salv-orlando has joined #openstack-nova | 04:50 | |
| *** shoutm_ has quit IRC | 04:50 | |
| *** alex_klimov has joined #openstack-nova | 04:50 | |
| *** unicell1 has quit IRC | 04:51 | |
| *** unicell has joined #openstack-nova | 04:51 | |
| *** shoutm has joined #openstack-nova | 04:52 | |
| *** salv-orlando has quit IRC | 04:57 | |
| *** unicell has quit IRC | 04:58 | |
| *** haomaiwa_ has quit IRC | 05:01 | |
| *** haomaiw__ has joined #openstack-nova | 05:04 | |
| openstackgerrit | Richard Jones proposed openstack/python-novaclient: Encode unicode filter arguments for server list https://review.openstack.org/224422 | 05:06 |
| *** shoutm has quit IRC | 05:06 | |
| *** shyu has joined #openstack-nova | 05:09 | |
| *** shoutm has joined #openstack-nova | 05:10 | |
| *** penick has quit IRC | 05:10 | |
| *** irenab has quit IRC | 05:12 | |
| *** haomaiw__ has quit IRC | 05:15 | |
| *** haomaiwang has joined #openstack-nova | 05:15 | |
| *** lpetrut has joined #openstack-nova | 05:20 | |
| *** achanda has joined #openstack-nova | 05:20 | |
| *** chenzeng has quit IRC | 05:21 | |
| *** EinstCrazy has quit IRC | 05:24 | |
| *** EinstCrazy has joined #openstack-nova | 05:28 | |
| *** unicell has joined #openstack-nova | 05:28 | |
| *** stevemar has quit IRC | 05:28 | |
| *** stevemar has joined #openstack-nova | 05:29 | |
| *** ijw has quit IRC | 05:30 | |
| openstackgerrit | Richard Jones proposed openstack/nova: Allow filtering using unicode characters https://review.openstack.org/224431 | 05:31 |
| *** ijw has joined #openstack-nova | 05:31 | |
| *** yuntongjin has joined #openstack-nova | 05:35 | |
| *** gsilvis has quit IRC | 05:37 | |
| *** moshele has joined #openstack-nova | 05:40 | |
| *** shoutm has quit IRC | 05:41 | |
| *** irenab has joined #openstack-nova | 05:43 | |
| *** shoutm has joined #openstack-nova | 05:43 | |
| *** dims_ has joined #openstack-nova | 05:45 | |
| *** mehdi has joined #openstack-nova | 05:47 | |
| *** mehdi has quit IRC | 05:51 | |
| *** dims_ has quit IRC | 05:53 | |
| *** asettle has quit IRC | 05:57 | |
| *** Jeffrey4l has quit IRC | 06:00 | |
| *** haomaiwang has quit IRC | 06:01 | |
| *** lsmola has joined #openstack-nova | 06:01 | |
| *** haomaiwang has joined #openstack-nova | 06:01 | |
| *** chenzeng has joined #openstack-nova | 06:02 | |
| *** chenzeng has quit IRC | 06:03 | |
| *** lyanchih has joined #openstack-nova | 06:03 | |
| *** nkrinner_afk is now known as nkrinner | 06:03 | |
| openstackgerrit | Eli Qiao proposed openstack/nova: Give instance default hostname if hostname is empty https://review.openstack.org/224438 | 06:04 |
| *** yuntongjin has quit IRC | 06:06 | |
| *** gsilvis has joined #openstack-nova | 06:10 | |
| *** alex_klimov has quit IRC | 06:14 | |
| *** Jeffrey4l has joined #openstack-nova | 06:14 | |
| *** armax has quit IRC | 06:15 | |
| *** Jeffrey4l has quit IRC | 06:20 | |
| *** lyanchih has quit IRC | 06:22 | |
| *** tpeoples has quit IRC | 06:23 | |
| *** lyanchih has joined #openstack-nova | 06:25 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/224449 | 06:27 |
| *** yamahata has joined #openstack-nova | 06:28 | |
| *** harlowja has quit IRC | 06:29 | |
| *** zhenguo has quit IRC | 06:30 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Add VIF extension in ext list for v21 comp mode https://review.openstack.org/224450 | 06:30 |
| *** zhenguo has joined #openstack-nova | 06:32 | |
| *** Jeffrey4l has joined #openstack-nova | 06:33 | |
| *** vilobhmm_11 has quit IRC | 06:36 | |
| *** salv-orlando has joined #openstack-nova | 06:37 | |
| *** chenzeng has joined #openstack-nova | 06:38 | |
| *** scheuran has joined #openstack-nova | 06:38 | |
| *** cfriesen has quit IRC | 06:43 | |
| *** salv-orlando has quit IRC | 06:43 | |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Add 'OS-EXT-VIF-NET:net_id' for v21 comptible mode https://review.openstack.org/224453 | 06:47 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Add 'OS-EXT-VIF-NET:net_id' for v21 compatible mode https://review.openstack.org/224453 | 06:48 |
| gmann | alex_xu_: johnthetubaguy : oomichi: for VIF v21 comp mode - https://review.openstack.org/#/c/224453/ , https://review.openstack.org/#/c/224450/1 | 06:49 |
| *** otter768 has joined #openstack-nova | 06:49 | |
| *** alex_klimov has joined #openstack-nova | 06:51 | |
| *** haomaiwang has quit IRC | 06:51 | |
| *** kumar has quit IRC | 06:52 | |
| *** haomaiwang has joined #openstack-nova | 06:53 | |
| *** markvoelker has quit IRC | 06:53 | |
| *** alex_klimov has quit IRC | 06:54 | |
| *** otter768 has quit IRC | 06:54 | |
| *** alex_klimov has joined #openstack-nova | 06:54 | |
| *** sahid has joined #openstack-nova | 06:54 | |
| *** alex_klimov has quit IRC | 06:55 | |
| *** alex_klimov has joined #openstack-nova | 06:55 | |
| *** e0ne has joined #openstack-nova | 06:56 | |
| *** yuntongjin has joined #openstack-nova | 06:57 | |
| *** claudiub has joined #openstack-nova | 06:58 | |
| *** haomaiwang has quit IRC | 07:01 | |
| *** haomaiwang has joined #openstack-nova | 07:01 | |
| *** matrohon has joined #openstack-nova | 07:03 | |
| *** e0ne has quit IRC | 07:03 | |
| *** tdurakov has joined #openstack-nova | 07:06 | |
| *** browne has quit IRC | 07:08 | |
| *** yuner027 has quit IRC | 07:09 | |
| *** stevemar has quit IRC | 07:10 | |
| *** ildikov has quit IRC | 07:10 | |
| *** yuner027 has joined #openstack-nova | 07:11 | |
| alex_xu_ | gmann: thanks | 07:12 |
| eliqiao | alex_xu: ping | 07:13 |
| *** rdopiera has joined #openstack-nova | 07:13 | |
| *** fawadkhaliq has quit IRC | 07:17 | |
| *** salv-orlando has joined #openstack-nova | 07:18 | |
| *** jichen has joined #openstack-nova | 07:19 | |
| *** markus_z has joined #openstack-nova | 07:19 | |
| *** markus_z has quit IRC | 07:24 | |
| *** yamahata has quit IRC | 07:26 | |
| *** claudiub has quit IRC | 07:31 | |
| alex_xu_ | eliqiao: pong | 07:31 |
| *** alex_xu_ is now known as alex_xu | 07:32 | |
| eliqiao | alex_xu: https://review.openstack.org/#/c/224438/ | 07:32 |
| eliqiao | ^^ this is for fixing hostname is empty | 07:33 |
| alex_xu | eliqiao: thanks a lot | 07:33 |
| alex_xu | eliqiao: will take a look at asap | 07:33 |
| eliqiao | alex_xu: no hurry, jenkins failed, I just do recheck. | 07:34 |
| alex_xu | eliqiao: ok, thanks | 07:34 |
| eliqiao | alex_xu: np | 07:34 |
| *** achanda has quit IRC | 07:38 | |
| *** jlanoux has joined #openstack-nova | 07:44 | |
| *** ildikov has joined #openstack-nova | 07:45 | |
| alex_xu | johnthetubaguy: oomichi, gmann, the hostname fix ^ which thanks to eliqiao | 07:45 |
| gmann | alex_xu: ok, Thanks eliqiao | 07:46 |
| gmann | alex_xu: i ran rate limit tests for v21 comp mode by adding RateLimitingMiddleware in paste.ini | 07:47 |
| gmann | alex_xu: test passing as you mentioned | 07:47 |
| alex_xu | gmann: thanks, cool :) | 07:47 |
| *** achanda has joined #openstack-nova | 07:47 | |
| gmann | alex_xu: our functional tests use noauth right? | 07:48 |
| * alex_xu relax an sign | 07:48 | |
| alex_xu | gmann: i guess so | 07:48 |
| gmann | alex_xu: searching where that is done..? | 07:48 |
| alex_xu | gmann: let me check | 07:49 |
| gmann | alex_xu: Thanks :) | 07:49 |
| *** achanda has quit IRC | 07:50 | |
| *** dims_ has joined #openstack-nova | 07:51 | |
| *** tdurakov has quit IRC | 07:51 | |
| *** fawadkhaliq has joined #openstack-nova | 07:51 | |
| *** tdurakov has joined #openstack-nova | 07:52 | |
| *** Marga_ has quit IRC | 07:53 | |
| *** tdurakov has quit IRC | 07:53 | |
| *** markvoelker has joined #openstack-nova | 07:54 | |
| alex_xu | gmann: I find it, it is here https://github.com/openstack/nova/blob/master/nova/test.py#L225 | 07:55 |
| *** shuquan has joined #openstack-nova | 07:55 | |
| gmann | alex_xu: yea, i see | 07:55 |
| alex_xu | gmann: if you checked ConfFixture, you find it use noauth2 for auth_strategy option | 07:55 |
| gmann | alex_xu: yea | 07:55 |
| gmann | alex_xu: for ratelimit i saw other code in legacy limit controller | 07:56 |
| alex_xu | gmann: what kind of code? | 07:56 |
| *** Daisy_ has joined #openstack-nova | 07:56 | |
| gmann | https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/legacy_v2/limits.py#L364 | 07:56 |
| gmann | https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/legacy_v2/limits.py#L265 | 07:57 |
| *** dims_ has quit IRC | 07:57 | |
| gmann | alex_xu: ^^ | 07:57 |
| alex_xu | gmann: which used by middleware? | 07:58 |
| *** markvoelker has quit IRC | 07:59 | |
| *** pixelb has joined #openstack-nova | 08:00 | |
| *** danpb has joined #openstack-nova | 08:00 | |
| *** haomaiwang has quit IRC | 08:01 | |
| *** mpaolino has joined #openstack-nova | 08:01 | |
| *** yangyp has joined #openstack-nova | 08:01 | |
| *** Daisy_ has quit IRC | 08:01 | |
| *** haomaiwa_ has joined #openstack-nova | 08:01 | |
| -openstackstatus- NOTICE: Gate is currently stuck, failing grenade upgrade tests due the release of oslo.utils 1.4.1 for Juno. | 08:02 | |
| johnthetubaguy | gmann: alex_xu: hey folks, seeing quite a few things that need reveweing | 08:02 |
| johnthetubaguy | well, no merging for us this morning I guess | 08:02 |
| alex_xu | johnthetubaguy: yea, gmann find something morehttps://bugs.launchpad.net/nova/+bug/1496664 | 08:02 |
| openstack | Launchpad bug 1496664 in OpenStack Compute (nova) "V2.1 comp mode behavior should be fixed for diff of v2 and v2.1 APIs" [Undecided,In progress] - Assigned to Ghanshyam Mann (ghanshyammann) | 08:02 |
| gmann | johnthetubaguy: yea | 08:02 |
| *** ijw has quit IRC | 08:03 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Allow leading/trailing spaces for name field in v2.1 compat mode https://review.openstack.org/220791 | 08:03 |
| gmann | johnthetubaguy: alex_xu : about to sumbit last one for ratelimit bit in paste.ini | 08:03 |
| alex_xu | gmann: ^ just address your comment, thanks for the review | 08:03 |
| gmann | alex_xu: Thanks i will check | 08:03 |
| johnthetubaguy | so I just made that bug block the RC release | 08:03 |
| *** ijw has joined #openstack-nova | 08:03 | |
| alex_xu | gmann: np | 08:03 |
| gmann | johnthetubaguy: RC is tomorrow right? | 08:04 |
| johnthetubaguy | gmann: you mean the ratelimit is not there, I actually quite like that | 08:04 |
| alex_xu | gmann: for the ratelimits, at least it didn't break it the API, in the legacy v2 api, you also can disable limit middleware | 08:04 |
| gmann | johnthetubaguy: its there we just need to add middleware in paste.ini for noauth2 and keystone | 08:05 |
| johnthetubaguy | gmann: so the dirty secret is its when we want it to be, but I am telling everyone Tuesday | 08:05 |
| johnthetubaguy | gmann: we are missing keystone and noauth2? | 08:05 |
| *** tdurakov has joined #openstack-nova | 08:05 | |
| johnthetubaguy | gmann: oh, those extra ones | 08:05 |
| gmann | johnthetubaguy: yea, m thinking of making something - http://paste.openstack.org/show/465924/ | 08:06 |
| *** shoutm_ has joined #openstack-nova | 08:06 | |
| * alex_xu happy if we needn't add ratelimit middleware back to api-paste.ini | 08:06 | |
| johnthetubaguy | yeah, honestly, I would leave off ratelimit | 08:06 |
| johnthetubaguy | and see if no one notices | 08:06 |
| johnthetubaguy | call it deprecated with v2.0 | 08:06 |
| gmann | alex_xu: johnthetubaguy and for noauth2 which is being used in sample tests | 08:07 |
| johnthetubaguy | noauth is cool | 08:07 |
| gmann | currently v21 comp mode tests use v2.1 sample file - https://review.openstack.org/#/c/224386/ | 08:07 |
| johnthetubaguy | I will just leave off ratelimit | 08:07 |
| johnthetubaguy | the thing doesn't really work, lets have that off by default and deprecate it | 08:08 |
| alex_xu | +1 | 08:08 |
| johnthetubaguy | well, at least I would like to find out if its useful! | 08:08 |
| johnthetubaguy | if your deployment is small enough for ratelimit to work, you don't need ratelimit | 08:08 |
| *** shoutm has quit IRC | 08:08 | |
| johnthetubaguy | unless I am missing something | 08:09 |
| johnthetubaguy | sorry, I went all rant-ey there | 08:09 |
| gmann | johnthetubaguy: ok, so deprecate for v2 also? | 08:09 |
| johnthetubaguy | gmann: do you have the bug for that, so I can tag it | 08:09 |
| alex_xu | it is useful, and we discussion that before, that is why we remove it in the v3 and v2.1 api | 08:09 |
| bauzas | morning Nova | 08:09 |
| alex_xu | s/it is useful/it isn't useful/... | 08:09 |
| *** tdurakov has quit IRC | 08:09 | |
| gmann | johnthetubaguy: i had that in https://bugs.launchpad.net/nova/+bug/1496664 | 08:09 |
| johnthetubaguy | gmann: erm, I guess I was going to leave it for v2.0, so it dies with v2.0 | 08:09 |
| openstack | Launchpad bug 1496664 in OpenStack Compute (nova) "V2.1 comp mode behavior should be fixed for diff of v2 and v2.1 APIs" [High,In progress] - Assigned to Ghanshyam Mann (ghanshyammann) | 08:09 |
| alex_xu | bauzas: morning | 08:09 |
| *** yuner027 has quit IRC | 08:10 | |
| johnthetubaguy | gmann: hmm, OK, although I am loosing track in that bug, its tempting to create a new one | 08:10 |
| johnthetubaguy | alex_xu: did we get any agreement on the hostname thing? | 08:10 |
| *** yuner027 has joined #openstack-nova | 08:10 | |
| *** tdurakov has joined #openstack-nova | 08:11 | |
| gmann | johnthetubaguy: ohk, so for tests we just add in noauth2 and run comp mode test on v2 sample file? | 08:11 |
| johnthetubaguy | just thinking if we have all these patches in https://etherpad.openstack.org/p/liberty-nova-priorities-tracking | 08:11 |
| alex_xu | johnthetubaguy: not yet, but I hope you and gmann can take a look the comment in #1495388 | 08:11 |
| gmann | johnthetubaguy: I think we can leave that also and have comp mode tests run on v21 sample | 08:11 |
| johnthetubaguy | gmann: yeah, I think it always did it that way, for some reason | 08:11 |
| alex_xu | johnthetubaguy: and eliqiao just help work out one patch for hostname https://review.openstack.org/#/c/224438/ | 08:11 |
| *** romainh has joined #openstack-nova | 08:12 | |
| gmann | johnthetubaguy: alex_xu : ok thats looks good for me to so leave ratelimit as it is in paste.ini | 08:12 |
| alex_xu | gmann: cool, thanks | 08:12 |
| alex_xu | johnthetubaguy: gmann for the hostname, I found if the input is Chinese or other language, you will get an empty hostname for guest. That is the only one I think we should fix | 08:13 |
| openstackgerrit | Andrea Rosa proposed openstack/nova: Add nova-manage command for cleaning attachment https://review.openstack.org/184537 | 08:14 |
| gmann | johnthetubaguy: alex_xu: added patches on etherpad | 08:14 |
| *** yassine has joined #openstack-nova | 08:14 | |
| alex_xu | johnthetubaguy: gmann the first step is we generate an unicode like 'server-{uuid}'. the second step is enable user to specify hostname when create new instance which need microversions and done in M | 08:15 |
| johnthetubaguy | alex_xu: yeah, totally | 08:15 |
| gmann | alex_xu: yea, +1 | 08:15 |
| johnthetubaguy | alex_xu: well, as an optional extra at least | 08:15 |
| johnthetubaguy | not if you change the display name, you don't change the hostname, if I remember correctly | 08:16 |
| johnthetubaguy | s/not/note/ | 08:16 |
| alex_xu | johnthetubaguy: yes | 08:16 |
| eliqiao | yeah | 08:16 |
| johnthetubaguy | alex_xu: is that default hostname too long, the uuid thing? | 08:16 |
| johnthetubaguy | I guess its the best we have | 08:17 |
| alex_xu | johnthetubaguy: the rfc said 255 allowed | 08:17 |
| johnthetubaguy | OK, good enough | 08:17 |
| gmann | yea its 255 as per rfc | 08:17 |
| johnthetubaguy | I know windows gets upset, but thats different | 08:17 |
| *** markus_z has joined #openstack-nova | 08:18 | |
| alex_xu | johnthetubaguy: ok, we can trunacate it as 64 | 08:18 |
| alex_xu | RFC952 said 64, RFC1123 relax to 255 | 08:18 |
| *** jistr has joined #openstack-nova | 08:20 | |
| *** lpetrut has quit IRC | 08:20 | |
| alex_xu | oops, no rfc 952 is 24 | 08:20 |
| johnthetubaguy | alex_xu: are you doing the thing to look for spaces in the middle of the name? | 08:20 |
| *** gsilvis has quit IRC | 08:20 | |
| gmann | alex_xu: or just appending ID instead of UUID? | 08:21 |
| johnthetubaguy | yeah 24 is too small, just ignore that | 08:21 |
| alex_xu | johnthetubaguy: the currently already strip the spaces for the hostname | 08:21 |
| johnthetubaguy | gmann: we shouldn't use the id, will cause problems for cells, well possibly | 08:21 |
| *** gsilvis has joined #openstack-nova | 08:21 | |
| johnthetubaguy | alex_xu: even in the middle of the hostname? | 08:21 |
| alex_xu | johnthetubaguy: yea | 08:21 |
| gmann | humm | 08:21 |
| johnthetubaguy | alex_xu: ah, cool, I think I had assumed we didn't do that | 08:22 |
| alex_xu | johnthetubaguy: the code works like 'a b' -> 'a--b' | 08:23 |
| alex_xu | '-' available for hostname | 08:23 |
| alex_xu | johnthetubaguy: do you know the windows support length of hostname? or eliqiao would like help to check that? | 08:24 |
| *** tdurakov has quit IRC | 08:24 | |
| *** tdurakov has joined #openstack-nova | 08:25 | |
| eliqiao | alex_xu: don't try windows yet | 08:25 |
| eliqiao | alex_xu: k, will do the google search first. | 08:26 |
| *** tdurakov has quit IRC | 08:26 | |
| johnthetubaguy | eliqiao: alex_xu: I can dig up the code I remember, one sec (I think 24 actually) | 08:27 |
| *** ndipanov has joined #openstack-nova | 08:27 | |
| alex_xu | eliqiao: thanks | 08:27 |
| johnthetubaguy | I would try google it anyways though! | 08:27 |
| alex_xu | johnthetubaguy: heh :) | 08:28 |
| gmann | that will be same case for _default_display_name | 08:28 |
| alex_xu | gmann: the _default_display_name use spaces in the middle 'Server {uuid}' | 08:28 |
| gmann | there also UUID is being appended | 08:28 |
| johnthetubaguy | eliqiao: https://github.com/openstack/nova/blob/master/nova/virt/xenapi/vmops.py#L1927 | 08:28 |
| gmann | alex_xu: so space we will be truncating right? | 08:28 |
| johnthetubaguy | alex_xu: oops, didn't spot that | 08:28 |
| eliqiao | NETBIOS limits the name length to 15 characters. You give the computer a longer hostname (which is fine with Active Directory since it uses DNS, not NETBIOS) but the NETBIOS name will be truncated. | 08:29 |
| alex_xu | gmann: yea...sounds good point | 08:29 |
| alex_xu | johnthetubaguy: cool, thanks | 08:29 |
| *** lucasagomes has joined #openstack-nova | 08:29 | |
| johnthetubaguy | eliqiao: yeah, that matches the code I just found | 08:29 |
| johnthetubaguy | eliqiao: but... its really tempting to just ignore that, I think cloudbase-init deals with that kind of thing, for example | 08:30 |
| gmann | then is it ok to truncate it auto ? | 08:30 |
| johnthetubaguy | I think we can ignore it actually | 08:30 |
| johnthetubaguy | its more about when you call sethostname in windows, I think | 08:30 |
| gmann | johnthetubaguy: yea, i also think so | 08:30 |
| johnthetubaguy | sorry for the "red herring" there | 08:31 |
| eliqiao | johnthetubaguy: haha ... herring, just searched, it is fish | 08:32 |
| *** tdurakov has joined #openstack-nova | 08:32 | |
| *** derekh has joined #openstack-nova | 08:33 | |
| johnthetubaguy | eliqiao: yeah, know idea how that expression came about, similar to "sending you up the garden path", we are full of crazy sayings over here | 08:34 |
| *** artom has quit IRC | 08:38 | |
| eliqiao | johnthetubaguy: thanks for the explaination, get your words :) | 08:39 |
| *** apuimedo has joined #openstack-nova | 08:39 | |
| *** Marga_ has joined #openstack-nova | 08:39 | |
| *** tdurakov has quit IRC | 08:40 | |
| EinstCrazy | Hello, everyone. I want to get some information of libvirt api，some doc like https://libvirt.org/sources/java/javadoc/ for python | 08:40 |
| *** tdurakov has joined #openstack-nova | 08:40 | |
| bauzas | johnthetubaguy: around for a chat re: https://review.openstack.org/#/c/180151/ ? | 08:41 |
| tdurakov | alaski, hi, are you around? | 08:41 |
| bauzas | tdurakov: alaski is East-Coast | 08:41 |
| bauzas | (Us) | 08:41 |
| bauzas | johnthetubaguy: since we have a bug re: live migration, I wonder the best opportunity to fix that | 08:42 |
| johnthetubaguy | bauzas: depends what the fix is | 08:42 |
| bauzas | johnthetubaguy: yeah, the problem is that we ported the memory check from the scheduler to the conductor when build_and_run_instance() was deprecated in favor of select_dest() | 08:43 |
| bauzas | johnthetubaguy: without considering the allocation ratios properlu | 08:43 |
| tdurakov | bauzas, hi:) I was thinking about using only casts for requests from api. What is the main point for that? Are we trying to reduce load on API layer? | 08:43 |
| tdurakov | what we are trying to improve? | 08:44 |
| bauzas | johnthetubaguy: so since we're now providing the allocation ratios on computes, I wonder if it would be better to check those and bypass the logic if not | 08:44 |
| bauzas | tdurakov: casts are non-blocking | 08:44 |
| tdurakov | baizas, i understand that | 08:45 |
| bauzas | compared to calls where you wait for the response | 08:45 |
| tdurakov | yep | 08:45 |
| bauzas | tdurakov: so, what's the misunderstanding ? | 08:45 |
| johnthetubaguy | bauzas: maybe... its for the case where the user specifies the host I guess, so failing on the compute is OK ish I supose | 08:45 |
| tdurakov | but what is the profit? | 08:45 |
| bauzas | johnthetubaguy: zactly | 08:45 |
| tdurakov | bauzas^ | 08:45 |
| bauzas | tdurakov: we don't want to have the API response being blocked by an RPC call | 08:46 |
| bauzas | tdurakov: the API contract implies a direct response | 08:46 |
| tdurakov | so, we are trying to decrease load? | 08:46 |
| bauzas | not load | 08:46 |
| johnthetubaguy | tdurakov: the problem with call is they timeout | 08:46 |
| bauzas | it's just SOA, you give a partial answer | 08:46 |
| johnthetubaguy | tdurakov: our general pattern, is, record the action in DB, send cast to something to work on it, return API request | 08:46 |
| bauzas | rather than waiting to pile the whole env | 08:47 |
| *** Marga_ has quit IRC | 08:47 | |
| tdurakov | bauzas, soa, hmm, where is the sofa here?:) | 08:47 |
| tdurakov | bauzas, soa, hmm, where is the soa here?:) | 08:47 |
| *** sudipto has joined #openstack-nova | 08:48 | |
| bauzas | okay, remove the idea of SOA, let me rather give you the tenets | 08:48 |
| johnthetubaguy | tdurakov: there is a load piece here, lots of open connections on the API server doesn't scale | 08:48 |
| bauzas | tdurakov: https://wiki.openstack.org/wiki/BasicDesignTenets | 08:48 |
| bauzas | point #3 | 08:48 |
| tdurakov | bauzas, ack:) | 08:49 |
| *** lyanchih has quit IRC | 08:49 | |
| bauzas | johnthetubaguy: just back to the problem I mentioned, it's all about having migrations that fail when instance creations do still work | 08:50 |
| *** lyanchih has joined #openstack-nova | 08:51 | |
| bauzas | johnthetubaguy: so, yeah, failing on the compute node is far better precise because of the claimls | 08:51 |
| openstackgerrit | Ghanshyam Mann proposed openstack/nova: Add 'OS-EXT-VIF-NET:net_id' for v21 compatible mode https://review.openstack.org/224450 | 08:51 |
| bauzas | johnthetubaguy: anyway, I'll see what I can do, tks | 08:52 |
| tdurakov | johnthetubaguy, bauzas, one more thing: https://bugs.launchpad.net/nova/+bug/1214943 - i have an idea about this, could we discuss it? | 08:52 |
| openstack | Launchpad bug 1214943 in OpenStack Compute (nova) "Live migration should use the same memory over subscription logic as instance boot" [High,Confirmed] - Assigned to Sylvain Bauza (sylvain-bauza) | 08:52 |
| johnthetubaguy | bauzas: we are moving a failure later, but it stops us failing in the wrong place, so I am OK with that I think | 08:52 |
| tdurakov | i'd like to add smth like verify_destination to scheduler | 08:53 |
| tdurakov | johnthetubaguy, bauzas ^ | 08:53 |
| bauzas | tdurakov: that's something I'm just discussing with johnthetubaguy | 08:53 |
| bauzas | :) | 08:53 |
| tdurakov | nice | 08:53 |
| bauzas | tdurakov: but the whole logic is broken and we're slightly fixing that | 08:53 |
| bauzas | tdurakov: I think you missed something | 08:53 |
| tdurakov | bauzas, if we remove all checks from conductor | 08:54 |
| bauzas | tdurakov: http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/check-destination-on-migrations.html | 08:54 |
| bauzas | tdurakov: ^ is the mid-term goal | 08:54 |
| tdurakov | and use filters of scheduler with new method - everything should be fine | 08:54 |
| bauzas | in the meantime, we have lots of problems that need some bugfixes | 08:55 |
| bauzas | http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/allocation-ratio-to-resource-tracker.html was a masterpiece for fixing https://review.openstack.org/#/c/180151/3 too | 08:55 |
| *** hanzhang has joined #openstack-nova | 08:55 | |
| *** alexpilotti has joined #openstack-nova | 08:55 | |
| bauzas | so yeah, we're trying to reduce the conductor logic to the bare minimum and prefer to schedule | 08:56 |
| bauzas | but until http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/check-destination-on-migrations.html is implemented, we have lots of problems we can possibly handle | 08:57 |
| *** shihanzhang has quit IRC | 08:57 | |
| *** oomichi has joined #openstack-nova | 08:58 | |
| *** moshele has quit IRC | 08:59 | |
| tdurakov | bauzas, what problem could be with host validation in conductor like def verify_destination(self, request_spec, host) ? | 09:00 |
| *** haomaiwa_ has quit IRC | 09:01 | |
| tdurakov | in scheduler i mean | 09:01 |
| markus_z | alex_xu: FYI: Bug 1491244 references to a TODO comment of you in the code. | 09:01 |
| openstack | bug 1491244 in OpenStack Compute (nova) "server details view can include empty string image" [Undecided,New] https://launchpad.net/bugs/1491244 | 09:01 |
| bauzas | tdurakov: you first need to persist the request spec | 09:01 |
| *** haomaiwang has joined #openstack-nova | 09:01 | |
| bauzas | tdurakov: because if not, the decision could be wrong | 09:01 |
| tdurakov | bauzas, could you provide example of a wrong decision? | 09:02 |
| markus_z | alex_xu: Maybe you already have an open commit for that. | 09:02 |
| bauzas | sure | 09:02 |
| bauzas | eg. I boot | 09:02 |
| bauzas | nova boot --same_host=uuid | 09:02 |
| bauzas | well, rather | 09:02 |
| bauzas | nova boot --different_host=uuid | 09:02 |
| bauzas | it will make sure that my instance will be placed anywhere but the host having the instance uuid | 09:03 |
| bauzas | I mean the other instance uuid | 09:03 |
| bauzas | now, when live migrating | 09:03 |
| bauzas | if not providing back that hint, I could be getting the destination to be the one I don't want | 09:04 |
| bauzas | which fails the enforcement | 09:04 |
| *** lyanchih has quit IRC | 09:04 | |
| tdurakov | :) yep, but what if user provide destination host? | 09:04 |
| bauzas | tdurakov: read the spec | 09:04 |
| bauzas | it changes the UX | 09:05 |
| *** moshele has joined #openstack-nova | 09:05 | |
| tdurakov | are you about force-migrate? | 09:05 |
| openstackgerrit | Shuquan Huang proposed openstack/nova: Fix nova configuration options description https://review.openstack.org/224500 | 09:06 |
| tdurakov | bauzas, i think we could do things in parallel, current conductor implementation is even worse, and it doesn't count --same_host at alll. | 09:09 |
| bauzas | tdurakov: I haven't said the contrary | 09:11 |
| tdurakov | if we get rid of conductor checks first, we could fix some bugs, without creating new, then when request_spec model would be ready we could use request_spec as a param for the new method | 09:11 |
| *** lpetrut has joined #openstack-nova | 09:11 | |
| bauzas | tdurakov: https://review.openstack.org/#/c/180151/3 is targeting the one that is blocking | 09:11 |
| tdurakov | by getting rid i mean creation of a new method in scheduler | 09:11 |
| tdurakov | bauzas, ^ | 09:12 |
| bauzas | tdurakov: but we could be a bit less aggressive, hence my discussion with johnthetubaguy above ^ | 09:12 |
| bauzas | tdurakov: well, all of that is targeted by http://specs.openstack.org/openstack/nova-specs/specs/liberty/approved/check-destination-on-migrations.html | 09:12 |
| bauzas | that's the consensus | 09:12 |
| tdurakov | bauzas, is there smth, i could implement? | 09:13 |
| bauzas | tdurakov: lxsli was the primary PoC for that | 09:14 |
| *** hanzhang has quit IRC | 09:14 | |
| *** lpetrut1 has joined #openstack-nova | 09:14 | |
| tdurakov | i worried that we have annoying high bug for 2 cycles really( | 09:14 |
| *** aix has quit IRC | 09:14 | |
| bauzas | tdurakov: well, there is a workaround | 09:15 |
| bauzas | tdurakov: you just have to not provide a dest host | 09:15 |
| bauzas | which will make bad decisions because of the miss of persisting the spec, but at least it will check the ratios and not fail because of that | 09:16 |
| *** lpetrut has quit IRC | 09:16 | |
| *** lpetrut1 is now known as lpetrut | 09:16 | |
| *** yuntongjin has quit IRC | 09:17 | |
| tdurakov | [12:14] <bauzas> tdurakov: lxsli was the primary PoC for that - what do you mean? could you rephrase? | 09:19 |
| bauzas | tdurakov: I deferred the implementation to lxsli | 09:19 |
| openstackgerrit | Wei Jiangang proposed openstack/nova: Remove unused variable 'context' https://review.openstack.org/224513 | 09:19 |
| tdurakov | bauzas, ok | 09:19 |
| bauzas | PoC = point of contact | 09:19 |
| *** lpetrut has quit IRC | 09:21 | |
| *** lpetrut has joined #openstack-nova | 09:21 | |
| tdurakov | bauzas, ok, PoC is a proof of concept either, that's why i asked:) | 09:22 |
| bauzas | yeah... | 09:23 |
| *** lyanchih has joined #openstack-nova | 09:24 | |
| openstackgerrit | Wei Jiangang proposed openstack/nova: Remove redundant variable 'context' https://review.openstack.org/224513 | 09:25 |
| *** ociuhandu has quit IRC | 09:25 | |
| *** flwang has quit IRC | 09:27 | |
| lxsli | hello | 09:27 |
| lxsli | off to a meeting | 09:29 |
| *** FL1SK has quit IRC | 09:32 | |
| *** shuquan has quit IRC | 09:35 | |
| *** Daisy has joined #openstack-nova | 09:39 | |
| *** bkopilov has quit IRC | 09:40 | |
| *** flwang has joined #openstack-nova | 09:41 | |
| *** shuquan has joined #openstack-nova | 09:42 | |
| *** fawadkhaliq has quit IRC | 09:42 | |
| *** fawadkhaliq has joined #openstack-nova | 09:43 | |
| *** Daisy has quit IRC | 09:43 | |
| * johnthetubaguy lost internet there for a moment, not sure why | 09:43 | |
| ndipanov | johnthetubaguy, they're spying on you! | 09:44 |
| *** beagles has joined #openstack-nova | 09:44 | |
| *** beagles is now known as Guest73647 | 09:44 | |
| *** aix has joined #openstack-nova | 09:45 | |
| *** fawadkhaliq has quit IRC | 09:47 | |
| *** IanGovett has joined #openstack-nova | 09:47 | |
| *** bkopilov has joined #openstack-nova | 09:49 | |
| *** tdurakov has quit IRC | 09:50 | |
| *** e0ne has joined #openstack-nova | 09:51 | |
| *** gsilvis has quit IRC | 09:52 | |
| *** moshele has quit IRC | 09:55 | |
| *** markvoelker has joined #openstack-nova | 09:55 | |
| *** moshele has joined #openstack-nova | 09:56 | |
| *** yassine has quit IRC | 09:57 | |
| *** bkopilov has quit IRC | 09:58 | |
| *** gsilvis has joined #openstack-nova | 09:59 | |
| *** shuquan has quit IRC | 09:59 | |
| *** dims_ has joined #openstack-nova | 09:59 | |
| *** markvoelker has quit IRC | 10:00 | |
| *** e0ne has quit IRC | 10:00 | |
| *** haomaiwang has quit IRC | 10:01 | |
| *** shuquan has joined #openstack-nova | 10:01 | |
| *** haomaiwang has joined #openstack-nova | 10:01 | |
| *** shuquan has quit IRC | 10:02 | |
| *** fawadkhaliq has joined #openstack-nova | 10:02 | |
| *** cfarquhar has quit IRC | 10:06 | |
| *** cfarquhar has joined #openstack-nova | 10:08 | |
| *** cfarquhar has quit IRC | 10:08 | |
| *** cfarquhar has joined #openstack-nova | 10:08 | |
| *** bkopilov has joined #openstack-nova | 10:11 | |
| *** e0ne has joined #openstack-nova | 10:11 | |
| *** ndipanov has quit IRC | 10:11 | |
| *** Guest73647 is now known as b3nt_pin | 10:13 | |
| -openstackstatus- NOTICE: Gate back to normal, thanks to the backlisting of the problematic version | 10:15 | |
| *** haomaiwang has quit IRC | 10:18 | |
| *** haomaiwang has joined #openstack-nova | 10:18 | |
| *** gsilvis has quit IRC | 10:19 | |
| *** ndipanov has joined #openstack-nova | 10:20 | |
| *** gsilvis has joined #openstack-nova | 10:21 | |
| *** romainh has quit IRC | 10:21 | |
| *** tdurakov has joined #openstack-nova | 10:22 | |
| johnthetubaguy | ndipanov: :) | 10:22 |
| *** lyanchih has quit IRC | 10:27 | |
| *** salv-orlando has quit IRC | 10:27 | |
| *** salv-orlando has joined #openstack-nova | 10:38 | |
| *** zhenguo has quit IRC | 10:40 | |
| openstackgerrit | Andrea Rosa proposed openstack/nova: Add nova-manage command for cleaning attachment https://review.openstack.org/184537 | 10:42 |
| *** smatzek has joined #openstack-nova | 10:44 | |
| *** nd has joined #openstack-nova | 10:46 | |
| *** mpaolino has quit IRC | 10:47 | |
| *** ZZelle has quit IRC | 10:50 | |
| *** ZZelle has joined #openstack-nova | 10:51 | |
| *** fawadkhaliq has quit IRC | 10:52 | |
| *** gszasz has joined #openstack-nova | 10:55 | |
| *** fawadkhaliq has joined #openstack-nova | 10:55 | |
| *** haomaiwang has quit IRC | 11:01 | |
| *** haomaiwang has joined #openstack-nova | 11:01 | |
| *** lsmola has quit IRC | 11:01 | |
| *** timbyr_ has joined #openstack-nova | 11:02 | |
| *** lsmola has joined #openstack-nova | 11:02 | |
| *** timbyr_ has quit IRC | 11:03 | |
| *** ijw has quit IRC | 11:04 | |
| *** ferest has joined #openstack-nova | 11:06 | |
| *** yassine has joined #openstack-nova | 11:06 | |
| *** dguitarbite has quit IRC | 11:08 | |
| *** dguitarbite has joined #openstack-nova | 11:10 | |
| *** ferest has quit IRC | 11:10 | |
| dguitarbite | EinstCrazy: May be this would help you a bit ... its a draft (incomplete) http://libvirt.org/docs/libvirt-appdev-guide-python/en-US/html/ | 11:12 |
| openstackgerrit | Anthony PERARD proposed openstack/nova: libvirt-xen: Add xen_vif_bridge_script_path config option. https://review.openstack.org/199093 | 11:14 |
| *** romainh has joined #openstack-nova | 11:16 | |
| *** hparekh2 has joined #openstack-nova | 11:22 | |
| *** e0ne has quit IRC | 11:23 | |
| *** hparekh2 has quit IRC | 11:23 | |
| *** hparekh has joined #openstack-nova | 11:23 | |
| *** hparekh has left #openstack-nova | 11:24 | |
| *** oomichi has quit IRC | 11:24 | |
| *** ccesario has quit IRC | 11:25 | |
| *** markvoelker has joined #openstack-nova | 11:26 | |
| *** claudiub has joined #openstack-nova | 11:30 | |
| *** tdurakov_ has joined #openstack-nova | 11:30 | |
| *** shihanzhang has joined #openstack-nova | 11:31 | |
| *** piyanai has joined #openstack-nova | 11:31 | |
| *** markvoelker has quit IRC | 11:31 | |
| *** dave-mccowan has joined #openstack-nova | 11:32 | |
| markus_z | gmann: ping | 11:32 |
| *** FL1SK has joined #openstack-nova | 11:33 | |
| *** ccesario has joined #openstack-nova | 11:37 | |
| *** danieru has quit IRC | 11:37 | |
| openstackgerrit | Bartosz Fic proposed openstack/nova: Live migration failure in API doesn't set vm_state to Error https://review.openstack.org/168916 | 11:43 |
| *** markvoelker has joined #openstack-nova | 11:47 | |
| *** thorst has joined #openstack-nova | 11:50 | |
| *** haomaiwang has quit IRC | 11:54 | |
| *** fawadkhaliq has quit IRC | 12:01 | |
| *** baoli has joined #openstack-nova | 12:04 | |
| *** e0ne has joined #openstack-nova | 12:04 | |
| *** lyanchih has joined #openstack-nova | 12:09 | |
| *** bswartz has quit IRC | 12:10 | |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Specify api_version for functional tests https://review.openstack.org/224593 | 12:10 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Split functional tests for legacy(V2.1) and microversions https://review.openstack.org/224594 | 12:10 |
| *** lyanchih has quit IRC | 12:11 | |
| *** shyu has quit IRC | 12:13 | |
| EinstCrazy | @dguitarbite Thankyou | 12:14 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/224638 | 12:15 |
| *** fawadkhaliq has joined #openstack-nova | 12:15 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/python-novaclient: Updated from global requirements https://review.openstack.org/224680 | 12:16 |
| *** VW_ has joined #openstack-nova | 12:18 | |
| *** A110602877_ has joined #openstack-nova | 12:20 | |
| *** haomaiwang has joined #openstack-nova | 12:23 | |
| *** ociuhandu has joined #openstack-nova | 12:23 | |
| *** chenzeng has quit IRC | 12:23 | |
| *** burgerk has joined #openstack-nova | 12:25 | |
| *** markvoelker has quit IRC | 12:31 | |
| *** ociuhandu has quit IRC | 12:31 | |
| *** ccesario has quit IRC | 12:33 | |
| *** tdurakov_ has quit IRC | 12:33 | |
| *** ccesario has joined #openstack-nova | 12:34 | |
| *** aysyd has joined #openstack-nova | 12:35 | |
| *** EinstCrazy has quit IRC | 12:35 | |
| *** achanda has joined #openstack-nova | 12:35 | |
| *** zhangjn has quit IRC | 12:36 | |
| *** yangyp has quit IRC | 12:36 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Test both NoAuthMiddleware and NoAuthMiddlewareV3 https://review.openstack.org/223425 | 12:38 |
| *** vladikr has joined #openstack-nova | 12:38 | |
| *** promulo has joined #openstack-nova | 12:41 | |
| *** promulo__ has quit IRC | 12:41 | |
| *** VW_ has quit IRC | 12:44 | |
| *** VW_ has joined #openstack-nova | 12:44 | |
| *** shoutm_ has quit IRC | 12:46 | |
| *** rlrossit has joined #openstack-nova | 12:47 | |
| *** piyanai has quit IRC | 12:48 | |
| *** mdbooth has quit IRC | 12:48 | |
| *** VW_ has quit IRC | 12:49 | |
| *** mdbooth has joined #openstack-nova | 12:50 | |
| *** edmondsw has joined #openstack-nova | 12:51 | |
| *** tdurakov_ has joined #openstack-nova | 12:53 | |
| openstackgerrit | ZhuChunzhan proposed openstack/nova: Improve the performance of get instances https://review.openstack.org/214555 | 12:53 |
| *** Daisy has joined #openstack-nova | 12:55 | |
| *** piyanai has joined #openstack-nova | 12:56 | |
| *** aix has quit IRC | 12:56 | |
| *** dustins has joined #openstack-nova | 12:58 | |
| *** ociuhandu has joined #openstack-nova | 12:58 | |
| *** aix has joined #openstack-nova | 12:59 | |
| *** haomaiwang has quit IRC | 13:01 | |
| *** haomaiwang has joined #openstack-nova | 13:01 | |
| *** ccesario has quit IRC | 13:02 | |
| *** mdrabe has joined #openstack-nova | 13:02 | |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Split functional tests for legacy(V2.1) and microversions https://review.openstack.org/224594 | 13:02 |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: Specify api_version for functional tests https://review.openstack.org/224593 | 13:02 |
| *** xyang1 has joined #openstack-nova | 13:03 | |
| *** yangyp has joined #openstack-nova | 13:04 | |
| *** ccesario has joined #openstack-nova | 13:07 | |
| *** A110602877_ has quit IRC | 13:07 | |
| bauzas | johnthetubaguy: alex_xu: FWIW, I just saw that the Nova API meeting ICS is wrong | 13:11 |
| bauzas | http://eavesdrop.openstack.org/calendars/nova-api-meeting.ics | 13:11 |
| alex_xu | markus_z: just checked https://bugs.launchpad.net/nova/+bug/1491244 , we return '{}' in v3 api, but for turn the v3 to v2.1, we revert the change back to return "". This is something we can fix by Microversions in v2.1. | 13:12 |
| openstack | Launchpad bug 1491244 in OpenStack Compute (nova) "server details view can include empty string image" [Undecided,Confirmed] | 13:12 |
| bauzas | johnthetubaguy: alex_xu: could you please update http://git.openstack.org/cgit/openstack-infra/irc-meetings/tree/ ? | 13:12 |
| alex_xu | bauzas: thanks | 13:12 |
| alex_xu | bauzas: let me update | 13:12 |
| bauzas | ack | 13:13 |
| alex_xu | bauzas: this one http://git.openstack.org/cgit/openstack-infra/irc-meetings/tree/ looks right | 13:15 |
| *** fawadkhaliq has quit IRC | 13:16 | |
| *** ctrath has joined #openstack-nova | 13:17 | |
| *** jecarey has joined #openstack-nova | 13:17 | |
| alex_xu | bauzas: emm.. http://eavesdrop.openstack.org/calendars/nova-api-meeting.ics looks right also :) weekly Tuesday UTC1200 | 13:17 |
| bauzas | meh, my IRC bouncer is flappy | 13:18 |
| *** EinstCrazy has joined #openstack-nova | 13:19 | |
| *** zhenguo has joined #openstack-nova | 13:20 | |
| bauzas | alex_xu: oh ok, it was a PEBKAC sorry | 13:21 |
| *** jichen has quit IRC | 13:22 | |
| markus_z | alex_xu: Thanks for the info. So would you say it currently works as designed? | 13:22 |
| openstackgerrit | sahid proposed openstack/nova: libvirt: verify cpu bw policy capability for host https://review.openstack.org/224726 | 13:22 |
| alex_xu | markus_z: yea, the API works as that, and for back-comaptible we can fix it without microversions bump | 13:23 |
| *** casusbelli has joined #openstack-nova | 13:23 | |
| *** eharney has quit IRC | 13:24 | |
| *** fawadkhaliq has joined #openstack-nova | 13:25 | |
| markus_z | alex_xu: So it would make sense to keep the bug open as a reminder, right? I don't think that the bug reporter will file a spec for that. | 13:25 |
| *** READ10 has joined #openstack-nova | 13:26 | |
| alex_xu | markus_z: yea, it's worth keep the bug open. But when somebody want to fix that, the nova-spec need as the API change procedure | 13:27 |
| *** irenab has quit IRC | 13:29 | |
| *** zzzeek has joined #openstack-nova | 13:31 | |
| *** openstackgerrit has quit IRC | 13:31 | |
| *** tongli has joined #openstack-nova | 13:31 | |
| *** smatzek has quit IRC | 13:31 | |
| *** VW_ has joined #openstack-nova | 13:31 | |
| *** openstackgerrit has joined #openstack-nova | 13:32 | |
| *** shoutm has joined #openstack-nova | 13:32 | |
| *** mriedem has joined #openstack-nova | 13:32 | |
| *** mleroy has joined #openstack-nova | 13:35 | |
| *** VW__ has joined #openstack-nova | 13:35 | |
| *** VW_ has quit IRC | 13:36 | |
| *** ijuwang has joined #openstack-nova | 13:36 | |
| *** VW__ has quit IRC | 13:36 | |
| *** VW_ has joined #openstack-nova | 13:37 | |
| *** moshele has quit IRC | 13:39 | |
| *** EinstCrazy has quit IRC | 13:39 | |
| *** achanda has quit IRC | 13:40 | |
| *** VikasC has quit IRC | 13:41 | |
| *** thangp has joined #openstack-nova | 13:42 | |
| *** burgerk has quit IRC | 13:44 | |
| *** zhangjn has joined #openstack-nova | 13:45 | |
| *** atuvenie has joined #openstack-nova | 13:45 | |
| *** zhangjn has quit IRC | 13:45 | |
| *** piyanai has quit IRC | 13:46 | |
| *** zhangjn has joined #openstack-nova | 13:46 | |
| *** piyanai has joined #openstack-nova | 13:46 | |
| *** ccarmack has joined #openstack-nova | 13:47 | |
| *** pumaranikar has joined #openstack-nova | 13:47 | |
| openstackgerrit | LinZhang proposed openstack/nova: Promoting SPICE vm performance(video, sound and other) https://review.openstack.org/223611 | 13:49 |
| *** apuimedo has quit IRC | 13:50 | |
| *** cfriesen has joined #openstack-nova | 13:50 | |
| *** stevemar has joined #openstack-nova | 13:50 | |
| openstackgerrit | Andrey Kurilin proposed openstack/python-novaclient: [docs] Fix length of underline https://review.openstack.org/224733 | 13:50 |
| *** pumaranikar has quit IRC | 13:52 | |
| *** shoutm has quit IRC | 13:54 | |
| mriedem | claudiub: dansmith: mikal: stable/kilo backport that's been around since june https://review.openstack.org/#/c/193139/ | 13:55 |
| *** matrohon has quit IRC | 13:56 | |
| *** su_zhang has joined #openstack-nova | 13:56 | |
| dansmith | got it | 13:56 |
| *** romainh has quit IRC | 13:57 | |
| johnthetubaguy | mikal tjones cburgess jgrimm adrian_otto funzo mjturek jcookekhugen irina_pov krtaylor danpb alexpilotti flip214 jaypipes garyk edleafe dims moshele anteaya Nisha sileht claudiub lxsli neiljerram markus_z swamireddy alevine tonyb andreykurilin ndipanov sc68cal akuriata artom jlvillal mnestratov kashyap aloga rgeragnov bauzas xyang tpatil med_ nic scottda nagyz dannywilson belliott dguitarbite sdague jroll diana_clarke _diana_ nova | 13:57 |
| johnthetubaguy | meeting time | 13:57 |
| dguitarbite | hi | 13:57 |
| johnthetubaguy | lol, that list has got big | 13:57 |
| johnthetubaguy | meeting in #openstack-meeting | 13:58 |
| bauzas | meh, the meeting room is not empty yet | 13:58 |
| dansmith | maybe people should just get calendars? :) | 13:58 |
| edleafe | we are early | 13:58 |
| edleafe | dansmith: belt and suspenders | 13:58 |
| *** romainh has joined #openstack-nova | 13:58 | |
| *** mtanino has joined #openstack-nova | 13:58 | |
| bauzas | dansmith: I'd love to have someone always messenging me :) | 13:58 |
| *** artom has joined #openstack-nova | 13:59 | |
| dansmith | yeah, but it's lazy :) | 13:59 |
| tonyb | dansmith: if only there was an esy way to get some of those .... | 13:59 |
| edleafe | what's wrong with lazy?? | 13:59 |
| bauzas | I thought developers were the most lazy people ? | 13:59 |
| edleafe | bauzas: exactly | 13:59 |
| dansmith | that's sad :( | 13:59 |
| *** smatzek has joined #openstack-nova | 14:00 | |
| *** diogogmt has joined #openstack-nova | 14:00 | |
| kashyap | bauzas: Please speak for yourself and don't make a sweeping generalization :P | 14:00 |
| bauzas | dansmith: http://threevirtues.com/ | 14:00 |
| *** haomaiwang has quit IRC | 14:01 | |
| edleafe | dansmith: http://boardofwisdom.com/togo/Quotes/ShowQuote?msgid=161639#.VfrHpbMS3gg | 14:01 |
| rlrossit | mriedem: if I boot an instance using virsh, nova can't see it right? that's part of the whole out-of-band thing? | 14:01 |
| *** 5EXAB5K66 has joined #openstack-nova | 14:01 | |
| *** lyanchih has joined #openstack-nova | 14:01 | |
| *** e0ne has quit IRC | 14:01 | |
| dansmith | *eyeroll* | 14:01 |
| mriedem | rlrossit: correct | 14:01 |
| *** tpeoples has joined #openstack-nova | 14:02 | |
| *** moshele has joined #openstack-nova | 14:03 | |
| *** eharney has joined #openstack-nova | 14:05 | |
| *** pumaranikar has joined #openstack-nova | 14:05 | |
| *** lyanchih has quit IRC | 14:06 | |
| *** e0ne has joined #openstack-nova | 14:06 | |
| *** sneti has joined #openstack-nova | 14:07 | |
| kashyap | rlrossit: Yeah, you're going behind Nova's back | 14:07 |
| *** matrohon has joined #openstack-nova | 14:07 | |
| *** shoutm has joined #openstack-nova | 14:08 | |
| *** su_zhang has quit IRC | 14:09 | |
| *** Daisy has quit IRC | 14:11 | |
| *** artom has quit IRC | 14:11 | |
| *** burgerk has joined #openstack-nova | 14:11 | |
| *** diogogmt has quit IRC | 14:11 | |
| *** artom has joined #openstack-nova | 14:11 | |
| *** nikhil has quit IRC | 14:14 | |
| *** Piet has quit IRC | 14:14 | |
| *** burt has joined #openstack-nova | 14:15 | |
| *** electrocucaracha has joined #openstack-nova | 14:17 | |
| *** burgerk_ has joined #openstack-nova | 14:17 | |
| *** ccarmack has left #openstack-nova | 14:18 | |
| *** mmedvede has joined #openstack-nova | 14:19 | |
| *** burgerk has quit IRC | 14:20 | |
| *** mehdi___ has joined #openstack-nova | 14:20 | |
| *** Daisy has joined #openstack-nova | 14:21 | |
| *** achanda has joined #openstack-nova | 14:22 | |
| *** obedmr has joined #openstack-nova | 14:22 | |
| *** raildo-afk is now known as raildo | 14:23 | |
| openstackgerrit | John Garbutt proposed openstack/nova: Imported Translations from Zanata https://review.openstack.org/224449 | 14:25 |
| *** achanda has quit IRC | 14:25 | |
| *** markvoelker has joined #openstack-nova | 14:26 | |
| *** nd has quit IRC | 14:27 | |
| *** nd has joined #openstack-nova | 14:27 | |
| *** mehdi___ has quit IRC | 14:28 | |
| openstackgerrit | Balazs Gibizer proposed openstack/nova-specs: Versioned notification API https://review.openstack.org/224755 | 14:28 |
| *** garyk has quit IRC | 14:28 | |
| *** gholler has joined #openstack-nova | 14:31 | |
| *** mtanino has quit IRC | 14:31 | |
| *** mmedvede has quit IRC | 14:31 | |
| openstackgerrit | Alex Xu proposed openstack/nova: Allow leading/trailing spaces for name field in v2.1 compat mode https://review.openstack.org/220791 | 14:32 |
| *** piyanai has quit IRC | 14:33 | |
| *** ildikov has quit IRC | 14:33 | |
| johnthetubaguy | jroll: does that look better in the spreadsheet | 14:34 |
| jroll | johnthetubaguy: fine with me if it works for y'all | 14:34 |
| jroll | I assume most of you know what 'ironic' means | 14:34 |
| *** rfolco has joined #openstack-nova | 14:34 | |
| openstackgerrit | Roman Podoliaka proposed openstack/nova: libvirt: delete the last file link in _supports_direct_io() https://review.openstack.org/224764 | 14:36 |
| edleafe | jroll: we're not Alanis Morissette | 14:36 |
| jroll | :) | 14:36 |
| jroll | I still want to build an inventory management system for ironic called 'alanis' | 14:36 |
| edleafe | jroll: perfect! | 14:37 |
| *** tonytan4ever has joined #openstack-nova | 14:39 | |
| *** nd has quit IRC | 14:39 | |
| mriedem | jroll: as in "alanis morissette"? | 14:39 |
| mriedem | oh b/c of the song | 14:40 |
| mriedem | i get it | 14:40 |
| mriedem | but, then i'd have to think of her every time i heard it, which would be terrible | 14:40 |
| mriedem | this is the alanis morisette i prefer to think of https://www.youtube.com/watch?v=FFQkO7-NKQ0&list=RDFFQkO7-NKQ0#t=4 | 14:40 |
| dansmith | just...wow | 14:41 |
| *** Daisy has quit IRC | 14:42 | |
| *** rfolco has quit IRC | 14:42 | |
| jroll | mriedem: yeah, it'd be a list of everything ironic :P | 14:43 |
| jroll | hahaha | 14:43 |
| *** Daisy has joined #openstack-nova | 14:45 | |
| *** Daisy has quit IRC | 14:45 | |
| *** electroc_ has joined #openstack-nova | 14:46 | |
| edleafe | jroll: but then it wouldn't match | 14:47 |
| *** mmedvede has joined #openstack-nova | 14:47 | |
| edleafe | jroll: it would have to be a list of coincidences that have nothing to do with anything ironic :) | 14:47 |
| jroll | you get the idea :| | 14:47 |
| *** ildikov has joined #openstack-nova | 14:48 | |
| *** electrocucaracha has quit IRC | 14:48 | |
| *** lyanchih has joined #openstack-nova | 14:48 | |
| *** penick has joined #openstack-nova | 14:49 | |
| *** Daisy has joined #openstack-nova | 14:50 | |
| *** oomichi has joined #openstack-nova | 14:51 | |
| oomichi | alex_xu: hi | 14:51 |
| alex_xu | oomichi: hi | 14:51 |
| *** Piet has joined #openstack-nova | 14:51 | |
| oomichi | alex_xu: about https://review.openstack.org/#/c/220791/13/nova/api/openstack/compute/aggregates.py | 14:51 |
| alex_xu | oomichi: it's super late for you :) | 14:51 |
| oomichi | alex_xu: no, I am in USA for qa meetup | 14:52 |
| oomichi | here is 9am :) | 14:52 |
| alex_xu | oomichi: ah, I see | 14:52 |
| alex_xu | oomichi: please go ahead | 14:52 |
| oomichi | alex_xu: I am confused now about the patch. | 14:52 |
| *** achanda has joined #openstack-nova | 14:52 | |
| alex_xu | oomichi: yup | 14:53 |
| oomichi | alex_xu: legacy v2 aggregate doesn't strip() the names | 14:53 |
| alex_xu | oomichi: yes | 14:53 |
| oomichi | alex_xu: why the patch does? | 14:53 |
| oomichi | alex_xu: on v2.0 compatible api? | 14:53 |
| *** tpeoples has quit IRC | 14:53 | |
| *** dims_ has quit IRC | 14:53 | |
| oomichi | alex_xu: that was different from legacy v2 behavior | 14:54 |
| alex_xu | oomichi: because some of strip, and some of no strip, then a tradeoff just for strip them all, and most of people didn't care about this case | 14:54 |
| oomichi | alex_xu: !? | 14:54 |
| alex_xu | oomichi: or you want to totally match the v2 behavior? | 14:55 |
| *** dims_ has joined #openstack-nova | 14:55 | |
| oomichi | alex_xu: what? the main purpose of the patch is only for v2 compatibility. that means the behavior should be the same | 14:55 |
| oomichi | johnthetubaguy: ^^^ | 14:56 |
| *** rfolco has joined #openstack-nova | 14:56 | |
| openstackgerrit | Ed Leafe proposed openstack/nova: Replace os.path.join() for URLs https://review.openstack.org/218309 | 14:57 |
| oomichi | alex_xu: if we are saying "most of people didn't care about this case", why we need to strip() at all? | 14:57 |
| *** armax has joined #openstack-nova | 14:57 | |
| alex_xu | oomichi: yea, totally match v2 also make sense, the initial decision I guess is for fixing then both for v2 and v2.1 | 14:57 |
| *** whenry has quit IRC | 14:58 | |
| oomichi | alex_xu: I am really confused now... | 14:58 |
| *** penick has quit IRC | 14:58 | |
| *** Marga_ has joined #openstack-nova | 14:59 | |
| *** unicell1 has joined #openstack-nova | 15:00 | |
| oomichi | johnthetubaguy: are you knowing that? https://review.openstack.org/#/c/220791 is making v2 compatible mode different from legacy v2 code | 15:00 |
| *** unicell has quit IRC | 15:00 | |
| *** Marga_ has quit IRC | 15:00 | |
| *** 5EXAB5K66 has quit IRC | 15:01 | |
| *** Marga_ has joined #openstack-nova | 15:01 | |
| *** ccesario has quit IRC | 15:01 | |
| *** haomaiwang has joined #openstack-nova | 15:01 | |
| oomichi | alex_xu: ok, I will write comments on the patch | 15:01 |
| alex_xu | oomichi: ok, that indeed discussion on the meeting | 15:02 |
| *** tpeoples has joined #openstack-nova | 15:02 | |
| openstackgerrit | LinZhang proposed openstack/nova: Promoting SPICE vm performance(video, sound and other) https://review.openstack.org/223611 | 15:02 |
| *** emagana has joined #openstack-nova | 15:02 | |
| dansmith | danpb: wanna review this libvirt serial console fix? https://review.openstack.org/#/c/215102/12 | 15:03 |
| alex_xu | oomichi: but the desicsion make for fixing both for v2 and v2.1, as you point out, if we just back-compatible with v2 now. totally match the v2 is make sense. I'm ok update the patch | 15:03 |
| johnthetubaguy | oomichi: so the key bit for me, was, we accept the same requests, and they still work | 15:04 |
| oomichi | alex_xu: "v2" means legacy v2 code or v2 compatible? | 15:04 |
| *** moshele has quit IRC | 15:04 | |
| alex_xu | oomichi: v2.1 compat | 15:04 |
| danpb | dansmith: ok | 15:04 |
| dansmith | danpb: thanks. it came up in the meeting as something we'd like to land for rc1 if we can get it reviewed | 15:05 |
| johnthetubaguy | oomichi: when we discussed it, it seemed to make sense to constantly strip, given we allow the silly input, and some places do stip it | 15:05 |
| oomichi | alex_xu: v2 compat should work as the same as legacy v2. | 15:05 |
| *** ccesario has joined #openstack-nova | 15:06 | |
| oomichi | alex_xu: if making them different, users will face the comp problems. I guess that is reason why you and johnthetubaguy want to do that | 15:06 |
| *** browne has joined #openstack-nova | 15:06 | |
| *** diogogmt has joined #openstack-nova | 15:06 | |
| oomichi | alex_xu: but current patch is different from legacy v2 code. | 15:06 |
| oomichi | I am really confused | 15:06 |
| *** zhenguo has quit IRC | 15:06 | |
| *** garyk has joined #openstack-nova | 15:07 | |
| oomichi | johnthetubaguy: I completely agree with consistent/constant api behaviors are great, but the purpose of v2 comp is different IMO. | 15:08 |
| *** rook has joined #openstack-nova | 15:09 | |
| alex_xu | johnthetubaguy: oomichi I'm ok update the patch, if we think that is good. | 15:09 |
| oomichi | johnthetubaguy: if we add strip() to APIs which didn't do that on legacy v2 api and users face backward incompatible problems, difficult to explain the reason | 15:09 |
| *** achanda has quit IRC | 15:09 | |
| johnthetubaguy | oomichi: alex_xu: which are the cases that have changed from v2 in v2 comp? | 15:11 |
| alex_xu | johnthetubaguy: some of api not strip the name | 15:11 |
| alex_xu | in v2 | 15:11 |
| *** diogogmt has quit IRC | 15:11 | |
| oomichi | alex_xu: yes, right | 15:12 |
| oomichi | I am investigating all code of legacy api, and I found many differences about this | 15:12 |
| oomichi | at the airport | 15:13 |
| *** diogogmt has joined #openstack-nova | 15:13 | |
| johnthetubaguy | alex_xu: agreed, do you remember which ones? | 15:13 |
| alex_xu | johnthetubaguy: you can find from the commit message https://review.openstack.org/#/c/220791/ | 15:13 |
| johnthetubaguy | alex_xu: I am wondering about updating the legacy v2 code at the same time | 15:13 |
| alex_xu | johnthetubaguy: we probably have there cases in the v2 API | 15:14 |
| *** rushil has joined #openstack-nova | 15:14 | |
| johnthetubaguy | alex_xu: just if we are happy with the API change, I think we should make legacy match this slightly new behaviour | 15:14 |
| oomichi | johnthetubaguy: how about not stripping name on legacy api instead? | 15:15 |
| johnthetubaguy | oomichi: its an extreme edge case, thats the reality of this right, it seems bad to reject requests that were previously accepted, and this seemed like the best comprimise | 15:15 |
| johnthetubaguy | oomichi: so I like the stripping as it gives us the same range of possible values between the new API and the old API, post this change | 15:16 |
| oomichi | johnthetubaguy: yeah, edge case. agree | 15:16 |
| oomichi | johnthetubaguy: I agree with relaxing validation htig. | 15:16 |
| johnthetubaguy | htig? | 15:17 |
| oomichi | johnthetubaguy: but we don't find the reason of striping yet. | 15:17 |
| oomichi | thing, sorry | 15:17 |
| johnthetubaguy | so its more about display names with trailing spaces just being confusing, and its reasonable that v2.1 outlaws thoses | 15:18 |
| *** nkrinner is now known as nkrinner_afk | 15:18 | |
| oomichi | johnthetubaguy: but input validation works before stripping and the range is fixed like 55 | 15:18 |
| oomichi | s/55/255/ | 15:18 |
| johnthetubaguy | so we didn't want to stop stripping the ones that are already stripped | 15:18 |
| *** sudipto has quit IRC | 15:18 | |
| *** tongli has quit IRC | 15:19 | |
| oomichi | due to json-schema validation which works before entering api method | 15:19 |
| alex_xu | oomichi: there are some initial discussion https://review.openstack.org/#/c/220279/ | 15:19 |
| johnthetubaguy | oomichi: yeah, I don't really mind about that | 15:19 |
| *** alexpilotti has quit IRC | 15:19 | |
| markus_z | bug 1495834 seems to be suspicious. nova-compute can be brought down. | 15:20 |
| openstack | bug 1495834 in OpenStack Compute (nova) "[VMware] Launching an instance with large image size crashes nova-compute" [Undecided,New] https://launchpad.net/bugs/1495834 | 15:20 |
| oomichi | johnthetubaguy: if strip works before validation, strip has meaning. but v2.1 doesn't do like that | 15:20 |
| *** mdrabe has quit IRC | 15:20 | |
| *** angdraug has joined #openstack-nova | 15:20 | |
| *** moshele has joined #openstack-nova | 15:20 | |
| *** hemnafk is now known as hemna | 15:21 | |
| johnthetubaguy | oomichi: sorry, not sure I understand what you mean. the strip does nothing for v2.1, thats the idea at least | 15:22 |
| *** rpodolyaka1 has joined #openstack-nova | 15:22 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Correct memory validation for live migration https://review.openstack.org/180151 | 15:23 |
| *** alexpilotti has joined #openstack-nova | 15:24 | |
| oomichi | johnthetubaguy: I think basic idea of strip() is that "spaces of start/end are meaningless" and "it is nice to remove them before storing name to DB." | 15:24 |
| johnthetubaguy | oomichi: yes | 15:25 |
| oomichi | johnthetubaguy: based on this idea, strip() should work before input validation, IIUC | 15:25 |
| *** zhangjn has quit IRC | 15:25 | |
| johnthetubaguy | oomichi: ideally, yes, but that proved messy I was told | 15:25 |
| alex_xu | oomichi: you talk about v2.1 compat or v2.1? | 15:26 |
| *** piyanai has joined #openstack-nova | 15:26 | |
| oomichi | alex_xu: v2.1 compat | 15:26 |
| *** zhangjn has joined #openstack-nova | 15:26 | |
| oomichi | johnthetubaguy: the checking string length come from DB definition like 255 | 15:26 |
| oomichi | johnthetubaguy: and json-schema definition contains 255 for each param length | 15:27 |
| oomichi | johnthetubaguy: but json-schema validation works before stripping. so the above idea doesn't work on v2.1/v2.1compat apis | 15:27 |
| johnthetubaguy | oomichi: its messy, for sure, something with spaces that is 256 long will be rejected, even though after strip it would pass | 15:28 |
| oomichi | because v2.1/v2.1compat cannot check stripped string | 15:28 |
| alex_xu | yea...we shouldn't waste time for this...it is useless | 15:28 |
| johnthetubaguy | if we could do it before, I would vote for that patch over this one, but I would like a fix now that does 99% of the cases, rather than not fixing it for liberty | 15:29 |
| *** krtaylor has quit IRC | 15:29 | |
| johnthetubaguy | thats not usually true for API patches, but in this case, it seems like a good trade off | 15:30 |
| *** VW__ has joined #openstack-nova | 15:31 | |
| oomichi | johnthetubaguy: I still cannot understand what is the purpose of strip() now even if we are saying that is conner case and different from legacy v2 | 15:32 |
| *** rushil has quit IRC | 15:32 | |
| *** earlephilhower has joined #openstack-nova | 15:32 | |
| *** VW_ has quit IRC | 15:34 | |
| *** dave-mccowan has quit IRC | 15:35 | |
| alex_xu | oomichi: the purpose is just have consistent fix for all the api, because trailing/leading spaces is unclear use-case for real user. | 15:35 |
| alex_xu | s/consistent/consistent and simple/ | 15:36 |
| oomichi | alex_xu: I agree with "unclear use-case for real user" | 15:36 |
| oomichi | alex_xu: and consistent behavior also is great | 15:36 |
| *** mmedvede has quit IRC | 15:36 | |
| alex_xu | oomichi: but not allow trailing/leading indeed back-incompatible thing | 15:37 |
| *** mdrabe has joined #openstack-nova | 15:37 | |
| oomichi | alex_xu: but automatic strip is strange for me. | 15:37 |
| *** ccarmack has joined #openstack-nova | 15:37 | |
| alex_xu | oomichi: unclear use-case vs back-incompatible | 15:37 |
| alex_xu | oomichi: the strip make the api not break the user and works, and just simple and consistent fix | 15:37 |
| mriedem | garyk: fyi bug 1495834 | 15:38 |
| openstack | bug 1495834 in OpenStack Compute (nova) "[VMware] Launching an instance with large image size crashes nova-compute" [Undecided,New] https://launchpad.net/bugs/1495834 | 15:38 |
| mriedem | markus_z: ^ | 15:38 |
| oomichi | alex_xu: why you can say that? if user app passes " AAA " as the name and strip it to "AAA" and the app checks the name of the response, the app will be broken | 15:39 |
| *** tongli has joined #openstack-nova | 15:39 | |
| oomichi | alex_xu: if expecting " AAA " as the name | 15:39 |
| dansmith | is this about display name? | 15:39 |
| alex_xu | oomichi: I guess only test check the response | 15:40 |
| oomichi | alex_xu: that is conner case, but it is imaginable | 15:40 |
| *** tonytan4ever has quit IRC | 15:40 | |
| markus_z | mriedem: yeah, that one looks suspicious | 15:40 |
| alex_xu | dansmith: yea | 15:40 |
| alex_xu | dansmith: about how to fix leading/trailing spaces https://review.openstack.org/220791 | 15:40 |
| dansmith | alex_xu: oomichi: a few of us were discussing this the other day and I think the consensus was strip was proper | 15:40 |
| *** devlaps has joined #openstack-nova | 15:40 | |
| dansmith | it's true that an older app would be confused if that stopped working, | 15:41 |
| dansmith | but it's also *really* strange behavior to allow spaces like that | 15:41 |
| oomichi | alex_xu: we cannot know what kind of app exist in real world and that is reason why we are making the same behavior of v2 | 15:41 |
| dansmith | and we use that for DNS names.. it's never acceptable for people to expect a leading or trailing space in a DNS name... | 15:41 |
| oomichi | dansmith: yeah, right. but we need to help such apps also on v2 compat api | 15:41 |
| dansmith | I dunno | 15:41 |
| dansmith | I think this is likely the kind of thing we would have "fixed" on v2 without giving it much thought | 15:42 |
| bauzas | +1 | 15:42 |
| bauzas | why should we consider spaces as valid ? | 15:42 |
| oomichi | dansmith: relaxing additionalProperties validation also is for such apps | 15:42 |
| bauzas | even for v2 | 15:42 |
| dansmith | oomichi: yeah, but that's very different than strip()ing a name | 15:42 |
| *** aix has quit IRC | 15:43 | |
| oomichi | dansmith: yeah, right. strip()ing is very conner case | 15:43 |
| dansmith | alex_xu: your patch removes it from the schema and just strip()s the name without asking, is that right? | 15:45 |
| dansmith | alex_xu: if so, your title is confusing "allow leading/trailing spaces ..." implies that your patch is allowing them | 15:45 |
| dansmith | not filtering them | 15:45 |
| alex_xu | dansmith: oh, yea, you are right, the titile is wrong | 15:46 |
| dansmith | alex_xu: I will comment with a suggested alternative | 15:46 |
| alex_xu | dansmith: thanks | 15:46 |
| dansmith | johnthetubaguy: I didn't follow earlier when you were involved, but what is your feeling on this strip()ing? | 15:47 |
| *** annegentle has joined #openstack-nova | 15:47 | |
| * dansmith keeps adding () to avoid confusion about stripping | 15:48 | |
| *** nd has joined #openstack-nova | 15:48 | |
| johnthetubaguy | dansmith: sorry, pullled into a quick meeting, I am really just caring about v2.1 compat allowing in requests v2.0 legacy does | 15:48 |
| dansmith | johnthetubaguy: okay, so relaxing the schema but cleaning up the name is good because it doesn't return a 4xx, but may require the client to be less stupid about comparing the result to their non-stripped name, right? | 15:49 |
| openstackgerrit | Alex Xu proposed openstack/nova: Filter leading/trailing spaces for name field in v2.1 compat mode https://review.openstack.org/220791 | 15:49 |
| tpeoples | hi, can i please get reviews on https://review.openstack.org/#/c/220634/ if anyone has a second (small changeset) | 15:50 |
| johnthetubaguy | dansmith: yeah, we usually return the changed name, so I feel less bad about that | 15:50 |
| dansmith | yeah, okay, I think we're on the same page | 15:50 |
| *** electroc_ is now known as electrocucaracha | 15:50 | |
| *** mmedvede has joined #openstack-nova | 15:50 | |
| oomichi | sorry, I need to pass the airport gate now. see you | 15:51 |
| *** oomichi has quit IRC | 15:51 | |
| *** vishwanathj has joined #openstack-nova | 15:51 | |
| openstackgerrit | Sylvain Bauza proposed openstack/nova: Correct memory validation for live migration https://review.openstack.org/180151 | 15:52 |
| bauzas | tdurakov: I think you could be interested in ^ | 15:52 |
| *** annegentle has quit IRC | 15:52 | |
| tdurakov | bauzas, yep, thank you! left some comments. | 15:53 |
| *** annegentle has joined #openstack-nova | 15:53 | |
| *** yangyp has quit IRC | 15:55 | |
| bauzas | tdurakov: as said this morning, we had a consensus on the upgrade path | 15:57 |
| bauzas | tdurakov: I don't want to call the scheduler like that | 15:57 |
| *** krtaylor has joined #openstack-nova | 15:57 | |
| bauzas | tdurakov: before having the persisted object | 15:57 |
| *** markus_z has quit IRC | 15:57 | |
| * bauzas needs to drop off for 15 mins | 15:57 | |
| *** alex_klimov has quit IRC | 15:57 | |
| tdurakov | bauzas, reordering of work-items doesn't seem to be too critical, but it's up to you:) | 15:58 |
| *** e0ne has quit IRC | 15:59 | |
| *** vladikr has quit IRC | 15:59 | |
| *** e0ne has joined #openstack-nova | 16:00 | |
| *** haomaiwang has quit IRC | 16:01 | |
| *** haomaiwang has joined #openstack-nova | 16:01 | |
| *** VW__ has quit IRC | 16:02 | |
| *** hitalia has joined #openstack-nova | 16:03 | |
| *** VW_ has joined #openstack-nova | 16:03 | |
| openstackgerrit | lyanchih proposed openstack/nova: libvirt: serial console ports count upper limit needs to be checked https://review.openstack.org/207373 | 16:04 |
| *** VW_ has quit IRC | 16:04 | |
| *** VW_ has joined #openstack-nova | 16:04 | |
| *** nd has quit IRC | 16:04 | |
| *** ijw has joined #openstack-nova | 16:05 | |
| *** jlanoux has quit IRC | 16:05 | |
| *** shoutm has quit IRC | 16:05 | |
| *** moshele has quit IRC | 16:06 | |
| *** emagana has quit IRC | 16:06 | |
| johnthetubaguy | dansmith: yeah, I think we agreed | 16:06 |
| *** nd has joined #openstack-nova | 16:06 | |
| nd | Hi. Has anyone played around with schedulers and filters in nova? | 16:06 |
| johnthetubaguy | dansmith: I was thinking about us patching in the strip into legacy_v2 also, as you said, thats a good rule of thumb here | 16:06 |
| *** yassine has quit IRC | 16:07 | |
| dansmith | johnthetubaguy: not sure I get what you mean.. just to make the patch cleaner? | 16:07 |
| *** zhangjn has quit IRC | 16:07 | |
| johnthetubaguy | dansmith: well, we could have a patch that fixes up legacy_v2 to strip, then add that patch on top that fixes up v2.1 | 16:07 |
| *** MarcF has quit IRC | 16:07 | |
| *** unicell1 has quit IRC | 16:10 | |
| *** piyanai has quit IRC | 16:10 | |
| dansmith | you mean to strip for v2 and schema-fail for v2.1? | 16:10 |
| garyk | mriedem: thanks. saw that yesterday. it may require a fix to oslo.vmware which we are working on https://review.openstack.org/#/c/224558/ (gate is blocking at the moment) | 16:11 |
| *** piyanai has joined #openstack-nova | 16:11 | |
| *** emagana has joined #openstack-nova | 16:12 | |
| dansmith | johnthetubaguy: ^ | 16:13 |
| *** Daisy has quit IRC | 16:13 | |
| *** unicell has joined #openstack-nova | 16:13 | |
| *** Daisy has joined #openstack-nova | 16:14 | |
| *** Daisy has quit IRC | 16:14 | |
| *** tonytan4ever has joined #openstack-nova | 16:15 | |
| *** rushil has joined #openstack-nova | 16:15 | |
| *** e0ne has quit IRC | 16:15 | |
| bauzas | nd: what's your question ? | 16:16 |
| openstackgerrit | Lucas Alvares Gomes proposed openstack/nova: Ironic: Add support for microversion https://review.openstack.org/221848 | 16:16 |
| bauzas | tdurakov: well, lxsli is already working on the implementation, so we don't need to rework the items | 16:16 |
| *** bkopilov has quit IRC | 16:16 | |
| tdurakov | bauzas, ok | 16:17 |
| *** piyanai has quit IRC | 16:17 | |
| johnthetubaguy | dansmith: erm, I think thats the current plan, I was thinking, v2.0 legacy strip, v2.0 compat strip, v2.1 fail, I think the v2.0 legacy change is the different bit | 16:18 |
| *** rpodolyaka1 has quit IRC | 16:18 | |
| *** scheuran has quit IRC | 16:19 | |
| dansmith | johnthetubaguy: hmm, not sure the current patch makes it fail for v2.1, I think it's strip() for all versions... alex_xu ? | 16:20 |
| lxsli | tdurakov: if you want to help, you can already look at https://review.openstack.org/#/c/221014 and https://review.openstack.org/#/c/212510/ | 16:20 |
| *** karimb has joined #openstack-nova | 16:20 | |
| nd | bauzas: when defining the scheduler_default_filters where multiple filters can be mentioned - how does the nova scheduler actually consider these? - AND , OR or sequential which first matches? | 16:20 |
| johnthetubaguy | dansmith: I think its just v2.1 already fails with trailing spaces, due the extra validation there | 16:21 |
| lxsli | nd: all filters must match, in order, for a host to be weighed | 16:21 |
| bauzas | nd: sequentuially | 16:21 |
| bauzas | nd: http://docs.openstack.org/developer/nova/devref/filter_scheduler.html | 16:21 |
| lxsli | nd: it can be helpful to put your most powerful filters (like an AZ filter) early in the sequence | 16:22 |
| dansmith | johnthetubaguy: the commit message says "The V2.1 API restrict those name fields. It disallows any space in | 16:22 |
| dansmith | the name field anymore. For avoid breaking the user, this patch | 16:22 |
| dansmith | relax the validation of name field." | 16:22 |
| nd | bauzas: thnks | 16:22 |
| dansmith | johnthetubaguy: but maybe that's not really what it's doing, as I don't think I see a schema change there | 16:22 |
| *** unicell1 has joined #openstack-nova | 16:23 | |
| mriedem | alaski: some brave souls out there http://lists.openstack.org/pipermail/openstack-operators/2015-September/008135.html | 16:23 |
| tdurakov | lxsli, https://review.openstack.org/#/c/221014 - i left a question earlier, is it ok, that whole patch just rewriting tests from mox to mock? | 16:23 |
| *** bkopilov has joined #openstack-nova | 16:23 | |
| nd | lxsli: you say all filters must match - but say if the 1st filter is able to give some hosts - will the scheduler proceed to now considering weights for these filtered hosts or continue analysing all remaining filters mentioned? | 16:23 |
| lxsli | tdurakov: the alternative is to combine a massive refactoring with an actual change, which would make the review really hard | 16:24 |
| tdurakov | johnthetubaguy, could you make it clear?^ | 16:24 |
| lxsli | nd: the scheduler starts with all the hosts, then runs each filter in sequence to reduce that list | 16:24 |
| *** Daisy has joined #openstack-nova | 16:24 | |
| lxsli | nd: weights are only calculated on the survivors | 16:24 |
| *** unicell has quit IRC | 16:25 | |
| *** penick has joined #openstack-nova | 16:25 | |
| alex_xu | dansmith: the current patch only for v2.1 compat mode | 16:25 |
| *** rpodolyaka1 has joined #openstack-nova | 16:26 | |
| dansmith | alex_xu: okay, then you should correct that in the commit message too, as it's very confusing | 16:26 |
| nd | lxsli: ok, yes I understand that. but say 1st filter is mentioned as AZ and next is mentioned as say Ramfilter - will the scheduler go through all the filters till the end and which ever hosts pass through all the filters only these are weighed? | 16:26 |
| lxsli | nd: yes | 16:26 |
| alex_xu | dansmith: ok, let me correct it | 16:27 |
| *** lpetrut has quit IRC | 16:27 | |
| nd | lxsli: ohh ok. so even if say 1st filter gives some hosts - they are not straight away weighed - untill all filters have given the final list? | 16:27 |
| lxsli | Say you have hosts 1..20. First filter, LessThan10Filter, reduces to 1..10. Then OddFilter reduces to 1,3,5,7,9. | 16:28 |
| lxsli | 11 didn't make it past the LessThan10Filter, so the OddFilter never sees it | 16:28 |
| alex_xu | dansmith: emm... actually I mention is only for v2.1 compat mode, as the title is "Filter leading/trailing spaces for name field in v2.1 compat mode" | 16:28 |
| tdurakov | lxsli, what about splitting this: https://review.openstack.org/#/c/212510 into smaller parts? | 16:29 |
| dansmith | alex_xu: right, but later in the body, it sounds like v2.1 is being changed | 16:29 |
| *** Daisy has quit IRC | 16:29 | |
| alex_xu | dansmith: just see the irc log, I know the part you confused | 16:29 |
| lxsli | tdurakov: it could be possible, once I have some confidence it vaguely works I can split it up | 16:30 |
| nd | lxsli: ok, understand it now. thnks very much for the detailed explanation. | 16:30 |
| *** romainh has left #openstack-nova | 16:31 | |
| lxsli | nd: happy to help | 16:31 |
| *** fawadkhaliq has quit IRC | 16:31 | |
| nd | bauzas: thnks for the link. I had gone through that, but still it was not clear for me. but thanks to you and lxsli , now I understand it | 16:31 |
| *** unicell1 has quit IRC | 16:32 | |
| *** matrohon has quit IRC | 16:32 | |
| lxsli | tdurakov: I could mark it WIP but john's -2 and the CI result seem pretty clear markers to me :) | 16:34 |
| *** jistr has quit IRC | 16:35 | |
| *** david-lyle has quit IRC | 16:36 | |
| bauzas | lxsli: you mean, red CI like mature tomatoes ? | 16:36 |
| bauzas | :p | 16:36 |
| tdurakov | lxsli, i think, if you start to follow work-items from spec, live become easier. just decompose it into smaller chunks first, it allow you to easy maintain updates and fix merge-conflicts | 16:37 |
| johnthetubaguy | tdurakov: sorry, you asked my a question, is that about the mox -> mock stuff? | 16:37 |
| tdurakov | johnthetubaguy yes | 16:37 |
| lxsli | tdurakov: do you see the TODO? :) | 16:37 |
| *** kiran-r has joined #openstack-nova | 16:37 | |
| lxsli | oh you want separate patches for each work item | 16:37 |
| lxsli | well, it's possible but they'd be small | 16:38 |
| tdurakov | lxsli, it would be easy enough to maintain, no one-liners, i think | 16:38 |
| openstackgerrit | lyanchih proposed openstack/nova: libvirt - Unrescue will not remove rescue disk when image_type=rbd https://review.openstack.org/205766 | 16:39 |
| lxsli | tdurakov: OK I'll look at that tomorrow or next week | 16:39 |
| johnthetubaguy | tdurakov: sorry, not sure I understand the question full, what did you want me to make clear? | 16:39 |
| johnthetubaguy | s/full/fully/ | 16:40 |
| mriedem | this review is sort of stuck, i should have probably put it on the meeting agenda long ago https://review.openstack.org/#/c/173204/ | 16:40 |
| mriedem | to microversion or not to microversion | 16:40 |
| mriedem | it's one of those, this is busted behavior, but it's very old behavior | 16:41 |
| *** lyanchih has quit IRC | 16:41 | |
| mriedem | and you're only impacted if you're not doing security groups on ports in neutron | 16:41 |
| mriedem | which is >0 of people using neutron | 16:41 |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/nova: Updated from global requirements https://review.openstack.org/224638 | 16:41 |
| johnthetubaguy | mriedem: oh yeah, I did look at that and confused myself, I have a feeling its probably safe, given how long ago neutron started adding the default security group for you | 16:42 |
| tdurakov | johnthetubaguy, as I understand, we have policy, that allow to change mox to mock if and only if we touch test anyway. Am i right? | 16:42 |
| mriedem | tdurakov: generally | 16:42 |
| tdurakov | mriedem, execeptions? | 16:43 |
| *** penick has quit IRC | 16:43 | |
| tdurakov | mriedem, s/execeptions/exceptions? | 16:43 |
| mriedem | tdurakov: see #4 here http://docs.openstack.org/infra/manual/developers.html#peer-review | 16:43 |
| lxsli | The situation is that I want to change a bunch of Mox tests | 16:43 |
| mriedem | tdurakov: depends on the case i guess, but usually you could just write a new test purely in mock rather than rewrite an old test and convert from mox to mock | 16:43 |
| *** IanGovett has quit IRC | 16:43 | |
| openstackgerrit | Claudiu Belu proposed openstack/nova-specs: Adds spec for instance live resize https://review.openstack.org/141219 | 16:44 |
| mriedem | lxsli: those are always terrible to review | 16:44 |
| mriedem | which is why we generally don't want to do that | 16:44 |
| lxsli | if I do that in the same patch as unmoxifying them, I feel the review is unnecessarily difficult | 16:44 |
| mriedem | why do we need to convert them to mock? | 16:44 |
| lxsli | so I split the unmoxification to a separate patch chain, which tdurakov doesn't feel I should have done | 16:44 |
| *** IanGovett has joined #openstack-nova | 16:44 | |
| * mriedem feels he should have gone to lunch rather than reply | 16:45 | |
| tdurakov | mriedem, here is the patch we are talking about, https://review.openstack.org/#/c/221014 | 16:45 |
| tdurakov | mriedem, bon appetit, then:) | 16:45 |
| johnthetubaguy | so I prefer the un moxification being in a separate patch, than in one mega patch | 16:45 |
| * lxsli could have gone home already | 16:45 | |
| mriedem | they are both not fun | 16:45 |
| mriedem | as i remember mdbooth was doing something similar with conductor or shelve tests (which extended the base conductor tests) | 16:45 |
| mriedem | and it died on the vine in the same way | 16:46 |
| tdurakov | johnthetubaguy, if mega patch become splitted first...:-p | 16:46 |
| *** IanGovett1 has joined #openstack-nova | 16:46 | |
| * bauzas drops off to dinner | 16:46 | |
| *** ericksonsantos has joined #openstack-nova | 16:46 | |
| bauzas | tdurakov: lxsli: interesting questions and I should provide my thoughts, but I'm stomach-driven | 16:46 |
| mriedem | it's hard knowing if you're losing test coverage in some way, or if you're introducing bugs with the new test structure | 16:46 |
| tdurakov | stomach-driven-development:) | 16:47 |
| *** david-lyle has joined #openstack-nova | 16:47 | |
| bauzas | always a bad idea | 16:47 |
| lxsli | mriedem: the point #4 implies you should update tests if you're changing them though, I'm just following the advice | 16:47 |
| *** gszasz has quit IRC | 16:47 | |
| bauzas | don't discuss when you're hungry | 16:47 |
| tdurakov | bauzas, +1 | 16:47 |
| lxsli | although I admit I hate mox so I have motivation | 16:47 |
| mriedem | lxsli: in an extreme way | 16:47 |
| *** sahid has quit IRC | 16:48 | |
| *** mtanino has joined #openstack-nova | 16:48 | |
| tdurakov | lxsli, let me do some archeology | 16:48 |
| *** IanGovett has quit IRC | 16:48 | |
| tdurakov | lxsli, example for you https://review.openstack.org/#/c/159442/ | 16:49 |
| johnthetubaguy | tdurakov: this has a reason for change, thats the difference | 16:49 |
| johnthetubaguy | it all seems boarderline | 16:49 |
| lxsli | I feel damned either way | 16:49 |
| lxsli | If I combine them, it's dead because it's too hard to review | 16:50 |
| tdurakov | johnthetubaguy, so, decide:) | 16:50 |
| lxsli | If I split them, it's dead because policy | 16:50 |
| *** derekh has quit IRC | 16:50 | |
| alaski | mriedem: yeah, network notifications aren't going to work for them (re: operator email). It's an easy fix but will have to wait for M | 16:50 |
| johnthetubaguy | tdurakov: I am saying its OK, but there is a chance no one will review it | 16:50 |
| *** ndipanov has quit IRC | 16:50 | |
| mriedem | alaski: is it a feature change? | 16:51 |
| alaski | mriedem: it's an RPC addition | 16:51 |
| mriedem | oh | 16:51 |
| alaski | cells is missing the plumbing for that call | 16:51 |
| johnthetubaguy | lxsli: I think once folks look at the last patch, it will probably justify the test changes, and thats cool | 16:51 |
| mriedem | yeah it waits then i guess | 16:51 |
| johnthetubaguy | tdurakov: if we were moving mock tests to mox, I am totally against that | 16:51 |
| johnthetubaguy | tdurakov: if there was no reasons for change the tests, and it was a mox -> mock re-write, I would be against that too, although I would feel bad about it | 16:52 |
| tdurakov | johnthetubaguy, who is going to move from mock to mox?O_o | 16:52 |
| johnthetubaguy | tdurakov: I -1ed a patch earlier in the week :) | 16:52 |
| *** unicell has joined #openstack-nova | 16:52 | |
| johnthetubaguy | although that was a less clear cut case, but still | 16:53 |
| mriedem | i'm not sure that https://review.openstack.org/#/c/219761/ justifies the 4 changes before it, at least not as a priority | 16:53 |
| mriedem | certainly not for M | 16:53 |
| dansmith | alaski: mriedem yeah | 16:53 |
| mriedem | *L i mean | 16:53 |
| *** lucasagomes has quit IRC | 16:53 | |
| lxsli | mriedem: that change is in preparation for check-destination-on-migration | 16:54 |
| *** nic has joined #openstack-nova | 16:54 | |
| tdurakov | johnthetubaguy, if you ok with that, lets review all of these patched then | 16:54 |
| lxsli | mriedem: unfortunately that can't have two parents so I've elected to base off bauzas/alaski's work | 16:54 |
| johnthetubaguy | tdurakov: I would rather you reviewed bug fixes though | 16:54 |
| mriedem | i guess i just saw https://review.openstack.org/#/c/219761/ at the top of the series | 16:55 |
| *** changbl has joined #openstack-nova | 16:55 | |
| lxsli | mriedem: yes, it's an artefact of Gerrit | 16:55 |
| mriedem | lxsli: where is the change for check-destination-on-migration ? | 16:55 |
| lxsli | mriedem: https://review.openstack.org/#/c/212510/ | 16:55 |
| * johnthetubaguy bravely sneaks away, before he is crazy later for his rehearsal again | 16:56 | |
| lxsli | it's not ready for review yet but the whole other chain we're discussing grew out of wanting to clean up the conductor slightly so I could implement this spec | 16:56 |
| *** salv-orlando has quit IRC | 16:56 | |
| tdurakov | alaski, hi, got fast question about call-to-cast, how we are going to detect that request was send, but wasn't received? | 16:56 |
| mriedem | lxsli: yeah, so that's nowhere near something we are going to have in liberty | 16:56 |
| lxsli | mriedem: no absolutely not | 16:57 |
| *** fawadkhaliq has joined #openstack-nova | 16:57 | |
| *** sneti has quit IRC | 16:57 | |
| mriedem | so right now this series of unit test changes is a big distraction | 16:57 |
| lxsli | I'd be happy if it made Mitaka | 16:57 |
| *** _kiran_ has joined #openstack-nova | 16:57 | |
| mriedem | so, let's just table this | 16:57 |
| lxsli | right and I'm not begging anyone for attention | 16:57 |
| alaski | tdurakov: at the moment we can't | 16:57 |
| mriedem | lxsli: you can blame tdurakov | 16:57 |
| lxsli | I do entirely :) | 16:57 |
| *** su_zhang has joined #openstack-nova | 16:57 | |
| lxsli | OK I'm going to go home, laters o/ | 16:57 |
| tdurakov | alaski, in perspective i mean:) | 16:57 |
| *** pumaranikar has quit IRC | 16:58 | |
| mriedem | i'm inclinded to just -2 https://review.openstack.org/#/c/221014/ | 16:58 |
| mriedem | until mitaka opens | 16:58 |
| *** su_zhang has quit IRC | 16:58 | |
| dansmith | I'm inclined to agree | 16:58 |
| *** piyanai has joined #openstack-nova | 16:58 | |
| *** su_zhang has joined #openstack-nova | 16:58 | |
| alaski | tdurakov: eventually we'll want something to monitor that tasks that should be making progress are making progress, and time them out and retry them at some point | 16:59 |
| mriedem | dansmith: done | 16:59 |
| *** Jeffrey4l has quit IRC | 16:59 | |
| mriedem | now i'm inclined to get lunch | 16:59 |
| tdurakov | alaski, ack, thanks! | 16:59 |
| *** kiran-r has quit IRC | 16:59 | |
| *** haomaiwang has quit IRC | 17:01 | |
| *** moshele has joined #openstack-nova | 17:01 | |
| *** haomaiwa_ has joined #openstack-nova | 17:01 | |
| alex_xu | johnthetubaguy: dansmith I found one more issue, and hesitate how to deal with it. | 17:01 |
| *** dave-mccowan has joined #openstack-nova | 17:02 | |
| alex_xu | johnthetubaguy: dansmith the security_groups API is special case. It is the only API without json-schema, that means both v2.1 and v2.1 compat mode didn't strip the name and description field. https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/security_groups.py#L182 | 17:03 |
| *** david-lyle has quit IRC | 17:03 | |
| *** rdopiera has quit IRC | 17:03 | |
| alex_xu | johnthetubaguy: dansmith so I should fix them just for v2.1 compat mode, or fix both? | 17:03 |
| *** harlowja has joined #openstack-nova | 17:04 | |
| alex_xu | and the Ken'ichi will add json-schema for security-groups in v2.1 https://review.openstack.org/189544 | 17:04 |
| alex_xu | or I just leave the security_groups api alone? | 17:04 |
| *** salv-orlando has joined #openstack-nova | 17:05 | |
| dansmith | alex_xu: I dunno, probably leave it for now, and we can follow up with another patch for it when people are back for discussion | 17:05 |
| alex_xu | dansmith: ok, thanks, that is a way forward | 17:06 |
| openstackgerrit | Alex Xu proposed openstack/nova: Filter leading/trailing spaces for name field in v2.1 compat mode https://review.openstack.org/220791 | 17:09 |
| alex_xu | dansmith: just correct the commit message ^ hope that better | 17:09 |
| dansmith | alex_xu: will look in a sec | 17:09 |
| alex_xu | dansmith: thanks | 17:09 |
| *** ijw has quit IRC | 17:10 | |
| *** browne has quit IRC | 17:11 | |
| *** browne has joined #openstack-nova | 17:11 | |
| *** hitalia has quit IRC | 17:12 | |
| *** danpb has quit IRC | 17:15 | |
| *** Marga_ has quit IRC | 17:15 | |
| *** tdurakov has quit IRC | 17:16 | |
| *** Marga_ has joined #openstack-nova | 17:16 | |
| *** tdurakov_ has quit IRC | 17:17 | |
| *** piyanai has quit IRC | 17:17 | |
| *** angdraug has quit IRC | 17:17 | |
| *** lsmola has quit IRC | 17:17 | |
| *** karimb has quit IRC | 17:17 | |
| *** hitalia has joined #openstack-nova | 17:18 | |
| *** _kiran_ has quit IRC | 17:19 | |
| *** ijw has joined #openstack-nova | 17:19 | |
| *** piyanai has joined #openstack-nova | 17:20 | |
| *** tonytan4ever has quit IRC | 17:20 | |
| *** lbeliveau has quit IRC | 17:22 | |
| *** VW__ has joined #openstack-nova | 17:22 | |
| *** mriedem has quit IRC | 17:23 | |
| *** piyanai has quit IRC | 17:24 | |
| *** VW___ has joined #openstack-nova | 17:24 | |
| *** VW__ has quit IRC | 17:24 | |
| *** VW___ has quit IRC | 17:25 | |
| *** VW_ has quit IRC | 17:25 | |
| *** VW_ has joined #openstack-nova | 17:25 | |
| *** rpodolyaka1 has quit IRC | 17:26 | |
| *** erhudy has joined #openstack-nova | 17:26 | |
| *** mriedem has joined #openstack-nova | 17:27 | |
| *** haomaiwa_ has quit IRC | 17:28 | |
| *** piyanai has joined #openstack-nova | 17:29 | |
| *** ssurana has joined #openstack-nova | 17:29 | |
| *** tonytan4ever has joined #openstack-nova | 17:30 | |
| *** moshele has quit IRC | 17:31 | |
| *** penick has joined #openstack-nova | 17:34 | |
| *** alex_klimov has joined #openstack-nova | 17:34 | |
| mriedem | PaulMurray: is hpcloud running neutron and if so do you know what version of nova/neutron? | 17:37 |
| *** mragupat has joined #openstack-nova | 17:37 | |
| mriedem | PaulMurray: nvm | 17:37 |
| mriedem | infra is hitting a thing | 17:37 |
| *** achanda has joined #openstack-nova | 17:39 | |
| *** rlrossit has quit IRC | 17:41 | |
| *** sneti has joined #openstack-nova | 17:45 | |
| *** rlrossit has joined #openstack-nova | 17:46 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Use the oslo.utils.reflection to extract the class name https://review.openstack.org/224843 | 17:46 |
| hemna | mriedem, anyone had time to check this one out yet? https://bugs.launchpad.net/nova/+bug/1493512 | 17:49 |
| openstack | Launchpad bug 1493512 in OpenStack Compute (nova) "n-cpu misreports shared storage when attempting to block migrate" [Undecided,New] | 17:49 |
| hemna | our QA guy says it's still happening | 17:49 |
| *** su_zhang has quit IRC | 17:50 | |
| *** su_zhang has joined #openstack-nova | 17:50 | |
| *** ociuhandu has quit IRC | 17:50 | |
| *** nd has quit IRC | 17:52 | |
| *** sneti has quit IRC | 17:53 | |
| mriedem | hemna: nope i haven't at least | 17:54 |
| mriedem | hemna: i'd probably try to find ndipanov to take a look at that one, or ftersin | 17:55 |
| hemna | can we get someone assigned the bug to take a look at it ? | 17:56 |
| hemna | I guess I could change the assigned to, but didn't want to assume | 17:56 |
| *** ijw has quit IRC | 17:57 | |
| mriedem | hemna: i wouldn't assume | 17:57 |
| doffm | hemna: I'll take a look. | 17:59 |
| *** VW_ has quit IRC | 17:59 | |
| *** VW_ has joined #openstack-nova | 17:59 | |
| mriedem | hemna: sseverson: i see https://review.openstack.org/#/c/91722/ was called out, which is interesting, since that blew up upgrades in another case from what i remember | 17:59 |
| mriedem | dansmith: ^ remember? | 17:59 |
| *** vilobhmm_11 has joined #openstack-nova | 18:00 | |
| *** sneti has joined #openstack-nova | 18:00 | |
| dansmith | yes I fucking remember that | 18:00 |
| * dansmith steams | 18:00 | |
| hemna | doh | 18:00 |
| *** thangp has quit IRC | 18:00 | |
| mriedem | hemna: sseverson: https://review.openstack.org/#/c/108468/ came after that for live migration + NFS | 18:00 |
| mriedem | dansmith: heh, i figured you would | 18:01 |
| mriedem | fondly | 18:01 |
| sseverson | mriedem: So it's possible that another check is busted? | 18:02 |
| mriedem | dansmith: do you remember what the patch was that had to fix that upgrade issue? | 18:02 |
| mriedem | sseverson: yeah | 18:02 |
| mriedem | sseverson: well you said this used to work, but that must have been awhile ago b/c that change was made in juno | 18:02 |
| *** su_zhang_ has joined #openstack-nova | 18:02 | |
| dansmith | mriedem: the original patch just bombed out if the version was lower than the new one it added | 18:02 |
| dansmith | which is BS | 18:02 |
| *** pumaranikar has joined #openstack-nova | 18:02 | |
| dansmith | so the fix was to remove the bomb, log a warning when we had to use the old version | 18:03 |
| mriedem | oh i sort of remember now | 18:03 |
| *** promulo has quit IRC | 18:03 | |
| *** VW_ has quit IRC | 18:03 | |
| dansmith | the patch was fixing a safety issue, which had been around for a long time, which was good, but we had to honor the older version if that was required because otherwise people could never migrate to newer code | 18:03 |
| *** VW_ has joined #openstack-nova | 18:04 | |
| *** READ10 has quit IRC | 18:04 | |
| *** jamielennox is now known as jamielennox|away | 18:05 | |
| *** ctrath has quit IRC | 18:05 | |
| *** su_zhang has quit IRC | 18:05 | |
| *** ccarmack has quit IRC | 18:06 | |
| *** e0ne has joined #openstack-nova | 18:07 | |
| sseverson | mriedem: I can say for sure that it worked in Kilo. Juno saw some unrelated issues that kept me from trying this. | 18:08 |
| mriedem | sseverson: oh, hm | 18:08 |
| mriedem | weird | 18:08 |
| *** ctrath has joined #openstack-nova | 18:10 | |
| doffm | Stack traces for #1346385 (Fixed by https://review.openstack.org/#/c/108468/) and #1493512 look similar, but not the same. Could be superficial. | 18:11 |
| openstackgerrit | Ivan Mironov proposed openstack/nova: nova.console.websocketproxy: Ignore malformed cookies. https://review.openstack.org/224854 | 18:13 |
| *** ccarmack has joined #openstack-nova | 18:13 | |
| openstackgerrit | Ivan Mironov proposed openstack/nova: websocketproxy: Ignore malformed cookies. https://review.openstack.org/224854 | 18:13 |
| *** tonytan4ever has quit IRC | 18:15 | |
| *** b3nt_pin has quit IRC | 18:15 | |
| *** e0ne has quit IRC | 18:17 | |
| mriedem | looking for changes to shared storage stuff in the compute manager in liberty i see some things around rebuild but not live migration | 18:17 |
| mriedem | time to dig into the libvirt driver | 18:17 |
| *** beagles has joined #openstack-nova | 18:20 | |
| *** beagles is now known as Guest6478 | 18:20 | |
| *** ociuhandu has joined #openstack-nova | 18:20 | |
| *** e0ne has joined #openstack-nova | 18:21 | |
| mriedem | hemna: sseverson: https://review.openstack.org/#/c/195885/ | 18:23 |
| *** VW_ has quit IRC | 18:23 | |
| mriedem | or maybe you've already seen that | 18:23 |
| mriedem | but that's what i think makes this explode in liberty | 18:23 |
| *** VW_ has joined #openstack-nova | 18:23 | |
| mriedem | doffm: ^ | 18:24 |
| sseverson | mriedem: Looks like they were attempting to fix a similar case. Oh sweet irony. | 18:25 |
| hemna | yah that looks like an attempt to solve a similar issue | 18:25 |
| mriedem | so that might have fixed non-shared storage live migration of an instance booted from volume, but breaks live migration with shared storage in the process? | 18:26 |
| hemna | mriedem, I think sseverson's bug is when there is no shared storage | 18:26 |
| hemna | it raises an exception saying it can't block migrate "InvalidLocalStorage: cld5b12 is not on local storage: Block migration can not be used with shared storage." | 18:27 |
| *** VW_ has quit IRC | 18:27 | |
| *** VW_ has joined #openstack-nova | 18:27 | |
| sseverson | mriedem: Except that I did remove the shared storage as a possibility and still came across the "cannot be used with shared storage" error. | 18:28 |
| *** vladikr has joined #openstack-nova | 18:28 | |
| sseverson | But it definitely looks like we're on the right track. | 18:28 |
| mriedem | so someone was working on adding a boot from volume live migration test in tempest | 18:29 |
| hemna | sseverson, I think the use case you are running into is a boot from cinder volume, and a nova flavor that has no local storage. | 18:29 |
| *** vladikr_ has joined #openstack-nova | 18:29 | |
| *** rushil has quit IRC | 18:29 | |
| hemna | mriedem, yah my other coworker Arkady | 18:29 |
| mriedem | hemna: https://review.openstack.org/#/c/165233/ | 18:29 |
| mriedem | yeah | 18:30 |
| *** burt has quit IRC | 18:30 | |
| mriedem | so that was passing tests | 18:30 |
| hemna | was an intern that moved on last week | 18:30 |
| mriedem | what's the difference? | 18:30 |
| *** rushil has joined #openstack-nova | 18:30 | |
| *** mmedvede has quit IRC | 18:31 | |
| *** fawadkhaliq has quit IRC | 18:31 | |
| *** fawadkhaliq has joined #openstack-nova | 18:31 | |
| hemna | not sure | 18:32 |
| *** rushil has quit IRC | 18:32 | |
| hemna | the flavor comes frmo CONF.compute.flavor_ref | 18:32 |
| *** fawadkhaliq has quit IRC | 18:32 | |
| *** vladikr has quit IRC | 18:33 | |
| doffm | Is cinder the difference? Are the tests using different volumes? (LVM) | 18:33 |
| mriedem | hemna: which is https://github.com/openstack-dev/devstack/blob/master/lib/tempest#L215 | 18:33 |
| mriedem | doffm: the tempest runs are using lvm | 18:34 |
| hemna | does that flavor have local storage ? | 18:34 |
| *** su_zhang_ has quit IRC | 18:35 | |
| hemna | the tempest default looks like flavor_ref=1 https://github.com/openstack/tempest/blob/master/tempest/config.py#L210-L211 | 18:36 |
| *** penick has quit IRC | 18:41 | |
| sseverson | mriedem: From the test notes: "Make sure to not select block migration..." | 18:42 |
| *** pumaranikar has quit IRC | 18:42 | |
| openstackgerrit | Tomoki Sekiyama proposed openstack/nova: Fix multipath iSCSI encrypted volume attach failure https://review.openstack.org/196482 | 18:43 |
| *** pumaranikar has joined #openstack-nova | 18:44 | |
| *** mmedvede has joined #openstack-nova | 18:45 | |
| mriedem | hemna: sseverson: ok i've marked the bug as liberty-rc-potential to get it on the radar since not being able to do live migration would be, not good | 18:46 |
| hemna | ok thanks man | 18:47 |
| *** greghaynes has quit IRC | 18:49 | |
| mriedem | hemna: i guess tempest.api.compute.admin.test_live_migration.LiveBlockMigrationTestJSON.test_iscsi_volume ... SKIPPED: Block Live migration not configured for iSCSI is for block storage volume migration? | 18:50 |
| hemna | humm | 18:51 |
| *** penick has joined #openstack-nova | 18:51 | |
| *** penick has quit IRC | 18:51 | |
| *** tojuvone has joined #openstack-nova | 18:53 | |
| *** piyanai has quit IRC | 18:53 | |
| *** jichen has joined #openstack-nova | 18:54 | |
| hemna | https://github.com/openstack/tempest/blob/master/tempest/api/compute/admin/test_live_migration.py#L149-L152 | 18:54 |
| hemna | that setting is defaulted to False | 18:55 |
| mriedem | yar | 18:55 |
| mriedem | https://github.com/openstack/tempest/blob/master/tempest/config.py#L363 | 18:55 |
| *** burgerk__ has joined #openstack-nova | 18:55 | |
| mriedem | b/c of https://bugs.launchpad.net/nova/+bug/1398999 | 18:55 |
| openstack | Launchpad bug 1398999 in nova (Ubuntu Wily) "Block migrate with attached volumes copies volumes to themselves" [High,Triaged] | 18:55 |
| hemna | heh I was just looking at that. | 18:55 |
| mriedem | sseverson: hemna: https://review.openstack.org/#/c/139085/ | 18:57 |
| *** burgerk_ has quit IRC | 18:58 | |
| *** pixelb has quit IRC | 18:58 | |
| hemna | was also backported to juno | 18:58 |
| hemna | https://review.openstack.org/#/c/176768 | 18:59 |
| *** tonytan4ever has joined #openstack-nova | 18:59 | |
| *** pixelb has joined #openstack-nova | 18:59 | |
| *** burgerk__ has quit IRC | 19:00 | |
| sseverson | mriedem: It's interesting that I'm not getting that "mapped volumes" error. | 19:01 |
| mriedem | sseverson: maybe you just don't get that far | 19:02 |
| mriedem | because of https://review.openstack.org/#/c/195885/ ? | 19:02 |
| sseverson | Hmm, I could see that, since I"m still in the pre-migration phase. | 19:02 |
| *** Guest6478 has quit IRC | 19:02 | |
| *** beagles_ has joined #openstack-nova | 19:03 | |
| *** beagles_ is now known as b3nt_pin | 19:03 | |
| *** b3nt_pin is now known as Guest31268 | 19:04 | |
| openstackgerrit | Joshua Harlow proposed openstack/nova: Iteratively generate host states from 'get_all_host_states' https://review.openstack.org/224373 | 19:04 |
| mriedem | this is unrelated, but looks dumb | 19:04 |
| mriedem | http://git.openstack.org/cgit/openstack/nova/tree/nova/virt/libvirt/driver.py#n6213 | 19:04 |
| *** Guest31268 has quit IRC | 19:04 | |
| mriedem | in pre_live_migration we connect volumes and then in the next step if we're doing block migration we raise an error | 19:05 |
| mriedem | after we've attached the volumes | 19:05 |
| mriedem | i think the rollback detaches the volumes, but still, derp | 19:05 |
| *** beagles_ has joined #openstack-nova | 19:05 | |
| hemna | check_can_live_migrate_source is where the error is raised | 19:05 |
| hemna | I presume that's in pre_live_migration check | 19:05 |
| hemna | trying to find the order of operations | 19:05 |
| *** beagles_ has quit IRC | 19:06 | |
| hemna | looks like it's an rpc into the compute manager itself. so it's probably before the pre_live_migration phase ? | 19:08 |
| hemna | dunno | 19:08 |
| mriedem | yeah | 19:08 |
| *** devlaps has quit IRC | 19:08 | |
| mriedem | check_can_live_migrate_source is called from compute manager | 19:08 |
| hemna | so that's probably why we don't see the exception from https://review.openstack.org/#/c/139085 | 19:10 |
| *** rdxc_2 has joined #openstack-nova | 19:10 | |
| *** david-lyle has joined #openstack-nova | 19:11 | |
| *** piyanai has joined #openstack-nova | 19:12 | |
| mriedem | bauzas: i appreciated your comments on that PTL thread in the ML :) | 19:13 |
| *** rlrossit has quit IRC | 19:13 | |
| mriedem | bauzas: imagine, having to be responsible for knowing when the PTL elections come up when you want to run for PTL | 19:13 |
| mriedem | like the people that would be in the lab in college the night before a project is due when they've had a month to work on it, and then complaining that the teacher was unfair in giving the hard assignment in the first place | 19:15 |
| *** piyanai has quit IRC | 19:16 | |
| bauzas | mriedem: yeah... | 19:16 |
| bauzas | mriedem: I tried not to provide a grumpy email | 19:16 |
| mriedem | bauzas: i was able to provide that :) | 19:17 |
| bauzas | but I really dislike the discussion | 19:17 |
| bauzas | if you want to be a PTL, you have to understand the x-project stuff and all the stuff | 19:17 |
| mriedem | http://www.ww2aircraft.net/forum/attachments/off-topic-misc-/249186d1386106954-grumpy-asshole770346tv8.jpg | 19:17 |
| *** tdurakov has joined #openstack-nova | 19:17 | |
| dansmith | my feeling on this one way or the other: http://img.memecdn.com/no-f-k-to-give_o_2090489.webp | 19:18 |
| bauzas | mriedem: tbh, I also dislike having candidates providing each cycle a candidacy by the last 5 mins | 19:18 |
| hemna | lol | 19:18 |
| * bauzas trying not to fingerpoint | 19:18 | |
| mriedem | dansmith: but but but | 19:18 |
| bauzas | but all of that just pisses me off | 19:18 |
| dansmith | mriedem: sorry, no fucks available | 19:19 |
| mriedem | overquota on fucks?! | 19:19 |
| mriedem | c'mon | 19:19 |
| dansmith | HTTP 404: Fucks not found | 19:19 |
| mriedem | language, please | 19:19 |
| bauzas | heh | 19:19 |
| dansmith | I feel the same way about your language complaint, FYI | 19:19 |
| *** raildo is now known as raildo-afk | 19:19 | |
| vilobhmm_11 | dansmith, bauzas, melwitt : https://etherpad.openstack.org/p/servicegroup-refactoring please have a look at the etherpad..have included various options we discussed yesterday…also classified the liveliness vs non-liveliness API's | 19:20 |
| bauzas | vilobhmm_11: ack | 19:21 |
| dansmith | vilobhmm_11: heh, that one method is pretty much what I was saying ... :) | 19:21 |
| vilobhmm_11 | which method dansmith | 19:21 |
| vilobhmm_11 | min/max ? | 19:21 |
| dansmith | vilobhmm_11: the one liveliness method.. is_up | 19:21 |
| bauzas | vilobhmm_11: heh | 19:21 |
| *** matrohon has joined #openstack-nova | 19:21 | |
| bauzas | vilobhmm_11: yeah only is_up is needed IMHO | 19:22 |
| vilobhmm_11 | yup! | 19:22 |
| bauzas | but that leaves the discussion to be agreed | 19:22 |
| bauzas | vilobhmm_11: happy with providing a new spec for mitaka adding those details ? | 19:22 |
| bauzas | vilobhmm_11: I want to make sure that we have consensus on that, since it seems it was not the case | 19:23 |
| *** rlrossit has joined #openstack-nova | 19:23 | |
| bauzas | vilobhmm_11: also, MHO is that the SG API should be using the Service object as the Facade | 19:24 |
| bauzas | vilobhmm_11: and change all the calls to the SG API to go by the Service objec | 19:24 |
| bauzas | vilobhmm_11: like becoming Service.is_up() | 19:24 |
| dansmith | bauzas: I dunno about that necessarily.. I could see it multiple ways | 19:24 |
| dansmith | that feels tighter, but it also mixes the persistence with the is-it-up-ness which feel like different layers | 19:25 |
| *** beagles has joined #openstack-nova | 19:25 | |
| dansmith | so I dunno | 19:25 |
| *** david-lyle has quit IRC | 19:25 | |
| bauzas | dansmith: what I like with objects is that you have a Facade not necessarly needing the same model for persistence | 19:25 |
| *** diogogmt has quit IRC | 19:25 | |
| dansmith | but that's a relatively small detail | 19:25 |
| dansmith | yes | 19:25 |
| dansmith | agreed with that | 19:25 |
| *** beagles is now known as Guest49018 | 19:25 | |
| *** rpodolyaka1 has joined #openstack-nova | 19:25 | |
| dansmith | it also means that potentially the compute services would not need to know what we were using for service liveness from day to tday | 19:25 |
| bauzas | dansmith: like, if I had time, I would love to get rid of HostState and use the ComputeNode object | 19:26 |
| *** diogogmt has joined #openstack-nova | 19:26 | |
| dansmith | which would make it easier to upgrade from db to zk for example | 19:26 |
| *** Marga_ has quit IRC | 19:26 | |
| bauzas | dansmith: yeah, I mean | 19:26 |
| vilobhmm_11 | dansmith : +1 for "it also means that potentially the compute services would not need to know what we were using for service liveness from day to tday" | 19:26 |
| bauzas | dansmith: I don't necessarly want to expose which model to use | 19:26 |
| *** claudiub has quit IRC | 19:27 | |
| bauzas | dansmith: like, call a Facade, and leave that facade using the right model | 19:27 |
| *** diogogmt has quit IRC | 19:27 | |
| bauzas | that way, it would make the same interface for 2 models | 19:27 |
| dansmith | yeah, I get that | 19:27 |
| bauzas | which would help the db/zk stuff | 19:27 |
| bauzas | but that's MHO, that's an implementation detail that we can discuss later on | 19:28 |
| *** armax has quit IRC | 19:28 | |
| vilobhmm_11 | dansmith, bauzas : please check #42, #43 https://etherpad.openstack.org/p/servicegroup-refactoring thats the only thing which we need to care about if liveliness needs to go to zk/memcache | 19:29 |
| mriedem | hemna: sseverson: another related change https://review.openstack.org/#/c/135074/ | 19:30 |
| dansmith | vilobhmm_11: is there any reason to keep it in zk instead of the db? | 19:30 |
| bauzas | vilobhmm_11: I don't get your point | 19:30 |
| *** rpodolyaka1 has quit IRC | 19:31 | |
| *** rpodolyaka2 has joined #openstack-nova | 19:31 | |
| bauzas | vilobhmm_11: disabled is not a liveliness check, it's a status | 19:31 |
| *** rpodolyaka2 has quit IRC | 19:31 | |
| vilobhmm_11 | dansmith : because liveliness is tied to "disabled/enabled" | 19:31 |
| dansmith | no | 19:31 |
| bauzas | vilobhmm_11: where ? | 19:31 |
| dansmith | it's not | 19:31 |
| dansmith | disabled and is-up are completely orthogonal | 19:31 |
| bauzas | dansmith: maybe he's referring to the DB driver, I need to doublecheclk | 19:31 |
| dansmith | no, he's asking about whether we store the disabled in zk or the db, when using zk | 19:32 |
| dansmith | *disabled flag | 19:32 |
| bauzas | okay, it's notr | 19:32 |
| bauzas | not | 19:32 |
| vilobhmm_11 | ok | 19:32 |
| * bauzas has fat fingers again | 19:32 | |
| vilobhmm_11 | then it should be fine | 19:32 |
| bauzas | vilobhmm_11: disabled is not checked by the SG API | 19:32 |
| * dansmith is creeped out by the fact that bauzas' fingers change diameter | 19:33 | |
| vilobhmm_11 | bauzas : I get that | 19:33 |
| vilobhmm_11 | dansmith : :P | 19:33 |
| bauzas | I ate pringles | 19:33 |
| bauzas | and I'm working at home with no sports | 19:33 |
| bauzas | so I'd guess my fingers are changing size | 19:33 |
| bauzas | but I'm diverging | 19:34 |
| vilobhmm_11 | dansmith : cool…so have listed most of the options here should help us to have a consensus soon…will update the mitake spec with the etherpad link https://etherpad.openstack.org/p/servicegroup-refactoring | 19:34 |
| dansmith | lol | 19:34 |
| dansmith | cool with me | 19:34 |
| vilobhmm_11 | dansmith, bauzas : thanks ! | 19:34 |
| bauzas | vilobhmm_11: thanks to you, will look at your change once it's up | 19:34 |
| *** alexpilotti has quit IRC | 19:35 | |
| vilobhmm_11 | bauzas : https://review.openstack.org/#/c/222423/ here is the spec | 19:35 |
| vilobhmm_11 | not changing the spec without having a consensus on the etherpad option | 19:35 |
| bauzas | vilobhmm_11: ack | 19:35 |
| dansmith | vilobhmm_11: the spec is where we need to record consensus.. how about you copy and modify with a new spec so we can have both to look at and then you can drop the one we don't do? | 19:36 |
| dansmith | the alternative is we try to get all the interested parties in the same place at the same time and try to hash it out ahead of time, | 19:37 |
| dansmith | but that will be harder for sure | 19:37 |
| *** Guest49018 has quit IRC | 19:37 | |
| vilobhmm_11 | dansmith : would prefer the second option..I guess we need jaypipes, johnthetubaguy and melwitt | 19:38 |
| vilobhmm_11 | and both of your's agreement before this makes its way | 19:38 |
| dansmith | I think sdague too as I think he had some thoughts on this | 19:38 |
| dansmith | re: deployment complexity | 19:39 |
| vilobhmm_11 | for sure… | 19:39 |
| *** ZZelle_ has joined #openstack-nova | 19:39 | |
| *** Apoorva has joined #openstack-nova | 19:42 | |
| *** Piet has quit IRC | 19:47 | |
| doffm | sseverson: Could I get some clarification on how to repeat #1493512. My understanding. 1 No shared storage. 2 Boot from volume (LVM or Cinder) 3 using block migration. | 19:49 |
| *** pixelb has quit IRC | 19:49 | |
| bauzas | alaski: thanks for your comments on https://review.openstack.org/#/c/221280/2/doc/source/tasks.rst,cm | 19:50 |
| bauzas | alaski: when saying that you'd like to remove the proposals, I guess I should rather remove all the below details and just provide a quick direction where we want to go, right ? | 19:51 |
| *** derekh has joined #openstack-nova | 19:51 | |
| bauzas | you mentioned a summary, I think I got the idea | 19:51 |
| alaski | bauzas: that would be my preference | 19:51 |
| bauzas | alaski: makes sense | 19:51 |
| *** burt has joined #openstack-nova | 19:51 | |
| bauzas | alaski: you made a valid point, adding more details could bound the future discussions by the ones provided by the devref - or change the devref | 19:52 |
| *** raildo-afk is now known as raildo | 19:52 | |
| alaski | bauzas: right. there's a lot of good detail in there that should stay. I just think a different focus for it would be preferable | 19:53 |
| bauzas | gotcha | 19:53 |
| *** casusbelli has quit IRC | 19:53 | |
| *** su_zhang has joined #openstack-nova | 19:57 | |
| *** sneti has quit IRC | 19:57 | |
| *** sneti has joined #openstack-nova | 19:57 | |
| *** earlephilhower has quit IRC | 19:59 | |
| *** ssurana1 has joined #openstack-nova | 20:02 | |
| *** su_zhang has quit IRC | 20:02 | |
| *** ssurana has quit IRC | 20:02 | |
| *** atuvenie has quit IRC | 20:02 | |
| *** diogogmt has joined #openstack-nova | 20:03 | |
| *** beagles_ has joined #openstack-nova | 20:03 | |
| *** dustins has quit IRC | 20:06 | |
| *** Piet has joined #openstack-nova | 20:07 | |
| *** ijw has joined #openstack-nova | 20:08 | |
| *** beagles_ has quit IRC | 20:08 | |
| *** tdurakov has quit IRC | 20:11 | |
| *** Piet has quit IRC | 20:12 | |
| *** diogogmt has quit IRC | 20:12 | |
| *** diogogmt_ has joined #openstack-nova | 20:12 | |
| *** Piet has joined #openstack-nova | 20:13 | |
| *** Piet has quit IRC | 20:13 | |
| *** lucasagomes has joined #openstack-nova | 20:17 | |
| *** tdurakov has joined #openstack-nova | 20:18 | |
| *** gholler has quit IRC | 20:19 | |
| *** fifieldt has quit IRC | 20:19 | |
| *** tdurakov has quit IRC | 20:20 | |
| *** fifieldt has joined #openstack-nova | 20:21 | |
| *** penick has joined #openstack-nova | 20:21 | |
| *** derekh has quit IRC | 20:21 | |
| *** mmedvede has quit IRC | 20:22 | |
| *** smatzek has quit IRC | 20:22 | |
| *** Marga_ has joined #openstack-nova | 20:26 | |
| _diana_ | I'm a new OpenStack developer trying to set up a multi-node devstack (1 controller, 2 compute nodes) to test live-migration. | 20:27 |
| _diana_ | The controller appears to be working fine, but I don't think it knows about the 2 compute nodes ('nova hypervisor-list' only shows the controller host). It should show all 3 hosts, correct? | 20:27 |
| *** haomaiwang has joined #openstack-nova | 20:29 | |
| *** lucasagomes has quit IRC | 20:29 | |
| *** mragupat has quit IRC | 20:29 | |
| *** karimb has joined #openstack-nova | 20:29 | |
| *** gyee has joined #openstack-nova | 20:31 | |
| *** vladikr_ has quit IRC | 20:31 | |
| *** penick has quit IRC | 20:31 | |
| *** haomaiwang has quit IRC | 20:33 | |
| *** diogogmt_ has quit IRC | 20:35 | |
| *** mmedvede has joined #openstack-nova | 20:35 | |
| artom | _diana_, you're running devstack? Are all nodes connecting to the rabbitmq with no issues? | 20:36 |
| *** tonytan4ever has quit IRC | 20:39 | |
| _diana_ | artom: no (I have AMQP errors), and that's what I'm currently looking at. I did add 'iptables -I INPUT -p tcp -m tcp --dport 5672 -j ACCEPT', but perhaps I need to revisit a iptables how-to. | 20:40 |
| artom | _diana_, it might be something as stupid as rabbitmq not running. | 20:41 |
| _diana_ | artom: yeah... I'll dig through the logs some more ;) thx | 20:42 |
| *** ssurana1 has quit IRC | 20:44 | |
| *** lucasagomes has joined #openstack-nova | 20:44 | |
| *** ssurana has joined #openstack-nova | 20:44 | |
| *** krtaylor has quit IRC | 20:45 | |
| *** su_zhang has joined #openstack-nova | 20:46 | |
| *** baoli has quit IRC | 20:46 | |
| *** baoli has joined #openstack-nova | 20:46 | |
| sseverson | doffm: That's correct. | 20:47 |
| *** ssurana1 has joined #openstack-nova | 20:47 | |
| *** ssurana has quit IRC | 20:48 | |
| *** e0ne has quit IRC | 20:50 | |
| *** annegentle has quit IRC | 20:52 | |
| *** tongli has quit IRC | 20:55 | |
| dansmith | _diana_: you want service-list | 20:56 |
| dansmith | _diana_: if service-list shows the compute service on the controller, then rabbit is working | 20:57 |
| *** penick has joined #openstack-nova | 20:57 | |
| _diana_ | dansmith: hmm... I tried that an hour ago ('nova service-list' that is), and it was only showing the controller host. Now it's showing all three :) No idea what changed... perhaps live migration will magically work now :) Off to try... thx! | 20:58 |
| *** penick has quit IRC | 20:59 | |
| dansmith | _diana_: well, which services does it show? | 20:59 |
| dansmith | _diana_: for one controller, you'd have a compute, conductor, and maybe something else there | 20:59 |
| dansmith | which are all for the controller node | 20:59 |
| dansmith | maybe pastebin the output so we can see | 20:59 |
| *** matrohon has quit IRC | 20:59 | |
| _diana_ | dansmith: http://paste.openstack.org/show/466852/ | 21:01 |
| dansmith | okay yeah, so that looks good | 21:01 |
| _diana_ | dansmith: earlier, there were only three rows | 21:01 |
| *** thorst has quit IRC | 21:01 | |
| dansmith | three services on the controller node, including a compute there | 21:02 |
| dansmith | and then a compute on each of the extra hosts | 21:02 |
| dansmith | all up, | 21:02 |
| dansmith | which means they're reporting over rabbit | 21:02 |
| dansmith | _diana_: so you're failing to do a live migrate, right? you're giving it a destination hostname, right? | 21:03 |
| dansmith | like: nova live-migration foobar compute-1.localhost.localdomain | 21:03 |
| *** raildo is now known as raildo-afk | 21:03 | |
| *** david-lyle has joined #openstack-nova | 21:06 | |
| *** ijw has quit IRC | 21:06 | |
| _diana_ | dansmith: instance creation is failing now, so I haven't tried another migration yet.. soon, hopefully :) | 21:06 |
| dansmith | _diana_: okay, failing how? | 21:06 |
| _diana_ | dansmith: Build of instance 338038d6-54ae-4a48-beb6-d53b3dd52ade aborted: Could not clean up failed build, not rescheduling | 21:07 |
| _diana_ | so not sure yet, digging | 21:07 |
| dansmith | nova show foobar | 21:07 |
| dansmith | may just give you NoValidHost | 21:07 |
| *** rlrossit has left #openstack-nova | 21:07 | |
| dansmith | pastebin the output of show and I'll point out a few things to look for | 21:07 |
| *** bswartz has joined #openstack-nova | 21:08 | |
| *** david-lyle has quit IRC | 21:09 | |
| *** baoli has quit IRC | 21:09 | |
| *** pumaranikar has quit IRC | 21:09 | |
| *** baoli has joined #openstack-nova | 21:09 | |
| artom | dansmith, so presumably service-list post-dates hypervisor-list, but does the latter show bad information? Or just less detailed than the former? | 21:11 |
| dansmith | artom: no, they show different things | 21:11 |
| *** eharney has quit IRC | 21:11 | |
| dansmith | hypervisor-list is just listing compute nodes, I think, which doesn't really tell you much about whether the service is up and running | 21:12 |
| *** sneti has quit IRC | 21:12 | |
| dansmith | knowing if it's running is key to knowing if rabbit is working | 21:12 |
| *** Guest47613 has joined #openstack-nova | 21:12 | |
| *** aysyd has quit IRC | 21:12 | |
| *** baoli has quit IRC | 21:14 | |
| _diana_ | dansmith: I unstacked & then stacked each of the compute nodes, and now I can create instances again. | 21:16 |
| dansmith | alright | 21:16 |
| _diana_ | dansmith: live-migration fails though http://paste.openstack.org/show/466877/ | 21:16 |
| mriedem | ccarmack: http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul/layout.yaml#n5174 | 21:16 |
| artom | dansmith, so hypervisor-list just pulls information from the database. | 21:16 |
| mriedem | that shows the different CI jobs run against tempest and in which queues | 21:16 |
| dansmith | artom: well, so does service-list, but it's different information | 21:16 |
| mriedem | ccarmack: so these are the jobs run on the experimental queue for tempest http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul/layout.yaml#n5211 | 21:16 |
| mriedem | ccarmack: which includes gate-tempest-dsvm-cells | 21:17 |
| *** burt has quit IRC | 21:17 | |
| dansmith | _diana_: does service-list show that compute-1 is up? | 21:17 |
| artom | dansmith, https://github.com/openstack/nova/blob/master/nova/db/sqlalchemy/api.py#L588 | 21:17 |
| _diana_ | dansmith: and not specifying a host (so that one is chosen for me) results in: There are not enough hosts available. | 21:17 |
| dansmith | _diana_: because what it's telling you is that you can't live migrate there because it's not up | 21:17 |
| mriedem | ccarmack: so push your work in progress tempest change that skips security groups for that one scenario test, mark the change work in progress, and then post a comment of 'check experimental' on the tempest change to run the experimental queue jobs against it | 21:17 |
| artom | That says nothing about how that information *gets* in the database :P | 21:17 |
| dansmith | _diana_: maybe you tried too quickly after restacking or something else went wrong? | 21:17 |
| dansmith | artom: of course it does | 21:18 |
| mgagne | I'm trying to understand how Nova calls libvirt to tell it to migrate (or not) block devices (with --block-migrate). Can someone point me to the right code block/function? | 21:18 |
| _diana_ | dansmith: you are right! now the two compute nodes are down :( | 21:18 |
| _diana_ | they were totally up before! | 21:19 |
| artom | dansmith, it does o_O ? | 21:19 |
| dansmith | _diana_: so, don't restack, look at the log files and see what they're complaining about | 21:19 |
| artom | What's "it" in your sentence? ;) | 21:19 |
| dansmith | _diana_: the n-cpu logs on the two compute-only nodes | 21:19 |
| dansmith | artom: it's returning a straight up query of the compute_nodes table | 21:19 |
| *** tonytan4ever has joined #openstack-nova | 21:19 | |
| *** jichen has quit IRC | 21:19 | |
| *** tonytan4ever has quit IRC | 21:20 | |
| artom | dansmith, right, so how does that table get populated? Not through n-cpu on compute nodes "registering" themselves over rabbitmq? | 21:20 |
| _diana_ | dansmith: fixed the compute nodes... rabbit & iptables again | 21:21 |
| dansmith | artom: it does, but it just means that registration had happened at one point.. the services table is what keeps track of "can I talk to them *now*" | 21:21 |
| artom | I get what you're saying - service != compute node. But isn't there overlap? | 21:21 |
| dansmith | there's a relation there | 21:21 |
| _diana_ | dansmith: closer... "ERROR (BadRequest): controller.localhost.localdomain is not on shared storage: Live migration can not be used without shared storage except a booted from volume VM which does not have a local disk. (HTTP 400) (Request-ID: req-09bbebc9-1726-4b17-b999-9865c715f5db)" | 21:21 |
| dansmith | _diana_: yep, progress | 21:21 |
| _diana_ | dansmith: I have to go pick up my kid, but I'll play some more tonight. thanks a ton! | 21:22 |
| dansmith | _diana_: so either you need to use block migration, or put all of the compute nodes on shared storage for instance images | 21:22 |
| _diana_ | dansmith: progress indees | 21:22 |
| dansmith | _diana_: okay | 21:22 |
| *** ijw has joined #openstack-nova | 21:22 | |
| dansmith | _diana_: the latter is probably easier | 21:22 |
| _diana_ | dansmith: indeed | 21:22 |
| dansmith | _diana_: NFS mount the /var/lib/nova/instances on each compute node back to the controller | 21:22 |
| mgagne | dansmith: I'm trying to make the new migrate-disks feature from libvirt work | 21:23 |
| *** dims_ has quit IRC | 21:23 | |
| mgagne | dansmith: and I'm having issues understanding the somewhat spaghetti code (IMO) that is the live migration process. | 21:24 |
| *** armax has joined #openstack-nova | 21:24 | |
| dansmith | mgagne: yep, it's complicated | 21:24 |
| *** jamielennox|away is now known as jamielennox | 21:24 | |
| dansmith | mgagne: I could sit here and act like I'm not having to figure it out on my own, or you can talk to danpb tomorrow morning | 21:24 |
| mgagne | dansmith: libvirt has little to no doc about that feature except: pass in a list of disks and good luck. | 21:24 |
| dansmith | the latter will be faster and more pleasant for both of us | 21:24 |
| mgagne | alright, taking note | 21:24 |
| *** Guest47613 has quit IRC | 21:26 | |
| *** piyanai has joined #openstack-nova | 21:26 | |
| *** diogogmt has joined #openstack-nova | 21:27 | |
| *** erhudy has quit IRC | 21:29 | |
| mriedem | dansmith: _diana_: looks like you're talking about the exact same live migration bug that sseverson and hemna were looking at | 21:30 |
| openstackgerrit | Andrew Laski proposed openstack/nova: Devref: Document why conductor has a task api/manager https://review.openstack.org/224907 | 21:30 |
| dansmith | mriedem: well, she's just trying to get any live migration working, but okay | 21:30 |
| mriedem | bug 1493512 for reference | 21:30 |
| openstack | bug 1493512 in OpenStack Compute (nova) "n-cpu misreports shared storage when attempting to block migrate" [Undecided,New] https://launchpad.net/bugs/1493512 | 21:30 |
| mriedem | in the gate we just test with block migration, no volumes involved | 21:31 |
| mriedem | and there is a change in the works to test a volume-backed live migration w/o block migration | 21:31 |
| *** ccarmack has quit IRC | 21:31 | |
| mriedem | haven't seen anything for shared storage | 21:31 |
| *** pixelb has joined #openstack-nova | 21:33 | |
| *** ctrath has quit IRC | 21:33 | |
| *** mriedem has quit IRC | 21:33 | |
| *** tonytan4ever has joined #openstack-nova | 21:37 | |
| *** ccarmack has joined #openstack-nova | 21:37 | |
| *** ccarmack has left #openstack-nova | 21:38 | |
| hemna | another reason to get that update that adds volume backed LM working | 21:38 |
| *** dave-mccowan has quit IRC | 21:39 | |
| *** mdenny has joined #openstack-nova | 21:41 | |
| *** jaypipes has quit IRC | 21:47 | |
| *** dims_ has joined #openstack-nova | 21:49 | |
| *** apuimedo has joined #openstack-nova | 21:52 | |
| *** karimb has quit IRC | 21:53 | |
| *** jecarey has quit IRC | 21:54 | |
| *** mcarden has quit IRC | 22:04 | |
| *** mcarden has joined #openstack-nova | 22:04 | |
| *** vishwanathj has quit IRC | 22:04 | |
| *** stevemar has quit IRC | 22:05 | |
| *** tonytan4ever has quit IRC | 22:05 | |
| *** vishwanathj has joined #openstack-nova | 22:05 | |
| *** vishwana_ has joined #openstack-nova | 22:07 | |
| *** dave-mccowan has joined #openstack-nova | 22:08 | |
| *** annegentle has joined #openstack-nova | 22:08 | |
| su_zhang | is there a good way to obtain the map between server-group and availability zone through nova API? | 22:09 |
| *** vishwanathj has quit IRC | 22:10 | |
| *** achanda has quit IRC | 22:11 | |
| *** beagles has joined #openstack-nova | 22:11 | |
| *** stevemar has joined #openstack-nova | 22:11 | |
| *** beagles is now known as Guest62063 | 22:11 | |
| *** ZZelle_ has quit IRC | 22:14 | |
| *** stevemar has quit IRC | 22:14 | |
| *** dims_ has quit IRC | 22:17 | |
| *** Guest62063 has quit IRC | 22:17 | |
| *** dims_ has joined #openstack-nova | 22:20 | |
| *** armax has quit IRC | 22:21 | |
| *** beagles_ has joined #openstack-nova | 22:21 | |
| *** mdrabe has quit IRC | 22:25 | |
| *** ijw has quit IRC | 22:25 | |
| *** beagles_ has quit IRC | 22:26 | |
| *** achanda has joined #openstack-nova | 22:26 | |
| *** mriedem has joined #openstack-nova | 22:28 | |
| *** annegentle has quit IRC | 22:30 | |
| *** VW_ has quit IRC | 22:31 | |
| *** VW_ has joined #openstack-nova | 22:31 | |
| *** mriedem1 has joined #openstack-nova | 22:33 | |
| *** mriedem has quit IRC | 22:35 | |
| *** VW_ has quit IRC | 22:35 | |
| *** annegentle has joined #openstack-nova | 22:37 | |
| *** obedmr has quit IRC | 22:45 | |
| *** erlon has quit IRC | 22:48 | |
| *** diogogmt has quit IRC | 22:51 | |
| *** tpeoples has quit IRC | 22:51 | |
| *** piyanai has quit IRC | 22:52 | |
| *** Piet has joined #openstack-nova | 22:53 | |
| *** Piet has quit IRC | 22:54 | |
| *** ijuwang has quit IRC | 22:55 | |
| *** dims_ has quit IRC | 22:59 | |
| *** alex_klimov has quit IRC | 22:59 | |
| *** david-lyle has joined #openstack-nova | 22:59 | |
| *** erhudy1 has joined #openstack-nova | 23:00 | |
| *** krtaylor has joined #openstack-nova | 23:02 | |
| *** annegentle has quit IRC | 23:03 | |
| *** rfolco has quit IRC | 23:08 | |
| *** hemna is now known as hemnafk | 23:14 | |
| *** lucasagomes has quit IRC | 23:16 | |
| *** pixelb has quit IRC | 23:18 | |
| *** ssurana1 has quit IRC | 23:21 | |
| *** baoli has joined #openstack-nova | 23:25 | |
| *** IanGovett1 has quit IRC | 23:28 | |
| *** nic has quit IRC | 23:33 | |
| *** markvoelker has quit IRC | 23:36 | |
| *** penick has joined #openstack-nova | 23:38 | |
| gmann | markus_z: pong | 23:39 |
| *** artom has quit IRC | 23:44 | |
| *** mdenny has quit IRC | 23:47 | |
| *** emagana has quit IRC | 23:47 | |
| *** armax has joined #openstack-nova | 23:48 | |
| *** alexpilotti has joined #openstack-nova | 23:49 | |
| *** xyang1 has quit IRC | 23:49 | |
| *** shoutm has joined #openstack-nova | 23:56 | |
| *** baoli has quit IRC | 23:56 | |
| *** thorst has joined #openstack-nova | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
